Schedule for Saturday
8:00-8:30   Registration, coffee, light snack in the main lobby
8:30-10:50   You know it's time: The beneficial effects of temporal attention on perception Eye movement-related brain activity in perceptual and cognitive processing Sensory and Cognitive Plasticity Following Visual or Auditory Impairment New computational approaches to P300 research in humans Functional and structural determinants of lifespan differences in cognition
11:00-11:30   Welcome address, announcements in the Effectenbeurs zaal
12:00-13:30   Posters, lunch, sponsor booths in the Grote zaal
13:30-16:00   Open and reproducible neuroscience: What, Why and How? Contemplating conflict: the role of metacognition and performance monitoring in voluntary action Theta oscillations in the human medial temporal lobe -- From single units to MEG Perspectives on sensory prediction The Anterior Cingulate Cortex and Social Cognition: Evidence from anatomy, electrophysiology, computational and translational neuroscience
16:00-17:30   Posters, snacks/coffee, sponsor booths in the Grote zaal
17:30-18:30   Eleanor Maguire: How are memories represented in the human brain?
Keynote in the Effectenbeurs zaal
19:00-23:30   WELCOME RECEPTION: Hotel Krasnapolsky
Drinks, snacks, music, friendly banter, science, magic, and more!
Schedule for Sunday
8:00-8:30   Registration, coffee, light snack in the main lobby
8:30-10:50   How to organize open and reproducible science Neural oscillatory mechanisms in temporal prediction Dilemmas in the face of uncertainty: Computational approaches to adaptive behavior Hippocampal representations of realistic memories: event segmentation, integration and organisation of memories in naturalistic contexts Is visual cortex relevant for visual working memory?
11:00-12:00   Satu Palva: Functional significance of large-scale neuronal interactions and scale-free dynamics in human cognition
Keynote in the Effectenbeurs zaal
12:00-13:30   Posters, lunch, sponsor booths in the Grote zaal
13:30-16:00   Functional interactions between neural systems for attention and decision making Can we measure without influencing the measured: Rhythmic sensory stimulation in brain research The neural basis of social decision-making Toward a unified account of frontopolar function in higher-level cognition
16:00-17:30   Posters, snacks/coffee, sponsor booths in the Grote zaal
17:30-18:30   Enjoy Amsterdam outside the Beurs van Berlage!
Schedule for Monday
8:00-8:30   Registration, coffee, light snack in the main lobby
8:30-10:50   Deciding how to decide: when and how the brain takes short-cuts to make decisions Mnemonic priorities: dynamic interplays between attention and working memory Connectivity research in cognitive neuroscience: Hype or approach of the future? The temporal dynamics of response inhibition: neural markers, modes, and modulations See, smell, and taste this! The neuroscience of multimodal food perception
11:00-12:00   Daphna Shohamy: How Memory Guides Value-Based Decisions
Keynote in the Effectenbeurs zaal
12:00-13:30   Posters, lunch, sponsor booths in the Grote zaal
13:30-16:00   Oscillations in the broad alpha band: New Findings Predictions in perception and action: neural mechanisms, cognitive determinants, and computational architectures Theta-gamma coupling as a mnemonic mechanism in the human brain Dynamic Causal Modelling: a promising tool for studying the putative mechanisms underlying brain function? Leveraging electrophysiology to study visual perception across non-human and human primates
16:00-17:30   Posters, snacks/coffee, sponsor booths in the Grote zaal
17:30-18:30   Bids and votes for ICON2020!
Effectenbeurs zaal
Schedule for Tuesday
8:00-8:30   Registration, coffee, light snack in the main lobby
8:30-10:50   Principled computations of goal-directed perceptual inference: from the cortical column to the large-scale network Neurobiology of Motivation: Anatomy, pharmacology, pathology, and computation Models of category formation and adaptation: distributional learning of speech and language Modulation of perceptual processes by motor processes: the role of neuronal oscillations. Cognitive and neural mechanisms of attention development
11:00-12:00   John Duncan: Attentional episodes and cognitive control
Keynote in the Effectenbeurs zaal
12:00-13:30   Posters, lunch, sponsor booths in the Grote zaal
13:30-16:00   How can deep neural networks help us understand sensory processing in the human brain? The Influence of Cortical Field Potentials on Perception, Action, and Cognition Building Cognitive Architecture in Atypically Developing Populations: Assessing The Potential of Targeted Cognitive Training The role of expectancy in episodic memory encoding Towards a theory of Prefrontal Cortex: Throwing out the babies with the bathwater?
16:00-17:30   Posters, snacks/coffee, sponsor booths in the Grote zaal
17:30-18:30   Announcement of next ICON and closing remarks
Effectenbeurs zaal

Click an item above to see detailed information below.
Date and time: Saturday 8:30
Room: Graan
Format and topic: Symposium, Attention
Coordinator: Felix Ball
Title: You know it's time: The beneficial effects of temporal attention on perception
Symposium abstract: Continuously, our brain has to process and to deal with a vast amount of information. Not only do we have to integrate information into coherent percepts at any given time point, we also have to compare these information with past percepts to keep track of changes in our environment. This process is computationally demanding and requires optimization. One possibility to optimize the workflow of information processing is to prepare for and predict future events. It is well known that e.g. spatial predictions result in increased perceptual performance by allocating neural resources of the dorso-ventral network. However, knowing where relevant events will occur is not the end of the line. Ultimately, all events are determined by their temporal onset. As time continuously progresses, predicting events' onsets is crucial as past events are doomed to be unrecognized if they were not perceived. Hence, it is of utmost importance to guide temporal attention to time windows of interest, specifically, to when or before events occur (note that this is different to spatial attention in which a false prediction by 1 degree might not be problematic at all). Over the last 2 decades researchers started to investigate whether and how we can direct attention in time. Today, there is corroborating evidence that temporal predictability improves performance: it enhances detectability of targets, increases accuracy in discrimination tasks (e.g. frequency judgment), and decreases response times. However, as in any new field in science, this is only the start and much has to be done. In this symposium, we will highlight the recent advances in the field and cover all its relevant aspects including unisensory and multisensory temporal attention research, the neural origins of temporal expectations, and it's use for clinical applications (e.g. training of elderly).

Speaker: Freek van Ede
Title: An introduction into studying temporal attention
Abstract: Traditionally, research on attention has focused on three key aspects: space-based, object-based, and feature-based attention. However, another - equally important - aspect of attention has long been ignored: attending to a specific point or period in time. Being able to predict the identity and location of relevant objects also requires the prediction of when these objects will appear. Recently, several experimental paradigms have been created to study temporal attention. They have revealed multiple forms of temporal information that can influence perception and action: isochronous or complex rhythms of stimulus trains, temporal associations between stimuli, temporal probabilities of events occurring within a given context (hazard rates), and recurring temporal sequences of events. Most of the studies so far have been carried out in the visual domain. Whereas many of the principles of temporal attention studied in vision may be applicable to other sensory domains and cross-modally, there may be important differences in detail. Research on temporal attention is still at an early stage, but has already demonstrated conclusively that predictive temporal information has strong influences on our performance. Findings across paradigms show that temporal attention enhances perceptual sensitivity to detect and discriminate targets and speeds response times. Neural studies suggest that temporal attention influences several stages of information processing, starting at the perceptual level. Furthermore, temporal information can interact strongly with other, spatial and object-based, modulatory attention signals. The specific mechanisms at play are only beginning to be investigated, and different forms of temporal attention may operate in parallel, through non-overlapping mechanisms. In my talk, I will provide a brief overview of the state of research on temporal attention, drawing from recent research in my lab to highlight some of the emerging principles.

Speaker: Felix Ball
Title: The role of multisensory stimulation in temporal expectations
Abstract: In real life, we are exposed to a continuous stream of complex multisensory information. This information needs to be integrated to generate a reliable mental model of our world. There is converging evidence for several optimization mechanisms to integrate incoming information, among them are multisensory interplay (MSI) and temporal attention (TE). Previous research focused on the influence of temporal expectation on perceptual processing mainly in unisensory auditory, visual, and tactile contexts. Thus, it is currently unknown whether MSI and TE interact. Here we tested -- in a series of experiments -- whether temporal expectations can enhance perception in multisensory contexts and whether this enhancement differs from enhancements in unisensory contexts. In all experiments auditory (A) and/or visual (V) stimulus sequences were presented either alone or simultaneously. Participants always discriminated visual and/or auditory frequencies of deviant target stimuli (high/low). We used near-threshold targets embedded in a sequence of distractors and the likelihood of target occurrence (early or late) was manipulated block-wise. Furthermore, we tested whether spatial and modality-specific target uncertainty (i.e. predictable vs. unpredictable target position or modality) would affect temporal expectation. In all our experiments, hidden temporal regularities improved performance for expected multisensory targets. Multisensory enhancement was dependent on task difficulty, and increased with increasing noise. Remarkably, TE effects were also enhanced for multisensory relative to unisensory stimulation and TE effects for unisensory stimuli even vanished under high spatial uncertainty. Together, our results strongly suggest that participants benefit from multisensory stimulation (relative to unisensory stimulation) and that this effect is maximal if the stimulus-driven uncertainty is highest. We propose that enhanced informational content enables the robust extraction of temporal regularities, highlighting the need for of multisensory paradigms in future studies investigating temporal expectations.

Speaker: Jennifer T Coull
Title: The role of left inferior parietal cortex in predicting time
Abstract: Being able to predict when relevant events are likely to occur improves how quickly and accurately they are processed. In a series of fMRI investigations we have found that the behavioral benefits of temporal predictability implicate left-lateralised inferior parietal cortex, in a task-independent manner. In our studies, temporal cues allowed participants to predict a priori the time at which a target would appear. Yet the very passage of time itself also provides temporally predictive information that can be used to hone information processing in a more dynamic way. The longer we wait for an event to occur, the higher is the conditional probability, and hence temporal predictability, of its occurrence (the "hazard function"). Temporal predictability can therefore be fixed in advance by temporal cues (prior probability) or evolve dynamically as a function of the elapse of time itself (posterior probability). We have found that left inferior parietal cortex is engaged not only by the fixed temporal predictability of learned temporal cues but also by dynamic changes in temporal predictability over time. Activity in left inferior parietal cortex, in tandem with that in right prefrontal cortex, tracked the increasing temporal probability of target occurrence over time. This increase in activity was independent of changes in coincident cognitive processes that also evolve dynamically over time, such as motor preparation, sustained attention or perception of the elapse of time itself. While left inferior parietal cortex appears to play a fundamental, context-independent role in temporal predictability, the recruitment of right prefrontal cortex may reflect its more general role in the dynamic monitoring and updating of information in working memory.

Speaker: Theodore Zanto
Title: The role of temporal expectations for age-related performance
Abstract: The ability to form temporal expectations is a fundamental cognitive function that facilitates perceptual and higher order cognitive processes. In advanced age (e.g., 60+ years), many cognitive functions exhibit declines, ranging from perceptual to higher order functions such as inhibitory control and working memory. Here, data will be presented suggesting that many of these age-related cognitive declines may stem from deficient engagement of temporal expectations. Specifically, younger adults (aged 18-35 years), but not older adults, utilize predictive cues to enhance performance on tasks that assess perceptual (discrimination) ability, inhibitory control, and working memory. These age-related performance declines are preceded by decreased anticipatory neural activity in sensory cortical regions as well as lowered functional connectivity between sensory and prefrontal cortex. Current efforts will then be described using non-invasive brain stimulation and cognitive training as potential pathways to remediate age-related declines in expectation abilities and subsequently improve cognitive functions that utilize temporal expectations.

Speaker: Toemme Noesselt
Title: Concluding remarks

Date and time: Saturday 8:30
Room: Administratie
Format and topic: Symposium, Electrophysiology methods
Coordinator: Andrey Nikolaev
Title: Eye movement-related brain activity in perceptual and cognitive processing
Symposium abstract: Recent advances in eye-tracking technology have allowed researchers to use eye movements as markers for segmentation of ongoing brain activity into episodes relevant to sequential steps of information processing. Consequently, the simultaneous recording of brain activity and eye movements has increasingly become popular in various fields of vision research [1-7]. The co-registration of brain activity and eye movement is particularly advantageous for the investigation of processes associated with free visual exploration [2-7]. The proposed symposium will discuss the dynamical neural mechanisms underlying perception, attention, memory and reading during eye movement, especially, in unrestricted viewing conditions. In addition, the symposium will contribute to a better understanding of the range of research questions that can be approached by the co-registration, the requirements for experimentation, and methodological solutions for simultaneous EEG and eye movement recording and data processing. The target audience of the symposium will be neuroscientists and psychologists who are interested in neurophysiological correlates of brain processes associated with eye movement and who wish to extend their methodological arsenal in this field.

Speaker: Olaf Dimigen
Title: Combining eye-tracking and EEG: Advantages, Challenges, and Applications
Abstract: Although normal vision involves 3-4 saccadic eye movements per second, human electrophysiological data is typically recorded during prolonged visual fixation. An alternative approach to data analysis, summarized in the present talk, is to align the EEG signal not to passive stimulations, but to the on- or offsets of naturally occurring eye movements, yielding saccade- or fixation-related potentials (SRPs/FRPs; e.g. [1]). In the first half of this talk, I will briefly outline the potential advantages and historical development of the SRP/FRP technique and summarize some basic features of eye movement-related activity in the scalp EEG. I will then propose an integrated analysis framework (see also: [2]) to address four methodological issues that have so far complicated EEG recordings during natural vision: (i) the synchronization of EEG and eye-tracking data, (ii) the correction of ocular artifacts, (iii) the deconvolution of overlapping brain responses, (iv) and the modeling of low-level covariates that influence cortical activity in free viewing situations [1]. The second half of the talk will summarize some recent findings using the SRP/FRP approach. In one line of research, we have studied the impact of involuntary microsaccades on traditional EEG/ERP datasets recorded during steady fixation. Results show that cortical activity from microsaccades is not just a hidden measurement artifact [3], but that it contains valuable information about a subject's attentional state [4]. In another line of research, the technique was used to study visual cognition during natural reading and scene perception. Within-subject comparisons show that the timing and morphology of established ERP effects (e.g. N170, N400) can be substantially different in during saccadic vision as compared to passive presentation paradigms [5, 6].

Speaker: Christof Korner, Hannah Hiebel, Joe Miller, Margit Hofler, Anja Ischebeck
Title: Attention and Memory Effects in Overt Visual Search
Abstract: To date, little is known about the neural mechanisms that underlie attention and short-term memory mechanisms in overt visual search. In two experiments, we concurrently recorded EEG and eye movements while participants searched for two identical targets amongst a set of distractors. In such a multiple-target paradigm, participants must continue the search after finding the first target and memorize its location. We investigated neural correlates of target and distractor processing in different stages of the search using fixation-related potentials (FRPs) [1,5]. To investigate attentional mechanisms, in Experiment 1 we compared the fixation of the first target with the fixation of a distractor. This revealed a P3-like component for target fixations which was absent for distractor fixations [2-4]. To investigate memory mechanisms, we analyzed the effect of a target fixation on continued search. Detection of the first target influenced distractor FRPs: a sustained negativity was observed for distractor fixations following the first target fixation, likely indicating the involvement of memory. In Experiment 2, the size of the search display was manipulated systematically (10, 22, 30 items). Eye movement analysis showed that the average number of fixations preceding the first target detection varied as a function of display size. We present FRPs for target fixations and discuss the influence of the number of preceding distractor fixations on their morphology. In addition, we analyzed the post-target negativity over longer time intervals and report on its temporal properties and functional role. Taken together, our experiments provide a comprehensive description of the neural mechanisms underlying attention and memory in overt search.

Speaker: Michael Ploechl, University of Luebeck, German
Title: ICA component selection strategies for combined EEG and eye movement research
Abstract: In the context of EEG analyses, Independent Component Analysis (ICA) is primarily used for removing eye- and other non-neural artifacts from the signal measured at scalp level. While the efficacy of this approach has made it a standard procedure in EEG research, this talk aims at providing alternative and extended perspectives on ICA as an EEG analysis tool, particularly in the presence of eye and body movements. First, we will review strategies for identifying and classifying Independent Components (ICs) with respect to their signal sources (e.g. brain, eye, muscle). Then we will discuss how selecting rather than rejecting unequivocally identified ICs may prove useful in paradigms where eye- and body movements, as well as other non-neural sources, render standard EEG analyses difficult or even impossible. Examples will demonstrate that defined IC selection may further improve the signal-to- noise ratio and how this approach may also allow to extract information from sources that are usually considered artifacts. Finally, we will discuss limitations and the scope of potential information loss that may arise from an IC selection strategy. Altogether a different perspective on how to use ICA for EEG analyses may facilitate studying neural activity under more natural behavioral conditions.

Speaker: Andrey R. Nikolaev
Title: EEG-eye movement co-registration reveals memory processes in free viewing behavior
Abstract: Combining EEG recording with eye-tracking enables the use of eye-movements for segmenting ongoing EEG activity. This approach is gaining importance for studying visual information processing, in particular in free visual exploration of visual scenes [1-3, 5-7]. We employed this approach to investigate memory processes. To do so we had to address the problem that effects of subsequent eye movements overlap on the EEG signal [1, 2], which may lead to confounding between cognitive and eye movement effects [4]. Next, we studied visual memory encoding in naturalistic viewing conditions. In one study we considered change blindness as a result of visual memory encoding failure. Correct change detection ensued when saccades landing near a prospective change target had presaccadic EEG amplitude corresponding to the saccade length. When change blindness occurred, this correspondence failed, indicating a covert attentional shift away from the fixation target. The dissociation between overt and covert attention leads to encoding failure and thereby to change blindness [5]. Then we compared brain activity during memory encoding and retrieval in free viewing. In a combined visual search-change detection experiment, we analyzed the functional connectivity within fixation intervals. We found differences between encoding and retrieval conditions in the graph-theoretical properties of functional connectivity, such as mean path length, radius, closeness and eccentricity. Encoding involves a more segregated functional connectivity than retrieval [6].

Date and time: Saturday 8:30
Room: Berlage
Format and topic: Symposium, Sensory processing
Coordinator: Stephen Lomber
Title: Sensory and Cognitive Plasticity Following Visual or Auditory Impairment
Symposium abstract: A remarkable property of the brain is its capacity to respond to change. This neuroplastic process endows a complex nervous system with the ability to adapt itself to its environment but, at the same time, makes it vulnerable to impoverished sensory or developmental experiences. Adaptive, or compensatory plasticity is a part of this overall process resulting from the loss of a class (or modality) of sensory inputs that is accompanied by a corresponding expansion of the remaining systems. Not only does this process provide some substitute for the lost modality, but the additional circuitry also conveys enhanced abilities to the remaining systems. Developmental studies of the deaf and blind, as well as recent studies in mature subjects, demonstrate remarkable crossmodal plasticity throughout the cerebrum. Sensory substitution devices that transfer information from one sense into another also reveal the remarkable multisensory nature of the cerebrum and underlying crossmodal plasticity. Studies of sensory loss and restoration are changing traditional views of cortical organization. Integrating animal and human models, and insights from the study of blindness and deafness, in this symposium we will discuss mechanisms of crossmodal plasticity in visual and auditory cortices throughout the lifespan, the role of critical periods, and impact on perception and cognition. The proposed symposium brings together information from both human and animal studies examining functional compensations following deafness and blindness, and the changes that occur following the initiation of hearing and sight. The proposed symposium includes a broad representation of speakers in terms of institution (Canada (x2), Israel, USA) and country of origin (India, Israel, Italy, and USA). A board representation of experimental approaches will be included (psychophysics, electrophysiology, fMRI, MEG) in addition to both animal and human studies. The speakers range from mid-career to senior scientists.

Speaker: Franco Lepore, Department of Psychology, Universite de Montreal
Title: Assessment and Rehabilitation of Cortically Blind Individuals: Novel Methods to Evaluate and Stimulate Blindsight
Abstract: A number of studies have shown that unilateral loss of posterior visual cortex due to surgery or stroke leads to blindness in the contralateral field, referred to as homonymous hemianopia. However, in numerous cases, patients can unconsciously perceive visual stimuli presented in the blind field, a phenomenon generally known as blindsight. Unfortunately, this captivating residual ability is still not completely understood and insufficiently exploited within rehabilitation strategies. Indeed, there is no consensus either on techniques to improve the quality of life of hemianopic patients or on the understanding of the mechanisms underlying blindsight. The first objective of our rather ambitious program is to evaluate the residual visual capacities of hemianopic patients. Thus, we tested them in several forced-choice paradigms, including target detection, movement and direction discrimination, spatial localization, low and high-frequency sensitivity, reaching movement towards "unseen" targets as well as recognition of emotional stimuli. To understand the mechanisms underlying the ability to unconsciously perceive these stimuli, we measured brain activity using MRI and fMRI and recorded the electrophysiological activity by using three distinct paradigms: visual mismatch negativity (vMMN) in an oddball paradigm, steady state and emotional paradigms. Following these assessments, we trained the patients using paired visual/auditory stimulations, repetitive moving dots and low contrast gratings to compensate and improve vision. We further tested them before and after training in the behavioral and imaging tasks. Results indicated significant improvements and changes in activations.

Speaker: Krish Sathian, Department of Neurology, Emory University School of Medicine
Title: Spatial Imagery and Blindness
Abstract: Spatial imagery is relevant to the formation of cognitive maps. We devised a spatial imagery task requiring mental navigation through an imaginary grid, based on auditory cues and judgment of the shape of the path traced, in a one-back same-different discrimination. This task, using functional magnetic resonance imaging (fMRI) in normal volunteers, was shown to activate multiple regions in a dorsal frontoparietal network, as well as the right lateral occipital complex; activation magnitudes at a number of parietal foci correlated across individuals with accuracy on the task (1). Blindness acquired at or after age 6 significantly impaired accuracy on this task relative to performance in age-matched controls, with a negative influence of increasing age; task accuracy correlated with the Santa Barbara Sense of Direction Scale score, indicating the relevance of our task to real-world spatial skills (2). However, congenital blindness did not significantly affect accuracy on this task (3). Taken together, our findings imply that, while visual input is not necessary for the development of proficient spatial imagery, this ability declines after the loss of vision. Rehabilitative training of blind individuals should take into account these differences in spatial skills as a function of when vision is lost.

Speaker: Stephen Lomber, Brain and Mind Institute, University of Western Ontario
Title: Enhance Visual Cognition in the Deaf is Mediated by Ventral Auditory Cortex
Abstract: When the developing brain is deprived of input from one sensory modality, it often compensates with supernormal performance in one or more of the intact sensory systems. To test this hypothesis at a cognitive level, we examined the visual capabilities of adult congenitally deaf cats and adult hearing cats on a battery of visual cognitive tasks. The animals were tested on their abilities to both learn and recall seven different visual discriminations: simple patterns, complex patterns, simple objects, junk objects, natural scenes, and faces (both human and conspecific). Both deaf and hearing cats learned to discriminate simple patterns, complex patterns, simple objects, junk objects, and natural scenes at similar rates. However, deaf cats were significantly faster at learning (fewer errors to criterion) both the human and conspecific faces compared to hearing cats. Abilities to recall any of the visual discriminations were no different between the hearing and deaf cats. The second part of this study was to examine if cross-modal reorganization in auditory cortex may be contributing to the superior cognitive capabilities of the deaf cats. To accomplish this, we bilaterally placed cooling loops on A1, A2, the temporal auditory field (TAF), and insular cortex (area IN) to permit their individual deactivation. Bilateral deactivation of A1, A2, or area IN, did not alter learning rates for either the human or conspecific faces. However, bilateral deactivation of TAF resulted in the elimination of enhanced face (both conspecific and human) discrimination learning capabilities of the deaf cats and resulted in performance similar to hearing cats. Overall, our results show that enhanced visual cognition in perinatal deaf cats is caused by cross-modal reorganization within "deaf" auditory cortex and that it is possible to localize individual visual functions within cross-modally reorganized auditory cortex.

Speaker: Amir Amedi, Department of Medical Neurobiology, The Hebrew University of Jerusalem Medical School
Title: Visual and Multisensory Navigation in the Human Brain and its Dependence on Visual Experience
Abstract: I will describe the extent and timescale with which sensory cortices can be recruited and modified by inputs coming from various natural or artificial sensory input modalities or even when conveying high-level cognitive information like language and memory. Our approach uses longitudinal studies in individuals with various degrees of visual deprivation, ranging from sighted-blindfolded to lifelong deprivation in patients with undeveloped retinas. I will describe the two main types of plasticity that we observed in the brain: (1) task-switching plasticity; and (2) task-selective sensory-independent organization. I will propose possible mechanisms that might give rise to such brain (re)-organization. In addition, I will show how we recently expanded our theoretical framework to include possible developmental mechanisms and implications for clinical rehabilitation including the development of a multisensory approach to restore vision and hearing (e.g. the multisensory bionic eye). By presenting an overview of our findings I will question classical theories of 'critical periods' by showing that "visual" regions do maintain their specific typical functionality and functional connectivity patterns even if "reawakened" in later periods in life including adulthood. Overall, through our approach and findings, new insights will emerge into the effects of learning and training on the (re)-organization principles of the human brain. A zoom in will be on unpublished results on navigation. Vision is considered the dominant sense used by humans for navigation, and many nodes of our brain's navigation network lay in fact in the visual system. What happens when we're deprived of vision? Subjects interactively navigated virtual mazes during fMRI (3 groups: visually, congenitally-blind and sighted-blindfolded via the EyeCane (an auditory navigation tool) and were scanned before and after training. After training all groups recruited V1, pCU and the scene-selective regions. This demonstrates the network's robustness to sensory input and the strength of cross-modal plasticity even in V1.

Date and time: Saturday 8:30
Room: Veiling
Format and topic: Symposium, Cognitive modeling
Coordinator: Francisco Barcelo, Bruno Kopp
Title: New computational approaches to P300 research in humans
Symposium abstract: Computational modeling of brain function using probability laws holds great promise in cognitive neuroscience, yet its application to scalp recorded M/EEG indexes of cognition is still scarce. One prominent example is the late positive complex (LPC, P300) of the event-related brain potential (ERP), which has been empirically linked to many cognitive domains (from perception and attention to language and decision-making) for over four decades of intensive research. Traditionally, the P300 is decomposed in two anatomically distinct sub-components: one with an anterior scalp distribution (P3a), followed by a second centro-parietally distributed one (P3b) at variably delayed latencies. The so-called 'context updating' model is currently the most widely accepted cognitive hypothesis on P300 function (Donchin & Coles, 1988). This model accounts mostly for P3b data as related to strategic or meta-cognitive processes variously concerned with (1) the probabilistic representation of environmental contingencies, (2) deploying attention, and/or (3) setting priorities and biases. P3a data have often been described as the brain's orienting response to novel or salient, mostly task-irrelevant, stimuli. These two ERP sub-components have been variously linked to stimulus uncertainty (Sutton et al., 1965), contextual surprise (Donchin, 1981), and perceptual decision-making. However, to date there have been few attempts to mathematically operationalize these constructs, and a computational theory for P300 data is still lacking (cf., Friston, 2005). Here we present new insights derived from four computational models inspired by probabilistic hypotheses, Bayesian theory, and information theory, which are potentially capable to offer an overarching integrative account of P300 function and revive scientific interest in this primary neural substrate of human cognition.

Speaker: Bruno Kopp, Antonio Kolossa
Title: P300 and the Bayesian brain hypothesis
Abstract: Empirical support for the Bayesian brain hypothesis, although of major theoretical importance for cognitive neuroscience, is surprisingly scarce. This hypothesis posits simply that neural activities code and compute Bayesian probabilities. Here, we introduce an urn--ball paradigm to relate event-related potentials (ERPs) such as the P300 wave to Bayesian inference. Bayesian model comparison was conducted to compare various models in terms of their ability to explain trial-by-trial variations in ERP amplitudes at different points in time and over different regions of the scalp. Specifically, we were interested in dissociating specific ERP responses in terms of Bayesian updating and predictive surprise. Bayesian updating refers to changes in probability distributions given new observations, while predictive surprise equals the surprise about observations under current probability distributions. Components of the late positive complex (P3a, P3b, Slow Wave) provided dissociable measures of Bayesian updating and predictive surprise. Specifically, Bayesian updating yielded the best fit for the anteriorly distributed P3a. The updating of observational predictions accounted best for the posteriorly distributed Slow Wave. In addition, parietally distributed P3b responses were best fit by predictive surprise. These results indicate that the three components of the late positive complex reflect distinct neural computations. As such they are consistent with the Bayesian brain hypothesis, but all of these neural computations seem to be subject to non-linear probability weighting.

Speaker: Jelmer Borst
Title: Towards an understanding of the P300 by combining Hidden semi-Markov Models with MVPA-analysis
Abstract: In this talk I propose a new method for identifying processing stages in human information processing. Since the 1860s scientists have used different methods to identify processing stages, usually based on reaction time (RT) differences between conditions. To overcome the limitations of RT-based methods we used Hidden Semi-Markov Models (HSMMs) in combination with multi-variate pattern analysis (MVPA) to analyze EEG data. This MVPA-HSMM methodology can identify stages of processing and how they vary with experimental condition. By combining this information with the brain signatures of the identified stages one can infer their function, and deduce underlying cognitive processes. In addition, it sheds new light on known ERP components such as the P300. We demonstrate this method on two tasks: associative recognition and the Sternberg working memory task, where response confidence and working memory updating influenced parietal activation.

Speaker: Florent Meyniel
Title: Tracking probabilistic inference in M/EEG signals
Abstract: The brain constantly infers the causes of the inputs it receives and uses these inferences to generate statistical expectations about future observations. Experimental evidence includes the so-called "surprise signals" such as the P300 recorded in electrophysiology, whose amplitude typically reflects the extent to which the actual observations deviate from expectations. Here, we propose to reverse-engineer the inference process and to use these surprise signals to identify the statistical model from which expectations originate in the brain. We find that a parsimonious Bayesian model accounts for classic EEG datasets from the literature. This model infers the time-varying matrix of transition probabilities between the sequential stimulus that the observer receives. This model also makes predictions about new experimental conditions that we tested in a new magneto-encephalography study. Our data support the model and they further suggest that the tracking of transition probabilities actually unfolds across distinct brain processes that integrate information at different time scales.

Speaker: Francisco Barcelo, Patrick S. Cooper
Title: An Information theory model of the P300 in cognitive control
Abstract: The most popular cognitive account of the endogenous P300 ERP component has been the "context updating" model (Donchin and Coles 1988), which explains it in terms of working memory updating triggered by a mismatch between task events and their perceptual context. However, task cues that anticipate a switch in task rules, or stimulus-response (S-R) mappings, also elicit P3-like (latency 300-900 ms) potentials. To date, it remains unclear whether such "switch positivities" show similar scalp topography and index context-updating mechanisms akin to those posited for domain-general (i.e., classic P300) positivities in many task domains. To examine this, we measured ERPs while participants were intermittently cued to switch or repeat the categorization rule of Gabor gratings varying in color and thickness (switch task), or else they performed two visually identical control tasks (go/nogo and oddball) albeit with distinct cognitive demands each. A simple information theory model helped us gauge cognitive demands under distinct temporal and task contexts as low-level S-R updating and higher-order task rule updating operations. Topographic scalp analyses confirmed significant split-second changes in the configuration of neural sources for both domain-general and switch positivities as a function of the task and temporal (proactive vs. reactive) contexts. However, switch positivities showed a centroparietal scalp distribution compatible with a family of P3-like potentials seen in many task domains. The rostro-caudal topography and intensity of this extended family of P3-like potentials critically depended on the temporal context for goal-directed behavior, as well as on low- and higher-order sensorimotor demands, rather than on their perceptual context alone. Findings partly met information theory predictions, and are compatible with a family of P3-like potentials involved in a variety of cognitive operations within fronto-posterior "multiple demand" cortices during the preparation and execution of simple sensorimotor rules.

Date and time: Saturday 8:30
Room: Verwey
Format and topic: Symposium, Developmental
Coordinator: Linda Geerligs
Title: Functional and structural determinants of lifespan differences in cognition
Symposium abstract: Aging has a pronounced impact on both the structure and function of the brain. Nevertheless, in most studies, brain measures explain only a small part of the variance in cognitive scores, and the literature contains many contradictory findings. To further our understanding of brain-cognition relationships in aging, new approaches are required. In this symposium we outline the results of a number of studies that have aimed to advance the current methods and challenge some of the commonly held assumptions in this field. These studies are based on a population-representative dataset of 700 individuals aged 18-88, which is unique in the extensive range of cognitive, demographic, and brain measures obtained for each individual. The first talk challenges some commonly used assumptions in functional connectivity research and shows that improving our analysis methods can improve our ability to link brain and cognition, especially in older adults. The second talk investigates why some cognitive tasks typically show age-related performance declines while others do not. The results of this study do not support the compensatory framework of healthy aging but suggest that maintenance of 'youth-like' activation patterns is associated with good performance. The final two talks show the benefits of using multivariate approaches to link brain and cognition in old age. The third talk shows how aging affects effective connectivity within and between brain networks, with effective connectivity being a better predictor of cognitive performance than functional connectivity. In addition this speaker will use structural equation models (SEMs) to relate memory performance to specific white matter tracts. The final talk uses group-modulated SEMs to demonstrate that aging can change both associations among neural measures (dedifferentiation), as well as the associations between cognition and brain measures.

Speaker: Linda Geerligs
Title: Functional connectivity in the aging brain: how can we best measure it and what does it tell us about cognition?
Abstract: Cognitive functions are carried out within interacting brain networks. Therefore, adequate communication (connectivity) within and between different networks is important for maintaining cognitive performance throughout the life span. Many studies have investigated how functional connectivity changes with advancing age. These studies have found some overlapping patterns, such as a reduction in the segregation between different functional networks with age. However, other findings have not been replicated across studies and few studies have found convincing associations between resting state connectivity and cognition. In part, this may be due to some of the commonly held assumptions about functional connectivity and the methods that are used to measure it. One of these implicit assumptions is that functional connectivity measures a consistent trait of participants and that we can therefore link connectivity in one state to task performance in another state. In contrast, we have found that connectivity is shaped both by the task demands (states) well as by participants specific patterns (traits). By studying functional connectivity across a wider range of mental states, we might gain a better understanding of the changes in brain function that underlie important dimensions of individual differences. In addition to studying connectivity across states, methodological advances may be critical to improving knowledge about brain-cognition associations with advancing age. One way to obtain more reliable and robust results is to use multivariate, rather than univariate measures to investigate functional connectivity. Furthermore, confounding factors such as vascular health, can have a large impact on functional connectivity measures, especially in aging samples. These confounds can be accounted for by adjusting the standard pre-processing pipelines. Adapting our methods in this way leads to clearer associations between connectivity and cognition across the lifespan.

Speaker: Karen L Campbell, David Samu, Lorraine K Tyler
Title: Context-dependent shifts in network functionality with age
Abstract: Some cognitive abilities (e.g., attentional control, explicit memory) are thought to decline with age, while others (e.g., language comprehension, implicit memory) are relatively preserved. However, the neural underpinnings of these different age-related trajectories across domains is still not well understood. Preserved functions may simply reflect maintained functionality of domain-specific networks (such as the frontotemporal language network) in the face of age-related structural loss. Alternatively, preserved functions may be those which can be compensated for by other (domain-general) networks. We test these alternative explanations across two fMRI studies from a population representative cohort. a) In the first study, we show that natural task-free language comprehension only activates language-specific networks and responsivity of these networks does not differ with age. Older adults do not show additional compensatory activation outside the language network during task-free comprehension, suggesting that this network is maintained, not compensated for, with age. b) In a second study, we contrast task-based language comprehension with two tasks (fluid intelligence and verbal production) that decline with age. We show that the same domain-general networks are activated across all tasks, but the effect of age on these networks depends on the task context. Older adults show decreased frontoparietal responsivity during fluid intelligence and verbal production, but not during language comprehension, suggesting that even domain-general networks can show maintained functionality depending on the other systems with which they are co-activated. Again, we found no evidence for age-related compensation. Taken together, these studies suggest that some cognitive abilities are preserved with age because the neural systems underlying those functions are maintained. These resilient systems may even help shore up domain-general networks that fail within other task contexts.

Speaker: Richard Henson
Title: Relating resting-state effective connectivity and white-matter structural connectivity to cognition
Abstract: I will describe two multivariate approaches to relating brain connectivity data to cognitive data: 1) using canonical correlation analysis (CCA) to relate effective connectivity parameters from dynamic causal modelling (DCM) of resting-state fMRI data to a range of cognitive measures, and moderation of this relationship by age, and 2) using structural equation modelling (SEM) to relate mean fractional anisotropy (FA) within a priori tracts of interest to measures of memory and executive function, and its potential mediation of age-related differences in cognition. In the first example, using DCM is important for distinguishing the effect of age on neural versus vascular components of the fMRI response, and for estimating directionality of effective connectivity; considerations that may explain why traditional estimates of functional connectivity on the same data did not reveal a relationship with cognition. Interesting, the multivariate relationship between DCM connectivity parameters and the range of cognitive measures was stronger in older than younger people, suggesting that cognition in old age is relatively more dependent on resting-state connectivity. In the second example, SEM was used to relate an a priori set of white- and grey-matter ROIs to latent factors estimated from a memory task (involving associative memory, item memory and priming) and from two executive tasks (Cattell and Hotel tasks). In both cognitive domains, model comparison showed that white-matter integrity made unique contributions beyond gray-matter volume, and furthermore mediated certain relationships between age and cognition, again emphasizing the importance of multivariate brain connectivity for cognition.

Speaker: Rogier Kievit
Title: Evidence for neural and neurocognitive age differentiation in a large healthy cohort
Abstract: It is well-established that brain structures and cognitive functions change across the lifespan. A longstanding hypothesis posits that in additionally the relations among cognitive factors changed, a hypothesis called age differentiation. However, to date evidence for this age-related differentiation hypothesis is mixed, and no studies have simultaneously examined cognitive and neural age differentiation. Here we multi-group Structural Equation Modelling to study differences within and between cognitive and brain factors across the adult lifespan (18-89 years) in a large (N=707), population-based sample. We found evidence for differentiation of brain structure, with covariance between both grey- and white-matter regions decreasing with age. However, despite overall decline across the lifespan, we found no evidence for cognitive differentiation across language, memory and fluid intelligence, suggesting a relatively stable covariance between cognitive factors. Finally, we observed evidence for specific patterns of age differentiation between brain and cognitive factors, such that white-matter became less correlated with memory performance in later life. We consider a set of explanations for the observed differentiation effects across the lifespan, and discuss the implications for healthy and pathological ageing.

Date and time: Saturday 13:30
Room: Graan
Format and topic: Symposium, Open science and replication
Coordinator: Dorothy Bishop
Title: Open and reproducible neuroscience: What, Why and How?
Symposium abstract: Concern about poor reproducibility in science have been growing over the past few years, but neuroscience has been rather late to recognise the problems. Neuroscience has particular challenges, because of the large amounts of data involved, the complexity of data processing, and the expense of imaging, which can limit achievable sample sizes (Poldrack et al, 2017). It is vital that this issue is tackled: science can only advance if we can confidently build on prior research. Furthermore, poor reproducibility stokes the lack of public trust in science.

Speaker: Erin McKiernan
Title: How open science helps researchers succeed
Abstract: Open access, open data, and other open scholarship practices are growing in popularity and necessity. However, widespread adoption of these practices has not yet been achieved. One reason is that researchers are uncertain about how sharing their work will affect their careers. I'll review the evidence demonstrating that open research is associated with increases in citations, media attention, potential collaborators, job opportunities and funding opportunities. I'll also talk about my personal experiences as a researcher practicing open science.

Speaker: Kirstie Whitaker
Title: How to do reproducible research
Abstract: This talk will discuss the perceived and actual barriers experienced by researchers attempting to do reproducible research in neuroscience, and give practical guidance on how they can be overcome. A key step is to make code open and reproducible; use of github for this purpose will be demonstrated, as well as options in different programming languages.

Speaker: Hannah Hobson
Title: Publication through preregistration for early career researchers and graduate students
Abstract: Pre-registration has been a topic of hot debate in the fields of neuroscience and psychology (Chambers et al., 2013; Scott, 2013). As more and more journals begin to offer pre-registration as a method of publication, it is pertinent to consider what this may mean for early career scientists, for whom pre-registration could represent a great opportunity or a risky endeavour. This talk aims to consider what pre-registration can offer PhD students and post-docs. Drawing on my own experience of conducting and publishing a pre-registered report during my PhD, I reflect the benefits for ECRs, and some of the potential challenges. I aim to provide a balanced account, with practical advice for ECRs and their mentors/supervisors, who are considering pre-registering a study.

Speaker: Rik Henson
Title: Practical Steps towards Open Science at the MRC CBU
Abstract: I will describe recent initiatives at the MRC Cognition & Brain Sciences Unit (CBU) in Cambridge, UK, towards opening our science. This has included 1) setting up a data repository associated with every co-authored publication, 2) revising, in collaboration with a local ethics committee, participant consent forms for future open sharing, and 3) setting up a managed access website for data previously acquired with restricted consent. I will also describe possible future plans, such as cover journals and internal preregistration.

Date and time: Saturday 13:30
Room: Administratie
Format and topic: Symposium, Decision-making
Coordinator: Frederike Beyer
Title: Contemplating conflict: the role of metacognition and performance monitoring in voluntary action
Symposium abstract: Conflicts are an integral part of voluntary action and they can arise at any stage from action selection to outcome monitoring. Flexible adaptation of voluntary action requires the capacity to detect conflicts and correct behaviour accordingly. Increased conflict can lead to both immediate and long-term changes in behaviour, such as increasing cognitive control, slowing reactions, and increasing uncertainty about action outcomes. What's more, action conflict can also influence the subjective experience of agency over actions consequences . Notably, research on metacognition and behavioural adaptation has largely remained separate from research on the sense of agency. In this symposium, we aim to bridge this gap by highlighting the central role that metacognitive processes play in the flexible adaptation, and experience, of voluntary action. For this, we will integrate findings on action monitoring, adaptation to conflict and errors, sense of agency, and influences of social context on behaviour. We will explore the role of metacognition in different responses to action conflicts. We will present novel models on the relation between error monitoring, metacognition, and sense of agency in the long-term optimization of goal-directed action. Speaker 1 will discuss the relationship between different processes of error monitoring and their functional significance. Speaker 2 will present behavioural and electrophysiological evidence for a model of cognitive control that highlights the importance of metacognitive processes in behavioural adaptation to conflict. Speaker 3 will show how action-related conflict influences the sense of agency, and how it is integrated with retrospective processes related to outcome monitoring. Speaker 4 will present evidence that social contexts can function as a source of action conflict, affecting subsequent outcome monitoring and reducing sense of agency. Across these talks, we will see how the online metacognitive monitoring of voluntary action - from intention, to action, to outcome - influences our behaviour and our subjective experience of agency.

Speaker: Marco Steinhauser, Francesco di Gregorio, Martin E. Maier, Soren K. Andersen
Title: Dual systems of human error monitoring
Abstract: Errors in choice tasks have been shown to elicit two consecutive components in the human event-related potential - the error-related negativity (Ne/ERN) and the error positivity (Pe). The Ne/ERN occurs immediately after the error and has been suggested to reflect either a mismatch between correct and actual response, a post-response conflict, or a (reward) prediction error. In contrast, the Pe emerges between 200 and 400 ms post-response, and is frequently assumed to be related to conscious error processing. While research has mainly focused on the functional significance of each single component, little is known how the underlying mechanisms are related. A frequent assumption is that both components constitute a cascade of error processing, and that the earlier Ne/ERN delivers the input to the later Pe. In this talk, we report evidence for the relative independency of the two components. In a first study, we show that a Pe is still observed under conditions where the Ne/ERN is fully absent. In a second study, we demonstrate that the Ne/ERN elicits attentional adjustments already before the Pe has emerged. Based on these and other findings, we discuss the idea that human error monitoring relies on two dissociable systems with different functional significance.

Speaker: Kobe Desender, Filip Van Opstal, Eva Van den Bussche
Title: The role of metacognition in cognitive control
Abstract: Contemporary models of cognitive control assume that control is triggered by the detection of conflict. For example, in the influential conflict monitoring model (Botvinick et al., 2001), the degree of control depends directly on the amount of conflict detected in a response layer. The model remains silent, however, about the role of metacognition in cognitive control. When participants perform a conflict task, they subjectively experience that trials with conflict are more difficult to respond to than no-conflict trials. Theoretically, it is possible that control is triggered by these subjective experiences, rather than by the level of conflict. In this talk, I will present data that supports this possibility. Both the decision to invest cognitive control, as well as the decision to avoid investing control are shown to crucially depend upon metacognitive awareness. Moreover, using electrophysiological recordings, it was possible to demonstrate that the detection of response conflict and the subjective experience of response conflict have spatially and temporally dissociable neural correlates. Taken together, these findings support the idea that subjective experiences are crucial in directing strategic behavior, and unravel a challenge for future models of cognition to design a framework which captures the dynamic influence of subjective experience on behavior.

Speaker: Nura Sidarus, Matti Vuorre, Patrick Haggard, Valerian Chambon
Title: Neural correlates of response conflict and their relation to the sense of agency
Abstract: Human voluntary action is typically accompanied by a sense of agency (SoA), that is, the feeling of being in control of one's actions and, through them, of events in the outside world. Much research has show that the SoA depends on a retrospective matching between the expected and actual outcome of an action. However, recent studies have revealed an additional, prospective component to the SoA, related to monitoring conflicts in action selection. Although conflict monitoring has been well studied, this field has remained largely separate from research on outcome monitoring and SoA. Our research aims to bridge this gap by assessing the relation between conflict and outcome monitoring processes, and the relation between conflict monitoring and SoA. In an event-related potentials (ERPs) study, subliminal priming of actions was used to induce response conflict. Participants responded to imperative stimuli, and observed action outcomes. Subjective agency ratings were collected at the end of each trial. Results show that incompatible priming disrupted action selection, and led to a reduction in SoA over action outcomes, relative to compatible priming. ERPs revealed that signals associated with SoA emerged already at the time of the action. This indexed an action monitoring process that signalled disruptions in action selection, and was linked to a reduction in SoA. Thus, we show that action monitoring signals influence SoA prospectively, as they emerge long before the outcome is known. Additionally, outcome monitoring was also related to SoA. Yet, there was no interaction between the neural correlates of action and of outcome monitoring. Importantly, SoA is best understood as resulting from an integration of prospective signals, related to action monitoring, with retrospective signals, based on outcome monitoring. Yet, prospective and retrospective components may make independent contributions to SoA.

Speaker: Frederike Beyer, Nura Sidarus, Sofia Bonicalzi, Patrick Haggard
Title: Thinking about others: influence of social context on outcome monitoring and sense of agency.
Abstract: The emergence of a sense of agency is mostly investigated as an individual phenomenon, with other people mainly serving as a source of ambiguity of authorship. However, the presence of others has substantial effects on individual behaviour and therefore can be assumed to influence individual cognition and information processing as well. Here, we propose a model of diffusion of responsibility, according to which other people affect cognitive processes, outcome monitoring and sense of agency. Specifically, we suggest that other people act as a source of dysfluency in decision-making , disrupting the selection and execution of appropriate behaviours. In a series of experiments, participants made costly voluntary actions. We manipulated the alleged presence of another player, who could act instead of the participant. The objective responsibility for action outcomes was unambiguous: participants always knew whether they had caused an outcome. Nevertheless, participants reported reduced sense of agency over action outcomes in trials in which the co-player was present. In an ERP study, we showed that outcome monitoring, reflected in the amplitude of the feedback-related negativity, was reduced in trials in which participants acted in the presence of the co-player. In an MRI study, we investigated the neural correlates of cognitive processes affected by the presence of the co-player during the action itself, and their relationship with sense of agency ratings. Our findings support the notion that other people can function as a source of action conflict, affecting subsequent monitoring of action consequences and the emergence of a sense of agency.

Date and time: Saturday 13:30
Room: Berlage
Format and topic: Symposium, Memory
Coordinator: Simon Hanslmayr
Title: Theta oscillations in the human medial temporal lobe -- From single units to MEG
Symposium abstract: Theta oscillations are rhythmic fluctuations around 4-7 Hz and can be observed in the local field potential in the human medial temporal lobe (MTL). The interpretation of these oscillations is mostly guided by frameworks derived from rodent studies, suggesting several mechanisms for how theta oscillations regulate core cognitive functions such as memory, spatial navigation or conflict processing. Inspired by these studies several researchers set out to explore these mechanistic frameworks in humans using invasive techniques such as single unit recordings and local field potential recordings in epilepsy patients. Additionally, exciting recent developments suggest that theta oscillations in the human MTL can even be recorded non-invasively with MEG. The speakers in this symposium will present new data using these invasive and non-invasive techniques to study the function of MTL theta oscillations in humans. The first speaker will present data putative single unit and local field potential data during spatial navigation and memory in humans. The second speaker will then also present data from putative single units and local field potentials in humans during encoding of associative episodic memory, exploring the relationship between single-unit firing and oscillations during memory formation. The third speaker will then show how theta oscillations in the hippocampus and in the prefrontal cortex mediate conflict resolution. Interestingly, theta in the two regions seems to play different roles, which is in contrast to what has been observed in rodents. Finally, the last speaker will present MEG data from healthy subjects, as well as intracranial EEG data from patients showing that MEG is capable to reveal theta sources in the medial temporal lobe. Together, the speakers in this symposium will revisit convergences as well as differences with previous and current ideas on the function of theta oscillations, thus highlighting open questions for future research.

Speaker: Joshua Jacobs
Title: Neuronal and field-potential activity underlying human spatial navigation and memory
Abstract: The ability to remember spatial environments is critical for everyday life. To understand, with a high spatial and temporal precision, how the brain supports navigation and forms spatial memories, we examined direct brain recordings from neurosurgical patients as they played our virtual-navigation video game. We found several novel signals that reveal the neural basis of human spatial memory and differentiate us from simpler animals. Humans have several types of neurons that represent a person's current spatial location, including place, grid, and path-invariant cells, which show that the neural coding of spatial location is supported by multiple medial-temporal subregions that play complementary roles. In addition I will describe our work identifying the neural basis of spatial memory encoding in humans. We found two types of memory-related signals in the human MTL: theta oscillations and broadband power spectrum shifts. In key ways these signals differ significantly from patterns seen in animals, in particular with human memory-related theta occurring at a slower frequency than would be expected from earlier work. We also examine interactions between single-cell and network oscillatory activity. An emerging theme from our work is that in terms of spatial cognition the human brain has both shared and distinctive characteristics compared with animal models.

Speaker: Frederic Roux, Simon Hanslmayr
Title: Frequency-specific coupling between LFP and single cell activity during memory encoding in the human hippocampus
Abstract: During the formation of episodic memories, the hippocampus receives input from different areas in the cerebral cortex. Arguably, this activity from multiple cortical areas belonging to the same episode converges in the hippocampus, where it is bound into a permanent memory trace. Neuronal oscillations at theta (3-8Hz) and gamma (30-80Hz) frequencies are a prominent feature of hippocampal activity. Recent evidence from electrophysiological and simulation studies, suggests that these oscillations are a key mechanism for the formation of memories by grouping distributed cell ensembles via regulating synaptic plasticity. However, the mechanism by which these assemblies "emerge" at encoding remains unknown. Here we examine changes in synchronization of hippocampal single neuron and neuronal ensemble activity in humans during pair-associative learning. Specifically we investigate the hypothesis that the emergence of cell assemblies is orchestrated by the phase of theta activity through the synchronization of single neuron spiking activity. So far our preliminary findings show that oscillatory activity at theta frequencies and single unit firing rates are both modulated during memory encoding, thereby raising two important question: 1) are single unit responses coupled to the phase of hippocampal theta activity and 2) does this coupling vary with learning? Evidence from simultaneous LFP and putative single unit recordings in the human hippocampus will be put forward to address these questions.

Speaker: Nikolai Axmacher, Carina Oehrn, Conrad Baumann, Juergen Fell
Title: Differential contributions of hippocampal and prefrontal theta oscillations to conflict processing
Abstract: Traditionally, the hippocampus has been mainly related to declarative memory formation and spatial navigation. More recently several additional functions were suggested, including working memory maintenance and pattern separation and completion. In addition, some studies suggest an involvement of the hippocampus in approach-avoidance conflict. Here, we investigated whether the hippocampus is also recruited during the resolution of cognitive conflicts. We combined intracranial EEG recordings in epilepsy patients with region of interest-based analyses of fMRI data from healthy participants during an auditory version of the Stroop paradigm. Our data reveal converging evidence that the hippocampus is indeed recruited during processing of cognitive response conflict. Specifically, conflict processing was associated with an increase in hippocampal theta oscillations, and the magnitude of this effect predicted accurate performance and faster reaction times on the level of individual trials. Notably, recordings from the dorsomedial prefrontal cortex (DMPFC; including anterior cingulate cortex and pre-supplementary motor cortex) in another group of epilepsy patients revealed a different temporal profile of conflict-related theta oscillations in that area. Thus, hippocampal and DMPFC theta oscillations seem to be generated by different mechanisms and to support different cognitive functions. Together, these data suggest that hippocampal theta oscillations play a role for the resolution of cognitive conflicts, while DMPFC is more important for the initial detection of conflicts.

Speaker: Maite Crespo Garcia, Monika Zeiller, Claudia Leupold, Gernot Kreiselmeyer, Stefan Rampp, Hajo M. Hamer, Sarang S. Dalal
Title: Theta power decreases predict spatial accuracy of subsequent active navigation: a combined intracranial EEG/MEG study
Abstract: Numerous MEG and EEG studies have reported increases in MTL theta power during mnemonic encoding that were associated with better subsequent episodic memory and spatial navigation performance. However, recent invasive studies have shown that successful encoding of items and associations are mainly characterized by broad decreases in theta activity (3-8 Hz). To investigate this issue further, we performed a combined intracranial EEG/MEG study and explored how theta activity, time-locked to the encoding of item-place associations, relates to subsequent navigation accuracy. The subsequent memory (SM) task was implemented inside computer-simulated scenarios with distal cues and administered as a video game. Pictures of buildings were presented at given contextual locations that participants were asked to remember. MEG analyses in healthy participants showed that slow-theta (2--5 Hz) power negatively correlated with spatial accuracy for locations. Beamforming sources were localized in MTL and cortical structures involved in spatial cognition. To validate these effects, we also acquired intracranial EEG recordings while epilepsy surgery patients performed the task. Additionally, we had the rare opportunity to simultaneously record MEG from an epilepsy surgery patient who was implanted in the left parietal and temporal cortex. With this dataset, we could investigate how negative SM effects manifesting in hippocampal activity were related to hippocampal-cortical phase interactions referenced to either MEG sources or intracranial signals. Functional connectivity analyses provided crucial insights: during power decreases, slow-theta in right anterior hippocampus and left inferior frontal gyrus phase-led the left lateral temporal cortex and predicted spatial accuracy. Taken together, our findings indicate that decreased slow-theta activity reflect local and long-range neural mechanism underlying the encoding of detailed spatial information and item-context associations. They are in line with the view that local suppression of low-frequency activity is essential for more efficient processing of detailed information.

Date and time: Saturday 13:30
Room: Veiling
Format and topic: Symposium, Sensory processing
Coordinator: Juanita Todd
Title: Perspectives on sensory prediction
Symposium abstract: Pattern repetition enables us to form predictions about the environment that assist in organising sensory experience to make sense of the world. These predictions (or internal models of the world) have a significant impact on how we filter the relevance of incoming information, influencing the degree to which events engage our limited resources. Sensory evoked potentials can be used to investigate how predictive regularity representations are established, and how these predictive processes shape automatic decisions about stimulus relevance. The recent popularity of predictive processing accounts has driven a resurgence of publications in this area, and here we consider the merits and limitations of this explanatory framework. In this symposium each speaker will use a different experimental paradigm to report on how their data deepen our understanding of prediction, and highlight key factors to consider in study design and data interpretation. All three speakers utilize sound sequences structured to introduce a tension or competition between potential predictive models. Speaker 1 will discuss how predictive coding frameworks challenge traditional accounts of how the brain will group sounds based on discrete spectrotemporal patterns. Speaker 2 will address where computational models succeed and fall short in accounts of auditory scene segregation. Finally, speaker 3 will argue that the presence of predictability on different timescales in sound sequences reveals parsimony whereby high-order beliefs about the environment will determine how dynamically the system updates key model parameters.

Speaker: Sabine Grimm, Maria Bader, Annekatrin Weise, Erich Schroger
Title: Predictive representations of spectrotemporal patterns
Abstract: Since 25 years the Mismatch Negativity (MMN) elicited by violations of a regularity within a series of tonal patterns has served to study the nature of spectrotemporal representations of sounds. Contrary to the majority of MMN studies, in this line of research sound patterns are used, which contain at least two tonal elements differing in a dimension such as frequency or duration concatenated into a single (but structured) perceptual unit (for example, tone A and tone B into sound AB). If a violation of repeating AB sounds elicits an MMN (e.g., a rare occurrence of BA), it can be inferred that the respective pattern regularity has been represented by the auditory system. Studies found that the automatically encoded representations of such patterns can contain at least eight elements and can be formed on the basis of the individual tonal elements' features or their feature relations (e.g., a rise in frequency from A to B). In recent years, the predictive nature of the automatically established representations underlying MMN has been emphasized, that is, the MMN is assumed to indicate a difference between the predicted and the actual sound. Revisiting the literature of pattern MMN studies from this predicting coding perspective throws a new light on the spectrotemporal representations. Results suggest that the contiguity between successive elements entails temporarily directed associations, which are used to predict the forthcoming acoustic input on an element-by-element basis. Thus, the transitions between adjacent tonal elements seem to play a special role. However, several factors influence the establishment of the respective regularity representations (e.g., number of repetitions of the regularity, the type of the transition and the duration of the elements). Moreover, repetition positivity emerges as an additional index of regularity representation of tonal patterns albeit with somewhat different sensitivity to modulating factors.

Speaker: Istvan Winkler, Tuende B. Szabo, Susan L. Denham
Title: Auditory stream segregation: How far does the concept of predictive processing take us?
Abstract: Several studies, including some from our group, have found that the presence of separate sequential regularities within interleaved sound sequences, such as a repeating pattern or a melody helps auditory stream segregation. These findings are suggestive that auditory stream segregation utilizes predictive processing principles. Indeed, some current computational models have evoked predictive processes for explaining auditory stream segregation. However, the role of regularities (and thus prediction) in segregating streams is controversial. First, even those studies that found an advantage of separate regularities in stream segregation disagree on whether regularities can induce segregation or they only help to maintain streams, which have been initially separated by simpler cues, such as feature separation or feature change rate. Second, it is also unclear, whether or not the putative predictive processes involved in auditory stream segregation require attention. Finally, there are also studies showing that when the sounds of the two sound sequences overlap each other in time, as they often do in realistic environments, random variation in one stream promotes auditory stream segregation. In terms of computational models of auditory stream segregation, models based on assessing feature distributions or the temporal coherence of features (without including predictive processing principles) appear to be no less successful in modeling behavioral and/or neurophysiological data than those based on predictive principles; although direct comparison of the success of these models is difficult to evaluate due to their different goals and evidence base. In the talk, we shall show some new ambiguous data regarding the effects of repeating patterns on separating interleaved sound sequences and discuss the possible involvement and limitations of predictive processing in auditory stream segregation.

Speaker: Juanita Todd, Alexander Provost, Jade Frost, Kaitlin Fitzgerald, Daniel Mullens, Istvan Winkler
Title: Hierarchical inference impacts sensitivity to probabilistic change - When hysteresis reflects parsimony.
Abstract: The auditory system is adept at learning to predict repetitive patterns. The formation of internal "prediction" models is inferred from changed responsiveness to sound measured in auditory evoked potentials (AEPs). Models provide an automatic filter of event-relevance in that sounds conforming to predictions show diminution of many AEP components, while deviations from predictions elicit larger responses. Deviance sensitivity alerts us to changes that might be important, indicating that an update to the model is advantageous, and is proportional to model precision (estimated reliability of the data on which the model is based). However, sequences that contain patterning on multiple timescales suggest the presence of hierarchical models with predictions formed at higher levels distorting learning on a local timescale. Prediction models anchor to first impressions of a sequence resulting in primacy effects. The primacy effect creates a kind of hysteresis in that modulation of AEPs lags behind what would be expected based on emergent patterns on a local timescale. In sequences alternating between two equiprobable states (context A, P|ToneA>P|ToneB versus context B, P|ToneB>P|ToneA) at a predictable rate, the deviance sensitivity reflected in AEPs in context B takes longer to emerge than in context A. This effect does not diminish with repetition of the sequence, is not present if the alternation between states is unpredictable, and effect reverses if a regular alternation rate is violated (i.e., states change too early or too late). Data obtained across a number of studies will be used to present the thesis that the auditory system is slow to update internal model precision estimates based on local probability changes if higher-order predictability is present. More specifically, this parsimony reflects sensory beliefs about how stable the environment is likely to be, how likely that is to change, and how advantageous it is to track these changes.

Date and time: Saturday 13:30
Room: Verwey
Format and topic: Symposium, Anatomy
Coordinator: Joshua Balsters
Title: The Anterior Cingulate Cortex and Social Cognition: Evidence from anatomy, electrophysiology, computational and translational neuroscience
Symposium abstract: The anterior cingulate cortex (ACC) is implicated in a broad range of behaviours and cognitive processes, but it has been unclear what contribution, if any, the ACC makes to social behaviour. The aim of this symposium is to bring together converging evidence from multiple neuroscientific domains (neuroanatomy, primate electrophysiology, neuroimaging, computational neuroscience, and clinical science) all of which suggest that a specific sub-region of ACC - the gyral surface (ACCg) - plays a crucial role in processing social information. Speaker 1 will outline the role of the ACCg in social decision making. Providing an overview of the anatomical and functional properties that make the ACCg ideally placed to track another's motivation, as well as providing a computational framework which describes the neural processes involved in tracking other people's motivation. Speaker 2 will discuss the anatomy of the ACC, with particular reference to how inter-individual variability in the anatomy and connectivity of the ACC corresponds to social decision making in humans and animals. Speaker 3 provides electrophysiological evidence from primates during a social reward task, highlighting increased neural synchronization between the ACCg and amygdala during prosocial choices. Using a combination of fMRI, EEG, and pharmaceuticals, speaker 4 explores the computations of the ACCg from a Bayesian perspective and demonstrates how dopamine modulates social signals in the ACCg. Speaker 5 highlights the clinical utility of this research and provides the first evidence that aberrant ACCg activity is an indicator of social symptom severity in Autism Spectrum Disorder. Collectively, these data provide converging perspectives supporting a crucial role for the ACCg in social decision making. This symposium will provide a novel perspective on social decision making research, including; new anatomical targets, embedded within novel computational frameworks, generating testable hypotheses for investigating socio-cognitive abilities in health and disease.

Speaker: Matthew Apps
Title: A computational framework for ACCg: predicting value for others
Abstract: To successfully interact with other people, it is vital that we track how motivated they are in order to understand and predict their future decision-making. Whilst much research has examined how we empathise with others, process their mental states, and the kinematics of their actions, very little is known about how we process the degree to which someone else is motivated. In this talk, I put forward a computational framework characterizing the factors that need to be tracked in order to process the value of behavior for another, and thus how motivated they are. Specifically, the value (benefits-costs) of actions must be predicted, and error signals must code for when such expectations about value are erroneous (prediction errors[PEs]). Moreover, this information is processed in an other-oriented reference frame. Thus, if a brain area tracks the motivation of another, it must exclusively code value for others, and not similarly motivate our own actions. I highlight that the gyrus of the anterior cingulate (ACCg) has the anatomical and functional properties that make it ideally placed to track another's motivation. I present fMRI evidence that ACCg activity signals the key properties that would allow it to code the motivation of another. It codes the value of behaviours for others, and signals PEs when another's expectations are erroneous. Moreover, I show that the ACCg processes this information in an other-oriented reference frame, and does not process similar information about the value of our own actions. This computational approach may provide a powerful framework for understanding variability in social cognition abilities in health and disease.

Speaker: Jerome Sallet
Title: Anatomical organisation of the ACC: inter-individual variability and interspecies comparisons
Abstract: The anterior cingulate cortex (ACC) has been associated with supporting reward-guided decision and social cognition. However, the ACC is not a homogenous brain region supporting a single overarching function. Defining the function of a brain area requires first an understanding of its anatomical organization. Its function is indeed constrained by its connections with other brain areas.    Characterizing the role of the different subdivisions of the ACC is notably challenging for two reasons. First, the ACC is classically considered as one functional unit but it has been shown that it is in fact a collection of anatomically distinct subregions. Secondly, there is a huge inter-individual variability of the morphological sulcal patterns. Sulcal patterns could vary either in terms of presence/absence but also in terms of relative position to one from another. Functional neuroimaging studies are classically based on statistical inferences based on the average brain activity of a group of subjects and they ignore those differences. However, taking into account the variability in sulcal patterns has proven to be essential to guide the interpretation of neuroimaging studies. I will first present results regarding the organization of the ACC and the dissociation between sulcal and gyral ACC, but also the impact of the inter-individual variability of the sulcal pattern on its organization.   Secondly I will use neuroimaging tools to identify the intrinsic organization of the cingulate sulcal and gyral regions based on connectivity pattern analysis in animal models used to investigate neuroscience of decision-making and social cognition. Those results will be interpreted in light of the organization of the human ACC regions. Altogether, I aim to provide an overview of the morphology, connectivity and evolutional history of the ACC. 

Speaker: Olga Dal Monte
Title: The neuronal coupling between the anterior cingulate cortex and the amygdala reflects social decision outcomes
Abstract: Accumulating evidence suggests that neuronal coordination across multiple brain regions is important for guiding complex behaviors. One critical component of social processing is the computation of prosocial and antisocial decisions, and recent studies have begun to elucidate how neurons from individual brain regions such as the anterior cingulate gyrus (ACCg) and the basolateral amygdala (BLA) are engaged in social decision-making. Although strong reciprocal connections between BLA and ACCg suggest their functional interactions, the mechanisms by which BLA and ACCg are coordinated remain elusive. In a social reward allocation task in which an actor monkey chooses among delivering juice rewards to himself (Self), another monkey (Other), both himself and the other (Both), or a juice collection bottle (Neither), we recorded local field potential (LFP) activity from ACCg and BLA simultaneously to investigate their neuronal coordination. The actors preferred to deliver rewards to Other over Neither, but also preferred to deliver rewards to Self over Both, providing the behavioral contexts for examining the ACCg-BLA neuronal interaction with respect to prosocial and antisocial decisions. Field-field coherence analyses revealed enhanced coupling in the beta and gamma frequency bands between ACCg and BLA for prosocial reward outcomes (contrasting Other versus Neither) compared to antisocial reward outcomes (contrasting Self versus Both). Over multiple days of testing, we found a significant correlation between the preference to donate reward to Other compared to Neither (prosocial decision) and the degree of synchrony between ACCg and BLA in both beta and gamma bands, whereas this correlation was absent for preference to deliver juice to Self compared to Both (antisocial decision). Our results demonstrate that neuronal synchronization between ACCg and BLA carries unique signatures underlying the computations of prosocial and antisocial decisions.

Speaker: Andreea Diaconescu
Title: Hierarchical learning and dopaminergic mechanisms in social inference
Abstract: Bayesian theories of brain function emphasize the role of uncertainty (or its inverse, precision) and hierarchically organized prediction errors (PEs) in predicting sensory events and their underlying causes. They are particularly critical for social interactions as human intentions are only expressed indirectly requiring inference from observations of ambiguous behaviour. To examine the role of precision-weighted PEs in the social domain, we integrated hierarchical Bayesian modelling and genetic analysis with both functional magnetic resonance imaging (fMRI) and electroencephalography (EEG) in a deception-free task where participants tracked the intentions of an adviser who communicated via videos. In the spatial domain, we found that precision-weighted PEs about the adviser's fidelity were encoded in the dopaminergic midbrain, targets of its projections, including the anterior cingulate and dorsomedial prefrontal cortices, and the anterior temporal-parietal junction (TPJ). The same computational quantity was represented distinctly in time in a subset of these cortical areas: Advice PEs showed an increased positivity peaking at 152ms in the anterior TPJ, then extending to the middle cingulate gyrus between 300 and 450ms, and finally to the dorsomedial PFC at 480ms after outcome onset. Furthermore, to identify the role of dopamine (DA) in social learning, we used a placebo-controlled, double-blind, between-subject design with DA precursor levodopa in both EEG and fMRI. We found that DA potentiates the extent to which social PEs are incorporated to predict intentions via a specific mechanism: Modulating DA availability led to an enhanced representation of precision-weighted advice PEs in the middle cingulate gyrus in particular in Met/Met carriers of the catechol-O-methyltransferase (COMT) gene who show reduced efficacy of COMT and therefore (presumably) higher tonic DA levels. Taken together, these multimodal studies highlight the importance of precision-weighted PE-updates in social learning, and the key role that dopaminergic effects in the middle cingulate cortex play in inferring others' intentions.

Speaker: Joshua Henk Balsters
Title: Aberrant ACCg activity in Autism Spectrum Disorders
Abstract: It has been proposed that in typically developing (TD) individuals, the anterior cingulate gyrus (ACCg) signals prediction errors exclusively for others (social PEs). These social PEs tell us that another person's expectations are different from the actual outcome, allowing us to update our expectations of other people. Although it is well established that individuals with Autism Spectrum Disorder (ASD) often struggle to understand the perspectives of other people, the underlying neural and computational deficits remain unclear. Based on consistent evidence across multiple species and methodologies, we propose that individuals with ASD have a deficit in correctly representing social PEs when tracking the expectations of others, and this putative social deficit might be linked to abnormal activity in the ACCg. Here, we devised a novel fMRI paradigm that allowed us to investigate BOLD activity during expected and unexpected outcomes for three Agents (Self, Other, and Computer). This factorial design allowed us to identify brain regions eliciting social PEs, i.e. where the difference between expected and unexpected outcomes of other people was greater than the difference between expected and unexpected outcomes for the person in the MRI scanner (Self) and the Computer. While the ACCg signalled social PEs in TD individuals, this crucial social signal was altered in individuals with ASD. Importantly, the degree to which social PE signalling was aberrant correlated with social symptom severity as measured by diagnostic tools (Autism Diagnostic Observation Schedule; ADOS). Effective connectivity analyses further revealed that, in TD individuals but not in ASD, the magnitude of social prediction errors was driven by input from the ventromedial prefrontal cortex. These data provide a novel insight into the neural substrates underlying social symptom severity in ASD, and further research into the ACCg and ventromedial prefrontal cortex could provide more targeted therapies to help ameliorate social deficits in ASD.

Date and time: Saturday 17:30
Room: Effectenbeurs
Format: Keynote,
Keynote speaker: Eleanor Maguire
Title: How are memories represented in the human brain?

Date and time: Sunday 8:30
Room: Graan
Format and topic: Symposium, Open science and replication
Coordinator: Eric Maris
Title: How to organize open and reproducible science?
Symposium abstract: Awareness of the importance of open and reproducible science is increasing, as is clear from several initiatives by both formal organizations (funding agencies, journals, etc.) and informal groups of like-minded scientists. It is useful to group these initiatives according to whether they are mainly oriented towards the past or the future. Initiatives oriented towards the past mainly focus on replicating the key findings of a discipline. The motivation for these replication studies is to establish an empirical basis of reproducible phenomena onto which future science can build. Initiatives oriented towards the future mainly focus on sharing and documenting raw data and analysis pipelines. Compared to replication studies, these initiatives are more oriented towards data and analysis pipelines that can be re-used for multiple research questions. The outcomes of these initiatives are standards for organizing data and metadata, as well as computer platforms on which these are implemented. The motivation for these initiatives is to build a framework in which open science is possible within a community that is larger than the collaborating team that collects and analyses the data. An important strength of replication studies is that, after agreement on the protocol, it is very clear what has to be done. For the initiatives focused on sharing and documenting data and scripts, this is less clear: standards must be developed allowing for efficient reuse, but without a concrete list of research questions for this reuse. However, the strength of the sharing and documenting initiatives lies in its promise to produce new insights on the basis of existing data. The objective of this symposium is to present views on the future of open and reproducible science. We will address topics pertaining to the documentation of our results, standards for sharing data and analysis pipelines, and the organization of the workflow at the level of the research institute. The final part of the symposium (approx. 25 min.) is a discussion between the audience and the speakers on how open and reproducible science should be organized in cognitive neuroscience.

Speaker: Dorothy V M Bishop
Title: Reporting standards for neurogenetic studies of humans
Abstract: Together with a group of colleagues, I have been conducting a review of a set of 30 studies that focused on associations between genetic and neurological or behavioural variables and were published in top neuroscience journals. We had anticipated that we might find methodological problems in this field, because it is characterised by huge multivariate datasets that encourage p-hacking, and studies are likely to be underpowered. Although we found some evidence of such problems, we became aware of a deeper problem with studies in this field, concerning comprehensibility of studies. Two issues arose. First, key information was often missing: for instance, it could be hard to know how many genetic variants or brain regions had been studied, whether corrections had been conducted for multiple contrasts, and whether phenotypes were correlated. Computing effect sizes was often impossible. Second, both genetic and neuroscience studies often used highly complex and specialised techniques. When the two are put together, one can end up with a paper that few people have the skills to understand. This becomes a serious problem if the results of these studies cannot be properly evaluated and must be taken on trust. I will highlight instances of good practice that can help improve comprehensibility of papers, and recommend standard reporting methods that should help us move forward.

Speaker: Russell Poldrack
Title: Towards open and reproducible neuroimaging
Abstract: I will discuss a set of initiatives aimed to improve the reproducibility of neuroimaging research. The first centers on the open availability of shared data at multiple levels, from raw data (via the OpenNeuro project) to statistical maps (via the Neurovault project) to coordinate-based data (via the Neurosynth project). These efforts, alongside many others across the field, have demonstrated the utility of fully open data sharing. A second initiative focuses on the organization of data via the Brain Imaging Data Structure (BIDS) standard. The use of a common file organization and metadata framework both improves the usability of shared data, and makes sharing easier. The third initiative centers on the ability to reproduce the specific analysis operations performed on a dataset, known as computational reproducibility. Sharing of analysis code is only a starting point for computational reproducibility, because of complex software dependencies and platform-dependence that can lead to significant differences in results from the same code. I will discuss the use of software containers to address this problem, in particular via the BIDS-Apps project, which is developing software tools that can be easily applied to BIDS-formatted datasets without requiring installation of complex software dependencies. I will conclude by discussing how education needs to evolve to train the next generation of researchers in more reproducible analysis approaches.

Speaker: Robert Oostenveld
Title: Life-cycle Research Data Management (RDM) at the Institute Level
Abstract: Forces from both inside and outside the scientific community charge individual researchers to (a) improve the documentation of their research process, and (b) to share more than just the most interesting findings in a manuscript. Besides scientific manuscripts, also data and detailed analysis pipelines need to be shared. With the detailed documentation and the data in hand, informed colleagues should be capable of confirming the reported outcomes and extend them, by reusing the shared data with other methods, or reuse the shared methods with other data. The sharing of data and methods necessitates conforming to a style of working that is recognized within the scientific community. An efficient implementation furthermore requires an IT implementation and adoption into the scientific workflow. We propose an elementary protocol and describe the corresponding implementation. We distinguish between three collection types, each with their own goal: data acquisition collections (data preservation for internal reuse), research documentation collections (enhancing reproducibility of the published results), and data sharing collections (reusing a study's data and analysis pipeline for other research questions). To realize this workflow at the Donders Institute, we teamed up with researchers, management, IT staff and administrators and implemented the Donders Research Data Repository ( Our RDM repository brings together the expertise from external stakeholders with that of the researchers and enables them to improve their scientific dissemination and enhance their impact.

Date and time: Sunday 8:30
Room: Administratie
Format and topic: Symposium,
Coordinator: Assaf Breska, Leon Y. Deouell
Title: Neural oscillatory mechanisms in temporal predictionAttention
Symposium abstract: Whether in speech, music, biological motion, or traffic lights, our environment is abundant with predictive temporal structure. A growing body of research is concerned with how the brain tracks such environmental rhythms: How are they used to predict the dynamics of upcoming events and to optimize perception, cognition, and action? A central emerging concept is the role of entrainment of low-frequency neural oscillations to the sensory stream as a mechanism of selection and attention in time, but other mechanisms of temporal prediction have been investigated, and many questions remain before we will arrive at an integrative theory of temporal prioritization. This symposium brings together current lines of behavioral and neurophysiological research that address several fundamental questions. While the common denominator is the dynamic adjustment of different frequency bands to different time scales, each presentation adds layers to this theme: the role of the amount of periodicity; the parsing of multiple timescales in the absence of acoustic cues; bottom-up versus top down control; and the biological significance of various neural signatures of temporal predictions. Integration of these data will lead to novel characterizations of the functional and neural architecture of temporal prediction in rhythmic and a-rhythmic contexts.

Speaker: Lucia Melloni
Title: Cortical tracking of natural and artificial online speech
Abstract: Linguistic units such as syllables, words, phrases, and sentences simultaneously unravel during speech processing. How does the brain parse, track, and process these concurrent linguistic units? In this talk, I will discuss our recent efforts to understand whether and how natural speech is processed online, and what role brain rhythms might play in this process. Specifically, I will discuss studies demonstrating that rhythmic cortical activity entrains to the time course of large linguist units, even in the absence of any acoustic cues for the boundaries between phrases and sentences. Similar entrainment is also observed during second language acquisition, and can rapidly appear when learning to parse an artificial language. Together, these studies show that cortical entrainment to linguistic units reliably tracks online speech processing, offering endless possibilities to objectively assess language processing in children, difficult-to-test-populations (e.g., minimally conscious patients), as well as language precursors in animal preparations to allow for cross-species comparison.

Speaker: Jonas Obleser, Lorenz Fiedler, Malte Wostmann
Title: Neural filters for auditory attention in time and space
Abstract: To listen means to regulate auditory attention dynamically in time and space. Listeners must be able to follow the temporal structure of acoustic stimulation (i.e., "entrain" to the sensory environment), but they also require mechanisms to actively disengage (i.e., "functionally inhibit") brain areas processing task-irrelevant information at certain times. A guiding hypothesis of our work is that neural oscillatory entrainment (~1--5 Hz) indexes a listener's faithful tracking of "external" auditory events, while the power of alpha oscillations (~8--13 Hz) plays a sizeable role in phenomena of "internal" neural communication such as top-down gain control and prediction. I will present recent electro-/magnetoencephalography evidence from our lab on how listeners use both mechanisms simultaneously to regulate auditory attention. First, when listeners attend one of two spatially distinct, but temporally concurrent speech streams, the time courses of sensory entrainment and alpha power lateralization are both being coupled to the dynamics of on-going speech, but with a relative phase-lag of almost 180° to each other. Second, we also probe auditory attention in more ecologically valid scenarios, where a to-be-attended story is spatially fused with a to-be ignored story at a stochastically varying attend-to-ignore sound-pressure ratio. Here, neural "entrainment" allows to reconstruct attention allocation at the single-subject level in most participants. Concurrently and on a slower timescale, the time course of alpha power tracks and allows to reconstruct the sound pressure change in ignored speech. Notably, reconstruction accuracy of attentional focus from entrainment vs. from alpha power provides non-redundant information across individuals. In sum, these data demonstrate that auditory attention in time and space utilizes two distinct yet complementary neurobiological mechanisms of sensory entrainment versus functional inhibition.

Speaker: Assaf Breska, Leon Y. Deouell
Title: Distinct mechanisms of rhythm- and memory-based attention shifting in time
Abstract: The brain can take advantage of temporal structure in continuous streams to predictively allocate attention and motor preparation in time. Rhythmic streams, such as speech and biological motion, presumably enable temporal predictions by entraining low-frequency neural oscillations. However, temporal predictions can also be generated in isolation, based on prior knowledge of intervals between events. Are temporal predictions in rhythmic streams fully explained by memory-based prediction or are they unique, as predicted by entrainment models? I will address this issue using behavioral and EEG data, and computational modelling. In the first study, rhythmic streams were compared to streams that enable prediction based mainly on memorizing intervals. In contrast to the prediction of a computational entrainment model, I will argue that delta and alpha patterns do not differ between the two conditions. However, rhythm-based predictions uniquely result in resonance, expressed in preparation-related slow brain activity, and in predictive behavioral costs at off-beat times rather than facilitation in on-beat times. Based on a second study, in which rhythm on-beat or off-beat times were made relevant in separate blocks, I will show that the effects of relevance (appearing at relevant versus irrelevant times regardless of the beat) and rhythmicity (appearing on- versus off-beat with the rhythm) are behaviorally additive and are expressed in distinct processing stages. Finally, I will describe the behavior of individuals with degenerative cerebellar ataxia, known to impair explicit interval timing, who benefited from rhythmic cuing to a comparable level as did healthy age-matched controls, but were impaired in their ability to benefit from a memorized interval. Together, these findings support distinct mechanisms of rhythm- and interval-based temporal prediction, and signify resonance and increased costs of unexpected times as the unique signatures of rhythm-based temporal prediction.

Speaker: Christoph Kayser
Title: The role of rhythmic brain activity in encoding speech dynamics
Abstract: Rhythmic brain activity within the temporal lobe at the time scale below about 100ms is known to align with temporal and acoustic regularities in the auditory environment. This speech-to-brain entrainment supposedly reflects changes in cortical neural excitation that is aligned with the predictive structure of speech, and which may provide a mechanism that supports the parsing of continuous speech into individual segments. To understand how the predictive regularity inherent to speech affects speech-to-brain entrainment we systematically manipulated the speech's temporal predictive structure by parametrically altering the distribution of pauses between syllables or words. This manipulation renders the local speech rate irregular while preserving intelligibility and the envelope fluctuations of the acoustic signal. Recording EEG activity in human participants, we found that this manipulation did not alter neural processes reflecting the encoding of individual sound transients, such as evoked potentials. However, it significantly reduced the fidelity of auditory delta (but not theta) band entrainment to the speech envelope. These results show that speech-to-brain entrainment in distinct frequency bands reflects distinct processes and suggest that delta entrainment is under top-down control and likely reflects prefrontal processes that are sensitive to acoustical regularities rather than the bottom-up encoding of acoustic features.

Date and time: Sunday 8:30
Room: Berlage
Format and topic: Symposium, Decision-making
Coordinator: Zsuzsika Sjoerds
Title: Dilemmas in the face of uncertainty: Computational approaches to adaptive behavior
Symposium abstract: Optimizing our actions in a dynamic world typically involves a tradeoff, or dilemma, between two seemingly opposing control processes. This dilemma has been described using various terms, including 'exploration versus exploitation', 'goal-directed versus habitual', 'flexibility versus persistence', 'reflexive versus reflective', and 'speed versus accuracy'. Although this dilemma has been addressed from different angles, they all involve control processes aimed to optimize behavior, often under various forms of uncertainty. The question rises how the various perspectives on this dilemma differ from each other, and what their commonalities are. Computational modeling is a powerful tool to formalize and investigate these adaptive processes, enabling the quantification of underlying latent variables. In this symposium, we bring together an international group of scientists who applied computational modeling approaches to study adaptive behavior from different angles within the dilemma's spectrum. After a short introduction on the state of affairs by the symposium organizer, speaker 1 will present a theoretical framework: the 'Antagonistic Constraints and Dilemmas of Control'. This framework introduces a number of antagonistic challenges or control dilemmas that are thought to regulate goal-directed behavior in a continually changing environment. Next, speaker 2 will discuss advanced computational approaches to study multi-step decision-making, offering a novel resolution to the reflexive versus reflective dichotomy of behavior. Subsequently, speaker 3 will focus on the dilemma between maintaining stable beliefs versus belief updating, providing evidence for a causal role of the human catecholamine systems in the regulation of learning rate. Speaker 4 will end the symposium by leaving the lab environment, and showing real-life examples of how patterns of exploratory choice in a supermarket environment oppose predictions of normative models based on objective rewards. Together, the presentations in this symposium will provide converging insights in the cognitive and neural mechanisms underlying adaptive behavior.

Speaker: Thomas Goschke
Title: Control Dilemmas and the Dynamic Regulation of Complementary Control Modes
Abstract: In this presentation I will present a theoretical framework (ACDC: Antagonistic Constraints and Dilemmas of Control) which rests on the assumption that organisms pursuing goal-directed action in a continually changing environment face a number of antagonistic challenges or control dilemmas, which require a dynamic, context-dependent balancing of complementary control modes. On the one hand, agents must maintain goals and task-sets and shield them from distracting information; on the other hand, agents must switch rapidly and flexibly between goals in order to adapt to significant changes (stability-flexibility-dilemma). Likewise, while coherent action requires goal-directed focusing of attention on task-relevant stimuli, on the other hand, agents must monitor the environment for task-irrelevant but potentially significant stimuli, which may eventually trigger an updating of goal representations in working memory (selection-monitoring-dilemma). Such control dilemmas raise the fundamental question how the balance between complementary control modes (stable goal maintenance vs. flexible goal updating) and control parameters (e.g., updating threshold; attention breadth) is regulated and adapted dynamically to changing contexts. I will present empirical findings and computational (connectionist) modeling studies from the Dresden Collaborative Research Centre "Volition and Cognitive Control" on how control parameters and the balance between complementary control states are modulated by conflict, affect, and reward.

Speaker: Cristian Buc Calderon, Wim Gevers, Tom Verguts
Title: Probabilistic evidence integration in multi-step decision making
Abstract: Imagine leaving your house in search for food in the neighborhood. Outside, going left subsequently affords you a second left-right choice between Thaï and Italian food, whereas going right affords you another left-right choice between Mexican and Lebanese food. This typical multi-step decision making situation pervades daily life, but its underlying mechanisms remain obscure. In this talk, we will distinguish four prominent models of multi-step decision making, namely serial stage, hierarchical evidence integration (HEI), hierarchical leaky competing accumulation (HLCA), and probabilistic evidence integration (PEI). Each model makes distinct predictions regarding the dynamics of evidence integration at the first decision step. We will further discuss how we were able to disentangle these models using a novel two-step reward-based decision paradigm. Our paradigm allowed testing a unique prediction of the HCLA and PEI models. Strikingly, but as predicted by these models, we show that the first-step decision dynamics were initially attracted towards the choice representing the highest mean before being redirected towards the choice representing the maximal reward. Moreover, we disentangled HCLA from PEI by an analysis of the attraction dynamics of both models. Only the PEI turned out to be consistent with the empirical attraction dynamics. Our results demonstrate that multi-step evidence integration adheres to probabilistic inference principles where future outcomes of potential actions are progressively unraveled during the decision process. More generally, the PEI model proposes a novel twist on the reflexive versus reflective dichotomy of behavior. In particular, in this model, an agent starts off as acting in a reflexive way, but as time goes on behavior gradually becomes reflective, possibly even at subsecond time scale.

Speaker: Marieke Jepma, Sander Nieuwenhuis
Title: Catecholaminergic regulation of learning rate in a dynamic environment
Abstract: Making accurate predictions in a dynamic world requires flexibly adapting one's rate of learning to the rate of environmental change. Whereas beliefs should be stable in noisy but otherwise static environments, they should be updated following environmental change. Recent studies have examined the computational mechanisms by which various environmental factors determine the degree to which new outcomes drive belief updating (i.e., the 'learning rate'). However, the brain mechanisms, and in particular the neuromodulators, involved in this process are still largely unknown. The brain-wide neurophysiological effects of the catecholamines norepinephrine and dopamine on stimulus-evoked cortical responses suggest that the catecholamine systems are well positioned to regulate learning about change, but more direct evidence for this idea is scant. Here, we address this issue in two studies employing scalp electrophysiology, computational modeling, and pharmacology. In both studies, the P3 component of the EEG -- an electrophysiological index of outcome-evoked phasic catecholamine release in the cortex -- predicted learning rate, and mediated the effect of prediction-error magnitude on learning rate. P3 amplitude also mediated the effects of two computational variables, capturing the unexpectedness of outcomes and the uncertainty of preexisting beliefs, on learning rate. Furthermore, a pharmacological manipulation of catecholamine activity affected learning rate following unsignaled task changes, in a way that depended on participants' baseline learning rate. Our findings provide converging evidence for a causal role of the human catecholamine systems in the adaptation of learning rate as a function of environmental change.

Speaker: Bradley Love
Title: Different Modes of Exploration
Abstract: In uncertain environments, effective decision makers balance exploiting options that are currently preferred against exploring alternative options that may prove superior. For example, a honeybee foraging for nectar must decide whether to continue exploiting the current patch or move to a new location. When the relative reward of options changes over time, humans explore in a normatively correct fashion, exploring more often when they are uncertain about the relative value of competing options. This behaviour appears driven by a frontal dopamine system modulated by COMT enzyme activity. The prevailing view is that exploratory behaviour is uncertainty minimising, capacity-limited, and frontally mediated. However, rewards in these laboratory studies were objective (e.g., monetary payoff), whereas many real-world decision environments involve subjective evaluations of reward (e.g., satisfaction with food choice). In such cases, rather than choices following preferences, preferences may follow choices with subjective reward (i.e., value) to maximise coherency between preferences and behaviour. If so, increasing coherency would lessen the tendency to explore while uncertainty increases, contrary to previous findings. To evaluate this possibility, we examined the exploratory choices of more than 280,000 anonymised individuals in supermarkets over several years. Consumers' patterns of exploratory choice ran counter to normative models for objective rewards -- the longer the exploitation streak for a product, the less likely were people to explore an alternative. Furthermore, customers preferred coupons to explore alternative products when they have recently started an exploitation streak. These findings suggest interventions to promote healthy lifestyle choices.

Date and time: Sunday 8:30
Room: Veiling
Format and topic: Symposium, Memory
Coordinator: Branka Milivojevic
Title: Hippocampal representations of realistic memories: event segmentation, integration and organisation of memories in naturalistic contexts
Symposium abstract: Memories for individual events are not stored in isolation, but rather, form associative networks of memories. However, the neural mechanisms underlying formation of these large-scale memory networks, and their organising principles have remained elusive. Of particular interest is how individual events are extracted from the continuous stream of experience, and subsequently, how those events are integrated into cohesive memory networks. Recent advances in both technology and neuroimaging (analysis) methods have enabled researchers to utilise more realistic stimulus materials and more controlled encoding conditions with the aim of studying neural mechanisms underlying memory for every-day, realistic events. In this symposium, we present a series of talks which demonstrate advances to our understanding of hippocampal mechanisms involved in event integration during realistic memory formation. All three speakers have adapted dynamic material such as movies, life simulation games, and virtual reality as a way of simulating reality in neuroimaging research of episodic memory.  Speaker 1 will discuss how context shifts lead to segregation of individual events out of continuous stream of experience, and subsequently encoded into memory. Speaker 2 addresses the importance of spatial and temporal contexts, and their interaction, for organisation of episodic memories. And finally Speaker 3 will discuss how events linked through stories lead to formation of dynamic narrative networks in memory.

Speaker: Aya Ben-Yakov, Rik Henson
Title: Hippocampal sensitivity to event boundaries in the encoding of narrative episodes
Abstract: An extensive body of research has established that the hippocampus plays a pivotal role in the encoding of new associations. Yet it remains unclear how entire episodes that unfold over time are bound together in memory. Real-life episodes can be viewed as a sequence of interrelated episodic elements, and their encoding may be incremental, such that each element that is encountered is registered to memory. Conversely, the episode may be stored in a temporary buffer and registered to long-term memory as a cohesive unit when it has come to closure. Using short film clips as memoranda, we find that hippocampal encoding-related activity is time-locked to the offset of the event, potentially reflecting the encoding of a bound representation to long-term memory. Notably, when distinct clips were presented in immediate succession, the hippocampus responded at the offset of each event, suggesting hippocampal activity is triggered the occurrence of event boundaries (transition between events). However, while brief film clips mimic several aspects of real-life, they are still discrete events. To determine whether event boundaries drive hippocampal activity in an ongoing experience, we analysed brain activity of over 200 participants who viewed a naturalistic film and found that the hippocampus responded both reliably and specifically to shifts between scenes. Taken together with the previous findings, these results suggest that during encoding of a continuous experience, event boundaries drive hippocampal processing, potentially supporting the transformation of the continuous stream of information into distinct episodic representations.

Speaker: Lorena Deuker, Nikolai Axmacher, Christian Doeller
Title: The role of space and time in organisation of episodic memories
Abstract: For an adaptive memory system it is not sufficient to store an individual event, but it must also represent how this event relates to other events in memory. In recent years it has become clear that the hippocampus supports this function not only by storing associations between events, but also by representing the degree of connectedness between events. Events in episodic memory are effectively defined by their spatial and temporal context, and it is of considerable theoretical interest how the brain organises the relationships between events in space and time. We addressed this question in a series of studies where we used neuroimaging while participants navigated virtual environments or viewed rich visual material, to track the emergence of event associations in the brain while systematically varying either spatial or temporal relationships between events. We also examined the interaction between these two factors, by independently varying spatial and temporal relationships between events. Here, I will present our work which demonstrates that changes in hippocampal pattern similarity reflected not only the temporal order between events - modeled here by presenting participants with days in the life of an affluent virtual family - but also the number of hours between events. Furthermore, we also used VR to investigate specific relationships between events, such as a spatial layout, while keeping visual input constant, thereby allowing us to study brain activity in conditions in which participants have to rely solely on in their internal representation of the spatial relationship between events. And finally, using navigation in a virtual city environment, we demonstrated that a change in hippocampal pattern similarity reflected the spatial as well as the temporal distances between objects along a specific route as remembered by the participant. Taken together, these studies demonstrate the importance of spatial and temporal context in organising our memories for events.

Speaker: Branka Milivojevic, Silvy Collin, Alejandro Vicente Grabovetsky, Meryl Varadinov, Christian Doeller
Title: Dynamics of hierarchical mnemonic networks in the hippocampus
Abstract: Our experiences form networks of related memories, and associations between events can cut across spatial and temporal proximity to form a coherent narrative, which in turn may provide another type of context for organisation of episodic memories. We used a combination of realistic stimuli (movies and stimuli developed using The Sims 3 life-simulation game), fMRI and across-voxel pattern similarity to examine whether formation of narrative-based memory networks in humans relies on hippocampal mechanisms which also underlie formation of spatiotemporal contexts. First, we demonstrated that patterns of hippocampal activity can be used to differentiate between narratives and that these narrative-context representations diverge gradually over time akin to remapping-induced spatial maps represented by rodent place cells. Secondly, we showed that event representations are remarkably dynamic and follow hierarchical organisation in memory. In one study, we showed that hippocampal and medial prefrontal (mPFC) neural patterns became more similar once previously unrelated events became linked through a new event. Further, we have shown that the scale of these event networks increases along the long axis of the hippocampus, with anterior hippocampus providing coarser event representations corresponding to multi-event narratives. And finally, we used a hierarchical memory task which required narrative updating with new events. Here we demonstrated a threefold hippocampal involvement in updating: [1] representing new events, [2] integrating these new events with earlier events, and [3] strengthening of the representation of earlier events after updating. In combination, these results suggest that memories are organised into networks of related events based on narrative contexts, which are highly dynamic and flexible, and follow a similar hierarchical organisation like space.

Date and time: Sunday 8:30
Room: Verwey
Format and topic: Symposium, Memory
Coordinator: Annelinde Vandenbroucke
Title: Is visual cortex relevant for visual working memory?
Symposium abstract: Visual working memory is the mechanisms through which we maintain visual information over a short period of time. Several studies have demonstrated that visual cortex is not only involved in the initial processing of visual information, but also in the maintenance of this information. Monkey physiology experiments have shown increased spiking activity in early visual cortex during the delay of a working memory task (Super et al., 2001; Pasternak & Greenlee, 2005; van Kerkoerle, 2017). Human fMRI studies have even shown that the exact information that is maintained in working memory can be decoded from these areas (Harrison & Tong 2009; Serences et al., 2009). However, whether visual cortex is necessarily involved in visual working memory remains under debate (Saad et al., 2015; Bettencourt et al., 2015; Mendoza-Halliday et al., 2014). Especially when visual cortex is engaged in another task, such as for example the processing of new information (Bettencourt et al., 2015, van Kerkoerle et al., 2017), or when focusing on a different working memory representation (Rose et al., 2016), the precise involvement of visual cortex is not yet clear. In this symposium, we aim to provide insight into the role of the visual cortex by looking at evidence from different experimental approaches. Speaker 1 will discuss TMS studies showing that visual cortex has a dissociable role in working memory depending on how it is behaviorally measured. Speaker 2 will show how activity-silent memory traces in sensory areas can be reactivated using TMS in combination with EEG. Speaker 3 will discuss evidence from fMRI data in combination with computational modeling showing that visual cortex is also involved in the suppression of discarded working memory information. Speaker 4 will present monkey physiology data showing that there is evidence for feedback to V1 during a working memory delay.

Speaker: Christianne Jacobs
Title: Aphantasia: a new model to study the role of mental imagery in visual working memory
Abstract: Some individuals cannot generate images in front of their “mind’s eye”. This phenomenon has recently been described and the term aphantasia was coined to refer to this absence of mental imagery in individuals (Zeman et al., 2010; Zeman et al., 2015). Aphantasia can serve as a model to investigate the functional relevance of mental imagery in other processes of visual cognition. We were specifically interested in the relationship between mental imagery and visual working memory, two processes that have often been linked and that appear to originate from similar mental and neural representations (e.g. Albers et al., 2013). In this talk, I will present visual working memory data from an aphantasic individual in comparison to matched controls with intact mental imagery. Overall visual working memory performance in the aphantasic individual was comparable to controls. However, when working memory retrieval required a higher level of precision, her performance suffered. This could indicate that normal imagers employ active mental imagery when they need to retrieve more detailed information from visual working memory.

Speaker: Bradley R. Postle, Qing Yu, Olivia Gosseries, Nathan S. Rose
Title: Visual representation in visual working memory
Abstract: The sensorimotor recruitment hypothesis holds that the retention of stimulus information, in the absence of environmental input, is supported by the very same circuits and computations that are responsible for the real-time perception of these stimuli. This applies, in particular, to situations in which this information is held in the focus of attention (FoA) in anticipation of an imminent memory-guided response. This talk will consider the role of early visual cortex, in humans, in situations such as this, as well as for the retention of information that is temporarily held in working memory but outside the FoA (“unattended memory items” [UMI]). For information in the FoA, multivariate decoding and encoding analyses indicate sustained, albeit dynamic, stimulus representation in calcarine and extracalcarine cortex. Hints about the source of top-down control of these representations comes from intraparietal sulcus (IPS). Although nonspatial item-specific stimulus information cannot be decoded from IPS, its delay-period BOLD signal intensity scales negatively with occipital decoding performance (increased load = increased IPS BOLD = decreased occipital decoding), and the strength of IPS connectivity with thalamus and with occipital cortex is negatively related to behavioral precision. Turning to UMIs, these can be momentarily reactivated with a pulse of TMS targeting either occipital or parietal cortex. Our working hypothesis is that these are maintained in short-lived patterns of stimulus-specific strengthened connectivity that, although not “active”, are under cognitive control. We are actively exploring the hypothesis that UMI reactivation via parietal TMS reflects the state of trial-, but not stimulus-, specific information in a parietal salience map, the implementation of which may be effected by the above-mentioned IPS-occipital connectivity.

Speaker: Annelinde Vandenbroucke
Title: Dissociable neural mechanisms in visual cortex underlie currently-relevant, future-relevant and discarded working memory representations
Abstract: In daily life, we use visual working memory to guide our actions. While attending to currently-relevant information, we must simultaneously maintain future-relevant information, and discard future-irrelevant information. However, the neural mechanism(s) by which unattended, but future-relevant, working memory (WM) information is maintained, and future-irrelevant information is discarded from working memory, are not well understood. In this talk, I will present functional magnetic resonance imaging (fMRI) data in combination with multivariate pattern analysis and computational modeling of simulated data. Although attended, currently-relevant WM information was maintained through representations in visual, parietal and posterior frontal brain regions, future-relevant WM items that were not in the focus of attention were neither inhibited nor actively maintained in these areas. Interestingly, unattended WM items that needed to be discarded from WM were suppressed in early visual cortex. Computational modeling showed that this suppression can be explained by a change in tuning curve functions underlying the neural representation of the discarded WM item. These findings suggest that visual cortex is not only involved in the active maintenance of WM information, but also in the suppression of information that is no longer needed.

Speaker: Timo van Kerkoerle
Title: Feedback modulation of early visual areas in monkey primary visual cortex
Abstract: Working memory is essential for daily life, but the cortical mechanisms are largely unclear. While attention and working memory have been shown to robustly modulate prefrontal as well as early visual areas when measured using fMRI in humans, the effects of spiking activity in monkey early visual areas has only been shown consistently for selective attention while the effects of working memory have been weak or absent. As BOLD effects have been found to correlate most strongly with synaptic activity, this could indicate that working memory only has a consistent effect on synaptic activity in early visual areas while leaving spiking activity relatively unaffected. If true, it becomes unclear what the role of early visual areas in working memory could be. We investigated the cortical mechanisms of attention and working memory in monkey primary visual cortex (V1) by recording both spiking activity as well as the current-source density, corresponding to synaptic inputs. We found that the effect on spiking activity was significantly stronger for attention than for working memory, while the CSD showed a feedback signature that was similar in both profile as well as amplitude for attention and working memory. This indicates that feedback input corresponds to a gain factor that only gives a strong modulation of spiking activity in early visual areas in the presence of a feedforward drive. Furthermore, we showed that the working memory signal could be abolished with a visual mask, but reappeared at a later point in time. Our results bring together the results of attention and working memory in humans with those in non-human primates. They suggest that working memory representations in early visual areas are present but can be easily perturbed. The role of these fragile representation could be to provide detail when needed, while more robust representations are stored in higher cortical areas.

Date and time: Sunday 11:00
Room: Effectenbeurs
Format: Keynote,
Keynote speaker: Satu Palva
Title: Functional significance of large-scale neuronal interactions and scale-free dynamics in human cognition

Date and time: Sunday 13:30
Room: Graan
Format and topic: Symposium, Decision-making
Coordinator: Mark Bellgrove
Title: Functional interactions between neural systems for attention and decision making
Symposium abstract: The literatures on attention and decision making are two of the most well developed in modern cognitive neuroscience each of which has benefited from a major multidisciplinary effort across several decades. Research on attention has focused on neural systems that selectively allocate processing resources in accordance with current goals to facilitate perception. In parallel, research on decision making has delineated the core neural mechanisms through which sensory inputs are translated into action. Despite the fact that both fields are effectively concerned with the neural determinants of perceptual performance, there has been remarkably little integration of findings or paradigms. This symposium gathers together a group of speakers who are part of a recently emerging effort to bridge these conceptual and technical gaps. The talks will present data confirming that the neural processes governing attention and decision making are intimately linked and will highlight new methods and paradigms for understanding their critical interactions. Importantly, this symposium draws together speakers with differing perspectives who have used a variety of techniques (psychophysics, computational modeling, single-unit recordings, non-invasive recordings) to address a common problem.

Speaker: Floris De Lange
Title: Biased information sampling in perceptual decision-making
Abstract: Perceptual decision-making is the process of translating sensory signals into categorical decisions. Computational and neural models typically dissociate between the encoding of sensory information in sensory circuits and their temporal integration of these signals into a decision variable in decision areas. Whereas attention is thought to modify the gain in sensory circuits, decision biases may only arise at a later stage, in decision areas. In my talk, I will challenge this view. I will present neurophysiological and neuroimaging data that show how decision biases propagate to sensory circuits, and how biased sampling in sensory circuits drive decisions. These data are best captured by conceptualizing perceptual decision-making as a process of probabilistic inference in reciprocally connected sensory and integration circuits.

Speaker: Jacqueline Gottleib
Title: How does attention respond to the demands of a decision situation?
Abstract: Scores of behavioral and neurophysiological studies support the idea that our experience is constructed through acts of selective attention or, to quote William James' famous dictum, "my experience is what I agree to attend to". Compelling expressions of this fact are active sensing behaviors (whisking, listening or looking) through which animals sample information from actively selected cues. Strikingly however, despite their ubiquity and importance, active sampling mechanisms have yet to be incorporated in decision theories, which start instead from the simplistic premise that the evidence supporting a decision is given (by the environment or context) and outside the decision maker's control. This profound disconnect between studies of attention and decision making leaves large gaps in our understanding of top-down attention control -- specifically, the ways in which individuals organize their information sampling policies to serve a decision situation. I will describe novel paradigms that examine this question in humans and non-human primates, behavioral evidence that sampling decisions depend on the rewards and uncertainty of ongoing actions, and single-neuron correlates of uncertainty and reward-based control. I will discuss the implications of these findings for attention and decision making and, time permitting, for processes of curiosity and creativity that remain poorly understood.

Speaker: Ole Jensen
Title: Implicit biasing of computational resource allocation by prediction and reward
Abstract: It is well established that attention and perception can be biased by external cues. These biases are reflected by robust modulations in alpha and gamma oscillations in sensory regions. This oscillatory activity impacts the resources allocated to the processing of sensory information and thus decisions making. We have been conducting a set of experiments in which we introduced various implicit biases. In one experiment we paired visual stimuli with reward or loss. Stimuli associated with reward or loss resulted in stronger alpha depression when the stimuli were presented both as targets and distractors. This suggests that processing resources are allocated to stimuli with higher valence in early sensory regions. In another set of experiments we have investigated how implicit predictions modulate brain oscillations. This has been done using paradigms in which abstract stimuli or words could be predicted. Our core finding is that alpha is depressed in sensory regions prior to the onset of predicted stimuli. This suggests that processing resources are allocated to stimuli that can be predicted. We are currently in the process of uncovering the frontal structures implementing these top-down biases. Recent connectivity analysis suggests that left prefrontal gamma activity reflects the top-down drive. In conclusion we suggest that various factors implicitly biases the allocation of computational resources in sensory regions and this is reflected by modulation in alpha power. In future work we will identify the frontal and subcortical networks implementing these biases and how they impact decision making.

Speaker: Redmond O'Connell
Title: How does spatial attention modulate neural mechanisms of human perceptual choice?
Abstract: The field of decision neuroscience has made great progress in characterizing the core neural mechanisms necessary for making simple perceptual decisions. In particular, a great deal of computational and neurophysiological research has characterized evidence accumulation processes that account for the timing and accuracy of choices in a wide range of scenarios. To date however, there has been limited progress in establishing how these decision processes interface with the other brain systems that support choice behaviour. For example, the benefit to perception afforded by the direction of spatial attention to a target location is well established, yet this phenomenon is rarely considered in decision neuroscience. A series of experiments were conducted in which participants monitored stimuli at numerous locations for unpredictable feature changes. Our results reveal that under conditions of spatial uncertainty, early target selection signals modulate the onset and rate of neural evidence accumulation, and thereby the timing and rate of perceptual reports. We also show that the onsets of target selection and evidence accumulation mechanisms are highly sensitive to asymmetries in the distribution of spatial attention. Our work highlights important top-down influences that account for a significant portion of variability in neural and behavioural indices of perceptual decision making.

Date and time: Sunday 13:30
Room: Administratie
Format and topic: Symposium, Electrophysiology methods
Coordinator: Rasa Gulbinaite
Title: Can we measure without influencing the measured: Rhythmic sensory stimulation in brain research
Symposium abstract: In invasive and non-invasive brain recordings, responses to rhythmic sensory stimuli (lights, sounds, touch) are also rhythmic, with the frequency matching the stimulus or its harmonics. Simultaneously presented stimuli oscillating at different frequencies generate differentiable spectral signatures in brain recordings (termed steady-state evoked potentials, SSEPs), making it possible to "frequency-tag" and track processing of each stimulus separately. Frequency tagging is a powerful technique used in fundamental neuroscience to study low-level vision and higher-level cognitive processes (e.g. attention, face processing, working memory), applied in SSEP-based brain-computer interfaces, and used as a diagnostic tool in the clinical domain. However, there is an implicit assumption underlying the frequency tagging approach: Rhythmic stimuli do not interact with ongoing endogenous brain rhythms in a frequency-specific manner, and do not affect the neural process that is frequency-tagged. An alternative hypothesis is that exogenous rhythmic stimulation, at least at certain frequencies, can entrain intrinsic brain rhythms and thus have frequency-specific behavioral effects. In this symposium we will address these two controversial standpoints by taking a pragmatic approach. We will delineate the boundary conditions for using rhythmic stimulation for frequency-tagging, and for noninvasive modulation of intrinsic brain rhythms. Speaker 1 will introduce the frequency tagging approach, with a focus on selective attention and multisensory interaction studies, and will provide critical evaluation of rhythmic stimulation as means to entrain intrinsic brain rhythms. Speaker 2 will provide evidence for behaviorally relevant network-level interactions between flicker and intrinsic brain oscillations at resonant frequencies in different frequency bands. Speaker 3 will demonstrate the potential of flicker as a tool to study cognitive functions associated with specific oscillatory patterns, with a focus on alpha-band rhythm and perceptual processes. Speaker 4 will elucidate the importance of choosing the appropriate tagging frequency based on the inherent speed of the cognitive process of interest.

Speaker: Christian Keitel
Title: Driving brain rhythms through dynamic visual stimulation
Abstract: Soon after intrinsic brain rhythms were first described in Human EEG, researchers attempted to drive these rhythms externally through periodic visual stimulation [1]. This approach, later known as "photic driving", "frequency tagging" or "visual entrainment" spawned various lines of research aimed at studying visual perception and cognitive influences thereon. Although based on highly similar paradigms, frequency tagging and entrainment studies nowadays emphasize different aspects of driven brain rhythms. With a recent surge in entrainment studies some apparent inconsistencies have surfaced that warrant reconciliation [2]. Here, I will introduce the basics of rhythmic stimulus-driven brain responses from a frequency-tagging perspective and highlight some advantages of the technique. Recent experimental data will demonstrate its versatility in studying attentional allocation in multi-stimulus situations as well as multisensory interactions [3, 4]. I will then re-iterate where findings from frequency-tagging experiments contrast with the notion of entrainment. Finally, results of a current study using quasi-rhythmic visual stimulation will shed some light on the relationship of stimulus-driven and intrinsic brain rhythms [5].

Speaker: Rasa Gulbinaite
Title: Measuring selective attention using frequency tagging: Does the frequency matter?
Abstract: In a review paper published around 40 years ago, David Regan wrote: "If you wish to study, attention or orienting response, then steady-state evoked potentials (SSEPs) are most likely completely useless." [1] The reputation of SSEPs was restored in the 1990s, when Morgan et al. (1996) demonstrated that not only SSEP frequency follows the rhythm of the stimulus but also SSEP amplitude is enhanced when the stimulus is attended [2]. The range of frequencies used for "frequency-tagging" stimuli in the studies of selective attention varies considerably: From 3 Hz to ~30 Hz. Are all tagging frequencies equally "useful" for tagging selective attention? Is attentional modulation of SSEP amplitude constant across frequencies? We focused on visual flicker in the alpha-band brain rhythm (7-13 Hz), due to the established role of alpha oscillations in supporting attentional processes. In a simultaneous EEG and MEG study, we found that different flicker frequencies drive different brain networks, with the strongest effects of attention on the interaction between alpha-band flicker (10 Hz) and endogenous alpha oscillations. In a follow up study, we demonstrated that the effectiveness of spatial attention is most impaired when flicker frequency matches the participant's occipital alpha rhythm. Motivated by these findings, we sought to characterize attention effects on SSEP amplitude across a wide frequency range (3-80 Hz) using custom-built LED-based hardware to ensure high stimulus timing precision. We found that attentional modulation of SSEP amplitude is the most pronounced in alpha and gamma (30-80 Hz) frequency bands. Together, these findings speak against frequency neutrality when using frequency tagging to study selective attention. In other words, SSEPs are not completely useless, but some tagging frequencies are more useful than others.

Speaker: Sayeed A. D. Kizuk, Kyle E. Mathewson
Title: Visual entrainment at alpha frequencies interacts with endogenously modulated oscillations to influence perceptual sensitivity
Abstract: Rhythmic external sensory stimulation of the brain entrains oscillations of brain activity at many frequencies. The brain also operates at these rhythms endogenously, for instance modulating alpha oscillations for flexible inhibition of sensory processing. We and other have found evidence that alpha oscillations represent a pulsating inhibition [1], with sensory processing modulated by the phase of ongoing alpha oscillations [2]. Rhythmic visual stimulation phase-locks ongoing alpha oscillations, creating rhythms in target sensitivity of the same period [3-4]. Phase-locking is increased with no concomitant increase in power, evidence that entrainment influences ongoing oscillations. If entrainment influences endogenous activity, it should be more effective in parts of the visual field processed in retinotopic areas where oscillations are endogenously larger. To test this hypothesis, we had participants lateralize their alpha oscillations during a target detection task following a spatial cue indicating the likely lateralized location of the target [5]. This spatial cue lateralized alpha oscillations as predicted, with increased alpha power over the hemisphere contralateral to the unattended location. Target detection was poorer in the unattended visual field. Between the spatial cue and the target, we rhythmically stimulated bilateral visual fields where the target could subsequently appear. Lateralized targets appeared at different phases with respect to the preceding entrainment. Target detection in the unattended field was modulated by this phase, with better detection for targets in-time with the entrainers. No visual modulation was present in the attended hemifield processed by the hemisphere with lower alpha oscillations. These results show evidence that exogenous entrainment can modulate ongoing endogenous oscillatory activity, phase-locking periods of preferential processing. This technique allows for better understanding of the role of endogenous oscillatory activity in brain processing, perception, and cognition.

Speaker: Bruno Rossion
Title: Understanding high level brain function with frequency-tagging
Abstract: Visual stimulation at a rapid periodic frequency rate, e.g. a flickering light at 17 Hz, leads to periodic EEG responses exactly at that frequency [1]. This frequency-tagging approach is often carried out under the assumption that the exact same stimulus should be flickered at a sufficiently high frequency (e.g., above 10 Hz) to "entrain" neural populations and lead to "oscillatory" brain responses stable in amplitude and phase ("steady-state visual evoked potentials", SSVEPs, after [2]). However, since (1) high level visual regions adapt to repeated identical visual stimuli and (2) critical information flickering above 10 Hz is not processed beyond primary visual areas, the frequency-tagging approach has been essentially confined to study low-level sensory processes, including their modulation by selective and spatial attention. In this presentation I will show how changing high level visual properties at a periodic frequency rate and fitting this rate to the time-constant of the process of interest extends this highly sensitive and objective approach to understand high level brain functions, such as generic and individual face recognition [3-5], lexical processing [6] and perceptual shape integration ("visual binding", [7]).

Date and time: Sunday 13:30
Room: Berlage
Format and topic: Symposium, Decision-making
Coordinator: Jan Engelmann
Title: The neural basis of social decision-making
Symposium abstract: Decisions involving other human beings are pervasive and among the most important ones in our lives. Recent progress in social neuroeconomics has identified two interacting neural systems that are involved in social decision-making: a social cognition system (TPJ, DMPFC) that is involved in mentalizing, and a valuation system (VMPFC, VS, amygdala) that processes social rewards and guides learning about interaction partners. This symposium presents recent advances on the neurobiological basis of social decision-making, including an animal model of social reinforcement learning, the neural circuitry underlying unobservable motives of social decisions, the distortionary influences of emotions on social cognitive neural circuitry, and the causal role of TPJ in mentalizing computations during strategic interactions. People negatively value unequal outcomes for themselves, but also for other. This phenomenon is known as inequity aversion. The first speaker shows that rats also show such social preferences. Lesion and pharmacological studies suggest that social preferences are acquired in a social reinforcement learning context that operates via basolateral amygdala and involves serotonin neurotransmission. One important scientific puzzle is to identify the motives that guide human social interactions. The second speaker shows that different motives, despite similar behavioral output, can be identified by differential connectivity patterns between key structures implicated in decision-making (anterior cingulate, anterior insula and ventral striatum). Decisions are rarely made in an emotional vacuum. Despite the fact that emotions are ubiquitous and can influence choices in irrational ways, little is known about their influences on the neural circuitry of social decisions. The third speaker confirms that aversive emotions distort trust decisions and trust-specific activity and connectivity in TPJ. The final talk investigates interactions between social cognition and valuation systems during competitive interactions. Combining inhibitory transcranial magnetic stimulation (cTBS) with model-based fMRI, the speaker shows that disrupting TPJ reduces behavioral and neural indices of mentalizing-related computations.

Speaker: Tobias Kalenscher
Title: Psychoneuropharmacological basis of inequity aversion in rats
Abstract: Inequity aversion is a behavioral, motivational and emotional response to an unfair reward distribution, given equal efforts to obtain rewards. Disadvantageous inequity aversion can be caused by a reward distribution that leaves the decision-maker worse off than a partner, advantageous inequity aversion can result from a reward distribution in which the decision-maker is better off than a partner. Both types of inequity aversion have been shown in humans and non-human primates, but it remains elusive if they evolved earlier in the phylogenetic history. In my talk, I will provide evidence that rats show disadvantageous and advantageous inequity aversion. I will argue that the rats' social preferences are the consequence of social reinforcement learning in which acoustic social signals emitted by the two interacting rats orchestrate their preferences for equal reward outcomes. I will present lesion and psychopharmacological data highlighting the importance of basolateral amygdala, and serotonin action in amygdala, in developing mutual reward preferences -- the presumed motive underlying advantageous inequity aversion.

Speaker: Grit Hein
Title: Uncovering the motives that drive social decisions
Abstract: Goal-directed human behaviors, including social decisions, are driven by motives. Thus, in order to fully understand a person's decision, it is crucial to understand the underlying motives. Identifying the motives behind a decision is difficult, because motives are not directly observable, and the same decision can be based on very different motives (for example the decision to help can be driven by empathy or a social norm such as reciprocity). In my talk I will present recent work that illustrates how neuroimaging methods such as functional magnetic resonance imaging (fMRI) can be used to investigate social motives, and their modulation through learning. The potentials and limitation of this approach will be discussed and important questions for future research will be outlined.

Speaker: Jan B. Engelmann
Title: Incidental emotions distort the neural circuitry of trust decisions
Abstract: In everyday life, we rarely make decisions in an emotional vacuum. That emotions can influence our decisions has been well-established by prior research. Emotions can both guide choices, for instance in the form of identifying what is relevant, but they can also influence choices in irrational ways, for instance by distorting value and introducing biases. Particularly aversive emotions, such as fear, are likely to be a key source of irrational human decision-making. Despite these considerations underlining the importance of emotions in decision-making, relatively little is known about how emotions influence the neural circuitry of choice. I will discuss recent findings that show that aversive emotions distort trust decisions and cause significant changes in the associated neural circuitry. Specifically, aversive emotions reduce trust and suppress trust-specific activity in left temporoparietal junction (TPJ). In addition, aversive emotions reduce the functional connectivity between TPJ and emotion-related regions such as the amygdala. We also identify a network of regions whose connectivity strength with TPJ is modulated specifically by transfer amounts in the trust game. Of particular interest is the activation pattern of the posterior superior temporal sulcus (pSTS), which suggests that pSTS plays a key role in mediating the impact of aversive emotions on brain-behavior relationships: Functional connectivity of right pSTS with left TPJ not only predicts mean trust taking in the absence of negative emotions, but aversive emotions also largely remove this association between TPJ-pSTS connectivity and behavioral trust. These findings may be useful for a better understanding of the neural circuitry of affective distortions and may thus help identify the neural basis of psychiatric diseases that are associated with emotion-related psychological and behavioral dysfunctions.

Speaker: Christian Ruff
Title: A causal brain network account of mentalizing during strategic social interactions
Abstract: During competitive interactions, humans have to estimate the impact of their own actions on their opponent's strategy. In this presentation, I will provide evidence that neural computations in the right temporo-parietal junction (rTPJ) and interconnected structures are causally involved in this process. By combining inhibitory continuous theta-burst transcranial magnetic stimulation (cTBS) with model-based functional magnetic resonance imaging (fMRI), we show that disrupting neural excitability in the rTPJ reduces behavioral and neural indices of mentalizing-related computations, as well as functional connectivity of the rTPJ with ventral and dorsal parts of the medial prefrontal cortex. These results provide a causal demonstration that neural computations instantiated in the rTPJ are neurobiological prerequisites for the ability to integrate opponent beliefs into strategic choice, through system-level interaction within the valuation and mentalizing network. In a second study, we examine whether these mentalizing-related computations in the TPJ are triggered by the social context of interacting with another human or whether they rather reflect specific algorithmic demands associated with interactions with a reactive process, irrespective of whether this occurs in a social context or not. Interestingly, neural activity in the rTPJ appears to track the algorithmic demands of the choice problem, but this activity influences choice and associated neural processes more strongly when participants believe they face a human opponent. This argues that neural computations in the TPJ are not functionally specialized for "social behavior" per se but rather reflect specialized computations that detect and deal with interactive processes, which are specifically choice-relevant during interactions with other human beings. Taken together, the two studies presented here thus move towards a mechanistic understanding of how neural computations in the TPJ and interconnected structures contribute to the control of strategic behavior during social interactions.

Date and time: Sunday 13:30
Room: Veiling
Format and topic: Symposium, Cognitive modeling
Coordinator: Micah Goldwater
Title: Toward a unified account of frontopolar function in higher-level cognition
Symposium abstract: The frontopolar cortex (FPC), the most anterior portion of the prefrontal cortex, shows the largest increase in size of any area of the brain compared to our closest primate cousins. Not coincidentally, the FPC appears to be crucial across a wide-range of tasks that all involve the higher-order thinking abilities that define human cognition, from metacognitive reflection to analogical problem solving. Further, it seems critical to a healthy aging brain, as variation in size of the FPC is uniquely predictive of the rate of decline in fluid intelligence during aging. While it is quite clear the FPC is central to human cognition, in comparisons to other regions of the prefrontal cortex, its specific function or role in these various higher-order cognitive abilities is relatively unknown. One potential reason for the relatively poor understanding is that often researchers working in these different areas of cognitive neuroscience operate entirely independent of each other, with little direct communication and collaboration. This symposium will attempt to bring some of these subareas of cognitive neuroscience together to see if we 1. can get closer to a more general understanding of the function of FPC, and 2. clearly identify lines of novel research to further move us towards this goal. Specifically, we will present research on fluid intelligence, category learning, and metacognition, each having previously revealed unique roles of FPC. After the three presentations, we will bring all the speakers together for a panel discussion and elicit integrative questions from the audience.

Speaker: Stephen Flemming
Title: The reflective mind: A role for frontopolar cortex in human metacognition
Abstract: The frontopolar cortex (FPC) occupies a large swathe of the human frontal lobes and yet its function remains poorly understood. Recently it has been discovered that metacognition - the ability to reflect on cognition and behaviour - is associated with markers of frontopolar structure and function obtained from neuroimaging. Individuals with more accurate metacognition on simple visual decision tasks have greater grey matter volume, white matter integrity and functional connectivity in FPC. I will present new studies from our lab that zero in on the computational role of FPC in metacognition. We have designed paradigms that allow the decoupling of sense of confidence in our performance from stimulus strength and reward. We find that while that posterior frontal areas tracked the integration of evidence for or against a decision, FPC activity was instead coupled to metacognitive judgments of confidence in performance. Leveraging mediation analysis we show that only FPC and precuneus met statistical criteria for mediating the effect of new evidence on changes in subjective confidence. In a complementary study we have used multivoxel pattern analysis to show that the representations of confidence in FPC are task-specific. Together our findings indicate a working model in which FPC supports a fine-grained representation of the reliability of current task performance, supporting both strategic control and the communication of metacognitive estimates to others.

Speaker: Micah Goldwater, Gareth Roberts
Title: Frontopolar cortex is critical for constructing and then adapting relational representations
Abstract: Representing the structural relations among objects, events, and ideas is central to higher-level cognition. These relational representations are structured hierarchically, such that objects are bound by how they relate (e.g., the sun is more massive than the earth; the earth revolves around the sun) and then higher-order relations bind these lower-order relations (e.g., sun as more massive than the earth causes the earth to revolve around the sun). It is these higher-order relations that are specifically critical to our most complex cognitive processes, such as analogical problem solving which applies the relations from one conceptual domain to another (e.g., when Rutherford applied the relations among the sun and earth to the atomic nucleus and electrons, respectively). One of the more prominent theories of the frontal pole is that it is specifically crucial for integrating multiple relations into a higher-order relational representation. This talk will review recent research from analogical thinking, cognitive control, mathematical problem solving, category learning, and fluid intelligence that all suggest that FPC is important for both initially constructing a higher-order relational representation, and then modifying these structural relations to adapt to changes in stimuli or task conditions. The talk will primarily focus on two sets of results. First from category learning showing that the FPC is uniquely active during 1. the initial learning of relational categories ( i.e., categories defined by the relations among objects, such as barrier, predator, same or different), and then 2. later generalising the category to novel exemplars. Second, results from fluid intelligence testing show that the FPC does not specifically respond to the complexity of the relations among the problem elements, but the novelty of these relationships. Taken together, the findings suggest that a key function of the frontal pole is the construction and modification of higher-order structured representations.

Speaker: Tyler Davis, Dmitriy Paniukov, Timothy Kelley, Benjamin England, Michael Serra
Title: The frontopolar cortex builds rules for sorting our external and internal worlds
Abstract: Bird or mammal? Will I pass or fail? Our brain constructs concepts that allow us to interpret both our external and internal worlds. Recently, FPC has been associated with both the learning of novel concepts and metacognitive judgments, raising the question of whether concepts that point outwardly and inwardly engage a common mechanism. Two possible candidates for this common mechanism are representational building and representational switching. From a representational building account, FPC is involved in creating structured representations that establish how observations are connected causally by rules or laws. Representational building thus suggests that FPC is involved in acquiring structured representations when forming rules to sort the external world, and when establishing higher-level metacognitive rules to explain our own behavior. Contrastingly, representational switching accounts focus on the role of the FPC in cognitive control, and suggest that FPC is involved in switching between such structured representations. In category learning, representational switching occurs during testing different hypotheses or candidate rules. In metacognition, switching may be involved when self judgments involve simultaneous attempts to apply an external rule and internally evaluate its success. We evaluate these two accounts in two rule-based category learning studies. In Study 1, we find that FPC activation persists as participants begin to apply uncertain rules, but beyond when participants are actively switching, suggesting that FPC continues to build rule-representations as subjects acquire more evidence for a rule. In Study 2, we find that FPC reflects higher-level metacognitive judgments of rule-based performance when isolating judgments from simultaneous attempts to classify a stimulus, suggesting that FPC function is not specific to switching between external and internal rules. Together, these results are supportive of the hypothesis that the FPC is involved with building and evaluating representations of the internal and external world to guide rule-driven behavior.

Date and time: Monday 8:30
Room: Graan
Format and topic: Symposium, Decision-making
Coordinator: Hanneke den Ouden
Title: Deciding how to decide: when and how the brain takes short-cuts to make decisions
Symposium abstract: In the past decades, the field of decision neuroscience has made huge advances in understanding the neural and computational basis of simple decisions such as choosing between two options or navigate simple mazes. However, humans routinely formulate plans in domains so complex that even the most powerful computers are taxed. To do so, they seem to avail themselves of many strategies and heuristics that efficiently simplify and approximate these difficult decisions [1--3]. This symposium will spotlight the work of five emerging researchers taking complementary approaches to understanding how our brains help us to navigate and reduce the computational complexity of the world around us.  Speaker 1 and 2 will focus on likely the brain's most economical yet least flexible decision heuristic, namely hardwired Pavlovian affective response tendencies. Speaker 1 will assess these heuristics across the life-span, and use PET to directly test the hypothesised role of striatal dopamine in driving these biases. Speaker 2 will then present pharmacology and EEG work in combination with computational modelling to uncover separable mechanisms of Pavlovian response biases and biased credit assignment in shaping affective decision-making. Speaker 3 will present work using fMRI that uncovers different contingent and non-contingent learning mechanisms operating in parallel, that are behaviourally and anatomically separable. Speaker 4 will directly address how the brain may allocate control between the different decision strategies, based on their relative costs and benefits, combining a number of novel tasks, computational models, and fMRI. Finally, speaker 5 will present invasive neurophysiological data from non-human primates performing a naturalistic reward-guided decision making task, in which information has to be sequentially acquired about different attributes of two alternative options. By using new population analysis techniques such as representational similarity analysis, he will demonstrate a clear triple dissociation of function across different subregions of prefrontal cortex.

Speaker: Marc Guitart-Masip
Title: Lifespan trajectories and dopaminergic modulation of Pavlovian biases during learning
Abstract: The instrumental selection of actions, and the vigor with which they are executed, are influenced by the affective valence of predicted outcomes [1]. Previous studies have demonstrated that reward facilitates learning of active choices, whereas punishment facilitates learning of passive choices [2]. This learning bias has been computationally characterized as a Pavlovian heuristic that speeds up learning in those circumstances most commonly encountered by decision-makers, and is modulated by pharmacological manipulations of the dopaminergic system [3,4]. However, the locus of the dopaminergic influences in the brain has remained elusive. Furthermore, although this learning bias influences appropriate and inappropriate choices, and is supposedly implicated in the expression of psychiatric and neurological abnormalities, most evidence supporting the existence of this bias has been obtained in younger adults. Here, I will first present data from a cross-sectional study quantifying this learning bias in participants between 7 and 78 years of age. These data demonstrate that this learning bias is present across the life span and tends to increase in middle and older age compared to children, adolescents and younger adults. Children and adolescents show a Pavlovian bias equivalent to that observed in younger adults, but have difficulties in learning to inhibit their responses regardless of the affective valence of predicted outcomes. I will also present data from a PET study in which we measured the binding potential (BP) of the SCH23390 radioligand to D1 dopamine (DA) receptors across the whole brain in 30 younger adults (18-30 years old) along with the same learning bias. These data demonstrate that higher D1 BP in caudate nucleus was associated with stronger Pavlovian bias during learning. No such association was found for prefrontal cortex. These data suggest that striatum is the locus of dopaminergic modulation of the Pavlovian bias.

Speaker: Hanneke den Ouden, Jennifer Swart
Title: Mechanisms of motivational biases in learning and choice
Abstract: Pavlovian conditioning is perhaps the most primitive and computationally parsimonious form of learning. Pavlovian responses may help reduce computational load by shaping our actions in an evolutionarily hardwired manner. Instrumental learning systems, in contrast, allow organisms to adaptively learn which actions are good in specific environments. Pavlovian and instrumental systems are often presented as a dichotomy, respectively driving cue-based motivational biases versus adaptive 'rational' choice. For example, a number of recent studies on motivational biases of action (e.g. appetitive activation / aversive inhibition) have interpreted these responses in terms of Pavlovian response biases [1--3]. An alternative possibility is that adaptive systems have evolved to take into account prior likelihoods of certain response-outcome associations, and learn these preferentially to minimize computational expenses, putatively subserved by striatal D1/D2 pathways [4]. Thus, motivational biases of behavioural activation may additionally arise from biased instrumental learning, i.e. credit assignment. Such a learning bias would provide a cognitively efficient mechanism for rapid learning of high-probability action-outcome associations whilst protecting against learning spurious associations. I will present recent work in which we combine computational modelling with a novel paradigm to disentangle the contribution of Pavlovian choice heuristics and motivational biases in instrumental learning, to well-established reward/punishment biases of motivated action. Next, we probed the pharmaco-computational mechanisms that may subserve these biases, using a catecholaminergic challenge (methylphenidate). Together, this work uncovers two distinct mechanisms by which motivation impacts behaviour, and helps refine current models of catecholaminergic modulation of motivated action.

Speaker: Gerhard Jocham
Title: Parallel mechanisms for reward-guided learning
Abstract: It is crucial for adaptive success to repeat choices that lead to success. Reinforcement learning theories have provided a powerful explanatory framework for such learning from experience [1]. However, they rely on the simple yet fundamental assumption that agents are able to attribute outcomes to the choices that caused these outcomes. This assumption, while true in simple laboratory choice scenarios, may often be not met in real-life scenarios, where consequences of actions are often delayed with many intervening irrelevant actions. Furthermore, many important outcomes may not even be consequences of the agent's own behavior. Recent work suggest that learning is possible even when this most fundamental assumption of knowing which choice led to which outcome is not met [2]. I will present two fMRI studies using novel reward-guided learning paradigms to show that humans deploy several independent learning mechanisms operating in parallel. First, I will show that, in a time-continuous environment, humans are still highly efficient at correctly associating rewards with choices that have caused these outcomes. This contingent learning appears to rely on the lateral orbitofrontal cortex. Crucially, however, subjects' choices were also guided by additional non-contingent mechanisms. These exploited either statistical dependencies, or temporal proximity between choices and rewards. Statistical learning relied on spreading credit for rewards to the average history of choices and outcomes and was related to amygdala activity following choices. Learning on the basis of close temporal coincidence was associated with timing-dependent reward responses in the sensorimotor striatum. These data show that humans deploy several learning mechanisms operating in parallel. Only one of these mechanisms exploited knowledge about the causal structure of the task. The contribution of the non-contingent mechanisms to behaviour may be even more pronounced in real-life scenarios, where the complexity of the causal structure imposes stronger computational demands on contingent learning.

Speaker: Wouter Kool, Fiery A. Cushman, Samuel J. Gershman
Title: Metacontrol in reinforcement learning
Abstract: Human behavior is sometimes determined by habit and other times by goal-directed planning. Recent work in reinforcement learning formalizes this distinction as a competition between a computationally cheap but inaccurate "model-free" system and an expensive but accurate "model-based" system [1,2]. However, it is unclear how the brain chooses to allocate control between them [3]. I will present five behavioral studies, one neuroimaging study, and a new computational model, which together suggest that arbitration occurs by comparing each systems' benefits discounted by an intrinsic effort cost for model-based planning. Our work employs a class of decision-making tasks, known as two-step tasks, which behaviorally dissociate model-free from model-based control [1]. First, I describe a novel version of this task, where, unlike previous versions, enhanced model-based planning yields increased reward. Next, we test our cost-benefit account by adding a "stakes" manipulation to these tasks: on certain trials, a cue signals that subsequent rewards will be amplified. Consistent with our hypothesis, model-based control increased on high-stakes trials in the novel task. Moreover, we found no sensitivity to increased stakes on the original two-step task, where habit and planning yield equivalent accuracy. Finally, we present a new reinforcement-learning model which adaptively arbitrates between the two systems. This allocation is guided by 'controller values' which are updated according to a 'controller prediction error', which integrate the system's costs and benefits. Our model provides a superior behavioral fit compared to previous developed models. In addition, a new neuroimaging study reveals a set of regions in frontal cortex, commonly associated with valuation and cognitive control, which encode this controller prediction error. Together, our results propose that the brain flexibly and adaptively integrates the costs and benefits of model-free and model-based control in order to guide arbitration between multiple reinforcement-learning systems.

Speaker: Laurence Hunt
Title: Neural mechanisms of sequential information sampling and reward-guided choice
Abstract: Decision-making often appears irrational. A major challenge is to explain why apparently irrational behavior occurs and what potential benefits it might have conferred for our evolutionary ancestors. One well-studied behavior in experimental psychology is "confirmation bias," where we sample information that simply confirms what we already believe. We have recently shown that a factor triggering such information sampling biases in human behaviour is Pavlovian approach: our natural tendency to approach items that are associated with reward [1]. How then might these biases shape decision making? Many accounts of reward-guided choice posit that agents simultaneously compare the expected value of alternatives in a choice set [2,3]. Yet real-world choices typically demand sequential shifts in attention between choice alternatives and their underlying attributes [4,5]. We hypothesise that approach-induced biases in information sampling allow these decisions may be solved by accepting or rejecting each option in turn. Neuroimaging [6,7] and lesion [8] studies indicate the importance of prefrontal cortex in supporting such decision processes, yet their neurophysiological basis currently remains unclear. We therefore recorded 548 single neurons from three subregions of prefrontal cortex during a sequential information search and decision task. We reveal a triple dissociation of attention, valuation and belief updating across macaque prefrontal cortex during sequential economic choice. In dorsolateral prefrontal cortex (DLPFC), neural population responses reflect deployment of spatial attention to different choice items; in orbitofrontal cortex (OFC), they reflect relative valuation of the currently attended choice option versus an alternative stored in working memory; in anterior cingulate cortex (ACC), a neural signature of belief updating indicates acceptance or rejection of the currently favoured alternative as new evidence is revealed. These results provide a neurophysiological mechanism for the sequential acceptance or rejection of choice alternatives during naturalistic information search and choice.

Date and time: Monday 8:30
Room: Administratie
Format and topic: Symposium, Memory
Coordinator: Nahid Zokaei
Title: Mnemonic priorities: dynamic interplays between attention and working memory
Symposium abstract: Contemporary research suggests that working memory (WM) -- our ability to maintain information over a short time period towards a behavioural goal-- is much more flexible and dynamic than traditionally conceived. Items are prioritized and updated continuously based on our changing goals and expectations. Accordingly, emerging contemporary models propose that the state of information in WM is critically shaped by an interaction between attention and memory. Here we outline the new wave of WM research that primarily focuses on understanding the nature of these interactions, on the basis of a wide range of complementary techniques, including computational modelling, EEG decoding and cutting-edge behavioral quantifications. Together they provide evidence for an updated theoretical and neural conceptualization of WM that incorporates a central and multifaceted role for attention. Specifically, in the first part of this symposium we will be exploring how temporal expectations may have a central role in guiding attentional prioritisation in WM (speaker 1), and how such dynamic prioritization may help guide visual search (speaker 2). These studies will also highlight a critical role for posterior alpha oscillations in attentional dynamics in visual working memory. Speaker 3 will discuss evidence for so-called 'hidden states' in working memory, and discuss their interaction with flexible attentional demands. Finally, speaker 4 will be exploring the principles that govern binding -- another function that has often been related to attention -- in visual WM. Together; this is thus anticipated to build toward an updated theoretical and neural conceptualization of WM, with attention as a central ingredient that has multifaceted utility.

Speaker: Freek van Ede
Title: Dynamic prioritisation in visual working memory by temporal expectations and spontaneous brain states
Abstract: The process of prioritising information that is most relevant to current goals is not only key for adaptive perception, but also working memory. In line with everyday working memory demands, I will first show that prioritisation in working memory can be dynamically and reversibly steered by internally guided temporal expectations (i.e. expectations regarding when individual mnemonic items are most likely to be probed). On the basis of EEG measurements in healthy humans, I will go on to show that such dynamic prioritisation is supported by the spatially and temporally specific attenuation of posterior alpha oscillations that, moreover, predicts working memory access times on a trial-by-trial basis. I will finally show that such lateralised alpha states immediately prior to the memory probe (which is always presented centrally) even predict access times when the two mnemonic items -- that are presented in opposite hemifields at encoding -- are equally relevant. Thus, even when there is no experimental incentive to prioritise one mnemonic item over another, individual items nevertheless appear to compete for a 'single-item focus of attention'. These data suggest that the dynamics of prioritisation in working memory must be understood in relation to temporal expectations as well as to "spontaneous" brain states, and link posterior alpha oscillations to these dynamics during visual working memory.

Speaker: Ingmar E.J. de Vries, Joram van Driel, Christian N.L. Olivers
Title: EEG correlates of priority switches in working memory-driven visual search
Abstract: Models of visual search assume that search is guided by an active visual working memory representation of what we are looking for. This attentional template for the currently relevant stimulus can be dissociated from accessory memory representations that are only needed prospectively, for a future task, and that should be prevented from guiding attention. Little is known about how the brain sequentially prioritizes two consecutive tasks. We report evidence showing that which item is prioritized for search is captured by posterior alpha EEG dynamics. We measured EEG while observers performed two consecutive working memory-guided visual search tasks. Prior to the first search task, a cue instructed observers which item to look for first (current template), and which second (prospective template). During the first delay, leading up to the first of the two searches, posterior alpha power (8--14 Hz) was suppressed contralateral relative to ipsilateral to the memory item. Importantly, these lateralized alpha dynamics were stronger for the item that was the imminent search target (the current template), than for the item that was going to be the subsequent search target (the prospective template). In a second study, we more closely investigated the electrophysiological mechanisms underlying the switch in priority between the first and the second search. At this point the first template can be dropped, while the second template needs to be prioritized. Interestingly, dropping the first template resulted in clear contralateral alpha enhancement, while switching priority to the prospective template resulted in amplified contralateral alpha suppression. Switching occurred within 175 - 225 ms after the first search response. Together, these findings reveal a crucial role for alpha-band oscillations in the prioritizing and switching of prioritized working memory representations.

Speaker: Michael J. Wolff, Janina Jochim, Elkan G. Akyuerek, Mark G. Stokes
Title: Dynamic hidden states underlying working memory guided behavior
Abstract: Recent theoretical models propose that working memory is mediated by rapid transitions in 'activity-silent' neural states (e.g., short-term synaptic plasticity). According to the dynamic coding framework, memory-related changes in such hidden states flexibly configure memory networks for memory-guided behaviour, and dissolve them equally fast to allow forgetting. We developed a novel perturbation approach to measure mnemonic hidden states in electroencephalogram (EEG). By 'pinging the brain' during maintenance, we show that memory item-specific information is decodable from the impulse response, even in the absence of attention and lingering delay activity. Moreover, hidden memories are remarkably flexible: An instruction cue that directs people to forget one item is sufficient to wipe the corresponding trace from the hidden state. In contrast, temporarily unattended items remain robustly coded in the hidden state, decoupling attentional focus from cue-directed forgetting. Finally, the strength of hidden-state coding predicts the accuracy of working memory guided behaviour, including memory precision.

Speaker: Paul Bays
Title: Neural architecture for feature binding in visual working memory
Abstract: Binding refers to the operation that groups different features together into objects. We propose a neural architecture for feature binding in visual working memory that employs populations of neurons with conjunction responses. We tested this model using cued recall tasks, in which subjects had to memorise object arrays composed of simple visual features (colour, orientation, and location). After a brief delay, one feature of one item was given as a cue, and the observer had to report, on a continuous scale, one or two other features of the cued item. Binding failure in this task is associated with swap errors, in which observers report an item other than the one indicated by the cue. We observed that the probability of swapping two items strongly correlated with the items' similarity in the cue feature dimension, and found a strong correlation between swap errors occurring in spatial and non-spatial report. The neural model explains both swap errors and response variability as results of decoding noisy neural activity, and can account for the behavioural results in quantitative detail. We then used the model to compare alternative mechanisms for binding non-spatial features. We found the behavioural results fully consistent with a model in which non-spatial features are bound exclusively via their shared location, with no indication of direct binding between colour and orientation. These results provide evidence for a special role of location in feature binding, and the model offers an explanation of how this special role could be realised in the neural system.

Date and time: Monday 8:30
Room: Berlage
Format and topic: Symposium, Connectivity/networks
Coordinator: Rogier Mars
Title: Connectivity research in cognitive neuroscience: Hype or approach of the future?
Symposium abstract: Diffusion MRI and resting state functional MRI have developed over the last couple of years from highly specialized tools to easily available approaches. Indeed, collecting connectivity data is now so common it is often considered as standard as collecting a T1 structural. Tools for the analysis of connectivity data are part of most freely available MRI analysis packages. Large-scale collaborative projects have been initiated to relate connectivity data to behavior, genetics, and various measures of individuality (e.g., Van Essen et al., 2013; Miller et al., 2016). However, with this general availability also comes the question of what new insights this flood of data has brought to the field of cognitive neuroscience. In this symposium, three young researchers will provide their views on this question. Focusing on the relationship between connectivity and functional data (Speaker 1), the relationship between brain connectivity and brain disease (Speaker 2), and connectivity as a tool for comparative cognitive neuroscience (Speaker 3), each will demonstrate how connectivity has been applied in their respective (sub)field. They will demonstrate that incorporating connectivity as a measure of interest has allowed each of them to reach conclusions that could not have been obtained in another way. However, they will also demonstrate how connectivity is complementary to other approaches, rather than a replacement. References: - Van Essen DC, Smith SM, Barch DM, Behrens TE, Yacoub E, Ugurbil K, WU-Minn HCP Consortium (2013) The WU-Minn Human Connectome Project: An overview. NeuroImage 80:62-79 - Miller KL, Alfaro-Almagro F, Bangerter NK, Thomas DL, Yacoub E, Xu J, Bartsch AJ, Jbabdi S, Sotiropoulos SN, Andersson JL, Griffanti L, Douaud G, Okell TW, Weale P, Dragonu I, Garratt S, Hudson S, Collins R, Jenkinson M, Matthews PM, - Smith SM (2016) Multimodal population brain imaging in the UK Biobank prospective epidemiological study. Nat Neurosci 19:1523-1536

Speaker: Alexandros Goulas
Title: Uncovering the connectional architecture of the cortex with network analysis tools: Implications for cognitive models
Abstract: The functional modus operandi of the cortex is commonly assumed to be closely related to its underlying macroscale structural organization. Consequently, functional models of the cortex, at a global whole-cortex or lobe-wise level, often incorporate insights pertaining to its connectional architecture. However, this usually involves relying on qualitative observations that might be intuitive but are not rigorously verified. Here, we will discuss work that aims to explicitly bridge the levels of structural brain organization and cognitive functioning, so that each can benefit from the other while respecting the intricacies of each level. Aspects of the so-called hierarchical organization of the cortex and its ramifications for cognitive models will be evaluated in the light of network analysis of a non-human primate, that is, the macaque monkey (e.g., Goulas et al., 2014). First, the structural connectivity assumptions of the anterior-posterior model advocated for the prefrontal cortex will be presented and addressed. Second, the serial versus parallel structural connectivity architecture of the cortex will be evaluated.

Speaker: Michel Thiebaut de Schotten
Title: Clinical application of connectivity: Advanced lesion symptom mapping analyses
Abstract: Patients with brain lesions are a unique opportunity to understand the functioning of the human mind. However, damages to the brain do not affect lesioned areas alone, but also remotely connected regions. As such, lesions to different areas can have similar behavioral expressions. Tractography and resting state functional connectivity can provide several complementary solutions to measure brain deafferentation effects. We have developed suite of approaches to study the long-range effects of focal cortical lesions (Thiebaut de Schotten et al., 2014; Thiebaut de Schotten et al., 2015). Here, we will present data showing the long-range effects of specific lesions in a large group of 37 patients with local brain regions affecting language production. Our results revealed a large network of brain areas all contributing significantly to the production of words of the same category. We confirmed that the deafferented areas are classically considered as functionally engaged in tasks related to verbal fluency and categorisation. These deafferented areas also belonged to larger functional networks, particularly the left ventral fronto-parietal network whose cortical thickness proportionally decreased with the performance in categorical fluency. Hence, the methods presented during this presentation allows for the capture of brain lesions' remote effects as well as the understanding of brain lesions' remote effects relationship with cognitive and behavioural measures.

Speaker: Rogier Mars
Title: Brain connectivity in comparative neuroscience: What can it tell us?
Abstract: What makes the human brain unique? We know that the human brain is larger than that of any primate, but precisely how its internal organization is special remains largely unknown. It has been suggested that the human brain contains more areas than that of other primates and research on the neural basis of language suggests that some connections between temporal and frontal cortex have expended in the human lineage (e.g., Rilling et al., 2008). However, a direct comparison between the architecture of different primate species' brains is difficult to realize, since anatomical techniques are expensive and often lethal for the experimental subject, making them unsuitable for use in humans. Comparison between anatomical data obtained in non-human primates and neuroimaging data obtained in humans suffers for difficulties in comparison between methods. However, the rise of MRI-based methods to study whole-brain connectivity might provide our best chance yet for a large-scale primate comparative neuroscience. Over the last few years we have used a variety of neuroimaging techniques to study the connections between areas of the brain, developing methods to quantitatively compare the architecture of different species' brains (Mars et al., 2016). Using these techniques we were able to parcellate the human association cortex into distinct regions and compare these to known areas in the macaque brain (Mars et al., 2011; Sallet et al., 2013; Neubert et al., 2014). The results show a greater than expected similarity between the two species in parietal and lateral frontal cortex, but also some interesting differences in neural circuits mediating, in the human, language and social cognition. More recently, we are developing techniques to quantitatively compare the organization of the whole brain across species, allowing us to identify hotspots of change between species. Using post-mortem data obtained from zoos and research institutes, we are able to study a larger range of species than previously possible.

Date and time: Monday 8:30
Room: Veiling
Format and topic: Symposium, Attention
Coordinator: Rene Huster
Title: The temporal dynamics of response inhibition: neural markers, modes, and modulations
Symposium abstract: The ability to stop behavior or cognitive processes supports the goal-oriented adaptation to a fast-changing environment and is considered a hallmark of cognitive control. Research suggests that its most common experimental operationalization, namely response inhibition, rests on the interaction of a triad of neural systems: regions associated with inhibitory control (putatively inferior frontal and pre-supplementary motor cortices), inhibitory actors (several distinct basal ganglia circuits), as well as inhibitory targets (e.g., within the motor cortex). Despite significant progress in the delineation of the brain circuits involved in inhibition, much remains unknown with respect to its temporal dynamics. At its core, this symposium focuses on work utilizing the high temporal resolution of electroencephalography (EEG) to delineate the temporal dynamics of sensory and higher cognitive processes supporting inhibition, their corresponding EEG markers, as well as changes in their course of action in different inhibitory modes. The presentations highlight certain intricacies related to the experimental design of response inhibition tasks, the interpretation of early and late electroencephalographic events (such as the N1, P2, P3, as well as alpha, theta, or delta-oscillations), the association of such events with neurotransmitter systems, and their spatio-temporal mapping to the aforementioned inhibitory triad as inferred from multimodal measurements (concurrent EEG, functional magnetic resonance tomography, and electromyography). The presented data not only significantly advance our understanding of the neural implementation of response inhibition, but further stress the utility of temporal information for differential clinical diagnostics.

Speaker: Jan R. Wessel
Title: The Dos/Do-Nots of Go/No-Go: A neurophysiological perspective
Abstract: Inhibitory control enables humans to stop prepotent motor activity. This ability is commonly studied using Go/Nogo or stop-signal tasks. In stop-signal tasks, prepotent motor activity is elicited by delaying stop-signals relative to go-signals. In Go/Nogo-tasks, however, trials include only one signal -- "go" or "nogo". Hence, prepotent motor activity has to be ensured differently; for example, by using rare Nogo-trials by presenting stimuli at rapid pace. However, a literature survey shows that ~40% of studies use equiprobable Go/Nogo-trials, and ~20% use very long stimulus-stimulus intervals (>4s). It is unclear whether such slow-paced, equiprobable Go/Nogo-tasks elicit prepotent motor activity -- and hence, whether they probe inhibitory control. We recorded EEG during four Go/Nogo-tasks, which varied in Nogo-probability and trial pace. We quantified prepotent motor activity on successfully-inhibited Nogo-trials using the lateralized-readiness potential. Only fast-paced Go/Nogo-tasks with rare Nogo-trials reliably evoked such activity. In each participant, we also used a stop-signal task and independent component analysis to isolate an established neural signature of inhibitory control, and investigated this signatures activity in the Go/Nogo-tasks. Increased prepotent motor activity on Nogo-trials was accompanied by greater fronto-central P3 amplitudes, thereby confirming it as an index of inhibitory control. Crucially, this inhibition-related activity was reduced by more than 75% in slow-paced, equiprobable Go/Nogo-tasks compared to fast-paced, rare-Nogo versions. Therefore, our results show that many common Go/Nogo-task configurations do not reliably evoke prepotent motor activity. Consequently, their inhibitory requirements are greatly reduced. This has major implications for the design and usage of Go/Nogo-tasks in psychological experiments.

Speaker: Ann-Kathrin Stock
Title: On the difference between correlates and predictors of response inhibition
Abstract: Over the last decades, response inhibition has been subject to extensive research and the EEG community has significantly contributed to our understanding of the neuronal processes constituting this essential cognitive faculty. Yet still, the nexus between behavior, cognitive sub-processes and neurophysiological measures has remained mostly correlative in nature. Against this background, we investigated whether classical neurophysiological correlates of inhibition sub-processes (i.e. the N2 and P3 ERPs) merely correlate with behavioral performance or actually provide information expedient to the prediction of behavior on a single-subject level. To tackle this question, we resorted to different data-driven classification approaches in a sample of n = 262 healthy young subjects who completed a standard Go/Nogo task while an EEG was recorded. Based on a median split of behavioral performance data, subjects were classified as displaying either good behavioral inhibition ("accurate/slow" group) or rather bad behavioral inhibition ("less accurate/fast" group). As expected, behavioral group differences were associated with significant amplitude variations in classical electrophysiological correlates of response inhibition (i.e., N2 and P3). Yet still, N2 and P3 amplitudes were not predictive for group membership on a single-subject level. Instead, amplitude differences in the Go-P2 originating in the precuneus (BA7) were shown to predict group membership on a single-subject level with up to 64% accuracy. These findings strongly suggest that the behavioral outcome of response inhibition greatly depends on the amount of cognitive resources allocated to early stages of stimulus-response activation during responding. This suggests that research should focus more on early processing steps during responding when trying to understand the origin of interindividual differences in response inhibition processes.

Speaker: J. Leon Kenemans
Title: Electrophysiology of response inhibition: Sensory-specific stop-related N1 and generic stop-related frontal P3
Abstract: Inhibition concerns the capacity to suppress on-going response tendencies. A well-established method to assess inhibitory capacity is to present 'stop-signals' during the execution of a reaction-time task. Various sources of evidence point to two distinct human electrophysiological mechanisms implicated in processing of the stop signal, termed stop N1 and stop P3, which have been related to proactive and reactive aspects of inhibition, respectively. The short-latency (100 ms) stop N1 most likely reflects enhanced processing in sensory cortex with successful inhibition, whereas the longer-latency stop P3 reflects similar enhancement in association cortex. The current presentation discusses evidence that stop-N1 is dependent on stimulus affordances: It is observed only when the perceptual separation between go (reaction-time task) and stop signals is sufficient, as when go and stop signals differ in sensory modality, or substantially within one sensory modality. Stop-P3 is far less dependent on such sensory-specific aspects, and also much more generic in other respects. As a more generic 'frontal P3' (fP3) it can be elicited also outside the context of explicit inhibition, for example by the salience of task-irrelevant but potentially distracting events. This idea is supported by the results of comparative source analysis of fP3 as elicited in a stop-signal task and the frontal P3 that is commonly observed in response to unique novel stimuli in the context of a novelty oddball paradigm. Finally, pharmacological dissociations between the two mechanisms are discussed. While stop P3 is sensitive to norepinephrenergic and cholinergic but not to dopaminergic manipulation, for stop N1 the reverse pattern seems to hold.

Speaker: Rene J. Huster
Title: On the temporal dynamics of selective and global inhibition
Abstract: Behavioral adjustments often require interactions between distinct modes of cognitive control and response inhibition. Hypothetically, fast and global inhibition is exerted in the reactive control mode, whereas proactive control enables the preparation of more selective inhibitory pathways in advance, but rather relies on a slower inhibitory system. Data will be presented from a series of studies that combined group-level blind source separation of single-trial EEG data in both the time and frequency domain, trial-wise prediction of behavioral performance from EEG phenomena, temporal cascading of EEG events and their effective connectivity, as well as their association with different modules of the cortico-basal-ganglia circuits and electromyographic (EMG) activity at effector muscles. Our data corroborate strong stopping-related effects with respect to the N2 and P3 as well as corresponding activity in the theta and delta frequency range. More importantly though, early parieto-occipital alpha activity, putatively reflecting the processing of task-relevant stimuli, is highly predictive of inhibition performance and seems to exert strong causal influence on later EEG phenomena (e.g., delta-activity, the P300). The analysis of subthreshold EMG and EEG activity in successful stop trials suggests that the overall pattern of muscular and neural events varies between conditions targeting different inhibitory modes. Yet, the suppression of responses does occur as early as 150 ms for both global and selective inhibition, which challenges the temporal basis for the dichotomy between reactive global and proactive selective inhibitory pathways.

Date and time: Monday 8:30
Room: Verwey
Format and topic: Symposium, Sensory processing
Coordinator: Kathrin Ohla
Title: See, smell, and taste this! The neuroscience of multimodal food perception
Symposium abstract: Overconsumption of energy-rich foods contributes to the obesity epidemic. Food sensory attributes like appearance, smell, and taste, as well as the experienced reward upon consumption stimulate intake behavior. Prior experiences, e.g. the sweet taste and pleasure of eating ice cream, elicit expectations about sensory and hedonic properties of similar foods, so that future encounter of a cone can trigger appetite and drive approach behavior. Biases in the regulation of these perceptual processes are thought to foster deviant eating behavior: while decreased taste and smell sensitivity may lead to malnutrition and associated weight loss, decreased valuation has been suggested to contribute to excess food intake and associated weight gain. Despite the pivotal role of chemosensation in food-related decision-making the neural mechanisms of their perceptual and evaluative processing as well as their interaction with other sensory information have only recently received scientific attention; fundamental differences in the neural organisation of these senses begin to emerge. Speaker 1 will elaborate on the workings of the gustatory processing pathway, the gateway to nutrient sensing, and present evidence for a link between taste-related decision-making and the predictive value of gustatory brain response patterns, thereby highlighting the flexibility of gustatory coding. Speaker 2 will address the integration of odor into food object perception during anticipation of eating (odor and appearance) and consumption (odor and taste) and demonstrate that integration with its multisensory context is a key feature of olfactory food perception that likely contributes to facilitated object memory retrieval and thus appetite elicitation. Speaker 3 will shed light at multisensory food perception in the context of interpersonal differences that may help to understand vulnerability for deviant food intake behavior. She will also discuss how plasticity in the neural networks underlying food perception may be adaptive for the regulation of food intake.

Speaker: Kathrin Ohla
Title: Flexible coding of taste categories in the human brain
Abstract: The sense of taste is the key to distinction of potentially nutritious or harmful food constituents and thereby for the acceptance of food. While taste processing in the periphery and, more recently, also the central nervous system has gained considerable attention in animal models, these processes are still to be investigated in humans. We asked how, when, and where taste information is processed. Using time-resolved multivariate pattern analyses (MVPA) of large-scale electrophysiological brain responses obtained from non-invasive scalp recordings, we investigated spatio-temporal properties of taste representations and asked whether these representations are used for perceptual decisions during different tasks. In this talk, I will present data to show that neuronal response patterns predict which of four tastants (salty, sweet, sour, bitter) participants tasted on a given trial (1) and, in line with recent findings from local field potentials in the rat (2), that this taste quality information is largely contained to the delta frequency band. The onset of taste prediction coincides with the earliest taste-evoked responses highlighting that quality is encoded early in the central gustatory processing cascade. Importantly, response patterns correlate with perceptual decisions and behavioral goals: tastes that participants discriminate less accurately also evoke less discriminated brain response patterns and quality-specific differences in decoding onset are more pronounced during fast-paced as compared to delayed response tasks with decoding onset predicting behavioral response latencies. Together, the results evidence a link between taste-related decision-making and the predictive value of gustatory brain response patterns thereby highlighting the flexibility of gustatory coding.

Speaker: Janina Seubert
Title: Multisensory influences on olfactory perception during food consumption
Abstract: Identification of familiar food and avoidance of rotten or contaminated food is critical for human survival. While odors play a key role in this perceptual process, surprisingly few studies have studied the neural basis of olfactory influences on object formation. The series of studies presented here address the integration of odor into the food object perception during two separate stages of the eating process, and two distinct sensory pathways: the anticipatory stage, where odors are perceived by sniffing (orthonasally) in combination with a visual image of the food, and the consummatory stage, where the taste of a food object is bound together with the odor that reaches the nose through the passageways of the throat (retronasally). I will present findings indicating that concurrent orthonasal olfactory input enhances visual cortex activity to food objects in a way that is linearly linked to the level of semantic overlap between the two modalities, and that this enhancement is mediated by a top-down relay linking olfactory and visual cortex. I will then discuss the perceptional phenomena arising from a concurrent presentation of semantically matched or mismatched sensory information during consummatory evaluation of food. I will demonstrate the impact of varying semantic overlap between smell and taste in a beverage on various indicators of sensory integration, and present evidence that these perceptual phenomena arise in parallel rather than in direct dependence of one another. Finally, I will present a study that directly compares the neural activation to food odors arising from a source outside the mouth, to taste within the mouth, and to the combination of taste and retronasal odor during naturalistic food consumption to demonstrate that integration with its multisensory context is a key feature of olfactory food perception that likely contributes to facilitated object memory retrieval and thus appetite elicitation.

Speaker: Sanne Boesveldt
Title: Plasticity in neural flavor processing
Abstract: Food and flavor perception is a multisensory experience, including the sense of smell and taste, but also texture, visual and even auditory input. These sensory inputs are sent to the brain where they may evoke separate responses and are integrated to form a unique combined flavor percept. However, the neural signature of flavor representation may change and is dependent on individual flavor preferences, internal (metabolic state) or other factors (experience, sensory dysfunction). Recent work by the author has shown that subjective flavor preference reveals activations in OFC, ACC, and striatum; using multivariate analysis, a broader activation pattern (additionally including anterior insula) was identified that could successfully discriminate likers of a certain flavor stimulus from dislikers (1). In another study, we examined how food protein intake is regulated by internal state, and neural food cue responsivity. We found that protein status modulates brain responses in reward regions to savory food cues, suggesting that diet can alter brain and behavioral responses towards bodily needs (2). Moreover, it has been shown that neural responses to food cues may differ between overweight and normal-weight individuals, and that weight loss may also affect this. Roux-en-Y gastric bypass (RYGB) surgery is a highly effective weight-loss intervention, and patients frequently report a shift in food preferences away from high-energy foods. We investigated changes in neural reward responses to food pictures and odors, as well as neural inhibitory control processing, before and after RYGB. Lastly, we combined standard general linear model analysis (GLM) and functional connectivity analyses to investigate the neuronal bases of olfactory disorders. We identified three networks related to sniffing and olfactory function, and show that severity of olfactory disorder was reflected in decreased piriform cortex activity during odor processing. Together these results highlight the multilayered plasticity of neural flavor processing.

Date and time: Monday 11:00
Room: Effectenbeurs
Format: Keynote,
Keynote speaker: Daphna Shohamy
Title: How Memory Guides Value-Based Decisions
Abstract: Learning is central to adaptive behavior. From robots to humans, the ability to learn from experience turns a rigid response system into a flexible, adaptive one. How are decisions shaped by past experience? What are the neurobiological and cognitive mechanisms that allow experiences to change the way we perceive and act in the world? I will present evidence for a critical role for memory mechanisms in the hippocampus in biasing value-based decisions, focusing on two distinct mechanisms. The first concerns the integration of information across discrete past events to support generalization of past experience towards novel decisions. The second concerns the retrieval and use of memories for rare, “one shot” events when making decisions about reward. Finally, I will discuss how results emerging from this work challenge the traditional view of learning systems and advance understanding of how memory biases decisions in both adaptive and maladaptive ways.

Date and time: Monday 13:30
Room: Graan
Format and topic: Symposium, Attention
Coordinator: Manuel Schabus
Title: Oscillations in the broad alpha band: New Findings
Symposium abstract: Alpha oscillations cover a broad frequency range from about 7 or 8 to 12 or 13 Hz. Although there is little doubt that alpha oscillations play an important role for cognitive processes, the mechanisms by which alpha activity contributes to cognition remains largely an open question, which will be discussed from different viewpoints. In one contribution a new theory is presented in which alpha is part of a binary hierarchy frequency architecture of brain and body oscillations. Variations in alpha are explained by frequency shifts to enable maximal coupling or decoupling with neighboring frequencies, which are achieved by harmonic or golden mean (g = 1.618) frequency ratios respectively. As an example, alpha may shift from 10 to 8 Hz, to obtain separation from theta (8/1.618 = 5 Hz), or may stay at 10 Hz to enable optimal coupling with theta. In a second contribution, empirical findings from transcranial neurostimulation studies are reported that shed new light on how cognition and neural communication can be achieved via oscillatory activity in the human brain. In a third contribution, the focus is on the functional meaning of frequencies at the upper and lower frequency borders of alpha. The role of the 12-15Hz ('sensorimotor-rhythm') for hyperarousal will be discussed and its applicability for neurofeedback studies will be critically evaluated. The final contribution reports new findings for the lower alpha band, which suggest that oscillations at around 8 Hz are involved in memory replay.
13:30 - 13:40: Introduction
13:40 - 14:10 Paul SAUSENG
14:10 - 14:40 Manuel SCHABUS
14:40 - 15:10 Simon HANSLMAYR
15:10 - 15:40 Wolfgang KLIMESCH
15:40 - 16:00 Final Discussion Round

Speaker: Wolfgang Klimesch
Title: Beyond Alpha: A Binary Hierarchy of Brain-Body Oscillations
Abstract: Research on selected oscillations, such as human alpha or theta, is important to yield insight into their functional meaning. But what are the properties of the EEG frequency architecture, how many frequencies are there, and what are their coupling principles? A new theory, the "binary hierarchy oscillation theory" will be presented in an attempt to answer these questions. Its basic assumptions are a doubling/halving relationship between neighboring frequencies and an individual scaling of all frequencies f(i) by a factors ‘s’ that represents individual heart frequency in Hz: f(i) = s*(2i). As an example, for frequencies f(i) with i = 1,2 ... 5, the traditional EEG frequencies can be predicted as 2.5, 5, 10, 20, and 40 Hz reflecting delta, theta, alpha, beta, and gamma respectively. Slower frequencies with i < 0 predict frequencies for muscle contraction involved in inhaling/exhaling, the respiratory cycle (around 0.30 Hz) and the typical peak frequency of heart rate variability at around 0.15 Hz. It is argued that the phase of these center frequencies f(i) is the basis for m:n phase to phase coupling but also m:n phase to (amplitude) envelope coupling. The conclusion is that brain and body oscillations do not vary arbitrarily but are scaled (aligned) to each other to form a single system where m:n coupling represents the ‘cross talk' within brain and body oscillations as well as between them.

Speaker: Paul Sauseng
Title: Messing with the vibe: What transcranial neurostimulation can tell us about the functional meaning of brain oscillations
Abstract: Today, most cognitive neuroscientist would agree that slow brain oscillations such as theta and alpha are not a mere epiphenomenon of neural activity, but that they have a certain functional meaning. Determining the exact function of for instance alpha oscillations, however, is the tricky part. To do so, non-invasive, transcranial neurostimulation methods can come handy. They allow gearing into intrinsic and dominant brain oscillations from outside the brain. Here, I will present three major approaches by which transcranial neurostimulation can widen our understanding of oscillatory brain activity: (i) entrainment; the reinforcement of intrinsic brain oscillations in amplitude; (ii) modulation; the alteration of frequency, amplitude or phase of oscillations due to online or offline neurostimulation; (iii) probing; eliciting brain responses at time intervals at which particular rhythmical brain activation patterns are measureable. These approaches help us understand how cognition and neural communication and coordination can be achieved via oscillatory activity in the human brain.

Speaker: Manuel Schabus
Title: Entraining the brain to improve sleep: A double-blind study using 12-15Hz Neurofeedback
Abstract: Neurofeedback training builds upon the simple concept of instrumental conditioning, i.e. behavior that is rewarded is more likely to reoccur. In the case of neurofeedback, information about specific EEG activity is fed back to the participant who is rewarded whenever the desired EEG-pattern is generated. If some kind of hyperarousal needs to be addressed, the neurofeedback community considers 12-15Hz sensorimotor-rhythm neurofeedback as the gold standard. Earlier treatment approaches indicated that this training could reduce attention-deficit/hyperactivity disorder and epilepsy symptoms and improve sleep quality by enhancing sleep spindle activity. We sought to critically test whether earlier findings on the positive effect of sensorimotor-rhythm neurofeedback on sleep and memory could also be replicated in a double-blind placebo-controlled study. Insomnia patients spent nine polysomnography nights and twelve sessions of neurofeedback and placebo-feedback training in our laboratory. Crucially, we found both neurofeedback and placebo-feedback to be equally effective for improving subjective sleep quality suggesting that the observed improvements were due to unspecific factors such as experiencing trust and receiving care from experimenters. Importantly, these improvements were not reflected in any objective EEG-derived measures of sleep quality. Furthermore, objective EEG measures that potentially reflected mechanisms underlying the efficacy of neurofeedback such as increase in 12-15Hz oscillations following training or enhancement of sleep spindles remained unchanged following training. We conclude that -- at least -- for the treatment of primary insomnia, neurofeedback does not have a specific efficacy beyond unspecific placebo effects. Although, the study provides support for the principle of NFT, i.e. that participants can learn to control neural processes it highlights that sustained changes of oscillatory activity are difficult to achieve, especially for individuals with learning difficulties.

Speaker: Simon Hanslmayr
Title: Decoding the content of episodic memories from desynchronized alpha oscillations
Abstract: During the reinstatement of episodic memories a pronounced decrease in low frequency oscillations in the alpha range (~7 - 14 Hz) can be observed. However, it is unclear what such a decrease in synchronization entails mechanistically. In a recently proposed framework the speaker suggested that power decreases enable the cortex to represent information in a rich and detailed manner, thus allowing the re-experience of previous episodes. In this talk results from two recent studies will be presented which support this idea. In both experiments, subjects were asked to first associate short movie sequences or sound clips with a unique word and then to vividly replay the associated movies or sounds in a later recall test. In the first experiment, EEG was recorded and pattern similarity analysis in conjunction with source analysis was conducted. The results show that memory reinstatement is accompanied by a decrease of low-frequency (8 Hz) power, which carries a temporal phase signature of the replayed stimulus. These replay effects were evident in the visual as well as in the auditory domain and were localized to sensory-specific regions. In a second experiment, we recorded simultaneous EEG-fMRI and tested whether the reinstatement of stimulus specific spatial patterns correlates with power decreases in the alpha frequency range. Together, our results suggest that power decreases in the alpha frequency range allow for stimulus specific temporal and spatial patterns to re-occur during dynamic memory replay in humans.

Date and time: Monday 13:30
Room: Administratie
Format and topic: Symposium, Sensory processing
Coordinator: Ryszard Auksztulewicz
Title: Predictions in perception and action: neural mechanisms, cognitive determinants, and computational architectures
Symposium abstract: The notion that the brain generates predictions to optimise behaviour is now well established [1,2]. However, the scope and implementation of predictions in perception and action remain disputed [3]. While neural mechanisms mediating prediction signalling have been proposed [4], direct evidence remains scarce, and it is unclear whether identical networks and mechanisms subserve different domains, complexity levels, and contexts. Here we bring together evidence from invasive and non-invasive electrophysiology, neuroimaging, causal manipulations, and modelling, to elucidate the determinants and mechanisms of neural prediction signalling. Using fMRI and modelling, Speaker 1 et al. demonstrate that learning complex predictive structures rests on different strategies supported by distinct neural circuits. These include visual cortico-striatal circuits for matching sequence statistics, and executive/motor cortico-striatal circuits for maximizing the most probable outcome. Speaker 2 et al. provide novel direct evidence for sensory prediction errors in the macaque area ML during statistical learning of face processing. Furthermore, they show that these lower-level mismatch signals are shaped by tuning properties of higher-level regions, consistent with recurrent computational architectures such as predictive coding. Speaker 3 demonstrates that humans perform Bayesian inference while categorizing objects, with different computational components (prior probability and stimulus precision) linked to distinct oscillatory correlates. Manipulating this activity using tACS has dissociable behavioural effects, supporting the causal role of neural oscillations in Bayesian perceptual inference. Speaker 4 et al. combine electrophysiology and modelling to infer dissociable neural mechanisms subserving prediction and attention, as well as predictability of different stimulus features. Furthermore, they demonstrate that behavioural and neural effects of predictions are strongly modulated by contextual relevance. Taken together, we provide convergent evidence that the brain performs perceptual learning and inference in recurrent neural architectures, mediated by oscillatory activity. However, our results also suggest further research avenues into specific implementations of predictions across learning stages, stimulus domains, and behavioural contexts.

Speaker: Zoe Kourtzi, Rui Wang, Yuan Shen, Peter Tino, Andrew Welchman
Title: Learning predictive statistics of increasing complexity
Abstract: When immersed in a new environment (e.g., navigating a new city or surrounded by speakers of an unknown language) we are challenged to make sense of an incomprehensible stream of incoming sensory information. At first, it seems like a befuddling cacophony that leaves us completely unprepared for what will happen next. And yet, quite rapidly, the brain is able to find structure and meaning in the incoming signals, helping us to predict and prepare ourselves for future actions. Natural human experience involves event structures of increasing complexity from simple repetition (e.g. tones) to complex probabilistic combinations (e.g. melodies). To investigate the brain mechanisms that mediate learning of predictive structures, we expose participants to temporal sequences of increasing complexity (from simple regularities to complex probabilistic contingencies, as determined by Markov models) and compare their ability to predict upcoming events with patterns of brain activity. To interrogate the learning process, we quantify the decision strategies that individuals use to learn complex structures and predict future events. At the behavioural level, we demonstrate that individuals adapt to the environment's statistics; that is they are able to extract structures that change in their complexity over time. Further, individual learning of structures relates to decision strategy. We show that learning complex structures relates to selecting the most probable outcomes in a given context (i.e. maximizing) rather than the exact sequence statistics (i.e. matching). At the neural level, we show a distinction in the circuitry that supports these different strategies. Matching engages brain regions in visual cortico-striatal circuits (including the hippocampus), while maximizing exploits interactions between executive and motor cortico-striatal areas. Our results demonstrate that engaging this later circuit relates to learning more complex structures. In sum, our findings provide evidence for dissociated cortico-striatal mechanisms for learning increasingly complex structures that are predictive of upcoming events.

Speaker: Caspar M. Schwiedrzik, Winrich A. Freiwald
Title: Predictive feedback shapes hierarchical face-processing
Abstract: Recognition of visual objects from retinal inputs requires feed-forward information processing through multiple, hierarchically organised brain areas, transforming representations into increasingly complex, object-centered formats. Feedback processing, in contrast, remains enigmatic. Theoretical models such as Predictive Coding suggest that feedback connections serve to transmit predictions formed by higher-level brain areas back to earlier ones. When incoming sensory information and prediction mismatch, a prediction error is generated. This functional architecture implies that prediction errors, although generated in a lower-level area, reflect the tuning properties of cells that generate the prediction, i.e., of higher-level areas. This would constitute a major departure from the idea of strict modularity of feedforward architectures, where each stage performs a discreet operation on its inputs without external contributions. Here we show that the macaque face-processing network feeds face representations from high-level areas back to a lower-level area for stimulus prediction. During feedforward processing, this three-level hierarchy transforms early, picture-based representations in area ML into increasingly identity-selective and less head orientation-specific representations in subsequent stages AL and AM. We find that after statistical learning, ML signals the difference between predicted and actual inputs. But these signals do not reflect local, image-based tuning to head orientation characteristic of ML. Rather, they are shaped by tuning properties of high-level face areas AL and AM, exhibiting sensitivity to facial identity. Thus, feedback enables lower-level areas to process information at a higher level of abstraction than afforded by the feedforward sweep. Our results suggest a specific functional role for feedback connections and provide direct evidence for computational architectures such as Predictive Coding which rely on recurrent processing.

Speaker: Markus Bauer
Title: Brain rhythms shape Bayesian perceptual inference
Abstract: Perception is ambiguous due to the complexity of the world and the limited information provided by sensory signals. The state of the world has to be inferred from these signals, without the existence of a unique solution. Research predominantly into multisensory cue combination has provided strong evidence that brains compute Bayesian statistics to limit this ambiguity, but the universality of such a mechanism has been questioned and the underlying neuronal processes remain hypothetical. In a visual 2AFC task, we show that humans integrate the prior probability of visual object category (face/house) with stimulus information in a Bayesian sense, weighted by stimulus precision. The implicitly learned prior probabilities affect perception through modulation of anticipatory occipital alpha-/beta-oscillations, even without cues or participants' explicit awareness of this probability manipulation. By contrast, the precision of face/house representations is encoded in visually induced gamma-oscillations. The causal relevance of these mechanisms was established through transcranial alternating current stimulation (tACS, sinusoidal at 1mA), applied in different sessions at participants' individual alpha- and gamma-frequencies whilst they performed the perceptual task. Crucially, acute alpha-tACS enhanced the influence of prior probabilities on perceptual decisions, whereas acute gamma-tACS enhanced the impact of sensory stimuli, congruent with the electrophysiological results. Furthermore, psychophysically, we provide additional evidence on Bayesian perception from tasks using continuous stimulus judgements and manipulation of full prior probability distributions. Neurobiologically, further insight into the neurobiological mechanisms underlying Bayesian perception and the implicit learning of prior probability is provided by functional and effective connectivity analyses of induced and evoked responses. Taken together, these results provide evidence on the causal role of alpha-oscillations in relaying predictive feedback signals and gamma-oscillations forwarding sensory signals whilst participants integrate precision weighted sensory evidence with prior probability for Bayesian inference of visual object identity.

Speaker: Ryszard Auksztulewicz, Lucia Melloni, Karl Friston, Anna Christina Nobre
Title: Sensory predictions are modulated by attention and task relevance
Abstract: The brain is thought to generate internal predictions to optimise behaviour. However, it is unclear to what extent these predictions are modulated by other top-down factors such as attention and task demands. In this talk I will present results of three studies combining human electrophysiology and computational modelling to identify the neural mechanisms of sensory predictions and their interactions with current context. First, using magnetoencephalography (MEG) and dynamic causal modelling (DCM), sensory predictions and temporal attention were orthogonally manipulated in an auditory mismatch paradigm, revealing interactive effects on evoked response amplitude. This interaction effect was modelled in a canonical microcircuit using DCM. While mismatch responses were explained by recursive interplay of sensory predictions and prediction errors, their attentional modulation was linked to increased early sensory gain. Second, we analysed electrocorticographic data recorded from patients performing a task orthogonally manipulating "what" and "when" predictability of auditory targets. The two predictability types modulated evoked responses in different cortical regions and at dissociable latencies. DCM served to disambiguate between models of stimulus predictability in terms of top-down processing and gain modulation: "what" predictability increased auditory short-term plasticity, while "when" predictability increased putative synaptic gain in motor areas. This suggests that distinct predictions are mediated by qualitatively different neural mechanisms. Finally, we independently manipulated the spatial/temporal predictability of visual targets, and the relevance of spatial/temporal information provided by auditory cues. Relevance modulated the influence of predictability on task performance. To explain these effects, we estimated our participants' subjective predictions using a Hierarchical Gaussian Filter (HGF). Model-based time-series of predictions and prediction errors were linked to dissociable induced activity measured with MEG. Predictions correlated with beta-band activity, while prediction errors were signalled by increased gamma and decreased alpha-band activity. Crucially, these oscillatory correlates were modulated by task relevance, suggesting that current goals influence prediction signalling.

Date and time: Monday 13:30
Room: Berlage
Format and topic: Symposium, Memory
Coordinator: Moritz Koester
Title: Theta-gamma coupling as a mnemonic mechanism in the human brain
Symposium abstract: It is assumed that the interplay between theta and gamma neuronal oscillations form a general coding scheme in the human brain, with specific roles for working memory and episodic memory processing. First empirical findings on theta-gamma coupling during mnemonic processes in humans have been challenged with regard to their correlative nature and possible confounds in measures of theta-gamma coupling. This symposium comprises recent empirical and theoretical advances in the field. In the first presentation, Speaker 1 and colleagues report empirical support for the theoretical model that theta-gamma coupling plays a significant role in the ordering of multiple items during working memory encoding and maintenance in intracranial (ECoG) recordings. In the same vein, Speaker 2 presents findings from recordings in the hippocampus of epileptic patients, which demonstrate increased cross-frequency coupling between the theta and beta/low gamma range during working memory maintenance. In a follow-up study, this coupling pattern varied at a lower frequency (2 Hz), suggesting an even more complex hierarchy of oscillatory processes during working memory maintenance. Regarding Episodic memory formation, Speaker 3 presents empirical evidence the proposal that hippocampal theta/gamma coupling and neocortical alpha/beta desynchronisation co-occur during episodic memory formation, assessed in epileptic patients. Finally, in line with this theoretical model, Speaker 4 experimentally manipulated theta and alpha oscillatory processes during episodic encoding, using steady-state visually evoked potentials. Subsequent memory performance was enhanced for evoked theta oscillations and gamma oscillations induced by the theta rhythm, compared to evoked alpha oscillatory processes. Taken together, these results provide strong empirical evidence for a pivotal role of theta-gamma coupling in human mnemonic processing. The findings will be discussed in the light of recent theoretical models, methodological challenges in the field, and with regard to possible links between working memory and episodic memory processes.

Speaker: Ole Jensen, Ali Bahramisharif, Josh Jacobs, John Lisman
Title: Serial representation of working memory items during maintenance as revealed in ECoG data
Abstract: It has been hypothesized that coupled oscillations support a mechanism for maintaining multiple working memory presentations. According to one model (Lisman/Idiart/Jensen; LIJ model) interacting oscillations in the gamma range (>30 Hz) and theta (5--15 Hz) band organize multi-item working memory. To test this hypothesis we analyzed ECoG data from patients performing a modified Sternberg task. To locate sites involved in working memory, we were able to identify sites where an increase in gamma (band activity depended on which letter was presented). As expected, gamma and power were modulated during the encoding and maintenance periods of the task. Importantly cross-frequency coupling (CFC) was present during both encoding and maintenance. Our key finding was the theta phase of letter-induced gamma power reflected the order of letters presentation during maintenance. These results demonstrate that working memory is maintained by temporal code organized by coupled theta and gamma oscillations. Our findings provide strong support for theories suggesting that serial representation of items are held in working memory and organized in time by coupled theta and gamma oscillations.

Speaker: Nikolai Axmacher
Title: Cross-frequency coupling during working memory in the human medial temporal lobe
Abstract: According to an influential computer model, working memory (WM) depends on an interaction between neural oscillations of different frequencies. This model assumes that individual items are represented by local neural assemblies whose activity is reflected by bursts of gamma-band activity. During maintenance of multiple items, these representations are reactivated during specific phases of simultaneously occurring oscillations at low frequencies, e.g. in the theta frequency range. One basic prediction of this model is that gamma band activity is locked to specific phase ranges of low-frequency oscillations, resulting in phase-amplitude cross-frequency coupling (CFC) between theta oscillations and gamma-band activity. In a first study, we tested this prediction in human epilepsy patients implanted with intracranial EEG electrodes in the hippocampus. The hippocampus was chosen as the target region because of neuroimaging as well as neuropsychological results suggesting that the hippocampus plays a role for WM maintenance of multiple items. Indeed, we found that CFC during a WM task was increased as compared to baseline periods in the inter-trial interval. CFC was most pronounced between the phase of oscillations at 7 Hz and the amplitude of activity at 28 Hz, i.e. corresponding to activity in the theta and beta / low gamma range, respectively. We further observed that the properties of CFC change with increasing WM load. In a follow-up study, we found evidence for an even more complex model: Time periods exhibiting pronounced CFC were not sustained throughout the entire maintenance period, but occurred in rhythmic alternation (at a frequency of 2Hz) with other time periods showing load-dependent disinhibition of the hippocampus. These data suggest an oscillatory hierarchy of different oscillatory processes during multi-item WM in the human hippocampus.

Speaker: Benjamin Griffiths, Simon Hanslmayr
Title: Hippocampal theta/gamma synchronisation and neocortical alpha/beta desynchronisation co-occur during episodic memory formation
Abstract: Episodic memory formation hinges upon two cognitive processes: 1) the processing of incoming information about the to-be-encoded event, and 2) the binding of this multimodal information into a coherent episode. A recent opinion paper proposed that a division of labour between the neocortex and the medial temporal lobe could form the neural mechanism underlying these processes (Hanslmayr, Staresina & Bowman, Trends in Neurosciences, 2016). Specifically, desynchronised alpha/beta neocortical activity would facilitate the processing of incoming information while synchronised gamma activity, coupled to theta phase, serves to bind this information within the hippocampus. In an empirical test of the framework, epileptic patients with hippocampal depth electrodes learnt video-word and sound-word pairs, then were later cued with the word and asked to recall the associated video or sound. Preliminary analysis revealed greater hippocampal theta-gamma phase-amplitude coupling for later remembered items than those which were forgotten. Furthermore, gamma power appeared to peak at an earlier phase of the theta cycle for later remembered items relative to later forgotten items. While speculative, this could be viewed as phase precision during memory formation. Conforming to the framework, a significant negative correlation between hippocampal theta/gamma power and neocortical low-frequency power was identified, demonstrating that as hippocampal synchronisation increases, neocortical desynchronisation increases. This finding supports the theory that hippocampal synchrony and neocortical desynchrony co-occur during episodic memory formation.

Speaker: Moritz Koster, Thomas Gruber
Title: Evoked theta oscillations and theta induced gamma oscillations during encoding enhance subsequent memory performance
Abstract: The formation of episodic memories is associated with increases in theta and gamma power co-occurring with reduced alpha power in the human electro- and magnetoencephalogram (EEG and MEG). Furthermore, recent studies suggest a critical role of theta-gamma phase-to-amplitude coupling (PAC) for successful episodic encoding. These findings fit well within contemporary theoretical frameworks on memory formation processes. However, due the correlative nature of neurophysiological measures, these findings remain inconclusive about the functional role of these frequency bands during memory formation. In the present study we experimentally manipulated oscillatory processes during encoding, by presented pictorial stimuli at different driving frequencies. Namely, stimuli were flickered at an individually determined theta or alpha frequency to elicit visually evoked potentials in these specific frequencies (steady state visually evoked potentials; SSVEPs). First, this was to test whether evoked theta oscillations would facilitate memory formation, compared to evoked oscillations in the alpha band. Second, our analysis focuses on subsequent memory effects of gamma oscillations induced by individual theta SSVEPs, compared to induced alpha SSVEPs, to scrutinize the functional role of the interplay between theta and gamma oscillations. Behavioral results confirm the proposal that theta SSVEPs led to higher memory performance compared to alpha SSVEPs indicating a functional difference in memory formation processes between these two frequencies. Preliminary analyses also confirm that theta induced gamma oscillations are more predictive for subsequent memory performance compared to alpha induced gamma oscillations, suggesting a specific functional role for theta-gamma PAC in episodic memory formation processes. These findings provide strong evidence for a functional difference of theta and alpha oscillations and highlight the specific function of theta-gamma coupling processes in episodic memory formation.

Date and time: Monday 13:30
Room: Veiling
Format and topic: Symposium, MRI methods
Coordinator: Elinor Tzvi
Title: Dynamic Causal Modelling: a promising tool for studying the putative mechanisms underlying brain function?
Symposium abstract: While the early neuroimaging era has focused mainly on localizing cognitive processes such as perception, cognition and action in specific brain areas (functional specialization), the last decade has seen a fundamental shift towards a functional integration view of brain function. This has led researchers to explore connectivity as the fundament of functional integration and, thus, cognitive processing. Dynamic causal modelling (DCM) represents a tool that allows one to infer on these directed interactions in the brain from measured brain activity by modeling the brain as a deterministic input-output system. While DCM has been frequently used in the neuroscientific community and has provided valuable insights into brain organization, questions may still remain regarding the utility and additive value of these generative modeling approaches. The goal of this symposium is to familiarize the cognitive neuroscience community with DCM as a promising framework for understanding the putative mechanisms underlying cognitive functions. Our symposium will present four speakers who will shed light on the role of network interactions in shaping various cognitive processes in healthy subjects as well as in patients with cognitive deficits. Speaker 1 will give a tutorial-like introduction to DCM, explaining the basic concept in simple terms and giving practical examples that illustrate the type of questions one can address with DCM. This ensures that the symposium will be maximally informative for the entire audience (including people less familiar with DCM). Speakers 2-4 will then demonstrate how DCM may be used to test the putative mechanisms underlying hemispheric lateralization of cognitive functions, language recovery in post-stroke patients with aphasia, and cortico-striato-cerebellar interactions during motor learning. After presenting our results, we will devote time to a panel discussion to evaluate the promises (and pitfalls) of DCM for the cognitive neurosciences in a dialogue with the audience.

Speaker: Hanneke Den Ouden
Title: Estimating neuronal connectivity using dynamic causal modelling
Abstract: Over the past decades, neuroimaging research has undergone a fundamental paradigm shift, where the focus has broadened from solely localizing cognitive processes in specific brain areas (functional specialization) to the study of interactions within dynamic networks of brain regions (functional integration). In this first talk of the proposed symposium, I will give a basic introduction to Dynamic Causal Modelling (DCM), one of the most frequently used approaches for estimating neuronal connectivity from functional magnetic resonance imaging (fMRI) data (Friston, Harrison, and Penny 2003; Stephan et al. 2010; Kahan and Foltynie 2013). I will explain the generative model that underlies DCM for fMRI, explaining the underlying neuronal state equations in simple terms, and illustrating their dynamics using simulations. I will discuss the types of questions that may be answered using DCM, afforded by the two main levels of inference in DCM: i) model comparison, to quantify evidence in favour of competing network hypotheses, and ii) parameter estimation, which provides physiologically informed summary statistics of neuronal networks. I will finish by presenting three exemplary studies that illustrate the power of this modelling framework: First, we assessed how altered connectivity explains different sensory experience in different types of synesthesia (van Leeuwen, den Ouden, and Hagoort 2011). Next, we investigated how individual differences in impulsivity predict the effect of dopaminergic (ant)agonism on intra-striatal connectivity (Piray et al. 2015). Finally, I will show how DCM parameter estimates enable dissecting the schizophrenia spectrum into clinically relevant subgroups (Brodersen et al. 2011). The goal of this tutorial-like introduction is to enable listeners with limited experience in DCM (and generative modelling in general) to familiarize themselves with the basic concepts of this framework and to follow the ensuing talks of our symposium. This way, we hope to make this symposium most informative for the entire audience.

Speaker: Stefan Fraessle
Title: Towards a Mechanistic Understanding of Hemispheric Lateralization in the Face Perception Network
Abstract: A widely-used approach for studying effective (directed) connectivity in the human brain is dynamic causal modeling (DCM), a generative modeling framework for inferring hidden neuronal states from neuroimaging data. We used DCM for functional magnetic resonance imaging (fMRI) to investigate the neural mechanisms underlying a fundamental property of the human brain -- that is, the hemispheric lateralization of cognitive functions. Hemispheric lateralization implies that a cognitive process typically engages one hemisphere more strongly than the other. Although previous studies have demonstrated hemispheric lateralization for the majority of cognitive tasks, we currently lack a mechanistic understanding of the putative processes that underlie this phenomenon. In our studies, we introduced a generative modeling framework for assessing effective connectivity in the face perception network, for which right-hemispheric lateralization is well established. We showed that the right-hemispheric dominance of the network arises from an asymmetric face-specific interhemispheric recruitment at the hierarchically early level of the occipital face area (OFA). Notably, this interhemispheric recruitment was correlated with OFA's gray matter volume and the strength of pupil constrictions to faces. These results point to a decisive role of the interhemispheric integration among bilateral OFA and suggest a relation to individual variability in processing strategies for faces (holistic vs. feature-based). In addition, we mapped how other lateralized processes (e.g., handedness) affect these mechanisms by altering intra- and interhemispheric integration among core regions of the face perception network. In this talk, I will introduce a principled DCM framework for testing the neural mechanisms underlying hemispheric lateralization in the human brain and show an example of how this approach provides deeper insight into the functional organization of the face perception network. I will also provide an outlook on an entirely new methodological variant of DCM that brings effective connectivity analyses into the realm of whole-brain connectomics, offering exciting new possibilities.

Speaker: Tali Bitan
Title: Interhemispheric connectivity in post-stroke patients with aphasia: compensatory or maladaptive?
Abstract: The role of the right hemisphere (RH) in compensating for Left hemisphere (LH) damage during language processing in patients with aphasia, has been called into question in the last decade. According to the hypothesis the increase in RH activation is due to a release from LH inhibition and the RH may play a maladaptive role in language recovery by exerting transcallosal suppression on homologues LH regions. I will present two studies that directly examine interhemispheric connections in patients with aphasia using fMRI, to shed light on the role of the RH in recovery from aphasia. In study #1 we assessed effective connectivity using dynamic causal modelling during sentence comprehension in 7 post-stroke patients with aphasia. The results showed that all changes in inter-hemispheric connection in patients compared to controls were associated with poor language performance, inconsistent with a compensatory role for the RH. However, no transcallosal suppression was evident between homologous regions in neither patients nor controls, and the maladaptive involvement was a result of both excitatory and inhibitory right-to-left influences. In study #2, a chronic patient with aphasia following LH damage received melody based intervention. The patient's improvement in the production of trained and untrained phrases was accompanied by an increase in resting state functional connectivity between right frontal language areas and the supplementary motor area involved in motor planning. Altogether these results demonstrate the complex role of RH regions in recovery from aphasia which depends on the performed task, among other factors, and can be gleaned from the analysis of interhemispheric connectivity. While an increase in both excitatory and inhibitory right-to-left connections in patients may be detrimental for a sentence comprehension task, an intervention that emphasizes intact RH functions may recruit right frontal areas for compensation in a sentence production task.

Speaker: Elinor Tzvi
Title: Dynamic causal modelling of motor learning networks
Abstract: Acquiring and retaining novel motor skills plays a fundamental role in human behavior. This process which is characterized by quick and efficient performance of movements underlies a range of behaviors from tying a shoe lace to riding a bicycle. Imaging studies in healthy volunteers have shown that basal ganglia structures as well as motor cortical areas, parietal cortex, and cerebellum are all, to some degree, involved in motor learning (Hardwick et al., 2013). Theoretical models (Doyon et al., 2003; Penhune and Steele, 2012) suggest that learning motor skills is mediated through interactions between motor cortical areas, striatum and cerebellum. However, only few studies have used connectivity-based methods to specifically address these hypotheses. In this talk, I will present results from fMRI studies we conducted in healthy participants and in patients suffering from spinocerebellar ataxia while they performed in a motor learning task. Our aim was to use dynamic causal modelling (DCM) to specifically investigate dynamic interactions within the theoretical cortico-striato-cerebellar network. In the healthy cohort, we found that causal interactions from motor cortex to cerebellum play an important role in the early encoding phase of motor learning. On the other hand, putamen-cerebellar connections contribute to the slow learning phase which follows over-night consolidation. These findings have led us to investigate the important role cerebellum plays in motor learning using a network approach in a patient cohort. We studied patients with cerebellar degeneration, a disease that has been shown to cause specific motor learning deficits. We found pathological learning-related increase in activity as well as modulation of cortico-cortical connections, important for the learning process, which were compromised by cerebellar disease and thus led to motor learning deficits. These findings provide not only a better understanding of motor learning networks but also give insight to mechanisms underlying cognitive deficits in cerebellar disease.

Date and time: Monday 13:30
Room: Verwey
Format and topic: Ask-the-experts, Sensory processing
Coordinator: Vinitha Rangarajan
Title: Leveraging electrophysiology to study visual perception across non-human and human primates
Symposium abstract: Visual perception is a fundamental aspect of sensory processing that has been investigated using numerous methodologies over the past several decades. The strikingly complex nature of visual perception leaves many topics of enduring debate. This panel leverages the expertise of key members of the field to bring together varied lines of electrophysiology research. Dr. Malach utilizes intracranial electrophysiology to study categorical perception in humans; Dr. Leopold combines short-term and longitudinal intracranial recordings in primates to investigate the perception of natural images; Dr. Self utilizes single unit recordings in both humans and primates to study how perceptual organization modulates activity in early visual areas. The panelists' overlapping interests, yet nuanced expertise, will allow us to explore how electrophysiology can be used to address key questions in vision: How homogenous are activity profiles of neighboring cells in visual cortex and how do their profiles segregate? How do we compare visual activity across homologs and what limitations must be taken into account? What roles do top-down processes play in the activity of early and higher-order visual regions? We will discuss how regions of the brain interact to create perception from sensory information, fundamental to our understanding of how we experience and interact with our environment.

Speaker: Rafael Malach
Title: .
Abstract: Dr. Rafael Malach is the Morris and Barbara Levinson Professor of Brain Research at the Weizmann Institute of Science, Rehovot, Israel. His innovative work combining non-invasive (fMRI) and intracranial methods has led to numerous discoveries in the areas of categorical visual perception in humans. Dr. Malach has also trained numerous successful visual neuroscientists across the world.

Speaker: David Leopold
Title: .
Abstract: Dr. David Leopold is the Senior Investigator and Chief of the Section on Cognitive Neurophysiology and Imaging in the Laboratory of Neuropsychology in the National Institute of Mental Health. He is also the Director of the Neurophysiology Imaging Facility. Dr. Leopold has studied visual perception for over two decades and has directed pioneering work in short-term and longitudinal studies of vision in primates utilizing functional MRI, intracranial electrode recordings, and reversible inactivation.

Speaker: Matthew Self
Title: .
Abstract: Dr. Matthew Self is a senior researcher in the Roelfsema Research Group at the Netherlands Institute of Neuroscience. His researched is focused on the neural mechanisms of perceptual organization. He has published several novel studies investigating how attention and figure-ground segregation modulate neural activity in the early visual system of primates and mice. He also leads a team of researchers recording single-unit activity from intracranial depth electrodes in human patients.

Date and time: Tuesday 8:30
Room: Graan
Format and topic: Symposium, Decision-making
Coordinator: Saskia Haegens
Title: Principled computations of goal-directed perceptual inference: from the cortical column to the large-scale network
Symposium abstract: Half a century. This is how long it took computational models to confirm Hubel & Wiesel's original observations: a feed-forward neural network can be extremely efficient at mapping sensory information to invariant representations. However, such feed-forward architecture is, by definition, remarkably inefficient at modulating its own processing as a function of changing goals. What is the micro- and macroscopic neural architecture responsible for goal-directed behaviour? Here, we will present a series of electrophysiological studies identifying the mechanisms responsible for the dynamic modulation and routing of sensory information within neocortex. Our results -- based on (intracranial) recordings in both macaques and human subjects, using isolated laminar probes as well as local population recordings and widespread coverage over the cortical surface -- reveal a consistent relationship between specific oscillatory and evoked responses, and the computations presumably responsible for goal-directed perceptual processing. Specifically, we will address the following key questions: * What are the neuronal mechanisms responsible for the modulation of sensory processing as a function of current task and context? * What is the link between the patterns of neuronal activity identified at the micro-, meso- and macroscopic scales? * Is there a principled relationship between neuronal markers (e.g., alpha or gamma band activity) and putative canonical computations (e.g., likelihood or prior estimation)? By identifying the neural dynamics invariant across several experimental setups, our lines of work pave the way to identify the architecture of the mammalian cortex allowing goal-directed processing.

Speaker: Timo van Kerkoerle
Title: Influence of attention, working memory and predictive coding on each layer of monkey visual cortex
Abstract: Our perception is an inherently active process. We dynamically select information that is behaviorally relevant for us at given moments in time, enabling us to interact effectively with the world around us. This ability is thought to depend on feedback from higher cortical areas to early sensory areas, but this process remains poorly understood. One issue in investigating feedback processing is that it is generally activated at the same time as feedforward processing, making it hard to distinguish one from the other. However, feedforward and feedback connections target separate layers in the cortex, providing a way to disentangle the two processing streams. We trained monkeys on different cognitive tasks and recorded simultaneously in the different layers of monkey primary visual cortex (V1). First, we found a robust effect of both attention and working memory on spiking activity in V1. The working memory signal could be abolished with a mask, but reappeared at a later point in time. This suggests that V1 is not only involved in the selection of task relevant information, but also in the maintenance of that information when the stimulus is no longer present. Furthermore, we found that attention and working memory gave rise to feedback signatures corresponding to the modulation of spiking activity and synaptic input to V1. Finally, we trained monkeys to recognize a pattern of visual stimuli. This also allowed us to investigate the effect on different cortical layers when a stimulus was expected to arrive, and the error signals when this expectation was violated. These findings provide new insights in the laminar circuits involved in top-down modulation of activity in early visual cortex when a visual stimulus is present, when a stimulus was present, and when a stimulus is expected to appear.

Speaker: Lucia Melloni
Title: Utilization of brain rhythms in predictive coding revealed by psychophysics, MEG, ECOG and laminar recordings
Abstract: Previous experience allows the brain to predict what comes next. This entails predictions about when a stimulus is likely to occur, as well as which stimulus it will be. How both types of predictions are implemented is currently poorly understood. Here, we investigated the mechanisms, cortical circuits and oscillatory responses involved in perceptual inference about time and content in a series of studies employing invasive electrocorticography and laminar recordings, acquired from epilepsy patients implanted for clinical purposes, as well as from healthy subjects using noninvasive magnetoencephalography. Our results show that prestimulus entrainment to the temporal structure of the task in sensory and motor areas in the delta band allows subjects to process predictable stimuli more efficiently. In contrast, predictions about content are carried by prestimulus alpha oscillations. Interestingly, these alpha oscillations occur not only in directly task-relevant brain areas, but in a wider, multisensory network, and interact with the degree of stimulus degradation in affecting visibility. Furthermore, prestimulus alpha correlates with the amplitude of early sensory components (P1/N1m complex), suggesting a role in the selective amplification of predicted information. Together, our results indicate that low-frequency oscillations in different frequency bands can serve as mechanisms to carry and test sensory predictions about time and content.

Speaker: Saskia Haegens
Title: Rhythmic facilitation of temporal attention as revealed by psychophysics, MEG and ECOG recordings
Abstract: In daily life, we are continuously trying to predict and anticipate what's coming next, in order to give our actions a head start and focus our limited resources on behaviourally relevant input. Here, I will discuss anticipatory attention in terms of temporal predictions: when a relevant event is expected to occur at a certain moment in time, how does the brain optimally prepare for the processing of this input? What are the neural mechanisms forming these predictions and underlying their top-down control? I will talk about the oscillatory dynamics involved in temporal attention, specifically, neuronal entrainment to slow frequency rhythms. First, I will present a series of psychophysics experiments, showing rhythmic entrainment as a mechanism for focusing attention on relevant input. Subjects performed an auditory discrimination task, in which they received a temporal cue, which was either informative (rhythmic condition) or not informative about the likely timing of the target (random condition). We found that when a target is presented in-phase with the cued rhythm, performance is better as compared to both the out-of-phase (unexpected) and the random-mode (uninformative) condition. Furthermore, we found that this holds for a range of rhythms, and that subjects can pick up rhythmic structure both explicitly and implicitly. Second, I will present an accompanying MEG study where we looked into the neural correlate of this effect. We found increased delta entrainment in the rhythmic (but not the random) condition, which was sustained after the cue, and correlated with task performance. Third, I will present data from an ongoing study where we record intracranially in human epilepsy patients, complementing the MEG study. Combined, these experiments show that the brain can entrain to relevant rhythms in sensory input, even in the absence of ongoing rhythmic stimulation, that this rhythmic entrainment is behaviourally relevant, and under top-down control.

Speaker: Jean-Remi King
Title: Decoding the representation, selection and maintenance of invisible stimuli along the visual hierarchy
Abstract: The computational architectures discovered in vision neuroscience and developed in computer vision are remarkably converging. For example, it is now clear that visual processes are implemented through a long hierarchy of processing stages, structured by multiple semantic (sensory to abstract) and spatial gradients (small to large receptive fields). However, the dynamics of these architectures remain under-specified. Here, we show how the neural architecture of visual processes, as well as the evoked and oscillatory dynamics of each processing stage, can be automatically identified with time-resolved neuroimaging and machine learning. In a series of four studies, human subjects were either presented with low (Gabor patch) or high level (digit, letter) stimuli that had to be either detected, integrated, memorized or ignored. We show that a rich spectrum of visual features are simultaneously and automatically encoded in the early visual cortices independently of the task. By contrast, our results reveal i) that only relevant features are maintained in associative cortices and ii) that these high-level representations specifically correlate with subjective reports. We show how these results call for a partial revision of current theories of visual perception. More generally, our studies illustrate how machine learning and time-resolved neuroimaging techniques can automatically reconstruct the computational architecture of complex cognitive processing.

Date and time: Tuesday 8:30
Room: Administratie
Format and topic: Symposium, Decision-making
Coordinator: Matthew Apps
Title: Neurobiology of Motivation: Anatomy, pharmacology, pathology, and computation
Symposium abstract: Motivation is key to the successful performance of many perceptual, cognitive and motor acts. Yet, motivation can vary considerably between healthy people, with clinical levels of motivational impairment having some of the most dramatic impacts on quality of life in both psychiatric and neurological conditions. Yet, typically such impairments and variability in motivation are measured using self-report, psychometric scales, which fail to characterize the underlying psychological or biological mechanisms. In contrast, cognitive neuroscience approaches have the potential to provide a much richer understanding of the key mechanisms that drive variability in motivation and can inform therapeutic interventions. Recently such approaches are beginning to characterize and precisely quantify motivation using cost-benefit decision-making as a framework. In this approach, the costs that must be incurred (e.g effort) are subjectively traded off against the benefits (e.g. rewards) of acting. In this symposium the speakers will present some of the latest developments in behavioural and cognitive neuroscience research using cost-benefit decision-making to probe questions at the core of motivation: Why are some people 'workers' and others 'slackers'? How are different types of cost (e.g. cognitive, physical or social) evaluated? What are the computational mechanisms underlying the willingness to work? And why does motivation change over-time? These questions are explored across species, health, and disease. This symposium brings together research using dopaminergic, serotonergic and cholinergic manipulations, brain imaging, computational modelling and clinical studies. The aim of this diverse, multisdisciplinary set of speakers is to outline the current thinking on what cognitive, computational and biological mechanisms lead to cost-benefit decisions being so highly subjective. Moreover, several speakers will present novel evidence of key signatures of motivational impairments in neurological and psychiatric conditions.

Speaker: Mathias Pessiglione
Title: Why not trying harder? Computational approach to motivation deficits in neuro-psychiatric diseases
Abstract: Motivation deficits such as apathy are pervasive in both neurological and psychiatric diseases. They are currently assessed with psychometric scales that do not give any mechanistic insight susceptible to guide therapeutic intervention. Another approach has emerged lately that consists in phenotyping the behavior of patients in motivation tests, using computational models. Motivation can be defined as the function that orients and activates the behavior according to two attributes: a content (the goal) and a quantity (the goal value). Decision theory offers a way to quantify motivation, as the cost that patients would accept to endure in order to get the benefit of achieving their goal. In recent studies, cost has been typically measured as the amount of effort that patients would exert to obtain a particular reward. The trade-off between effort and reward was found to involve specific cortical, subcortical and neuromodulatory systems. Ideally, there would be a one-to-one mapping between specific neural components and distinct computational variables of the decision model. Thus, fitting computational models to patients' behavior would allow inferring the dysfunctional mechanism in both cognitive terms (e.g., hyposensitivity to reward) and neural terms (e.g., lack of dopamine). In this talk, I will present neuroimaging, pharmacological and patient studies that provide proofs of concepts for such computational approach of motivation disorders. I will start with the case of dopamine and sensitivity to reward, and then establish further links between computational variables and the brain systems that are targets of diseases and medications.

Speaker: Catherine Winstanley
Title: Deciphering decision making: Exploring interactions between neural and neurochemical systems underlying cognitive effort in rats
Abstract: The degree to which we are willing to select options that require more cognitive effort but which have the potential for greater rewards has far-reaching consequences for our economic and personal success. However, relatively little is known regarding the neurobiology governing the adjudication and application of cognitive effort in the decision-making process. We therefore developed a decision-making paradigm for rats which requires animals to choose between two options that differ in the degree of cognitive effort required to attain success. In this rat cognitive effort task (rCET), animals decide at the start of each trial whether to perform an easy or difficult attentional challenge. In the easy condition, rats must correctly localize a visuospatial target which is illuminated for 1.0s, whereas on hard trials, the target is only presented for 0.2s. Hard trials are more attentionally demanding, but accurate performance is rewarded with double the number of sugar pellets. We have observed that rats differ dramatically in their preference for the hard option, independent of their attentional ability, leading to their classification as either "workers" or "slackers". Through a series of pharmacological inactivation experiments, we have begun to characterize a network of regions within the affective corticostriatal loop that are involved in determining choice. We have also determined that, unlike decisions based on physical effort costs, choice on the rCET is not dopamine-dependent, but is very sensitive to the effects of cholinergic drugs. Unlike many other decision-making assays, choice of the harder option is also very sensitive to modulation by by Δ9- tetrahydrocannabinol (THC), such that even low doses cause rats to "slack off", even though this drug does not impair animals' ability to actually perform the more cognitively demanding challenge. These studies indicate that the evaluation versus employment of cognitive effort are regulated by somewhat unique, dissociable neurobiological mechanisms

Speaker: Matt Apps
Title: The dynamics of work: Neural and computational mechanisms of effort and fatigue
Abstract: Human behavior is often limited by motivation. Impairments to motivation are reported to be one of the greatest factors influencing quality of life in patients with Parkinson's disease (PD). Yet, the neural mechanisms underlying variability in motivation in healthy people, or in neurological conditions, are still poorly understood. Recent accounts cast motivation as a decision-making problem, where the costs of exerting effort are subjectively traded off against the potential benefits of obtaining rewards. Here, we use brain imaging and cost-benefit decision-making tasks in healthy people and PD patients, to examine the computational and neural basis of the willingness to exert effort. I present two sets of studies that probe distinct aspects of motivation. First, we examine the neural basis of processing the subjective value of rewards associated with either cognitive or physical effort. We find that the dorsal anterior cingulate cortex (dACC), dorsolateral prefrontal cortex and anterior insula play crucial roles in subjectively valuing rewards regardless of whether they are associated with cognitive and physical costs. However, we show that the Amygdala may have a unique role in valuing rewards associated with cognitive but not physical costs. Second, we examine how people's willingness to exert effort changes over-time as they become fatigued. We highlight some of the key factors that may underlie such dynamic changes in motivation in healthy people, and reveal a computational signature for increased levels of motivational impairment in PD due to fatigue. These findings pave the way for understanding the multi-dimensional nature of motivation and the mechanisms underlying motivational impairments in health and disease.

Speaker: Eliana Vassena
Title: Worth the effort? Neurobiology and computational modeling of effort-related behavior.
Abstract: The most obvious question popping up in your mind before undertaking any sort of life endeavor is a very obvious one: is it worth it? The answer to this question implies a cost-benefit trade-off: the course of action will be undertaken if the final reward will outweigh the cost required to obtain it. Such cost typically involves a certain degree of mental or physical effort. Prospective rewards and effort requirements are computed by the human brain, and inform subsequent decision-making and task-preparation. Neuroimaging evidence shows that effort is coded by medial and dorsolateral Prefrontal Cortex (MPFC, DLPFC). However, how such coding relates to existing accounts of PFC is unclear. We propose a theoretical-computational account of how these regions contribute to effort-based behavior. This account succeeds in including effort-related decision-making and performance in existing accounts of PFC functioning based on prediction and prediction error principles (Alexander & Brown 2011, 2015). We argue that effort and reward coding in PFC stems from predictability of task-difficulty and success rate, and occurs in a serial fashion. In a series of experiments we show that altering order of presentation of effort and reward cues affects performance and alters decision-making. Besides reconciling effort-coding evidence with prediction error literature on PFC, these results open up promising experimental avenues to investigate the plasticity of effort investment, particularly relevant for pathological conditions characterized by impairments of motivation.

Date and time: Tuesday 8:30
Room: Berlage
Format and topic: Symposium, Cognitive modeling
Coordinator: Katerina Chladkova
Title: Models of category formation and adaptation: distributional learning of speech and language
Symposium abstract: Humans are efficient statistical learners who can track the probabilities of events in their environment in order to form categories. This sensitivity to distributional information is found across modalities (Love, 2003; Goudbeek et al., 2009; Maye et al., 2002) and is reflected in neural responses (Garrido et al., 2016). The domain of speech and language constitutes perhaps the most intricate learning problem in that learners not only need to develop all the phonological categories remarkably fast (within their first year) but they also need to continuously adapt these categories throughout their lifetime (encountering new talkers and languages). In the lab, infants and adults often can learn speech categories through distributional training but the degree of success differs largely across studies: as a result, the learning mechanism has not yet been fully understood. This symposium contributes to our knowledge of the cognitive processes and constraints associated with distributional learning. Four experts introduce their computational models that employ distributional learning for speech and language development (see also e.g. Feldman et al., 2009; Thiessen & Pavlik, 2013). Speaker 1 introduces his deep-learning model – a multilevel neural network – that forms phonological categories on the basis of distributional bottom-up evidence alone, with top-down modulations when the bottom-up information is insufficient. Speaker 2 presents a Weighted Gaussian mixture model of phonological development throughout the lifespan that integrates multiple sources of auditory and visual information and factors out irrelevant context. The ideal adapter model developed by speaker 3 shows that hierarchical distributional learning plays a key role in speech processing by adults who constantly deal with enormous between-talker variation in the input. Finally, speaker 4 uses the linear ballistic accumulator model to analyse and reveal the mechanisms that humans employ during phoneme categorization. The factors that mediate the success of learning, as well as the biological and cognitive plausibility of the models are discussed.

Speaker: Paul Boersma
Title: Deep learning models of human bottom-up and top-down acquisition of phonological categories
Abstract: Several kinds of multilevel artificial neural networks that are similar to the "deep" networks that have been so successful lately in machine learning techniques, are also capable of modelling human language acquisition. This talk shows that these networks are capable of developing categorical behaviour when confronted with purely bottom-up distributional auditory evidence for incoming phonological categories, and that in cases where the distributional evidence is too scant for direct distributional learning, top-down information from meaning will still lead to the development of appropriate phonological categories.

Speaker: Joe Toscano
Title: Using statistical learning to acquire speech categories: A computational framework for studying phonological development
Abstract: Statistical learning offers a powerful mechanism for explaining how human infants acquire speech sound categories. For example, previous work has demonstrated that, with only a brief exposure, infants can use the distributional statistics of acoustic-phonetic cues to infer phonological categories (Maye, Werker, & Gerken, 2002, Cognition), a process that can be instantiated as a computational model of the developmental process (McMurray, Aslin, & Toscano, 2009, Developmental Science). How far can distributional learning take us? Does it provide a universal mechanism for explaining phonological development? In order to answer these questions, one of the primary issues that must be addressed is the fact that there is no one-to-one mapping between acoustic cues in the speech signal and phonological categories. Thus, any model of phonological development must explain how multiple cues are combined and how contextual factors (such as differences in talker identity) are handled. Moreover, it must do this in a way that is developmentally realistic (i.e., learning without labeled inputs telling the model what the "correct" answer is). I will present recent computational modeling work from our group using weighted Gaussian mixture models (WGMM) to study (1) how multiple acoustic cues are integrated; (2) how contextual differences can be factored out using unsupervised learning; (3) how integrated audio-visual representations are learned; and (4) how cue weights change over development and across the lifespan. Overall, we find that statistical learning provides a viable mechanism for describing these processes, and that, by studying the developmental process, we can understand principles about speech perception more broadly (e.g, whether or not listeners learn optimal representations of speech sound categories). I will also discuss areas where the WGMM framework must be extended to provide a more complete model, such as incorporating information from higher-level representations (e.g., lexical information) that may influence development.

Speaker: Dave Kleinschmidt
Title: The ideal adapter: Life-long, hierarchical distributional learning
Abstract: The ideal adapter is a computational-level theory of how human speech perception copes with variability at various levels. Even for a single talker, speech is highly variable: linguistic units (phonetic categories, words, etc.) are _distributions_ of acoustic cues. This means that speech perception is a problem of inference under uncertainty. When the listener knows the distribution of cues for each category, they can work backwards from a particular observed cue value to _infer_ how likely each category was. As distributional learning theories of language acquisition have long acknowledged, these distributions must be _learned_ during acquisition. The ideal adapter, however, further implicates distributional learning in adult language processing, in two ways. First, speech is variable not only _within_ a talker but also _between_ talkers. Thus, the cue distribution for each phonetic category (or other linguistic unit) changes from one situation to the next, and listeners need to continuously update their beliefs about these cue distributions via distributional learning. Indeed, listeners rapidly adapt to unfamiliar talkers, and this adaptation is quantitatively well predicted by a simple belief updating model based on the principles of distributional learning. Second, for an ideal adapter distributional learning is _hierarchical_. A critical difference between acquisition and adult language processing is that adults have substantial experience with other talkers. This experience is useful to an ideal adapter because it makes some accents (cue distributions) more likely than others, which provides an informative starting point for distributional learning. However, for this experience to be useful, listeners need to generalize from _particular_ prior experiences. This, in turn requires that they learn (implicitly or explicitly) the _distribution_ of talkers' accents, at multiple levels (e.g., all speakers of English, females, people from Detroit). Such hierarchical distributional learning is how an ideal adapter balances stability and plasticity, and is supported qualitatively by listeners' ability to recognize (and benefit from) familiar talkers, generalize to similar talkers, and adapt to novel talkers.

Speaker: Don van Ravenzwaaij, Gabriel Tillman, Titia Benders, Scott D. Brown
Title: Acoustic Cue Weighting in Vowel Perception, A Process Model Account
Abstract: Listeners rely on multiple acoustic cues to recognize any phoneme. The relative contribution of these cues to listeners' perception is typically inferred from listeners' categorization of sounds in a two-alternative forced-choice task. Here we advocate the use of an evidence accumulation model to analyze categorization as well as response time data from such cue weighting paradigms in terms of the processes that underlie the listeners' categorization. We tested 30 Dutch listeners on their categorization of speech sounds that varied between typical /A/ and /a:/ in vowel quality (F1 and F2) and duration. Using the linear ballistic accumulator model, we found that the changes in spectral quality and duration lead to changes in the speed of information processing, and the effects were larger for spectral quality. In addition, for stimuli with atypical spectral information, listeners accumulate evidence faster for /A/ compared to /a:/. Finally, longer durations of sounds did not produce longer estimates of perceptual encoding time. Our results demonstrate the utility of evidence accumulation models for learning about the latent processes that underlie phoneme categorization.

Date and time: Tuesday 8:30
Room: Veiling
Format and topic: Symposium, Sensory processing
Coordinator: Alice Tomassini
Title: Modulation of perceptual processes by motor processes: the role of neuronal oscillations.
Symposium abstract: A consistent body of research in the last decade has highlighted the role of oscillatory activity in sensory processing, leading to the notion that perception is inherently discrete and periodic. More recently, the focus has also turned to the role of motor processes in the proposed periodic nature of perception. This goes far beyond the long-standing idea that externally-triggered movements might be dominated by rhythmic components. The motor system might actually be capable of exerting endogenous control of oscillatory activity that entails perceptual consequences. Motor-related modulations of perceptually-relevant oscillations have been shown when sensory information is predictable in time – suggesting that the motor system orchestrates the temporal tuning of attention, optimizing information extraction during active exploration. Furthermore, oscillations in perceptual performance time-locked to the execution of voluntary movements have been shown for stimuli that are unpredictable and, importantly, unrelated to the motor task, pointing to a rather automatic form of sensory-motor coupling. In this symposium, we will present novel findings that reveal a leading role of the motor system in the modulation of sensory processing through rhythmic brain activity, and discuss their implications for action-perception coupling mechanisms.

Time schedule
8.30 Alice Tomassini - General Introduction
8.42 Luc Arnal
9.15 Benjamin Morillon
9.48 Alice Tomassini
10.20 David Acunzo

Speaker: Alice Tomassini, Pieter Medendorp, Eric Maris
Title: Theta oscillations synchronize perception with motor intention
Abstract: Our motor system orchestrates the sampling of sensory information by orienting our receptor organs in space and time. Mounting evidence further suggests that motor signals also contribute to the actual analysis of the incoming sensory data, thereby shaping the perceptual outcome. I will present behavioral and neurophysiological evidence showing that the motor system can modulate low-level sensory function, even when the movement does not involve the receptor organ and the stimuli are irrelevant for the motor performance. Crucially, this motor-related modulation has a rhythmic signature, occurs in an anticipatory fashion -- i.e., during movement planning - and entails perceptual consequences. We found rhythmic fluctuations of visual perception which are time-locked to the execution of voluntary hand movements and emerge before movement onset. This movement-locked rhythmicity in perceptual performance is predicted by the phase of EEG theta oscillations (~4 Hz) very early during movement planning (> 1 s before movement). Moreover, theta oscillations are phase-locked to the onset of the movement. Remarkably, the alignment of theta phase and its perceptual relevance unfold with similar non-monotonic profiles, indicating their relatedness. Altogether, these results show that the same brain rhythm that modulates perception also aligns to the future intended movement. Action planning hence seems to be accompanied by an endogenous phase adjustment of perceptually-relevant brain oscillations. The present work suggests that action and perception may be bound in an automatic way since the very early processing stages through neuronal oscillatory activity in the theta range.

Speaker: Benjamin Morillon
Title: Motor Origin of Temporal Predictions in Auditory Attention
Abstract: Temporal predictions are increasingly recognized as fundamental instruments for optimizing performance, allowing humans to exploit regularities in the world. It is proposed that the motor system instantiates predictive timing mechanisms, helping to synchronize temporal fluctuations of attention with the timing of events in a task-relevant stream, thus facilitating sensory selection. I will present a neurophysiological account for this theory in a paradigm where participants track a slow reference beat while extracting auditory target tones delivered on-beat and interleaved with distractors. At the behavioral level I will show that overt rhythmic movements sharpen the temporal selection of auditory stimuli, thereby improving performance. Capitalizing on magnetoencephalography recordings I will provide evidence that temporal predictions are reflected in Beta-band (~20Hz) energy fluctuations in sensorimotor cortex. While temporal predictions modulate the encoding of auditory information in bilateral auditory and fronto-parietal regions, overt movements further amplify this modulation in right-lateralized fronto-parietal associative regions and optimize behavior. Together, these findings are compatible with Active Sensing theories, which emphasize the prominent role of motor activity in sensory processing. Finally, I will dissociate the notion of temporal predictions from the idea of entrainment to periodic sensory inputs, and show in a behavioral experiment that periodic stimulation specifically fastens motor responses, whereas temporal predictions improve the precision of auditory processing.

Speaker: David J Acunzo, David Melcher, Andreas Wutz
Title: Alignment of perceptual and neural oscillations to voluntary saccades
Abstract: Visual perception involves the active selection of both where and when to move our eyes. There is extensive neurophysiological evidence that saccadic eye movements interact with ongoing neural oscillations. This link between motor actions and neural oscillations might reflect an optimal strategy for visual exploration. In a series of studies, we have explored the nature of the link between saccadic eye movements, neural oscillations and perceptual performance in tasks requiring temporal integration/segregation. We have found that whether two stimuli are temporally integrated or segregated is influenced by ongoing oscillations. In some cases, these perceptual alternations are aligned to stimulus or eye fixation onset. Moreover, even when making saccades in near darkness, we found consistent and dramatic changes in oscillatory activity in oculomotor and visual areas. Together, these results provide evidence that active perception involves a close alignment of perceptual and neural oscillations with saccadic eye movements.

Speaker: Luc H. Arnal
Title: A proactive sense of time in action
Abstract: The Predictive Coding theory and other Bayesian accounts of brain functioning provide a unifying standpoint about perception and action. In these frameworks, internal models make it possible to predict either forthcoming sensory input or the sensory consequences of an action. In both cases, the brain not only predicts the causes of sensory inputs ('what' is going to happen), but also 'when' they are most likely to happen. Precise temporal anticipation at the time-scale of seconds is critical, as a large number of phenomena pertaining to perception and action (i.e., speech, movement, object trajectories etc.) occur at this scale. However, whether the neural systems and neurophysiological mechanisms underlying predictive timing in perception and action rely on similar systems, computations and neurophysiological mechanisms remains unclear. In this presentation, I will discuss recent behavioral and neurophysiological evidence suggesting that endogenous (motor) signals in the delta and beta frequency-bands play a useful role in temporal prediction. I will discuss the "top-down" nature and putative motor origin of such signals and will argue that these endogenous neural rhythms offer adapted computational solutions to precisely anticipate upcoming events, keep-track of elapsing time at the time-scale of seconds and contribute to efficiently process complex quasi-periodic signals such as speech or musical rhythms.

Date and time: Tuesday 8:30
Room: Verwey
Format and topic: Symposium, Attention
Coordinator: Joe Bathelt
Title: Cognitive and neural mechanisms of attention development
Symposium abstract: Attention is the ability to flexibly select and prioritise information that is relevant to the task at hand. This skill is important throughout the lifespan, contributing to school readiness, academic achievement, and career success. Differences in attention have also been proposed as a potential mechanism for the development of later cognitive and behavioural problems, in the context of neurodevelopmental disorders such as attention deficit hyperactivity disorder (ADHD) and autism spectrum disorder (ASD). Research over the last decade has seen considerable interest in understanding the underlying mechanisms of attention development. On one hand, brain imaging has highlighted the importance of distributed networks and their dynamic interplay for attention development. On the other hand, cutting-edge behavioural studies have emphasised the importance of cross-syndrome comparisons to differentiate mechanisms that are specific to diagnostic categories. This symposium will introduce the range of current endeavours to understand attention in children and adolescence. Each speaker will describe how novel behavioural and neuroimaging approaches shed light on mechanisms of typical or atypical attention development.

Speaker: Christiane Rohr, Anish Arora, Ivy Cho, Prayash Katlariwala, Dennis Dimond, Deborah Dewey, Signe Bray
Title: Functional Network Integration and Attention Skills in Young Children
Abstract: Early childhood is a period of profound neural development and remodeling during which cognitive skills undergo rapid maturation. Over the past decade, functional magnetic resonance imaging (fMRI) has enabled the examination of functional brain networks as children mature. Due to the challenges associated with scanning young children, however, studies in early childhood are sparse. Here, we examined the relationship between age, attention and functional connectivity in five cognitive networks in 61 female children between 4-7y (mean=5.35; SD=0.78). Children's attention skills were assessed using the Early Childhood Attention Battery, which measures three components of attention: sustained, selective and executive;. these correlated with age r=.55, r=.5 and r=.3, respectively. Following training in an MRI simulator, children freely watched a children's TV show in a 3T GE750w scanner. 35 components were extracted from their fMRI data using FSL's MELODIC. Among these, the default mode, salience, frontoparietal, ventral and dorsal attention networks were identified as networks of interest and subjected to a dual regression, which resulted in a set of participant-specific spatial maps for each network. We then tested for linear effects of age and attention in these networks, thresholded at p<0.01 family-wise error corrected. All five networks showed increasing connectivity with age, particularly in the core regions of each network. Attention components were associated with changes in select networks after controlling for age. Selective attention, for instance, was related to an increase in connectivity in the dorsal attention network. Our findings thus highlight the profound network integration occurring across this age range, and add to our understanding of how changes in brain networks may relate to the development of attention skills in the important window of early childhood.

Speaker: Joe Bathelt, Sally Butterfield, Susan E. Gathercole, Duncan E. Astle
Title: Distinguishing subtypes of attention problems in children using network science and structural connectivity
Abstract: Diagnosed attention deficits in childhood and adolescence are a common problem with an estimated prevalence of between 5.9–7.1%. But problems of attention are common in the wider population, with variable severity, and are associated with worse school performance and poorer longer-term prospects in adulthood. Given their prevalence, considerable research efforts have been spent uncovering mechanisms that lead to attention deficits, with a view to their subsequent remediation. However, results are largely inconsistent. Heterogeneity within attention deficit groups, both with and without formal diagnosis, may be a potential reason. The Diagnostic and Statistical Manual of Mental Disorders 5th edition (DSM-V) distinguishes inattentive, hyperactive/impulsive, and combined subtypes with severities ranging from mild to severe. These subtypes show distinct behavioural profiles as well as subtype-specific neurological and genetic markers. However, these symptom dimensions are not unique to the ADHD subtypes. Attention impairments are also common across a range of neurodevelopmental disorders that are highly comorbid with ADHD, like learning impairment, oppositional defiant disorder, autism spectrum disorder and others. Instead of the traditional grouping of disorder categories, more salient dimensions may exist that take the range of comorbidities into account and relate more closely to underlying biological mechanisms.

Speaker: Hilde M. Geurts, Leo Kannerhuis
Title: ADHD & ASD: What is known regarding cognitive convergence and divergence?
Abstract: After a short history of studying convergence and divergence regarding cognitive development in children with ADHD and ASD, recent research avenues will be discussed. The focus will be on executive control/attention as this cognitive domain has gained the most research attention when focusing on the convergence between the aforementioned two neurodevelopmental disorders. Since the recent changes in the DSM-5, in which also the combined diagnosis of ADHD and ASD is now included, this research field seems to be booming, but in order to ensure it will be blooming it is of importance to be aware of the challenges in this field. Therefore, in this talk findings (e.g., Hartman et al., 2016) regarding developmental changes in symptomatic overlap will be discussed which are illustrate of specific measurement issues. Moreover, recent cognitive findings (e.g., Andersen et al., 2014) will be used to illustrate how crucial it is to include a developmental perspective and to be careful with general statements regarding cognition in ADHD and ASD.

Speaker: Grainne McLoughlin, Jason Palmer, Charlotte Tye
Title: Interaction between pre-stimulus and post-stimulus oscillations differentiates ADHD and ASD
Abstract: Oscillatory activity in the brain has been shown to be a key factor in regulating attention. Neural oscillations measured using EEG can index attentional processes both before and after a stimulus appears. Accumulating evidence suggests that activity occurring during the pre-stimulus period is a reliable index of attentional engagement and plays a significant role in modulating post-stimulus behavioural and neurophysiological responses. Attentional impairments have been proposed as a common mechanism for the development of autism spectrum disorder (ASD) and attention deficit hyperactivity disorder (ADHD). In this talk, I will introduce concepts underlying pre-stimulus oscillatory activity in relation to attentional processes and present findings that demonstrate differential associations between pre-stimulus alpha, task performance and post-stimulus oscillatory activity and event-related potentials, in a cross-disorder study of children with ASD and ADHD. Alpha (9-12 Hz) power and phase at stimulus onset were measured in the pre-stimulus period, and compared to behavioural responses, amplitude of early event-related potentials (ERPs; P1, N2) and post-stimulus theta (5-8 Hz), derived from EEG sources separated using independent component analysis (ICA). Children with ASD showed increased attentional engagement, or preparation for the upcoming stimulus (greater alpha desynchronisation) compared to children with ADHD. Children with ADHD showed impaired attention and cognitive control (N2 and P3) compared to ASD and TDC. A clear distinction between the disorders was found in the relationship between desynchronisation of pre-stimulus alpha and early ERP components and post-stimulus theta. Alpha desynchronisation was positively correlated with both P1 amplitude and theta synchronisation in ADHD and TDC, but negatively correlated in ASD. That is, in ASD, greater preparation for the upcoming stimulus unexpectedly resulted in poorer attentional processing post-stimulus. These findings may explain inconsistent attention deficits reported in previous studies in ASD and demonstrate the importance of objective brain measures in distinguishing pathophysiological mechanisms between ADHD and ASD.

Date and time: Tuesday 11:00
Room: Effectenbeurs
Format: Keynote,
Keynote speaker: John Duncan
Title: Attentional episodes and cognitive control
Abstract: All human cognition is controlled in a series of attentional episodes, breaking complex problems into simpler, more solvable sub-problems. In human fMRI studies, a common or multiple-demand (MD) pattern of frontal and parietal activity is associated with diverse cognitive demands, and with standard tests of fluid intelligence. Based on behavioural, neuropsychological, fMRI and single unit data, I suggest that the core function of MD cortex is to control complex cognition in a structured sequence of attentional episodes. At the same time, I consider the complementary control functions of MD and "default mode" networks.

Date and time: Tuesday 13:30
Room: Graan
Format and topic: Symposium, Sensory processing
Coordinator: Iris Groen
Title: How can deep neural networks help us understand sensory processing in the human brain?
Symposium abstract: Sensory processing in the human brain is dedicated to extracting information from real-world environments to enable adaptive behavior, and is mediated by multi-stage neural activity in a hierarchical network of cortical regions. Understanding this process requires a predictive, quantitative model that captures the complexity of this network and its underlying spatio-temporal dynamics and explains human behavior. In recent years, deep neural networks (DNNs) have emerged as promising computational models for understanding sensory processes. The reason for their increased popularity is that DNNs perform with unprecedented, human-level accuracy on visual object recognition tasks (Russakovsky et al., 2015). Moreover, recent research comparing visual representations in DNNs and primate brains has demonstrated a striking correspondence, creating much excitement in visual neuroscience and beyond (Kriegeskorte, 2015; Yamins & DiCarlo, 2016; Marblestone et al., 2016). The goal of this symposium is to examine how we can push this approach beyond establishing mere correspondences between models and brain data towards generating truly novel insight into the sensory representations underlying adaptive behavior. To introduce the approach for a wide audience, Speaker 1 will start with a brief introduction to DNNs and the established findings, and then move on to propose an integrated theoretical framework for how the use of DNNs in neuroscience can help achieve a new kind of understanding of visual processing. Speakers 2 and 3 will discuss how DNNs can elucidate the functional organization of visual cortex, including the development of multiple visual pathways and the dynamic routing of information within those pathways. Finally, Speaker 4 will focus on the comparative capacity of DNNs to simultaneously predict behavioral and neural measurements. A common theme that will emerge from this symposium is that considering and manipulating the task used for DNN training is critical to generate novel insights into sensory processing in human brains.

Speaker: Radoslaw Cichy
Title: What can we learn from artificial deep neural networks about biological brain function?
Abstract: Understanding the neural basis of complex cognitive tasks requires answering three questions: what is happening when and where in the human brain when we carry out those tasks? The talk has two parts. In the first part, I will present recent empirical work in vision science that addresses these questions by integrating deep neural networks, magnetoencephalography (MEG) and functional magnetic resonance imaging (fMRI) in a common analysis framework. Convergent results from several groups show a hierarchical correspondence between brain activity and DNNs in both space and time. Investigating the origin of the correspondence, I will discuss the importance of teasing apart factors architecture, task, and training procedure. The second part of the talk will be theoretical in nature. It will start with a general interpretation of the results, spelling out what the comparison with DNNs taught us about human brain function: it provides a new kind of understanding, a novel form of human brain mapping, it is at a sweet spot of what current non-invasive recording techniques in humans can accomplish, and it provides a general framework for further investigation. Next, I will point out and relate to other research efforts that used DNNs to understand brain function, e.g. of the auditory system, and suggest how yoking efforts in computer science and neuroscience could provide further insight.

Speaker: H. Steven Scholte, Max M. Losch, Sander M. Bohte
Title: Visual pathways from the perspective of cost functions and deep learning
Abstract: The insight that we can understand higher-tier visual cortex from the perspective of multiple pathways with different goals has been seminal in shaping vision research. The remarkable similarities between deep neural networks (DNNs) and the human brain make it an attractive idea to try to use them in the endeavor to understand why there are multiple pathways in the brain. DNNs learn from data under optimization of cost functions, which map the state of the system to a measure of performance. Using such an optimization procedure results in a network with units in the lower tier layers more tuned to the statistical structure of the input and units in higher tier layers more tuned to the cost function used for training the task. One could argue that the single loss function optimized in nature is the fitness and abundance of progeny. Such a problem is easier to solve when it is decomposed in subtasks with their own, uncorrelated, cost function. We will argue that "vision for perception" and "vision for action", central to the dual-pathway theory of the visual system, could be examples of such a decomposition. If a localization and identification task are performed on the same input image, and this problem is solved by a network that consists of multiple layers it can be imagined that the units in the lower-tier layers can be used to solve both tasks. Conversely, the units in the higher-tier layers are expected to be tuned to a specific output layer. From this it would follow that task as "vision for perception" can be further decomposed into sub-tasks such as shape discrimination and surface segmentation, each with their own cost-function. To support this theory we will review examples of recent DNN's architectures and compare them to the functional anatomy of the visual system.

Speaker: Jonas Kubilius, Kohitij Kar, Daniel L. K. Yamins, James J. DiCarlo
Title: Building temporal convolutional neural networks: How predictability and parsimony can help each other 
Abstract: In the recent years, feedforward deep neural networks have surpassed other classes of models in predicting neural responses in the primate inferior temporal (IT) cortex (Yamins et al., 2014) and in providing response patterns consistent with primate behavior in several object judgment tasks (e.g, Kubilius et al., 2016). However, despite a strong promise for bringing better predictive models for many phenomena in cognitive sciences, deep nets remain poorly adopted by cognitive researchers. A common argument against using them is that deep nets are too complex and do not provide an adequate understanding of the processes occurring in the system. In this talk, I will argue that predictability and understanding, or perhaps more concretely, the parsimony of the models that we build are not necessarily inconsistent goals and can in fact enhance each other. To illustrate this idea, I will describe how investigating the performance patterns of object identification in deep nets in our recent work lead to the predictions of response decoding latencies in the monkey IT cortex. Specifically, we observed that the images that deep nets found hard to identify lead to the longer response decoding latencies in the IT cortex, presumably reflecting the lack of recurrent connections in these feedforward architectures. Next, I will demonstrate how we used these empirical observations to inform and constrain new classes of models. I will discuss how we built a general-purpose temporal convolutional neural network architecture that can be defined for any network topology, including within-layer recurrence, feedback and bypass connections. Such a multi-pathway architecture and its time-varying outputs, while built out of the needs of a particular task, also provide a more plausible model of the visual system, opening a possibility for investigating dynamic processes in machines and primates using deep architectures.

Speaker: Iris IA Groen, Michelle R Greene, Chris Baldassano, Li Fei-Fei, Diane M Beck, Chris I Baker
Title: Mind the gap: comparing DNN and functional models of scene representation in brain and behavior
Abstract: An important question when considering DNNs as models of sensory processing is whether these models use similar features to solve visual tasks as humans do. In this talk, I will argue that addressing this question requires a careful comparative approach that quantifies the degree of shared information across multiple models in both brain and behavior. I will illustrate this argument using behavioral and functional magnetic resonance imaging (fMRI) measurements of real-world scene categorization. The popularity of DNNs is based on their outstanding performance on object recognition tasks, but their performance on other visual tasks, such as scene recognition, is being explored increasingly (Zhou et al., 2014; Cichy et al., 2016; Malcolm et al., 2016). Recent behavioral evidence suggests that humans may use other features than objects for scene recognition, such as the functions afforded by a scene ("Could I prepare food here?"). Specifically, scene functions were capable of predicting additional, unique variance in behavioral scene categorization that was not explained by object labels or an object-recognition trained DNN (Greene et al., 2016). However, comparing these models directly against fMRI responses indicated that the DNN model best predicted responses in higher-level, scene-selective visual cortex, with limited additional contribution from functions. Together, these findings demonstrate a discrepancy between the brain and behavioral results, whereby scene functions explain scene categorization behavior better than DNN representations, while DNN representations predict brain activity better than scene functions. Interestingly, a place-recognition trained DNN improved both behavioral and fMRI prediction performance, leading to increased shared information between the DNN and functional representations. Together, these results highlight that while DNNs hold great promise as models of sensory processing, their ability to predict brain measurements does not necessarily transfer directly to predicting human behavior, and may require learning representations that are more task-specific.

Date and time: Tuesday 13:30
Room: Administratie
Format and topic: Symposium, Cognitive modeling
Coordinator: Jason Palmer
Title: The Influence of Cortical Field Potentials on Perception, Action, and Cognition
Symposium abstract: There is a growing body of research demonstrating a link between physiological factors involving brain potentials and oscillations, and psychological factors of perception, action, and cognition, or cognitive control. In particular, several studies have demonstrated that the phase of alpha oscillations in the occipital cortex influence the efficacy of visual perception. Potentials in the parietal cortex have also been shown to correlate with perceptual timing. More recent studies have shown that the phase of theta oscillations is directly related to the timing of the initiation of motor response in the motor cortex. Theta oscillations have also been proposed to underlie cognitive coordination more generally, possibly forming a global brain mechanism enabling cognitive control. In this symposium, prominent researchers present recent work demonstrating these various links and discuss future prospects in this exciting research area.

Speaker: Laura Dugue
Title: The rhythm of attentional exploration
Abstract: A growing number of researchers support the idea that neural oscillations allow dynamic changes in perception and cognition. This topic is gathering interest in the scientific community as compelling evidence accumulates. A variety of brain functions, such as attention, perception and memory, have been associated with oscillations in specific brain areas and frequency bands. For example, in a 2011 study, we showed that the phase of occipital alpha oscillations predicts phosphene perception (illusory flash induced by TMS applied over the visual cortex). In this talk, I will present experiments exploring covert spatial attention, the ability to enhance visual processing without shifting one's gaze. In the first series of experiments, we use TMS and EEG to investigate temporal dynamics of covert attention during difficult visual search tasks. We show that occipital theta oscillations (4-7Hz) modulate attentional search performance periodically. Successful and unsuccessful searches were associated with different amounts of post-stimulus phase-locking, as well as opposite pre-stimulus oscillatory phases. Additionally, TMS applied over the occipital pole demonstrated a periodic pattern of interference at the same frequency. In another set of experiments using psychophysics, we explored the link between attentional periodicity in time, and attentional exploration of the visual space during difficult search tasks. We propose that covert spatial attention is non-uniformly allocated to different stimuli presented in a search array, and that this deployment is periodically modulated in time at the theta frequency. Finally, in a TMS study, we explicitly manipulated attention shift from an unattended location to a target location, revealing periodic sampling of visual information in time at the theta frequency. Using complementary methodologies, we were able to further our understanding of the spatio-temporal behavior of attention. Together, the results suggest that attentional exploration is a rhythmic process, supported by oscillations of brain activity in the theta frequency band.

Speaker: Chris Saville
Title: The relative contribution of single-trial stimulus and response-locked P3b latencies to reaction time -- insights from psychopharmacology, clinical neuroscience, and experimental psychology
Abstract: The P3b is unusual amongst classical event-related potential (ERP) components in that it is prominent in both he stimulus-locked and response-locked average ERP. This has been argued to reflect a role in bridging between perception and action. While there is something approaching consensus on what the stimulus-locked P3b latency reflects, the functional significance of response-locked latencies is less clear. Innovations in measurement of single-trial ERPs permit a fresh perspective on these issues. This presentation will combine single-trial analyses of data from psychopharmacological, clinical neuroscientific, and experimental psychology studies to explore the overlapping but distinct functional significance of stimulus and response-locked P3b latencies

Speaker: Grainne McLoughlin, Jason Palmer
Title: Theta oscillations and performance in multiple delay flanker and go/nogo paradigms
Abstract: Theta oscillations have recently been closely associated with cognitive control in a variety of contexts. Theta power has been shown to be positively correlated with performance in tasks requiring cognitive discernment. Recently we have shown that central midline theta phase measured in EEG, particularly its variability, is also closely related to attention regulation in a genetic study involving participants with ADHD performing an Arrow Flanker experiment. To further investigate the relation between theta phase and performance, we designed a novel modified Arrow Flanker task with multiple delays between the onset of the Flanker and Target stimuli, to investigate a potential link between the Flanker-Target delay, the central midline theta phase, and the probability of error. We compared this to a similar novel multiple delay Go/NoGo task. We show that in both of these paradigms a strong theta signal is observed in the central midline and the phase of the theta signal is highly predictive of cognitive preparation and timing of the response. We also found an "interference effect" between the Target-Flanker delay, and the probability of error, with a distinct peak in error percentage occurring at delays of approximately one theta cycle in both paradigms. We present a hypothesis on how this interference effect may be mediated by the theta oscillation.

Speaker: Jason Palmer, Grainne McLoughlin
Title: Theta as a potential means of global cognitive coordination
Abstract: Recently a number of papers have highlighted the importance of brain oscillations in controlling and coordinating a participant's behavioral performance in standard stimulus-response paradigms. In particular, theta oscillations have been proposed as an important factor possibly underlying the global coherence of effectively coordinated perception and action. In this talk, we review the recently proposed theories of theta function, as well as its physiological basis. We also present results from EEG experiments that demonstrate theta occurring in globally disparate brain regions with a definite pattern of phase delay. We present a theory as to the mechanism of global cognitive control in speeded stimulus-response tasks involving thalamo-cortico-striatal connections subserving a coordinated perception-action-cognition cycle. We also discuss neurocomputational models that have been proposed to model paradigms such as the Eriksen Flanker task and how these should be modified to incorporate oscillatory dynamics. Finally we suggest how the appropriateness or goodness of fit of these models might be assayed through the comparison of model simulation results with behavioral experimental data.

Date and time: Tuesday 13:30
Room: Berlage
Format and topic: Symposium, Developmental
Coordinator: Kim Cornish
Title: Building Cognitive Architecture in Atypically Developing Populations: Assessing The Potential of Targeted Cognitive Training
Symposium abstract: The efficacy of cognitive training programs has been one of the most widely debated and controversial topics in cognitive neuroscience in recent years. There are numerous commercialised products claiming to improve cognitive and brain function without any high quality scientific evidence, which has contributed to significant skepticism by cognitive neuroscientists. Regarding evidence based cognitive training programs, methodological limitations are common and few have used a longitudinal approach to capture subtle developmental changes in cognition. In this symposium, we highlight four recent longitudinal, randomised controlled trials (RCTs) that have been conducted in 'at-risk' child populations. We focus on two fundamental cognitive domains: attention and working memory, known to be 'gateways to learning'. Speaker 1 focused on children born extremely preterm, evaluating the effectiveness of a working memory training program using neuroimaging and neurocognitive assessments over a 2-year period. Speaker 2 studied children with neurodevelopmental disorders such as autism to determine if a novel interactive attention training program can improve separable components of attention e.g. alerting vs. control, and if so, can improvements be sustained over a 6-month period. In children with traumatic brain injury Speaker 3 examined whether a working memory training program improved separable components of working memory e.g. phonological vs central executive, over a 3 month period; and Speaker 4 tested in a population-based sample of children screened for low working memory, the effectiveness of a working memory training program in a school-setting for improving academic performance over a 2-year period. Collectively, these findings build a strong case for utilising cognitive training in targeted groups of children. It seems unrealistic to expect training to benefit all paediatric populations. Instead, this work builds a pathway to a greater understanding of which children benefit most and when.

Speaker: Peter J Anderson, Leona Pascoe, Elysha Josev, Gehan Roberts, Deanne K Thompson, Marc Seal, Katherine Lee, Chiara Nosarti, Susan Gathercole, Lex W Doyle
Title: Randomised Controlled Trial of Working Memory Training with Extremely Preterm Children
Abstract: Objective: Preterm children are at significant risk for a range of neurodevelopmental difficulties, including their capacity to learn and process new information. It has been suggested that inattention and impaired working memory are primary deficits for preterm children, contributing to their learning and behavioural problems. The aim of this trial was to assess the effectiveness of a cognitive training program in extremely preterm (EP)/extremely low birth weight (ELBW) 7-year-olds in improving working memory, attention and academic achievement compared with a placebo program. Design/Methods: Double-blind placebo-controlled, randomised trial of the Cogmed RM program with 7-year children born EP (<28 weeks' gestation)/ELBW (<1000g). Participants were randomised to receive Cogmed or a placebo training program. The Cogmed program involves intensive and adaptive training of working memory (20-25 training sessions of approx. 45 mins each over 5-7 weeks). The placebo program is identical but the tasks are set to a low complexity level so that working memory is not taxed. Working memory, attention and academic achievement were assessed pre-intervention and post-intervention (within 2 weeks, and 12 and 24 months later). Brain MRI scans were performed pre- and post-intervention to assess training induced neuroplasticity. Results: 45 children were randomised to Cogmed and 46 to the placebo program. The intervention and placebo groups were well-matched, and at baseline performed similarly on working memory and attention measures. No group differences were found immediately post training on tests of working memory, attention or academic achievement, and mixed model analyses found no group differences in change over time. No training induced neuroplasticity was observed. Longer term outcomes related to working memory training are ongoing. Conclusions: Cogmed was not effective in improving working memory, attention or academic achievement, and did not alter brain structure immediately following training in this cohort of EP/ELBW 7-year-olds.

Speaker: Natalie L Phillips, Anna Mandalis, Suzanne Benson, Louise Parry, Adrienne Epps, Angie Morrow, Suncica Lah
Title: Working Memory Training for Children with Moderate-Severe Traumatic Brain Injury: A Randomised Controlled Trial
Abstract: Objective: Paediatric traumatic brain injury (TBI) places children at risk of deficits in working memory (WM), which is a multi-component system comprising a central executive (CE), phonological loop (PL), and visuospatial sketchpad (VSSP). WM is strongly related to attention and academic skills in childhood. This study aimed to examine whether different components of WM can be improved following adaptive WM training (Cogmed) and whether improvements in WM generalise to other cognitive (attention) and academic skills (reading and mathematics) in children with TBI. Design/Methods: Double-blind, randomised, placebo-controlled trial. Twenty-seven children with moderate-severe TBI (8-15 years, >12 months post-TBI) were randomized to 5-7 weeks (20 -- 25 sessions) of adaptive WM training (Cogmed; n = 13) or non-adaptive WM training (active placebo; n = 14). Standardised tests of WM, attention, and academic skills were administered at pre-training, post-training, and 3-months follow-up. Results: Children in the adaptive group demonstrated significantly greater gains on select WM tasks (VSSP, but not PL or CE) from pre- to post-training (pre-post) and pre-training to follow-up (pre-follow-up). No gains were found on tests of attention. Adaptive training resulted in significantly greater gains on select academic skills (reading, but not mathematics): reading accuracy pre-follow-up. Conclusions: This first, to our knowledge, study to examine the efficacy of adaptive WM training for children with TBI provides preliminary evidence of near and far transfer of training to WM and academic skills, respectively.

Speaker: Hannah E Kirk, Kylie M Gray, Kirsten Ellis, John Taffe, Kim M Cornish
Title: Attention Training for Children with Neurodevelopmental Disabilities: A Randomised Controlled Trial
Abstract: Objective: Children with intellectual and developmental disabilities (IDD) experience heightened attention difficulties which have been linked to poorer cognitive, academic and social outcomes. Although, increasing research has focused on the potential of cognitive training in reducing attention problems, limited studies have assessed whether this intervention could be utilised for those with IDD. This study aimed to assess the efficacy of an attention training program in children with IDD in improving core attention skills, problem behaviours and academic outcomes (literacy and numeracy). Design/Methods: In a double blind randomised controlled trial, 76 children (IQ<75) aged 4 to 11 years were assigned to an adaptive attention training program (TALI) or a non-adaptive control program. Both programs were completed at home over a 5 week period and consisted of 25 sessions, each of 20 minutes duration. Attention, academic skills and problem behaviours were assessed at pre-training, post-training, and a 3 month follow-up. Results: Children in the attention training condition showed greater improvement in selective attention performance compared to children in the control condition. These improvements were maintained 3 months after training had ceased. The attention training program was not effective in promoting improvements in other aspects of attention, problem behaviours or academic skills at post-training; however at follow-up the training group showed greater improvements in numeracy skills. Conclusions: The findings suggest that although attention training may enhance some aspects of attention and learning in children with IDD further refinement of the intervention is needed to promote larger, more global improvements.

Speaker: Megan M Spencer-Smith, Jon Quach, Gehan Roberts, Fiona Mensah, Susan Gathercole, Melissa Wake, Peter J Anderson
Title: Individual Differences in Response to Working Memory Training: A Population-Based Randomised Controlled Trial
Abstract: Objective: Memory Maestros was the first population-based randomised trial of a working memory training program in children 6-7 years screened as having low working memory. A select benefit to verbal working memory of training vs usual teaching at 6 months post-randomisation diminished at 12 and 24 months, and did not transfer to academic performance. Given heterogeneity in at-risk populations, it is important to understand if subgroups might benefit more from training than others. This study aimed to examine individual differences in response to working memory training. Design/Methods: Design: Population-based RCT. Setting, Participants: All consenting Grade 1 students (6-7 years, n=1801) from 44 randomly selected primary schools in Melbourne, Australia were screened, with 452 assessed as having low working memory and enrolled in the trial. Intervention: Cogmed RM program involving 20-25 training sessions of 45 minutes over 5-7 weeks in small groups out of class. Control: Usual teaching. Outcomes: Individual change in verbal and visuospatial working memory standard scores from baseline to 6-months, categorised as improved (>1SD normative mean), stable (within 1SD) or declined (<1SD). Predictors, Moderators: Child sex, baseline inattention and health status. Results: In the intervention and control groups, most showed stable working memory (52-59% and 50-62%, respectively) a subgroup improved (27-46% and 35-40%) and a small number declined (3-11% and 5-7%); no differences in rates were evident between groups. None of the child factors significantly predicted improved working memory, nor was there evidence for interaction between intervention and any child factor for improved working memory. Conclusions: Child sex, baseline inattention and health status were not sensitive markers of whether a child improved after training, or benefitted more from training out of class or continuing usual teaching.

Date and time: Tuesday 13:30
Room: Veiling
Format and topic: Symposium, Memory
Coordinator: Andrea Greve
Title: The role of expectancy in episodic memory encoding
Symposium abstract: Episodic memories are rarely created in isolation but encoded in relation to one's contextual expectations and prior knowledge of an event. It is widely established that episodic memory is facilitated when novel information is congruent with our expectations. Recent work, however, also proposes superior memory for novel information that conflicts with pre-existing knowledge, i.e. causing a prediction error. This symposium focuses on the role of expectancy and prediction errors by presenting four talks that investigate the mechanisms by which prior expected/unexpected information modulates episodic learning. We start by showcasing evidence that violations of expectations directly influence episodic memory, in line with the 'Predictive Interactive Multiple Memory Signals' (PIMMS) framework. We then proceed to investigations that demonstrate expected events also improve episodic memory but only under specific circumstances, consistent with the 'Schema-Linked Interactions between Medial prefrontal and Medial temporal region' (SLIMM) framework. Furthermore, we present behavioural and pupillometry data to establish the success of different techniques to activate relevant knowledge in a way that improves memory performance. Finally, we turn to the neural mechanism supporting the integration of new experiences into existing knowledge structures and present electrophysiological evidence that early stages of episodic memory formation are modulated by the congruency between an event and prior knowledge. Taken together, this symposium provides recent insight into understanding the mnemonic role of expectancy. Collectively, the presentations outline theoretical approaches and future opportunities for how to directly link prior expectations to episodic learning.

Speaker: Andrea Greve, Elisa Cooper, Alex Kaula, Rik Henson
Title: Improved associative memory when expectations are violated
Abstract: I will describe three behavioural experiments that show new associations between a scene and an object are better remembered the more strongly they violate expectations established by prior presentations of those scenes and/or objects. This supports the Predictive Interactive Multiple Memory Signals (PIMMS) framework, in which memory encoding is driven by Prediction Error, which reflects the divergence between two probability distributions: one reflecting the prior probability (from previous experiences) and the other reflecting the sensory evidence (from the current experience). Prediction Error is therefore a useful concept not only in learning theories, but also one-shot declarative memory.

Speaker: Andrea Greve, Elisa Cooper, Roni Tibon, Rik Henson
Title: Improved associative memory when expectations are confirmed
Abstract: Encountering an episode which is highly congruent with our expectation, i.e. conforming to existing world knowledge or schemas, is known to improve subsequent memory for this event. Such memory advantages have been widely established in the past, but recent work also proposes improved memory for novel information that conflicts with pre-existing knowledge. Here we present three behavioural experiments to examine this seemingly paradoxical claim by investigating the mechanisms by which expected, unexpected and unrelated information influences episodic memory. Our data reveal superior performance for expected and unexpected compared to unrelated information, in line with the SLIMM ('Schema-Linked Interactions between Medial prefrontal and Medial temporal region') framework. We furthermore illustrate that the mnemonic advantage occurs for central but not peripheral details of expected information, confirming a prediction derived from SLIMM. Overall, we provide supporting evidence that expectancies and confirmatory or violating evidence play a central but distinct role in episodic memory. We will present an overall discussion of our findings and consider their broader theoretical implications.

Speaker: Garvin Brod, Marcus Hasselhorn, Silvia Bunge
Title: Does making a prediction improve memory?
Abstract: It is well known that activating prior knowledge during memory encoding improves memory performance. But what are simple techniques to reliably activate knowledge? In two experiments, we tested whether asking participants to make a prediction about a specific outcome is a viable technique to activate relevant prior knowledge in a way that improves memory performance. Experiment 1 examined how making a prediction as compared to post-hoc evaluation of the result of a soccer match affects episodic memory for results that conform or violate expectancies. We hypothesized that making a prediction will particularly benefit memory for events that violate expectancies, because making a prediction beforehand should yield a larger prediction error for the surprising outcome. We aimed to measure prediction error during memory encoding using the pupillary response to the presentation of the outcome. Initial behavioral results support our hypothesis that asking participants to make a prediction improves their otherwise poor memory for expectancy-violating events, as compared with post-hoc evaluation. Corresponding pupillometry data will be discussed. Experiment 2 extended the first experiment by testing whether asking participants to make a prediction also benefits the formation of conceptual knowledge structures. Gains in conceptual knowledge were assessed via a repeated hierarchy test that gauged participants' knowledge about population sizes of European countries. The study phase matched the one of Experiment 1, again comparing prediction and post-hoc evaluation. Initial results indicate that asking participants to make a prediction facilitated their learning of a conceptual knowledge structure. Across both experiments, then, making a prediction appears to boost learning. Thus, asking participants to make a prediction could be a viable technique to activate prior knowledge and improve memory performance.

Speaker: Lluis Fuentemilla, B Nicolas, A Rodriguez-Fornells, N Bunzeck, R de Diego-Balaguer, P Packard
Title: The encoding-dependent nature of memory formation for congruent and incongruent events
Abstract: As the stream of experience unfolds, our memory system rapidly transforms current inputs into long-lasting meaningful memories. A putative neural mechanism that strongly influences how input elements are transformed into meaningful memory codes relies on the ability to integrate them with existing structures of knowledge or schemas. However, it is not yet clear whether schema-related integration neural mechanisms occur during online encoding. Here, we present data from two studies including behavioral and electroencephalographic (EEG) data that examined the encoding-dependent nature of this phenomena in humans. In the first study, we show that an accelerated onset of the neural mechanisms supports enhanced memory for words with congruent semantic information but it also increases false recall for semantically related words in a later memory test. In the second study, we used a life-like sequence of picture events and showed a dissociated involvement of neural oscillatory responses to successful memory encoding of items that were congruent or incongruent with event contextual sequence. Altogether, our results show how meaning-based congruence between an event and previous information can influence processes during the earliest stages of episodic memory formation.

Date and time: Tuesday 13:30
Room: Verwey
Format and topic: Ask-the-experts, Cognitive modeling
Coordinator: Eliana Vassena
Title: Towards a theory of Prefrontal Cortex: Throwing out the babies with the bathwater?
Symposium abstract: Prefrontal Cortex (PFC) is one of the most studied regions of the brain. After a century of cross-discipline investigation, it is widely recognized as the locus of high-level cognitive processes. Paradoxically, the field is struggling to achieve a consensus on the neural implementation of these functions. The goal of this panel is to critically review the state-of-the-art, and set out a future agenda for the neuroscientist investigating PFC. First, we will discuss the importance of functional localization, from a hundred years of neuropsychological findings to more recent neuroimaging fallacies (eg. the "cingularity", with countless functions attributed to the Anterior Cingulate Cortex). Second, we will discuss the role of anatomical heterogeneity. Can lesion studies (diseased brain) and comparative studies (monkey brain) inform research on neurofunctional architecture of PFC in humans? Third, we will discuss how functional localization relates to interregional connectivity. Can information flow across regions unveil how representations are coded? Can connectomics provide the missing link between localized function and network-level activity? Fourth, animal and human data inspired computational models. Do current models of PFC satisfy principles of parsimony and falsifiability? Is it meaningful to look for a unifying computational principle or is this inconsistent with anatomical and functional constraints?

Speaker: Lesley Fellows
Title: .
Abstract: 'Speaker 1 studies the role of specific prefrontal sub-regions in component processes of decision-making and executive function. She is a cognitive neurologist with expertise in lesion methods in humans. She believes that robust cognitive neuroscience models will emerge from synthesis of evidence from converging methods in humans, and across species. Loss-of-function approaches are especially critical in a field largely dominated by correlational methods, and in newer research areas, such as social, affective and decision neuroscience, all of which implicate PFC.'

Speaker: William Alexander
Title: .
Abstract: 'Speaker 2 investigates the computational and neural processes underlying cognitive control and decision-making. Interconnected regions in prefrontal cortex (PFC) are frequently activated across a broad range of tasks, leading to a proliferation of competing accounts regarding their roles. By integrating empirical findings across levels of description and experimental contexts, Speaker 2 attempts to derive general computational principles of PFC function. His recent work suggests that a framework based on error computation and processing can account for a wide array of empirical effects observed in the frontal lobes.'

Speaker: Rogier Mars
Title: .
Abstract: 'Speaker 3 works on the intersection of comparative and cognitive neuroscience. He is interested in how the organization of human and non-human primate brains differs and how this relates to their different behavioural abilities. Recently, he has focused on developing techniques to use connectivity to quantitatively compare the human and macaque monkey brain. This work showed greater than expected similarity in frontal cortex organization between these species, although some intriguing differences were also noted.'

Speaker: Charlie Wilson
Title: .
Abstract: 'Speaker 4 studies how the dynamic interactions of the prefrontal cortex permit cognitive control and learning. He combines chronic neurophysiological recordings with the lesion approach, in monkeys performing cognitive tasks. He seeks to characterise the mechanisms of cortical interaction by studying local field measures, in particular local oscillations and the dynamic interactions between those oscillations in different cortical regions. He then tests how critical these mechanisms are to the task using interventions that selectively interrupt the interactions, using lesions, reversible inactivation, and recently using chemogenetic interventions.'

Date and time: Monday 8:30
Room: Keurs
Format and topic: Hackathon, Hackathons/tutorials
Coordinator: Johannes Fahrenfort
Title: The Amsterdam Decoding And Modeling toolbox (EEG/MEG)
Symposium abstract: Recent years have seen an enormous increase in time-resolved multivariate analysis of EEG/MEG data, but despite its popularity few easy-to-use software packages exist. We recently developed a Matlab toolbox based on standard EEG/MEG input formats, which will be publicly launched during this hackathon. The ADAM toolbox applies backward decoding and forward encoding models to any dataset in standard EEGLAB or Fieldtrip format. The toolbox visualizes multiple-comparison corrected group decoding and forward encoding results in a variety of ways, such as time-by-time generalization matrices, time by frequency matrices, cortical tuning functions and topographical maps of (forward-transformed) classifier weights. In this workshop we apply a backward decoding analysis of a publicly available MEG/EEG dataset involving the perception of familiar, unfamiliar and scrambled faces. We will go over the steps involved in single subject analysis, and show you how to perform and visualize a group-level statistical analysis. Furthermore, we will go into differences and similarities between MEG and EEG results. At the end of the workshop you will be able to (1) visualize and interpret group-level results of a decoding analysis and (2) apply multivariate decoding on your own data. Some knowledge of Matlab is beneficial, but not required to take part in this workshop. If you wish to participate in the workshop using your own laptop, you will need a working Matlab installation (>2012b) (desktop computers will not be available).

Date and time: Sunday 8:30
Room: Keurs
Format and topic: Hackathon, MRI methods
Coordinator: Steven Miletic
Title: A tutorial on multivoxel pattern analysis (MVPA) of fMRI data in Python
Symposium abstract: Stochastic models of neurons exhibit much more biologically plausible dynamics compared to their deterministic counterparts, but they are also equally more difficult and computationally expensive to simulate. In this hackathon we explore the Sim.DiffProc package for R by Guidoum and Boukhetala (2016) to simulate a number of stochastic differential equations and to extract key statistics like the first passage time. At the end of this hackathon, the participants will be able to simulate neuronal models, with exact or approximate numerical schemes, and to estimate parameters of a few stochastic neuronal models. Participants should have basic prior knowledge about programming and should have RStudio installed on their laptops at the beginning of the hackathon.

Date and time: Tuesday 8:30
Room: Keurs
Format and topic: Hackathon, Hackathons/tutorials
Coordinator: Jan Hendrik Kirchner
Title: Stochastic modelling for neurosciences in R
Symposium abstract: Stochastic models of neurons exhibit much more biologically plausible dynamics compared to their deterministic counterparts, but they are also equally more difficult and computationally expensive to simulate. In this hackathon we explore the Sim.DiffProc package for R by Guidoum and Boukhetala (2016) to simulate a number of stochastic differential equations and to extract key statistics like the first passage time. At the end of this hackathon, the participants will be able to simulate neuronal models, with exact or approximate numerical schemes, and to estimate parameters of a few stochastic neuronal models. Participants should have basic prior knowledge about programming and should have RStudio installed on their laptops at the beginning of the hackathon.

We are also happy to announce the following hackathons:

Date and time: Sunday 8:30
Room: Keurs
Format and topic: Hackathon, MRI methods
Coordinator: Steven Miletic
Title: A tutorial on multivoxel pattern analysis (MVPA) of fMRI data in Python
Symposium abstract: This tutorial/hackathon introduces participants to the most important aspects in scripting a multivoxel pattern analysis (MVPA) using the Python programming language, with a focus on using tools from the popular 'scikit-learn' package. The tutorial covers (1) data preparation and preprocessing, (2) feature selection and extraction methods, (3) model selection and evaluation, and (4) feature visualization. During the tutorial, participants work through interactive Jupyter Notebooks, which include some explanation of core concepts and exercises, and in which they gradually build a machine learning pipeline on a real fMRI dataset. By the end of the tutorial, participants will be able to use Python, and scikit-learn specifically, to build a complete, cross-validated, and efficient machine learning pipeline. The organizers have extensive experience in applying (multimodal) multivoxel pattern analyses on fMRI data and are the maintainers of the high-level MVPA package 'skbold' ( Additionally, they teach the course 'Neuroimaging: Pattern Analysis', which is part of the research master psychology at the University of Amsterdam.

Date and time: Monday 8:30
Room: Keurs
Format and topic: Hackathon, Hackathons/tutorials
Coordinator: Johannes Fahrenfort
Title: The Amsterdam Decoding And Modeling toolbox (EEG/MEG)
Symposium abstract: Recent years have seen an enormous increase in time-resolved multivariate analysis of EEG/MEG data, but despite its popularity few easy-to-use software packages exist. We recently developed a Matlab toolbox based on standard EEG/MEG input formats, which will be publicly launched during this hackathon. The ADAM toolbox applies backward decoding and forward encoding models to any dataset in standard EEGLAB or Fieldtrip format. The toolbox visualizes multiple-comparison corrected group decoding and forward encoding results in a variety of ways, such as time-by-time generalization matrices, time by frequency matrices, cortical tuning functions and topographical maps of (forward-transformed) classifier weights. In this workshop we apply a backward decoding analysis of a publicly available MEG/EEG dataset involving the perception of familiar, unfamiliar and scrambled faces. We will go over the steps involved in single subject analysis, and show you how to perform and visualize a group-level statistical analysis. Furthermore, we will go into differences and similarities between MEG and EEG results. At the end of the workshop you will be able to (1) visualize and interpret group-level results of a decoding analysis and (2) apply multivariate decoding on your own data. Some knowledge of Matlab is beneficial, but not required to take part in this workshop.

Date and time: Tuesday 8:30
Room: Keurs
Format and topic: Hackathon, Hackathons/tutorials
Coordinator: Jan Hendrik Kirchner
Title: Stochastic modelling for neurosciences in R
Symposium abstract: Stochastic models of neurons exhibit much more biologically plausible dynamics compared to their deterministic counterparts, but they are also equally more difficult and computationally expensive to simulate. In this hackathon we explore the Sim.DiffProc package for R by Guidoum and Boukhetala (2016) to simulate a number of stochastic differential equations and to extract key statistics like the first passage time. At the end of this hackathon, the participants will be able to simulate neuronal models, with exact or approximate numerical schemes, and to estimate parameters of a few stochastic neuronal models. Participants should have basic prior knowledge about programming and should have RStudio installed on their laptops at the beginning of the hackathon.

We are grateful for the generous support of:

Click to go to RR website!

For questions about sponsorships, contact

Search the ICON program database

To search through the program, type in the search fields below. Note that search terms are not case-sensitive and are combined using AND (no OR option). Note also that posters are not yet added to the database.

Search author names:
Select topic from list:

Search the titles:

Select presentation date from list:

Search the abstracts:

Select presentation type from list:

Day/time Location Type Topic Authors Title and abstract
Saturday 12.00 - 13.30 Grote zaal Poster #1 Neural modeling Jan Hendrik Kirchner
All authors & affiliationJan Hendrik Kirchner, Institute of Cognitive Science, University of Osnabruck, Wachsbleiche 27, Osnabruck, Germany, Bernstein Center Freiburg, Albert-Ludwigs Universitaet, Hansastrasse 9a, Freiburg, Germany
Large Deviation Theory for Parameter Estimation in Simple Neuron Models
Show/hide abstractTo investigate the complex dynamics of a biological neuron that is subject to small random perturbations we can use stochastic neuron models. While many techniques have already been developed to study properties of such models [1], especially the analysis of the (expected) first-passage time or (E)FPT remains difficult. In this work I applied the large deviation theory (LDT), which is already well-established in physics and finance [2], to the problem of determining the EFPT of the mean-reverting Ornstein-Uhlenbeck (OU) process. The OU process instantiates the Stochastic Leaky Integrate and Fire model and thus serves as an example of a biologically inspired mathematical neuron model. I derived several classical results using much simpler mathematics than the original publications [3] from neuroscience and I provided a few conceivable interpretations and perspectives on these derivations. Using these results I explored some possible applications for parameter estimation and I provided an additional mathematical justification for using a Poisson process as a small-noise approximation of the full model. Finally I performed several simulations to verify these results and to reveal systematic biases of this estimator.
Saturday 12.00 - 13.30 Grote zaal Poster #2 Sensory processing Talia Brandman
All authors & affiliationTalia Brandman, University of Trento; Chiara Avancini, University of Cambridge; Olga Leticevscaia, University of Trento; Marius Peelen, University of Trento.
Effects of auditory and semantic cues on the neural dynamics of visual object categorization
Show/hide abstractRecent electrophysiological studies have shown that object category is represented in the multivariate response pattern across the scalp, thereby probing the neural dynamics of visual object representation. Particularly, decoding of intact object category from magnetoencephalography (MEG) signal was found to peak between 100 to 300 ms after visual onset (Carlson et al., 2013; Cichy et al., 2014). Here, we asked how extra-visual information such as auditory and semantic cues would affect the neural dynamics of visual object categorization. To this end, we used a novel approach in MEG, in which we degraded object resolution in order to boost the facilitating effect of external cues. Participants viewed degraded objects or their mean-luminance equivalent. Each image was preceded by the onset of an auditory cue: a word, a sound or broadband noise. We used cross-decoding classification of object category (animate/inanimate) to compare the multivariate response patterns evoked by intact objects to those evoked by the main-experiment stimuli. Results revealed better classifier prediction of object category for degraded objects with informative auditory cues (sound, word) than with noise. Decoding was at chance for uniform gray images preceded by informative sounds, indicating that the sounds alone were not enough to evoke representations of object category that resembled those of intact objects. These results demonstrate an effect of auditory-driven facilitation of visual object representations in the MEG signal. This suggests that when visual detail is insufficient, extra-visual information may influence the neural dynamics of visual object representations.
Saturday 12.00 - 13.30 Grote zaal Poster #3 Social behavior Andrew Martin
All authors & affiliationAndrew Martin, University of Queensland * Peter Su, University of Queensland Marcus Meinzer, University of Queensland
Cultural background influences tDCS effects on social cognition
Show/hide abstractBackground: Social cognition is influenced by cultural upbringing as demonstrated across several behavioral and functional neuroimaging studies. However, no study has addressed differences in response to electrical brain stimulation between cultural groups. Method: 40 healthy participants (20 Australian "Westerners" and 20 Singaporean "East Asians", aged 18-35 yrs) completed a battery of sociocognitive tasks in a within-subjects, crossover, sham-controlled study. They received 1mA anodal stimulation for 20 minutes or sham stimulation over the dorsomedial prefrontal cortex (dmPFC) using high-definition tDCS (HD-tDCS) while completing three sociocognitive tests: 1) a low-level visual perspective taking task from the perspective of the self or other (VPT), 2) the Reading the Mind in the Eyes Test (RMET), and 3) an episodic memory task relying on either self or other encoded memories for mental attributes. Results: Baseline differences were identified on the VPT task with the East Asians integrating allocentric (external) information into the egocentric (self) to a greater extent than the Westerners. However, following anodal-tDCS to the dmPFC, the Westerners integrated allocentric perspectives into the egocentric perspective to the same extent as the East Asians. In the episodic memory task, anodal-tDCS removed the bias towards self-encoded memories in both groups. Additionally, in the East Asian group, overall memory was significantly impaired, whereas there was no overall memory cost in the Western group. Conclusion: Cultural influences across development result in differences in social cognition and influence the effect of transcranial direct current stimulation. Underlying differences in neural processes associated with social cognition may explain results and could be explored using simultaneous fMRI-tDCS. Future studies employing electrical brain stimulation to influence cognition need to consider cultural background.
Saturday 12.00 - 13.30 Grote zaal Poster #4 Language Stefan Frank
All authors & affiliationStefan Frank, Radboud University Jin-Biao Yang, NYU Shanghai
Lexical representation explains cortical entrainment during speech perception
Show/hide abstractThere is considerable debate on the precise role of hierarchical syntactic structure in sentence comprehension. Ding et al. (2016) presented evidence that cortical entrainment during speech perception reflects the neural tracking of hierarchical syntactic structure of simple sentences, which supports the view that hierarchical processing is inevitable. However, we demonstrate that the same results follow from a simple computational model that represents only lexical-level information. Ding et al.'s participants listened to syllable sequences at a fixed rate. Depending on experimental condition, each four-syllable subsequence could be analyzed hierarchically as containing linguistic units at one or two higher levels, or lacking meaningful structure. The resulting MEG signal showed spectral power peaks at precisely the presentation rates of linguistic units at the levels present in the analysis. Our model represents the same stimuli as sequences of high-dimensional vectors, assigned by a distributional semantics model. Words that share syntactic/semantic properties are encoded by similar vectors. Consequently, lexical properties that occur at a fixed rate in the stimulus sequence correspond to a recurring approximate numerical pattern in the time-series of vectors. For example, one of the conditions involved sentences like ‘dry fur rubs skin', with every second word a noun and every forth word a verb. Words from the same syntactic category tend to be represented by similar vectors, resulting in model-predicted spectral power peaks at exactly the occurrence rates of two-word phrases and four-word sentences. The model also correctly predicts power spectral in the other experimental conditions. Crucially, vectors represent only lexical information and the spectral power analysis is applied over a sequence of vectors that is not processed, integrated, or interpreted. Hence, the model shows how the cortical entrainment results can be explained by the stimuli's lexical properties, without recourse to syntactic structures or integrative processes.
Saturday 12.00 - 13.30 Grote zaal Poster #5 Sensory processing Mariya Manahova
All authors & affiliationMariya E. Manahova, Donders Institute, Radboud University, Nijmegen, Netherlands Pim Mostert, Donders Institute, Radboud University, Nijmegen, Netherlands Peter Kok, Department of Psychology, Princeton University, United States Jan-Mathijs Schoffelen, Donders Institute, Radboud University, Nijmegen, Netherlands Floris P. de Lange, Donders Institute, Radboud University, Nijmegen, Netherlands
Stimulus familiarity and expectation jointly suppress neural activity in lateral occipital cortex (LOC)
Show/hide abstractPrior knowledge about the visual world can change how a visual stimulus is processed. Two important types of knowledge are stimulus familiarity (i.e., whether a stimulus has been seen before) and stimulus expectation (i.e., whether a stimulus is expected to occur, based on the context). Neurophysiological studies in monkeys have shown suppression of spiking activity for both expected (Meyer and Olson, 2011) and for familiar (Li, Miller, and Desimone, 1993) items in object-selective inferotemporal cortex (IT). It is an open question, however, if and how these types of knowledge interact in modulating the sensory response. In order to address this issue, and to examine whether previous findings generalize to non-invasively measured neural activity in humans, we designed an experiment in which we separately manipulated familiarity and expectation, and we used magnetoencephalography (MEG) to record human brain activity. We found a modulation of neural activity specifically in the lateral occipital complex (LOC), the putative human homologue of monkey IT (Orban, van Essen, and Vanduffel, 2004) by both familiarity and expectation, such that familiar and expected stimuli showed strongest activity suppression. These results show that distinct types of sensory knowledge jointly determine the amount of neural resources dedicated to stimulus processing, and generalize previous findings from macaque to human vision.
Saturday 12.00 - 13.30 Grote zaal Poster #6 Decision-making Andrew Reid
All authors & affiliationMarcel van Gerven, Donders Centre for Cognition, Radboud University Nijmegen
A naturalistic driving simulation paradigm for probing cognitive function
Show/hide abstractCognitive science commonly employs well-constrained experimental paradigms with simplistic, fixation-bound stimuli. Such designs are advantageous because they allow researchers to test straightforward models of cognition, unencumbered by confounds such as eye movements, luminance changes, and environmental complexity. A potential drawback, however, is that such stimuli are difficult to generalize to human cognition as it occurs in complex, naturalistic settings. In particular, the function of neuromodulatory systems, such as norepinephrine and dopamine, is likely to be more robust under the continuous, unconstrained environmental conditions under which they evolved. As an attempt to investigate neuromodulatory processes under more realistic conditions, we have developed a novel 3D highway driving simulator, designed specifically for testing hypotheses about cognitive function in a continuous, naturalistic way. Our software is flexible and easy to customize, and provides a systematic way to evaluate variables in an event-driven manner. We are currently applying this simulation to assess the influence of norepinephrine in reinforcement learning, sequential decision making, and working memory, recording behavioural events simultaneous with eye tracking and EEG acquisition. This poster is intended to describe our Java-based driving simulator, including its XML-format specification that can be used to flexibly control the flow and behaviour of traffic. Furthermore, we will discuss the analytic challenges of this continuous approach, and present novel solutions for the analysis of observations made under this paradigm. In particular, we suggest the use of an event-based analysis, where subject-derived events such as saccades and button presses signifying the execution of a decision can be modeled [1], classified, clustered, and statistically analyzed. Finally, the confound of non-equiluminance will be discussed, along with potential ways to address this issue.
Saturday 12.00 - 13.30 Grote zaal Poster #7 Sensory processing Edden Gerber
All authors & affiliationEdden M. Gerber, The Edmond and Lily Safra Center for Brain Sciences, The Hebrew University of Jerusalem, Israel Leon Y. Deouell, the Edmond and Lily Safra Center for Brain Sciences and Department of Psychology, The Hebrew University of Jerusalem, Israel
Cortical representation of persistent visual stimuli: from the timescale of electrophysiology to the timescale of fMRI
Show/hide abstractWhen we gaze at a static visual image for a long duration (e.g. several seconds), will the same neuronal populations that responded to the stimulus onset continue to be active for the entire stimulus duration -- or will a different neural representation emerge beyond the initial, onset-driven response? In a previous study using intracranial recordings in humans (iEEG), we found that the fidelity of sustained visual activity gradually decreased across the ventral stream. That is, the duration of the stimulus could be accurately decoded from single-trial responses in early visual cortex but not in anterior, category-selective regions. In the current study, we used fMRI to examine this question and generalize it from the sparse anatomical sampling of iEEG to the whole cortex, and from brief (0.3-1.5 sec) to longer stimulus durations. Fifteen participants were shown grayscale images of faces or houses, with each stimulus presented for either 9 or 15 seconds (34.5 sec SOA), while performing a simple visual vigilance task, using a slow event-related design in a 3T scanner. Seemingly in contradiction to our iEEG findings, we found robust sustained responses across the entire ventral stream hierarchy and no evidence for an anatomical gradient of decreasing sustained response fidelity. The two sets of results can be reconciled by taking into account the different temporal resolutions of BOLD signal and iEEG. Sporadically-bursting neural activity, which is not informative about the presence of the stimulus on a sub-second timescale, can still be informative about the stimulus time course with a coarser resolution such as that of the hemodynamic response, which serves as a low-pass filter over the population activity. The combined findings raise new questions about which temporal resolution is relevant for subjective perception.
Saturday 12.00 - 13.30 Grote zaal Poster #8 Memory Marcin Leszczynski
All authors & affiliationMarcin Leszczynski, Department of Neurological Surgery, Columbia University Medical Center, New York, USA Juergen Fell, Department of Epileptology, University of Bonn, Bonn, Germany Ole Jensen, Centre for Human Brain Health, University of Birmingham, Birmingham, UK Nikolai Axmacher, Department of Neuropsychology, Ruhr University Bochum, Bochum, Germany
Alpha activity in the ventral and dorsal visual stream controls information flow during working memory
Show/hide abstractSuccessful performance in a working memory (WM) task requires both maintenance of relevant and inhibition of irrelevant information. Although considerable evidence indicates that the ventral and dorsal visual stream (VVS and DVS, respectively) play a critical role in maintenance of item identity and spatial information, the electrophysiological mechanisms underlying these processes remain elusive. Here we used electrocorticography recordings from VVS and DVS and from the prefrontal cortex in epilepsy patients while they were performing a delayed match-to-sample task. We studied mechanisms involving alpha oscillations supporting WM at the local and global network level because previous studies in humans and nonhuman primates suggested a role of alpha in functionally inhibiting task-irrelevant information. The experimental conditions (face identity vs. orientation) were designed to engage selectively either the VVS or DVS depending on task instructions. We observed a double dissociation between alpha power reductions in the ventral and dorsal visual stream. Alpha power was reduced in the VVS during maintenance of face identity and in the DVS during maintenance of spatial orientation of the very same stimuli. The phase of alpha oscillations modulated broadband high frequency gamma activity in both regions. Interestingly, gamma activity occurred during more diverse alpha phase ranges when task-relevant information was maintained, potentially reflecting longer excitable "duty cycles". Furthermore, the prefrontal cortex showed stronger alpha phase-based connectivity with the area relevant for maintenance of task-specific features. Our findings support a model in which the VVS and DVS is recruited by the prefrontal cortex via selective reduction of alpha power. As a result, excitable duty cycles in the two visual areas are extended. Furthermore, these results suggest that despite possible physiological differences in the generation of alpha oscillations in the VVS and DVS, the functional role of that rhythm is preserved across cortical areas.
Saturday 12.00 - 13.30 Grote zaal Poster #9 Clinical Yishul Wei
All authors & affiliationYishul Wei, Netherlands Institute for Neuroscience* Michele A Colombo, Netherlands Institute for Neuroscience Jennifer R Ramautar, Netherlands Institute for Neuroscience Ysbrand D van der Werf, VU University Medical Center Eus JW Van Someren, Netherlands Institute for Neuroscience and VU University Medical Center
Deconstructing the Eyes-Closed Resting State in Insomnia Disorder with High-Density Electroencephalography
Show/hide abstractObjective: People with Insomnia Disorder (ID) not only have sleep problems, but also complain about daytime functioning. It has been shown that neuropsychological processes during the eyes-closed wake resting state influence subsequent sleep onset latency and sleep quality. Focusing on pre-sleep eyes-closed wakefulness might therefore give insight into the pathophysiology of ID. The present study aimed at characterizing the neurophysiological profile of ID during evening eyes-closed wakefulness at a fine-grained level by utilizing the high spatiotemporal resolution of high-density electroencephalography (HD-EEG). Methods: 256-channel HD-EEG supplemented with lead-II electrocardiography (ECG) was recorded from 32 people with ID and 32 age- and sex-matched controls during evening wakeful rest of 5-min duration with eyes closed. The relative power of each frequency from 0.5 to 45 Hz in each scalp channel was estimated by Welch's method. Heartbeat-evoked potential (HEP) was obtained by averaging EEG segments time-locked to ECG R-waves. Duration and occurrence indices of four microstate classes were computed through temporal clustering of momentary EEG topographies. All derived measures were compared between groups by means of nonparametric statistics. Results: As compared to controls, people with ID showed widespread increased relative power within the beta frequency range, centered at 19 Hz. The late HEP amplitude 376--500 ms after the R-wave was larger in ID at the frontal region. The durations of microstate classes B and C were reduced in ID, while the occurrence of microstate class D was increased. Conclusion: With reference to previous links between each marker and psychological processes in the literature, the findings suggest that people with ID are characterized by heightened arousal, hypersensitivity to interoceptive inputs, and deficits in attention. Future works can probe the same processes in ID during sleep by extending the analyses to sleep EEG, or explore neurophysiology-based subtyping of ID using these markers.
Saturday 12.00 - 13.30 Grote zaal Poster #10 Sensory processing Jong-rok Do
All authors & affiliationJong-rok Do, Center for Cognition and Sociality, Institute for Basic Science Kang Yong Eo, Center for Cognition and Sociality, Institute for Basic Science Yee-Joon Kim, Center for Cognition and Sociality, Institute for Basic Science
Effect of stimulus variability on the mean computation of sequentially presented objects: An EEG study
Show/hide abstractPrior behavioral studies showed that human can extract statistical summaries of sequentially delivered objects. However, little is known about neural mechanisms supporting such statistical perception. Here we recorded EEG to investigate how stimulus variability influences the mean computation of sequentially presented objects. Observers viewed a sequence of ten randomly oriented Gabors presented centrally every 600 ms, and reported the orientation of their mean by adjusting a red probe bar, preceded by a blank period. Each sequence had either small or large orientation variance. In both variance conditions, observers put more weights on later items, consistent with a recency effect demonstrated in previous studies. However, time-frequency analysis revealed that stimulus-encoding steady-state visual evoked potentials (ssVEPs) were stronger throughout the time period of sequence in small variance condition than in large variance condition. The sequential linear mixed-effects modeling also showed that ssVEP amplitude locked to the last item was positively correlated with perceptual error size in large variance condition, whereas ssVEP amplitudes locked to items in the second half of the sequence were negatively correlated with perceptual error size in small variance condition. These results suggest that the recency effect arises differentially depending on stimulus variability. When observers have difficulty in integrating more variable items, the last item contributes most to the mean computation leading to a poor mean representation. However, when observers can easily integrate less variable items, the last item just confirms their already computed mean representation leading to the enhanced precision. Furthermore, we found that posterior alpha-band power as well as parieto-occipital alpha-band phase synchronization was stronger during blank period in small variance condition than in large variance condition. This suggests that alpha-band activities and their functional connectivity between occipital and parietal area can used as an index of the degree of integrating information.
Saturday 12.00 - 13.30 Grote zaal Poster #11 Attention Jennifer Coull
All authors & affiliationJennifer T Coull, Aix-Marseille University & CNRS, France * David Phrakornkham, Aix-Marseille University, France Inga Korolczuk, Jagiellonian University, Krakow, Poland Franck Vidal, Aix-Marseille University, France
When making predictions in time, left parietal cortex activity reflects engagement of the temporal prior
Show/hide abstractBeing able to predict when relevant events are likely to occur improves how quickly and accurately they are processed. Prior fMRI and EEG investigations have shown that the behavioral benefits of temporal predictability consistently implicate left-lateralised inferior parietal cortex in a variety of task contexts. However, the precise functional role of left parietal cortex in temporal predictability remains unknown. Is it involved in retrieving a stored representation of the predicted time prior to the event's appearance or, alternatively, in enhancing sensorimotor processing of the event when it appears at the predicted time? To address this question, we used fMRI to measure regional brain activity during performance of the temporal orienting of attention task. In this cued reaction time task, a visual cue either predicts (temporal cue) or not (neutral cue) the time at which an imperative target will appear (i.e. predicts target onset). We adapted this paradigm so that cues instead predicted the moment of target offset. This allowed us to include a long and variable interval between the cue and the target, meaning regional brain activity could be measured separately at the cue and target stages of the task. Behavioural analyses confirmed that participants made effective use of the temporal cue to predict target offset. fMRI analyses indicated that left inferior parietal cortex was activated by temporal predictability at the cue, rather than target, stage of the task. These data suggest that left inferior parietal cortex is activated when participants access a temporal template stored in memory (the "prior") to make temporal predictions. More generally, our results are consistent with an attentional framework in which left parietal cortex is the source of a top-down signal, which biases activity in task-specific brain regions at predicted moments in time.
Saturday 12.00 - 13.30 Grote zaal Poster #12 Clinical Elisa Ruohonen
All authors & affiliationPiia Astikainen, Department of psychology, University of Jyvaskyla Veera Alhainen, Department of psychology, University of Jyvaskyla
Show/hide abstractThere is an attentional bias in processing of negatively valenced stimuli in depression. However, it is not clear if pre-attentive processing is anomalous in depression and, if so, whether it reflects trait- or state-related impairment. To study this, we compared event-related potentials (ERPs) to changes in facial expressions in depressed (n = 28) and control (n = 22) participants. For the depressed group, the measurement was conducted three times: first measurement was conducted when all depressed participants had an acute episode of depression, the second 2 months later when some of the participants had received a brief psychological intervention for depression, and the third 36 months after the first measurement when all had received the intervention. In the depression group, depression was reduced from moderate to minimal depression comparing first and second/third measurement. The control group was measured once. The face pictures were presented in an oddball condition in two stimulus blocks where sad or happy faces were interspersed with frequently presented neutral faces. Peak values for P1 and N170 were extracted. Sad faces modulated the amplitude of P1 in the depressed but not in the control group. In the depressed group the P1 response amplitude correlated positively with depression symptoms in the first measurement. No group differences were found for N170 amplitude in either stimulus conditions. However, longer N170 latencies to sad than to neutral faces were found only in the depressed group. P1 modulation for sad faces was reduced from first measurement to second and third measurement, indicating that the processing of sad faces had normalized. The results suggest mood-dependent impairment in sad face processing in depression.
Saturday 12.00 - 13.30 Grote zaal Poster #13 Clinical Magnus Liebherr
All authors & affiliationMagnus Liebherr1,2, Patric Schubert1, Stephanie Kersten1, Christina Lutz3, Sarah Stickl1, Franziska Schaefer1, Johanna Moeller1, Christian T. Haas1 1 Hochschule Fresenius, University of Applied Sciences, Faculty of Health and Social Sciences, Institute of Complex Health Research, Idstein, Germany; 2 University of Duisburg-Essen, Department of General Psychology: Cognition, Duisburg, Germany; 3 LUNEX - International University of Health, Exercise and Sports, Faculty of Health Science, Luxembourg
Decision-making under motor demands in people with multiple sclerosis
Show/hide abstractIn this day and age, the simultaneous performance of two tasks must be seen as prerequisite in working life, social context, as well as daily living. However, the ability of successfully handling these situations depends on numerous factors and is adversely affected in people with multiple sclerosis (PwMS) (e.g. D'Esposito et al., 1996; Wajda et al., 2013). The present study aimed at systematically investigate the influence of cognitive task difficulty, the way of presentation as well as additional motor demands in PwMS. Therefore, five PwMS (56.8±6.2 years) and five healthy controls (56.6±6.6 years) were included. Six cognitive tasks with increasing task difficulty, from a simple reaction task to a mnemonic -- dichotomous choice and double inhibition task were presented both acoustically. Based on Lajoie et al. (1993), we chose sitting as non-demanding and standing as more-demanding motor task. Results showed an increasing reaction time from task one to task six in both acoustic (PwMS: task1/sitting: 437.15ms±124.38ms task1/standing: 516.39ms±100.12ms; task6/sitting: 1089.15ms±163.69ms task6/standing: 972.73ms±235.28ms; controls: task1/sitting: 280.76ms±225.24ms task1/standing: 300.25ms±273.85ms; task6/sitting: 626.42ms±587.30ms task6/standing: 604.01ms±507.51ms), and visual task-setting (PwMS: task1/sitting: 471.26ms±73.69ms task1/standing: 540.77ms±107.42ms; task6/sitting: 849.30ms±18.93ms task6/standing: 822.85ms±150.37ms; controls: task1/sitting: 360.06ms±29.85ms task1/standing: 386.55ms±67.59ms; task6/sitting: 791.08ms±130.89ms task6/standing: 788.60ms±200.52ms). While there is a clear differentiation between PwMs and healthy controls, no influence of the motor conditions could be revealed. Regarding the error rate, PwMS and healthy controls showed a similar pattern across all conditions. Future investigations should focus on further, more demanding, motor tasks as well as a larger sample size.
Saturday 12.00 - 13.30 Grote zaal Poster #14 Decision-making Gilles de Hollander
All authors & affiliation
7 Tesla fMRI shows little evidence for functional subdivisions in the subthalamic nucleus
Show/hide abstractThe subthalamic nucleus (STN) is a small subcortical structure, important in speeded decision-making. It is the main target of Deep Brain Stimulation (DBS) in Parkinson's Disease. However, the optimal target region for DBS within the STN is still a matter of debate. In particular, it has been proposed that the STN consists of a "motor", "associative", and "limbic" area and that the "motor" area in the dorsolateral tip of the STN is the preferred target for DBS. However, there is to date no direct evidence in healthy human subjects for a separate function of these different regions. We used an optimized 7 Tesla (T) functional magnetic resonance imaging (fMRI) protocol and individual anatomical masks of the STN, to probe different segments of the STN during a perceptual decision-making task. The task involved a "potential payoff cue", known to involve limbic cortical areas, as well as a difficulty manipulation, known to modulate activity in prefrontal cortex. We tested whether this task differentially activated STN subregions. All parts of the STN showed task-related activity. Furthermore, the ventromedial part of the STN was marginally more activated when a payoff cue versus a neutral cue was presented (BF10=1.68 for right STN; BF10=2.73 for right STN), but this effect was not significantly larger than in other parts of the STN. There were no differences between hard and easy trials. The dorsolateral part of both the left STN (BF10=1.53) and the right STN (BF10=1.35) were marginally correlated with the extent to which subjects modulate their behavior, as estimated by the diffusion decision model. In sum, our data suggest that the STN is involved in perceptual decision-making and in implementing responses biases. However, activity is not modulated by stimulus difficulty and different STN segments show similar activity profiles. Therefore, the debate about functional subbdivisions in the STN is still unresolved.
Saturday 12.00 - 13.30 Grote zaal Poster #15 Clinical Max Keuken
All authors & affiliation
Structural changes of the substantia nigra in Parkinson's disease.
Show/hide abstractParkinson's disease (PD) is a progressive neurodegenerative disorder which is challenging to diagnose in vivo. Using magnetic resonance imaging (MRI) several promising biomarkers, such as changes in volume and shape of the substantia nigra (SN) have been proposed. Notable post mortem changes in PD include an increase of neuroinflammatory processes in the SN and are considered to have a significant impact in SN degeneration in PD. In the present study we investigated several in vivo anatomical features of the human SN associated with PD using ultra-high field 7 Tesla structural MRI. To address the presence of inflammation, an independent post mortem study was conducted. The in vivo MRI results indicate a volumetric increase of the SN in the PD group accompanied by a change in shape at the rostral-ventral border. In addition, the resulting disease-specific probability atlas indicated a caudal shift of the SN in the PD group. The post mortem results confirmed a clear loss of dopaminergic neurons but provided no support for increased neuroinflammatory processes in the SN.
Saturday 12.00 - 13.30 Grote zaal Poster #16 Attention Anna Lena Biel
All authors & affiliationAnna Lena Biel, 1 Department of Psychology, Ludwig-Maximilians-Universitaet Muenchen, Munich, Germany, 2 Graduate School of Systemic Neurosciences, Ludwig-Maximilians-Universitaet Muenchen, Munich, Germany Tamas Minarik, 1 Department of Psychology, Ludwig-Maximilians-Universitaet Muenchen, Munich, Germany Barbara Berger, 1 Department of Psychology, Ludwig-Maximilians-Universitaet Muenchen, Munich, Germany Paul Sauseng, 1 Department of Psychology, Ludwig-Maximilians-Universitaet Muenchen, Munich, Germany, 2 Graduate School of Systemic Neurosciences, Ludwig-Maximilians-Universitaet Muenchen, Munich, Germany
Title Posterior EEG Theta-Gamma Phase Coupling as Signature of Predictive Coding
Show/hide abstractOur visual perception is strongly influenced by our expectancies about incoming sensory information. It is assumed that mental templates of expected sensory input are created that are compared to actual sensory input, which can be matching or not. Concerning the underlying neuronal mechanisms of such predictive coding, cross-frequency phase synchronization between theta and gamma band EEG oscillations in posterior parietal cortex has been proposed to reflect the integration of top-down controlled mental templates with bottom-up visual processing. In this study, we investigated how matching between sensory input and mental templates held in working memory is affected by the certainty about which activated template must be matched. In a visual search paradigm, we compared theta-gamma phase coupling for conditions where participants had to keep either one or multiple target templates in mind for successful search. We find that memory matching appeared as a transient posterior phase synchronization between EEG theta and gamma oscillations in an early time window around 150 ms after search display onset. Our results indicate stronger transient theta-gamma phase coupling over posterior sites contralateral to target location for conditions where one mental template was required than for multiple templates. This is understood to be in line with previous theoretical accounts, lending promising support for a transient posterior theta-gamma phase coupling as a neuronal correlate of matching between incoming sensory information and memory contents from working memory.
Saturday 12.00 - 13.30 Grote zaal Poster #17 Memory Tzu-Han Cheng
All authors & affiliationTzu-Han Cheng, Department of Psychology, National Taiwan University, Taipei, Taiwan* Chun-Hui Li, Department of Psychology, National Taiwan University, Taipei, Taiwan Bo-Cheng Kuo, Department of Psychology, National Taiwan University, Taipei, Taiwan
Expectation for response diffculty influences alpha activity in visual short-term memory
Show/hide abstractVisual short-term memory (VSTM) is an important function that allows us to briefly maintain a subset of relevant information for supporting higher level mental processes and guiding behaviors. Recent studies have revealed that neural oscillations, especially in the alpha band (8-15Hz), serve to suppress irrelevant distractions and to protect the relevant information in VSTM. However, it remains unclear whether expectation for response difficulty can influence VSTM performance and alpha activity. We addressed this issue using electroencephalography (EEG). In this study, participants were instructed to perform a delayed-response task with four levels of difficulty in a predictable context (e.g. they can predict response difficulty in the given block) and an unpredictable context (e.g. response difficulties were intermixed in the given block). We fit a general linear model to the power data on a trial-by-trial basis using the response difficulty level as the regressor for each time point and each frequency (8-15 Hz) of the epoched data for each task context at the individual level. We then tested whether the expectation for response difficulty can modulate the relationship between the strength of alpha activity and the levels of response difficulty at the group level. We showed significant effects during pre-stimulus and maintenance stages of VSTM. For both stages, we observed a stronger alpha activity in the predictable context compared to the unpredictable context. These results suggest that alpha oscillations play a critical role in anticipating upcoming stimuli prior to the encoding and protecting memorized items during maintenance stage. Taken together, our results highlighted the influences of context-driven response expectation on pre-stimulus and maintenance alpha activity in VSTM.
Saturday 12.00 - 13.30 Grote zaal Poster #18 Attention Rodika Sokoliuk
All authors & affiliationDr. Rodika Sokoliuk, School of Psychology, University of Birmingham; Dr. Stephen Mayhew, Centre for Human Brain Health, School of Psychology, University of Birmingham; Dr. Kevin Aquino, Sir Peter Mansfield Imaging Centre, School of Physics and Astronomy, University of Nottingham; Dr. Susan Francis, Sir Peter Mansfield Imaging Centre, School of Physics and Astronomy, University of Nottingham; Dr. Simon Hanslmayr, School of Psychology, University of Birmingham; Dr. Karen Mullinger, Sir Peter Mansfield Imaging Centre, School of Physics and Astronomy, University of Nottingham; Centre for Human Brain Health, School of Psychology, University of Birmingham
Alpha oscillations over sensory-specific areas mediate attentional shift between different targets in a multimodal paradigm
Show/hide abstractA continuous stream of sensory information enters our brain during daily life. Directing our attention facilitates selecting specific and inhibiting processing of unimportant information. Recent work revealed that alpha (8-13Hz) oscillations play an important role in attention, whereby an increase in alpha power leads to inhibition of brain areas that process irrelevant information. Most previous studies focus on the influence of alpha oscillations within one sensory modality, here we investigate the role of alpha oscillations when switching attention between sensory modalities. Is it one global, or several sensory-specific sources of inhibitory alpha power that mediate attention towards one or the other modality? We employed a Posner-type paradigm modulating attention between visual and somatosensory modalities (0/40/60/100%). Each trial started with a visual cue, to indicate the likelihood with which a target appears in each sensory modality followed by an inter-stimulus interval (ISI; 1.3-1.6s) during which attention was maintained before stimulus presentation. Participants had to discriminate between low/high frequency stimuli, ignoring middle frequency distractor stimuli in the other sensory domain. EEG data (128 electrodes) were recorded on 16 healthy participants. Data analyses were performed on the ISI period. Attention modulation effects were source localized using a beamformer. Results showed a decreased mean alpha power during the ISI over sensory brain regions specific to processing the attended stimuli and an increase over regions for the opposing sensory stimulus. Virtual electrodes in sensory-specific regions show a graded increase in alpha power with decreasing attention for the respective modality. In addition, parietal areas implicated in the attention network also appear to show modulations with task difficulty (i.e. 0/100% compare with 40/60% conditions) and warrant further investigation. This suggests that in multimodal attention paradigms, alpha oscillations in sensory-specific areas mediate attention to the respective target, moreover, parietal alpha oscillations seem to mediate task difficulty.
Saturday 12.00 - 13.30 Grote zaal Poster #19 Memory Elena Solesio-Jofre
All authors & affiliationElena Solesio-Jofre, Maria Hernandez-Lorca, Elisabet Rodriguez-Alzueta, Luis Carretie; Research Group on Brain, Affect and Cognition at the Psychology Department. University Autonoma of Madrid. Campus Cantoblanco, 28049, Madrid, Spain
Aging effects on the regulation of emotions and their underlying neural networks
Show/hide abstractIntroduction: An age-related increase in the preference for positive over negative information in memory, namely the positivity effect, seems to relate to a better cognitive functioning. However, several studies have challenged reliability of this effect. Objectives: First, to examine whether task demands affect the reliability of the positivity effect. Previous research points to a mitigation of the positivity effect for highly demanding tasks. However, the influence of task demands on this positivity effect, using a single study, still remains unexplored. Second, to examine the neural underpinnings of the positivity effect with electroencephalography (EEG). Only a few EEG studies to date have focused on the study of the positivity effect, and the most of them have used event-related potentials, without further localizing the sources of neural activity. Material and methods: 20 young and 20 older individuals performed a memory task with emotional material that manipulated low and high cognitive demands while they were recorded with a 128-channel EEG system. Behaviourally, we explored the presence of a positivity bias in older compared to young adults for both experimental conditions. Regarding EEG data, we are currently working on source reconstruction analyses with beamformers. Results: Older adults showed an increased positive preference compared to their younger counterparts just in the less demanding condition. Regarding our EEG results, we expect age-related differential networks for both low and higher demanding tasks. For the former task, we also predict such a network to be modulated by the presentation of positive information. Conclusions: Our data provide evidence that cognitive demands modulate the positivity effect in older individuals. We expect such a positivity bias to be associated with a time-varying neural network. These findings are extremely relevant for future interventions with older individuals as we have been able to identify the neural trace of the positivity effect.
Saturday 12.00 - 13.30 Grote zaal Poster #20 MRI methods Xiaoqing Gao
All authors & affiliationXiaoqing Gao, Psychological Sciences Research Institute, Institute of Neuroscience, University of Louvain, Belgium Francesco Gentile, Department of Cognitive Neuroscience, Faculty of Psychology and Neuroscience, Maastricht University, The Netherlands, Bruno Rossion, Psychological Sciences Research Institute, Institute of Neuroscience, University of Louvain, Belgium
Fast Periodic Stimulation (FPS): A highly effective approach in fMRI brain mapping
Show/hide abstractFunctional magnetic resonance imaging (fMRI) is an important technique for human brain mapping. However, current fMRI measurements may suffer from low signal-to-noise ratio (SNR) in detecting high-level neural responses in individual brains, leading to low test-retest reliability in spatial activation maps. Here, we present a highly effective novel fMRI approach to define category-selective brain regions. As in EEG frequency-tagging (Rossion et al., 2015), we presented a large variety of natural images at a fast rate (6 images/sec) throughout the entire experiment to stimulate the visual areas continuously. By introducing transient switches to a target category (faces) at a slow fixed frequency (every 9 secs, i.e., 0.111 Hz), we obtained a periodic differential neural response that directly reflects category selectivity. A model-free Fast Fourier Transform (FFT) of hemodynamic activity in this paradigm achieved a two-fold increase in sensitivity (peak SNR) in comparison to a conventional block-design approach, allowing us to map comprehensive extended face-selective areas including the anterior temporal lobe (ATL) in individual brains. Using diverse natural images, we created contrasts at both lower-level visual properties and higher-level category properties in successive images. While the lower-level contrasts were at random time points, the category contrast would only happen at a fixed frequency by design. Therefore, we effectively eliminated the influence of low-level visual cues and increased the specificity of category-selective response. As a result of high sensitivity and specificity, we achieved high test-retest reliability, which reached the highest values (80-90%) yet reported in this area of research. The power of a model-free fast periodic stimulation (FPS) approach with a slow temporal resolution method opens a new avenue for understanding brain mapping of visual categorization.
Saturday 12.00 - 13.30 Grote zaal Poster #21 Decision-making Leonie Balter
All authors & affiliationLeonie Balter, University of Birmingham & University of Amsterdam Sarah Aldred, University of Birmingham Jos A Bosch, University of Amsterdam Suzanne Higgs, University of Birmingham Jet JCS Veldhuijzen van Zanten, University of Birmingham Jane E Raymond, University of Birmingham
Experimentally induced inflammation decreases the ability to flexibly update reward-seeking behaviour
Show/hide abstractInflammation is a core feature of many chronic health conditions, as well as ageing and obesity; all of which have been associated with impairments in motivated behaviour. Animal studies plausibly demonstrate that inflammation can cause reorientation of the motivational state, but the human evidence has remained largely correlational. In this double-blinded randomised placebo-controlled crossover study salmonella typhoid vaccination (0.025 mg; Typhim Vi, Sanofi Pasteur, UK) was used to experimentally induce acute inflammation (N = 30 healthy males, mean age = 25). In both conditions, participants completed a novel monetary reward-learning task at the time of the peak inflammatory response (6 hours post-injection). This task taps into distinguishable aspects of motivational behaviour, including simple association learning, and behavioural flexibility. The task involved two phases: an acquisition phase wherein participants learned to associate abstract symbols with monetary rewards and punishments, and a contingency-change phase wherein an unannounced shift in reward contingencies required adaptation of action strategies (i.e., cognitive flexibility). During the acquisition phase, learning was clearly demonstrated in both conditions, although inflammation resulted in overall poorer performance (p < .05). Moreover, after an unannounced shift in reward-contingencies, inflammation led to slower adaptation (p < .05). Interestingly, choices during learning that involved simple good/bad options were unaffected by inflammation, whereas choices involving different values with the same valence were impaired. The current results showed, that inflammation selectively affects learning whenever choices are effortful, independent of outcome value. Moreover, we show here for the first time that the behavioural flexibility needed to cope with reward contingency change is also degraded by inflammation.
Saturday 12.00 - 13.30 Grote zaal Poster #22 Decision-making Hause Lin
All authors & affiliationSaunders Blair, University of Toronto Hutcherson Cendri, University of Toronto Michael Inzlicht, University of Toronto
Midfrontal theta and pupil dilation track subjective conflict in value-based decisions
Show/hide abstractPeople often experience decision conflict when choosing between competing options or responses, for example when deciding between a healthy yet bland snack versus an unhealthy but tasty one. Theta band neural oscillations (4--8 Hz) over the midfrontal cortex have been linked to objective conflict processing during cognitive control tasks (e.g., incongruent/high-conflict vs. congruent/low-conflict stimuli on Stroop task). Here we ask whether midfrontal theta also tracks subjective conflict, such as during value-guided decisions. We also test whether theta reflects conflict in a binary fashion or if it is sensitive to parametric changes in conflict. We further ask if pupil dilation—also associated with objective conflict processing—tracks subjective conflict. Subjective conflict was manipulated during an intertemporal choice task by presenting participant-specific model-derived delayed rewards (e.g., $19.23 in 10 days). We observe strongest midfrontal theta power and largest pupil dilation when the immediate and delayed rewards are equally desirable (i.e., most conflicting choice). Crucially, both signals reflect different gradations of conflict: As one reward becomes increasingly more desirable than the other, theta power and pupil dilation decrease. We therefore provide evidence for neurophysiological signals that compare subjective values and reflect indecision. Midfrontal theta and pupil dilation may be relevant for understanding how value-based decision making is implemented neurophysiologically.
Saturday 12.00 - 13.30 Grote zaal Poster #23 Developmental Michael Hahn
All authors & affiliationMichael Hahn, Laboratory for Sleep, Cognition and Consciousness Research, Department of Psychology, Centre for Cognitive Neuroscience, University of Salzburg, Salzburg, Austria Ann-Kathrin Joechner, Laboratory for Sleep, Cognition and Consciousness Research, Department of Psychology, Centre for Cognitive Neuroscience, University of Salzburg, Salzburg, Austria Dominik Heib, Laboratory for Sleep, Cognition and Consciousness Research, Department of Psychology, Centre for Cognitive Neuroscience, University of Salzburg, Salzburg, Austria Manuel Schabus, Laboratory for Sleep, Cognition and Consciousness Research, Department of Psychology, Centre for Cognitive Neuroscience, University of Salzburg, Salzburg, Austria Kerstin Hoedlmoser, Laboratory for Sleep, Cognition and Consciousness Research, Department of Psychology, Centre for Cognitive Neuroscience, University of Salzburg, Salzburg, Austria
Developmental changes of slow wave activity during NREM sleep reflect changes in cognitive abilities across puberty
Show/hide abstractTheoretical background: Puberty is considered as a phase of cortical restructuring due to synaptic pruning (Huttenlocher & Dabholkar, 1997). Developmental trajectories of slow waves during sleep follow the same time course and thus are believed to be an indicator of this restructuring process (Feinberg & Campbell, 2013). Objectives: The goal of this longitudinal approach was to investigate developmental changes of slow wave sleep activity from pre- to post-puberty and whether they are connected to individual differences in memory consolidation and general cognitive abilities. Methods: Sleep polysomnography, declarative learning (word pair association task) and general cognitive ability assessments were performed in 16 healthy subjects (7 male, 9 female) at pre-pubertal age (8-11) and 7 years later at post-pubertal age (14-18 years). Polysomnography was recorded ambulatory during 4 nights per subject (2 nights pre-pubertal, 2 nights post-pubertal). General cognitive abilities were assessed using the Wechsler Intelligence Scale (WISC). Preliminary results: At pre-pubertal age, slow wave amplitudes were higher with steeper slopes as compared to post puberty. In contrast, slow wave lengths increased from pre to post-pubertal age. Interestingly, our data revealed that the decrease of the centro-parietal slow wave slopes was associated with a performance increase in spatial intelligence (WISC), whereas the increase in central slow wave length was related to a boost in declarative memory capacity from pre- to post-pubertal age. Conclusions: Centro-parietal cortical activations are involved in working memory tasks. Thus the developmental changes of slow waves during NREM sleep could reflect the development of a more efficient and effective working memory network across puberty, which might be related to higher cognitive functioning. This study was funded by the Austrian Science Fund (T397-B02, P25000).
Saturday 12.00 - 13.30 Grote zaal Poster #24 Consciousness Andres Canales-Johnson
All authors & affiliationAndres Canales-Johnson, University of Amsterdam Lola Beerendonk, University of Amsterdam Alejandro Ezquerro-Nassar, University of Cambridge Tristan Bekinschtein, University of Cambridge Simon van Gaal, University of Amsterdam
Alertness level differentially modulate behavioural and neural markers of conflict and conflict adaptation.
Show/hide abstractInstantaneous conflict as well as across-trial conflict adaptation, two sub processes of cognitive control, can occur in the absence of conflict awareness. Conflict monitoring thus does not seem to depend on content of consciousness, but it remains unclear at which extent it relies on the level of consciousness. Here we use the transition from wakefulness to sleep as a model for studying the behavioral and neural markers of conflict under different levels of alertness. We used an auditory task in which participants listened to the words 'left' or 'right' presented either to the left or right ear while transitioning towards sleep. Participants had to press a button matching the meaning of the word (i.e. 'left' or 'right') with the corresponding hand (left or right), ignoring the location where the stimuli were presented. We hypothesized that instantaneous conflict is the result of an automatic process and would therefore be independent of conscious level. Contrarily, since conflict adaptation requires the integration of information over time, we expected it to depend on the level of consciousness. Regarding conflict processing, our results revealed longer reaction times and higher frontal theta-band power in incongruent trials (e.g. 'left' word coming from the right ear) compared to congruent trials (e.g. 'right' word coming from the right ear) during awake and drowsy states. However, these behavioural and neural markers were only observed during the awake state in the case of conflict adaptation. This results suggest that alertness levels differentially modulate the processing of conflictual information in the brain.
Saturday 12.00 - 13.30 Grote zaal Poster #25 Attention Liisa Raud
All authors & affiliationLiisa Raud, Department of Psychology, University of Oslo, Oslo, Norway Niamh Dooley, Cognitive Psychology, Faculty of Social Sciences, Leiden University, Leiden, the Netherlands Rene J. Huster, Department of Psychology, University of Oslo, Oslo, Norway
Differential dynamics of response selection and suppression in go/no-go and stop signal tasks
Show/hide abstractInhibitory control is critical for flexible behavior, enabling us to override predominant habits, temptations, or impulses. The most common experimental paradigms to investigate inhibition are the go/no-go task (GNGT) and the stop signal task (ssT). However, despite their largely similar neural responses, each task elicits additional unique activations, thereby questioning their equivalency. We compared the electromyography (EMG) and electroencephalography (EEG) indices of inhibition during both the GNGT and ssT in 33 participants. The primary task was to make speeded responses with either left or right hand as indicated by the imperative go stimulus, and to withhold or cancel the response in case of a no-go or stop signal, respectively. Using the frequency of the sub-threshold muscle twitches in the EMG of the suppressed hand as an index of inhibitory demand, we found substantially more twitches in stop than in no-go trials (41% and 17%, respectively). These results were paralleled by stronger frontocentral N200 and P300 in the ssT than in the GNGT, putatively reflecting the activation of a common inhibitory network, but stronger inhibitory demand in the ssT. This pattern was reversed, however, for go trials: measured from the non-responding hand, sub-threshold muscle twitches occurred in 24% of the go trials in the GNGT, but only in 10% in the ssT. This effect was paralleled by stronger parietal P300 in the GNGT than in the ssT. In the GNGT, this parietal P300 was even stronger in go than in no-go trials. Altogether, these observations indicate divergent neural dynamics during response selection and suppression, with the GNGT rather targeting the first and the ssT the latter. Further investigation is necessary to determine whether these selection and suppression processes rely on common or unique neural mechanisms in both tasks.
Saturday 12.00 - 13.30 Grote zaal Poster #26 Memory Lotte Talsma
All authors & affiliationLotte J Talsma, 1 Department Brain and Cognition, University of Amsterdam, Postbus 15915, 1001 NK Amsterdam Anouk M. van Loon, 2 Department of Experimental and Applied Psychology, and 3 Institute of Brain and Behavior Amsterdam, Vrije Universiteit Amsterdam, Amsterdam 1081 BT, The Netherlands H. Steven Scholte, 1 Julia.A. Broekhuizen, 1 J. Huisman, 1 Heleen A. Slagter, 1
Does cortical excitability in lDLPFC determine WM performance and tDCS response?
Show/hide abstractUpdating and maintaining information in working memory is crucial for goal-directed behavior. However, some people are better at this than others. transcranial Direct Current stimulation (tDCS), in which a small current is run between two scalp electrodes, is a promising method to enhance WM. Yet, large individual differences in tDCS responsiveness are also observed, with some individuals showing substantial WM improvements and others in fact impairments by the same stimulation. Cortical excitability (i.e., excitation/inhibition balance) may critically affect cortical functioning by determining neuronal response thresholds in a particular area. Here, we examined if cortical excitability levels in the left dorsolateral prefrontal cortex (lDLPFC, important for WM functioning and key target in tDCS studies) may predict interindividual differences in WM and tDCS response. Magnetic Resonance Spectroscopy (MRS) was used to quantify levels of the main inhibitory and excitatory neurotransmitters in the brain (both at rest and on-task): GABA and Glx and related to performance on two WM tasks (the N-back and the Sternberg task) (n=30). We expected that better WM performers would show low cortical excitability at rest but higher excitability during the WM task, i.e. reflecting better functional on-task disinhibition. Secondly, to relate cortical excitability to tDCS-induced improvements, a subset of subjects (N=19) came back for two tDCS sessions in which anodal or cathodal stimulation (20 minutes; 1 mA) was applied over lDLPFC concurrent with a WM task (letter N-back). We expected that depending on their initial excitability level, stimulation may push or pull individuals to or away from the point of optimal functioning (on an inverted U-curve). These findings are important to further elucidate the mechanisms underlying individual differences in WM functioning and may have significant practical implications with regard to the use of tDCS to optimize WM functioning in not just some, but all individuals.
Saturday 12.00 - 13.30 Grote zaal Poster #27 Connectivity/networks Tomas Knapen
All authors & affiliationTomas Knapen, VU Daan van Es, VU Martijn Barendregt, VU
Retinotopic Organization in the Default Mode Network
Show/hide abstractThe brain's default mode network (DMN) consists of a set of brain regions that consistently show decreases in BOLD signal during task engagement, when most areas show increases in BOLD signal. Despite being a well-known property of the DMN, the nature and function of deactivations in these regions remains unclear. Here we demonstrate that a fundamental neuroanatomical principle, retinotopy, exists within the human DMN. Specifically, BOLD decreases in the lateral parietal node of the DMN are specific to the appearance of a visual stimulus in a circumscribed region of retinotopic space. We estimated population receptive fields (pRFs) of negative amplitude from BOLD time courses in the DMN, and show that this region contains a coherent retinotopic map along the cortical surface. Moreover, this description of spatial preferences in the DMN, combined with ongoing activation patterns, allows us to reconstruct (decode) the position of a visual stimulus with a fidelity comparable to the known retinotopic maps of the intraparietal sulcus. Our results indicate that retinotopically specific activations and deactivations synergistically subserve the processing of visual information. As DMN regions have been shown to contain pRFs that selectively activate for social information, it is likely that representations in retinotopic and other reference frames coincide in these regions. This overlap would allow local computations to integrate multiple levels of information processing.
Saturday 12.00 - 13.30 Grote zaal Poster #28 Consciousness Maxine Sherman
All authors & affiliationMaxine T. Sherman, Sackler Centre for Consciousness Science, University of Sussex, UK Anil K. Seth, Sackler Centre for Consciousness Science, University of Sussex, UK Ryota Kanai, Department of Neuroinformatics, Araya Brain Imaging, Japan
Integration of perceptual priors into decision confidence in an occipital-frontal functional network
Show/hide abstractIt is clear that prior expectations shape perceptual decision-making, yet their contribution to the construction of subjective decision confidence remains largely unexplored. Using fMRI, we investigated the neural correlates of confidence and their modulation by top-down expectation. We recorded fMRI data while participants performed a target detection task. Participants also reported their subjective confidence in their perceptual judgments. The prior odds of the target being present versus absent was manipulated orthogonally to attention. Attention was manipulated by instructing participants to either ignore (single-task) or perform (dual-task) a concurrent visual search task. This factorial design enabled us to isolate effects of expectation from potentially confounding effects of attention. Behaviorally, confidence was higher when the perceptual decision had a high prior probability. This effect of expectation on confidence was mirrored in right inferior frontal gyrus (rIFG) BOLD activity. Specifically, we found perceptual ‘prediction error' responses (reflecting the mismatch between prior and inference) in rIFG that were negatively correlated with confidence. Connectivity analyses showed that relevant sensory signals for confidence were represented in intracalcarine sulcus. These analyses also showed that relevant priors for confidence were represented bilateral orbitofrontal cortex (OFC). Supporting the notion that OFC plays an important role in incorporating expectations into confidence, behavioral effects of expectation on confidence correlated with OFC white matter density. Together, our results indicate that predictive information is integrated into subjective confidence in rIFG. Our findings advance our understanding of the neural basis of subjective perceptual processes by revealing an occipital-frontal functional network that integrates top-down and bottom-up signals into the construction of confidence.
Saturday 12.00 - 13.30 Grote zaal Poster #29 Memory Mikolaj Magnuski
All authors & affiliationMikolaj Magnuski, Faculty of Psychology, University of Social Sciences and Humanities, Warsaw, Poland, Jagiellonian University in Cracow Aleksandra Kolodziej, Faculty of Psychology, University of Social Sciences and Humanities, Warsaw, Poland Aneta Brzezicka, Faculty of Psychology, University of Social Sciences and Humanities, Warsaw, Poland
Late middle frontal response to maintenance onset correlates negatively with load in Sternberg task.
Show/hide abstractIntroduction: In a typical Sternberg task (Sternberg, 1966) the participant is presented with a succession of digits to remember but does not know how many digits will be presented on a given trial. This uncertainty may influence the readiness to shift from encoding to maintenance mode once maintenance period starts. As more digits are presented the probability of maintenance period increases and the uncertainty is reduced. Therefore if maintenance period starts early the participant has to exert more cognitive control to switch to maintenance mode. In line with this reasoning a recent study (Kaminski et al., 2017) found that maintenance neurons in the medial frontal cortex (MFC) show increased activity to early start of maintenance period. We investigate this phenomenon with EEG. Methods: Thirty- three participants (aged 18-35) performed Sternberg task with 6 memory loads (from 2 to 7 digits to remember) when EEG was recorded. All digits were presented in fast succession (one digit every 0.5 seconds) and appearance of fixation dot started the maintenance period (which lasted randomly from 1.5 to 2.5 seconds). We compared event related potentials locked to the onset of the fixation dot separately for different loads. Results and Conclusions: We observe that late middle-frontal ERP response to maintenance start strongly depends on how early encoding period ends - displaying a significant cluster with negative relationship with load that starts around 250 ms after maintenance onset and continues at least up to 500 ms. The topography of this effect resembles the topography of middle frontal theta power in earlier stages (300 - 400 ms after maintenance onset) and later shifts more towards posterior regions (after 400 ms). We interpret this result as a manifestation of different probability of switching from encoding into maintenance phase with load.
Saturday 12.00 - 13.30 Grote zaal Poster #30 Neural modeling Bianca Trovò
All authors & affiliationTrovò Bianca, Neurospin, Inserm-Unicog Unit, CEA Saclay, Gif-sur-Yvette, FRANCE & Universite Pierre et Marie Curie, Paris 6 Iscan Zafer, Neurospin, Inserm-Unicog Unit, CEA Saclay, Gif-sur-Yvette, FRANCE Schurger Aaron, Neurospin, Inserm-Unicog, CEA Saclay, Gif-sur-Yvette, FRANCE & Ecole Polytechnique Federale de Lausanne, Geneve, SWITZERLAND
Spontaneous action initiation with temporal constraint
Show/hide abstractThe Readiness Potential (RP) is a slowly increasing surface-negative cortical potential that precedes spontaneous voluntary movements. A recent interpretation provided by the stochastic decision model [2] suggests that this slow exponential preceding the motor event could be the result of a time-locked average of ongoing sub-threshold fluctuations in neural activity. In fact, autocorrelated background activity seems to play a crucial role in the planning and preparation of actions. According to our model, slow fluctuations continuously drifts randomly closer to or farther from the decision-threshold for initiating action in a bounded-accumulation fashion where ‘noise' in the brain is integrated over time. 
In particular, the model predicts that the movement is more likely to happen at the 'crest' of these ongoing fluctuations, and less likely to do so if it coincides with a trough. In classical RP studies [1,3], though, the window of opportunity in which to perform the movement is almost unlimited. 
So we developed a new experimental paradigm in order to investigate the parametric variations of the RP shape as a function of increasing levels of temporal freedom (temporal uncertainty). We will perform a variant of the Libet task in which subjects, sitting inside the MEG, will be asked to initiate a finger tap within a given time window for each trial. Participants are free to make the movement whenever they want as long as they do it before the time has elapsed. The time limit variable, signalled by an animated clock, will change among blocks in a counterbalanced way across subjects. 
Our main prediction is that the movement-preceding activity in pre-motor areas of the frontal lobe will appear to begin earlier, and be more prominent in the time-locked average, as the window of time within which the subject is allowed to move becomes longer. The temporal constraint will then affect the Early but not the Late component of the RP.
Saturday 12.00 - 13.30 Grote zaal Poster #31 Developmental Danilka Castro Canizares
All authors & affiliationDanilka Castro Canizares, Centro de Investigacion Avanzada en Educacion (CIAE). Valeska Amor Diaz, Centro de Neurociencias de Cuba. David Maximiliano Gomez, CIAE Pablo Dartnell, Centro de Modelamiento Matematico & CIAE.
Contribution of Working Memory Components to Efficiency in Simple Arithmetic During School Age
Show/hide abstract At present there are multiple evidences regarding the relation between working memory (MT) and arithmetic performance. However the findings are still inconsistent as to what components of WM can actually to predict mathematical performance and which do not. Although the contribution of each WM component may vary according to age and the level of children in mathematics learning, this variable has not been systematically evaluated. The aim of this study was to evaluate the unique contribution of WM components (phonological loop, visuo-spatial sketchpad and central executive) to the variance of efficiency in simple arithmetic during school age. School age children of different grades, with and without difficulties in mathematical learning, were evaluated through verbal and visuo-spatial WM tasks with control of general (intelligence and age) and specific (basic numerical comparison skills: symbolic and non-simbolic) domain variables. The results show that the unique contribution of WM components to the variance of efficiency in simple arithmetic varies throughout the school development: with a greater involvement of spatial WM component in first years of schooling, and greater contribution of verbal WM component as school development increases. In addition, significant differences in WM tasks performance between children with and without difficulties in mathematics were found, suggesting that the origin of difficulties in this domain could be very associated to WM functioning. These results are especially relevant for pedagogical and neurocognitive intervention during school age.
Saturday 12.00 - 13.30 Grote zaal Poster #32 Clinical Young-Chul JUNG
All authors & affiliationDeokjong Lee, Yonsei University College of Medicine Young-Chul Jung, Yonsei University College of Medicine
Altered Cortico-Cerebellar Functional Connectivity and Impulsivity in Internet Gaming Disorder
Show/hide abstractObjective: Internet gaming disorder (IGD) is a type of behavioral addiction characterized by abnormal executive control, leading to loss of control over excessive gaming. Attention deficit and hyperactivity disorder (ADHD) is one of the most common comorbid disorders in IGD, involving delayed development of the executive control system, which could predispose individuals to gaming addiction. We investigated the influence of childhood ADHD on neural network features of IGD and hypothesize that IGD can develop via two distinct pathways according to subject history of childhood ADHD. Methods: Resting-state functional magnetic resonance imaging analysis was performed on 44 young, male IGD subjects with and without childhood ADHD and 19 age-matched, healthy male controls. Posterior cingulate cortex (PCC)-seeded connectivity was evaluated to assess abnormalities in default mode network (DMN) connectivity, which is associated with deficits in executive control. Results: IGD subjects without childhood ADHD exhibited expanded functional connectivity between the PCC and brain regions implicated in salience processing (anterior insula, medial prefrontal cortex). By contrast, IGD subjects with childhood ADHD showed expanded functional connectivity between the PCC and cerebellum (crus II), a region involved in executive control. The strength of connectivity between the PCC and cerebellum (crus II) correlated with stronger impulsiveness. Conclusion: Individuals with IGD showed altered PCC-based functional connectivity, the characteristics of which might be dependent upon history of childhood ADHD. Our findings suggest that altered neural networks for executive control in ADHD would be a predisposition for developing IGD.
Saturday 12.00 - 13.30 Grote zaal Poster #33 Sensory processing Madeline Klinger
All authors & affiliationMadeline Klinger, Brain Imaging Center, Goethe University Frankfut Christian Kell, Department of Neurology, Goethe University Frankfurt Danko Nikolic, Computer Sciences Corporation
Learning to Adapt: How Mechanisms of Color Constancy Learn
Show/hide abstractColor constancy refers to the ability of the visual system to quickly adapt to new illumination conditions, permitting accurate color perception even under unfavorable lighting. It is not currently known, however, whether the adaptive processes responsible for color constancy can be trained, or whether they are fixed in adulthood. We thus investigated whether an intensive color constancy training task can improve adaptive color-perception mechanisms. In Experiment 1, subjects were trained on continuous adjustment of color perceptions. They viewed a 30-minute film over which a single-color filter (e.g., magenta) faded in and out over a 15-second cycle, forcing continual adaptation to illumination conditions. After training, subjects tended to better discern the true colors of natural scenes viewed under the trained filter color than scenes viewed under a novel, untrained filter color (e.g., green). Because we altered the filter illumination continuously during training in Experiment 1, the time needed to adapt to a single illumination condition remained unclear. Hence, in the second and third experiment, we investigated the actual time needed to adapt. In Experiment 2, we induced color adaptation by presenting color-filtered natural scenes for 12 seconds; we then measured the perception of aftereffects and found that recovery from color adaptation required about 700-800 milliseconds. Finally, Experiment 3 investigated how this recovery time course was affected by color constancy training in Experiment 1. These findings indicate that adaptive visual processes indeed can be trained in adulthood. The results will be discussed in relation to the recently introduced theory of practopoiesis, which proposes a three-level adaptive organization of the nervous system: slow learning mechanisms adjust the properties of fast adaptive mechanisms, while these fast adaptive mechanisms in turn continually adjust the flow of neural activity within brain circuits.
Saturday 12.00 - 13.30 Grote zaal Poster #34 Decision-making Elena Sildatke
All authors & affiliationElena Sildatke, University Hospital Cologne, Department of Psychiatry and Psychotherapy, Cologne, Germany Thomas Schueller, University Hospital Cologne, Department of Psychiatry and Psychotherapy, Cologne, Germany Theo O.J. Gruendler, Otto-von-Guericke-University, Faculty for Economics and Management, Magdeburg, Germany, Center for Behavioral Brain Sciences, Magdeburg, Germany Markus Ullsperger, Otto-von-Guericke-University, Faculty for Economics and Management, Magdeburg, Germany, Center for Behavioral Brain Sciences, Magdeburg, Germany, Otto-von-Guericke University, Department of Psychology, Magdeburg, Germany Veerle Visser-Vandewalle, University Hospital Cologne, Department of Stereotactic and Functional Surgery, Cologne, Germany Daniel Huys, University Hospital Cologne, Department of Psychiatry and Psychotherapy, Cologne, Germany Jens Kuhn, University Hospital Cologne, Department of Psychiatry and Psychotherapy, Cologne, Germany, Johanniter Hospital Oberhausen, Department of Psychiatry, Psychotherapy and Psychosomatic, Oberhausen, Germany
Error Related Processing In Ncl. Accumbens Of Patients With Severe Heroin Addiction
Show/hide abstractPerformance monitoring incorporates error detection, a process that can be studied by its electrophysiological correlate the error-related negativity (ERN). ERN is a negative deflection following errors and is generated in the mediofontal cortex (MFC). It has further been associates with Ncl. accumbens (NAcc), a subcortical structure also involved in performance monitoring. Particularly, an ERN-like potential was observed in the NAcc following errors and cortical ERN was abolished by striatal lesions but enhanced by stimulating in the NAcc. However, there is little evidence about the relation of NAcc and ERN and its role in performance monitoring. In order to find correlates of performance monitoring in the NAcc, also with respect to MFC activity, we conducted recordings with patients (n=4) with severe heroin addiction who had implanted depth electrodes in the NAcc. Electrophysiological activity from scalp surface (FCz) and local field potentials (LFPs) from the NAcc were simultaneously recorded during a flanker task. For LFPs, statistical significance between conditions (error, correct) was assessed with single-trial permutation tests. A cross-correlation analysis between NAcc and FCz was performed. Data recorded at FCz were compared to a control group. We found no differences between patients and controls. NAcc activity differed between conditions during response, though no ERN-like potential was evident. Interestingly, this difference was significant only for (two) patients who also had a significant ERN, indicating a discrimination between false and correct responses in the NAcc and an association between ERN and NAcc. Later in trial, we observed enhanced negativity after errors (300-700ms) that might signal the need for adjustments. Cross-correlation indicated a top-down directed information flow (cortexNAcc). Our results show that NAcc activity is modulated by trial correctness during the early and the later stage of response processing, providing support for its role in performance monitoring and signaling the need for adjustments.
Saturday 12.00 - 13.30 Grote zaal Poster #35 Social behavior Michael Rojek Giffin
All authors & affiliationH. Steven Scholte, University of Amsterdam Frans van Winden, University of Amsterdam Richard Ridderinkhof, University of Amsterdam Carsten K.W. De Dreu, Leiden University
Probability of kill, reward and risk is differentially tracked in reward and fronto-parietal networks when humans compete to win versus not lose.
Show/hide abstractThis study combined functional magnetic resonance imaging (fMRI) with a novel economic contest called the "predator-prey" game (PPG) which models domineering, exploitative behavior in one player (the predator) and subservient, protective behavior in the other player (the prey). We calculated the expected reward, the expected kill/survive (the probability of an investment resulting in a victory over the other player), and expected risk for every trial, and examined how these various criteria modulated neural activity. In the predator role, participants exhibited significant neural modulation by expected kill/survive in the ventral striatum. This modulation was significantly greater in predator trials than in prey trials, and within predator trials was significantly greater for expected kill/survive than for both expected reward and expected risk. In the prey role, participants exhibited significant neural modulation by expected risk across an expansive cluster spanning frontal and parietal regions that are commonly associated with a range of higher order cognitive functions. This modulation was greater in prey trials than in predator trials, and within prey trials was greater for expected risk than for expected kill/survive, but not greater for expected risk than for expected reward. Taken together, when individuals are in a dominant position, the expectation that an act will subordinate the other player, regardless of that act's monetary reward, is tracked by the ventral striatum; when individuals are in a subservient position, the expected risk associated with behavior is tracked by a large fronto-parietal network.
Saturday 12.00 - 13.30 Grote zaal Poster #36 Clinical Sylvie Bernaerts
All authors & affiliationSylvie Bernaerts, Research Group for Neuromotor Rehabilitation, Department of Rehabilitation Sciences, University of Leuven (KU Leuven), Leuven, Belgium Claudia Dillen, Research Group for Neuromotor Rehabilitation, Department of Rehabilitation Sciences, University of Leuven (KU Leuven), Leuven, Belgium Jean Steyaert, Research Group Psychiatry, Department of Neurosciences, University of Leuven (KU Leuven), Leuven, Belgium Kaat Alaerts, Research Group for Neuromotor Rehabilitation, Department of Rehabilitation Sciences, University of Leuven (KU Leuven), Leuven, Belgium
Long-term effects of oxytocin on fronto-amygdala connectivity in autism spectrum disorders: a randomized-placebo controlled trial
Show/hide abstractAutism spectrum disorders (ASD) are characterized by impairments in social interaction and repetitive and restricted behaviors. To date, no pharmacological treatment exists targeting the core symptoms of ASD, yet the pharmacological use of the neuropeptide oxytocin (OT) has gained interest from the research community to explore its potential for elevating social deficits in ASD. We aimed to examine neural and behavioral effects of single- and multiple-dose OT treatment and the possibility of retention effects one month post-treatment using a double-blind, randomized, placebo-controlled, between-subject clinical trial including thirty-eight adult men with ASD. A key question was to determine whether similar neural effects are observed after single- versus multiple-dose treatment; whether the neural effects are paralleled by behavioral improvements on core ASD symptoms and specifically, whether the observed effects can outlast the period of actual administration. Doses of 24 IU oxytocin (Syntocinon®, Sigma-tau) or placebo nasal spray (saline natrium-chloride solution) were administered daily for four weeks. Resting-state fMRI neuroimaging was adopted to evaluate ‘intrinsic' OT-induced changes in functional connectivity of amygdala-prefrontal circuits at rest (medial prefrontal cortex and orbitofrontal cortex). Behavioral effects of OT treatment were primarily assessed on two core ASD symptoms via self-report questionnaires: restricted and repetitive behavior and social responsiveness, and secondarily, on self-reports of state attachment, trait attachment, and quality of life. Data showed that four-week OT-treatment (24 IU/day) reduced functional connectivity of the amygdala with prefrontal regions in the orbitofrontal cortex until one month post-treatment. Neural changes in amygdala-prefrontal coupling were shown to be paralleled by behavioral improvements in repetitive behavior, attachment avoidance and social motivation. These findings provide further insights into the biological mechanism by which oxytocin exerts its prosocial effects and the observation that the neural and behavioral effects outlasted the period of actual administration further supports the therapeutic potential of oxytocin interventions for ASD.
Saturday 12.00 - 13.30 Grote zaal Poster #37 Attention Pawel Strozak
All authors & affiliationPawel Strozak, The John Paul II Catholic University of Lublin Agnieszka Fudali-Czyz, The John Paul II Catholic University of Lublin Pawel Augustynowicz, The John Paul II Catholic University of Lublin Piotr Francuz, The John Paul II Catholic University of Lublin
Visual change detection after saccades during optokinetic stimulation
Show/hide abstractVisual change detection is a demanding task for human cognitive system, as demonstrated in many studies. However, little is known about how asymmetries in overt and covert distribution of visuo-spatial attention can govern the visual change detection and modulate its neural correlates. It is also unclear how covert and overt attentional shifts can interact in this domain. To this end, event-related potentials (ERPs) were recorded while 24 participants performed visual size change detection task in the S1-S2 matching paradigm. Each trial began after leftward or rightward saccadic eye movement and consisted of two sequentially presented circles (S1-S2), with S2 either slightly larger than S1 (difficult change condition), significantly larger than S2 (easy change condition), or the same as S1 (no change condition). The change detection task was embedded in optokinetic stimulation (OKS) consisting of black-and-white vertical stripes on the visual periphery that were moving leftward, rightward, or were static. We hypothesized that eye movements and OKS would trigger overt and covert shifts of attention, respectively. We also predicted that this would enhance the performance of visual change detection task and modulate ERPs recorded during that task. Subjects performed better in easy change condition than in difficult change condition. Mean proportion of hits was larger after rightward saccades, and there was a trend towards mean proportion of correct rejections being larger after leftward saccades. These results were reflected at the neural level. N270 amplitude difference between easy change and no change conditions was larger after rightward saccades, whereas LPC amplitude difference was larger after leftward saccades. Behavioral results were not affected by optokinetic stimulation. However, the magnitude of N270 difference wave was largest during static OKS after rightward saccades. The results are discussed in terms of asymmetries in overt and covert shifts of attention during visual change detection task.
Saturday 12.00 - 13.30 Grote zaal Poster #38 Language Kristof Strijkers
All authors & affiliationKristof Strijkers, AMU & CNRS, LPL. Valerie Chanoine, AMU, LPL. Dashiel Munding, AMU, LPC. Anne-Sophie Dubarry, AMU & CNRS, LPL. Agnes Trebuchon, AMU, INS. Jean-Michel Badier, AMU, INS Xavier Alario, AMU & CNRS, LPC.
The neural dynamics of syntactic predictions.
Show/hide abstractIn contemporary neuroscience and cognition prediction has become a key computational construct to explain the speed and efficiency of the brain in processing the complex mental world. In the domain of language, many studies have demonstrated that constraining language contexts can bias the upcoming availability of specific words. But what exactly is predicted, when, and how remains debated. The current study examined whether the brain utilizes grammatical information to anticipate syntactical structure in a continuous and automatic fashion, and which spatiotemporal dynamics would underpin such syntactical predictions. We conducted an MRI-constrained MEG experiment in which we contrasted the neuro-magnetic sources elicited by nouns and verbs when they were preceded by a predictive context (i.e., possessive pronouns for nouns, and personal pronouns for verbs) versus an unpredictive context (visually matched symbols). The results showed ultra-rapid (detected from ~80 ms onwards) noun-verb differences in the left and right inferior frontal gyrus (IFG), but only when those nouns and verbs were preceded by the syntactically predictive context (i.e. their corresponding pronoun). In addition, the contrast between possessive and personal pronouns that preceded the rapid noun-verb modulations in the IFG produced differences in source activation in various regions of the prefrontal cortex (the superior frontal and orbitofrontal cortex), previously associated with top-down processes such as goal-directed behaviour, planning and proactive biasing. Taken together, we conclude from these data that (a) predictive processing constitutes an integral part of syntactic parsing and (b) these syntactic predictions may be sustained by the dynamical binding of the LIFG with a domain-general prefrontal network responsible for proactively generating top-down biasing signals on to the linguistic system with which it is in synchrony.
Saturday 12.00 - 13.30 Grote zaal Poster #39 Sensory processing Lin-Yuan Tseng
All authors & affiliation Lin-Yuan Tseng, Research Center for Brain and Consciousness, Taipei Medical University, Taiwan; Research Center for Brain and Consciousness, Shuang-Ho Hospital, Taiwan Tzu-Yu Hsu, Research Center for Brain and Consciousness, Taipei Medical University, Taiwan; Research Center for Brain and Consciousness, Shuang-Ho Hospital, Taiwan; Graduate institution of Health and Biotechnology Law, Taipei Medical University, Taiwan
Asymmetry in sensitivity to temporal asynchrony of audio-visual ordering: An EEG study
Show/hide abstractWhen audio and visual stimuli are presented very close to each other in time, there are different perceptual effects regarding the synchronous/asynchronous perception is depending on which of the two is presented first. The mechanism underlying this asymmetric phenomenon and the time-course according to which asynchronous or synchronous perceptual judgments are made, however, are still unclear. To address these questions, we combined an audio-visual simultaneity judgment task paradigm and EEG. The order of visual and auditory stimuli (i.e. visual stimulus leading auditory, VA; auditory stimulus leading visual, AV) and the time interval between these two stimuli were manipulated. The aims of the study were thus two-fold: 1) do VA and AV conditions affect the judgment of synchrony given the same time interval? 2) do subjective synchronous and asynchronous responses reflect different neural processes for the given conditions? In line with prior studies, our behavioural data suggested that different stimulus-onset asynchronies and the order of the stimuli modulate perceptual experiences. Our results also showed an asymmetric phenomenon where the VA ordering results in a wider time window in which the stimuli are judged to be synchronous/asynchronous than the AV condition. N1 and P1 amplitudes in response to the first stimulus were found to be higher during AV trials in which participants reported perceiving asynchronous audio and visual stimuli. This pattern was also seen in the amplitude of the N1 and P1 responses to the second stimulus for both conditions. These results suggest that where the first stimulus is auditory, the waveform induced by it already contains information as to the relative temporal order of the stimuli and thus ultimate perceptual experience.
Saturday 12.00 - 13.30 Grote zaal Poster #40 Developmental Isabel C. Duarte
All authors & affiliationGilberto Silva, CNC.IBILI, Institute for Biomedical Imaging and Life Sciences, University of Coimbra Isabel Catarina Duarte, ICNAS, Institute for Nuclear Sciences Applied to Health, University of Coimbra Inês Bernardino, CNC.IBILI, Institute for Biomedical Imaging and Life Sciences, University of Coimbra Tânia Marques, CNC.IBILI, Institute for Biomedical Imaging and Life Sciences, University of Coimbra Miguel Castelo-Branco, ICNAS, Institute for Nuclear Sciences Applied to Health, University of Coimbra
Rhythmic Motor Performance in Neurofibromatosis Type 1 as assessed by EEG, fMRI and GABA spectroscopy
Show/hide abstractNeurofibromatosis type1 (NF1) is associated to extensive clinical and behavioural difficulties, and impaired inhibition has been postulated in this disorder. Inhibition is involved in the function of motor central pattern generators and in the control of speed and stability of motor rhythms, so we explored the link between behavioural motor control, GABA levels, brain rhythms and brain activity in NF1. We recruited 21 adults with NF1 and 20 age and gender-matched controls. Participants underwent fMRI, GABA spectroscopy and EEG recordings. During fMRI scans and during EEG, participants performed an audio-paced finger tapping at 1, 3 or 5Hz. GABA was measured in motor cortex, after functional localization of M1. The power at the tapping frequency (measure of performance) showed differences between groups at 1Hz and 3Hz (Mann-Whitney, p=0.003 and p=0.043). Time-frequency results at the beta sub-band (20-26Hz) mirrored those findings, showing a less pronounced cyclic pattern for the NF1 group. However, significant differences were only found for behavioural patterns. A correlation between GABA and the power of the beta sub-band for the 1Hz condition was found only for the control group. For fMRI, the control group showed a higher recruitment of the extrapyramidal motor system (putamen, cerebellum and red nucleus) during the fastest condition (t(190)>3.09, p<0.05, FDR). It is been suggested that the abnormal neurodevelopment in NF1 results from inhibitory/excitatory imbalance with consequences in multiple cognitive domains. Our results corroborates this view by showing: 1) that task performance significantly discriminated NF1 and controls, which can be related to the behaviour of inhibitory interneurons that influence beta band modulations; 2) no correlation between GABA levels in M1 with the beta power in NF1, unlike the control group and as described before in healthy participant cohorts; 3) decreased recruitment of cerebellum, where inhibitory interneurons are essential regulators of its function in rhythmic synchronization.
Saturday 12.00 - 13.30 Grote zaal Poster #41 Attention Tom Marshall
All authors & affiliationTom Marshall, University of Oxford Jill O'Reilly, University of Oxford
Stable and competitive neuronal dynamics within the human dorsal attentional network
Show/hide abstractThe dorsal attention network -- consisting of frontal eye fields (FEF) and intraparietal sulci (IPS) -- is implicated in goal-directed shifting of attention in space(1). However, in human fMRI studies of attention these regions typically display concurrent activation, and little is known about the distinct contributions FEF and IPS make to directing attention. Primate recordings from FEF and LIP (the putative analogue of IPS) also show that both regions contain single neurons that demonstrate evidence accumulation, increasing spike rate until a perceptual decision is made(2,3). We hypothesize that distinct roles of these two regions can be identified at the intermediate level of neuronal dynamics. Specifically, we contend that FEF displays point-attractor dynamics; whereby populations representing different choice outcomes inhibit one another in a ‘winner-take-all' fashion, and that IPS displays bump-attractor dynamics; integrating information about past events and constructing a stable ‘salience map' of space. To test this we will combine computational modelling with magnetoencephalography. We developed a moving dots task with two competing sets of coherent dots to tease apart the effects of competition and evidence strength. This task requires integration of evidence for two competing alternatives both within a trial - where rightward and leftward motion vary independently - and across trials - in a ‘reversal learning' context where one motion direction occurs more frequently within a block. We will use a biophysical neural mass model instantiating competitive dynamics(4), in combination with a Bayesian learning model, to generate predictions about within- and across-trial activity respectively. Preliminary data suggest a strong competitive dynamic in behaviour (reaction times) that is compatible with the neural mass model. This study will allow us to go beyond characterising the dorsal attentional network as a single entity, and delineate the sub-processes carried out in distinct network nodes from which the process of directing attention is built.
Saturday 12.00 - 13.30 Grote zaal Poster #42 Social behavior Sally Grace
All authors & affiliationSally A. Grace, Centre for Mental Health, Swinburne University, Melbourne, Australia; Izelle Labuschagne, Australian Catholic University, Melbourne, Australia; Catarina Kordachia, Australian Catholic University, Melbourne, Australia; Susan L. Rossell, Centre for Mental Health, Swinburne University, Melbourne, Australia; Psychiatry, St Vincent's Hospital, Melbourne, Australia; Monash Alfred Psychiatry Research Centre Monash University, Melbourne Australia.
Oxytocin and brain activity in humans: A systematic review and coordinate-based meta-analysis of functional MRI studies
Show/hide abstractOxytocin (OXT) is a neuropeptide that has a critical role in human social behaviour and cognition. Research investigating the role of the OXT on functional brain changes in humans has often used task paradigms that probe socioemotional processes. Preliminary evidence suggests a central role of the amygdala in the social cognitive effects of OXT, however, inconsistencies in task-design and analysis methods have led to inconclusive findings regarding a cohesive model of the neural mechanisms underlying OXT's actions. The aim of this meta-analysis was to systematically investigate these findings, and inform the literature on the neural correlates of intranasal OXT. We employed a coordinate-based meta-analysis for task-based neuroimaging literature using activation likelihood estimation (ALE). A systematic search was conducted for whole-brain task-based fMRI studies that compare OXT to placebo. The authors searched PubMed, PsycINFO, and Scopus databases for fMRI studies on intranasal oxytocin up to February 2017. From each study, coordinates were extracted from clusters with significant differences in OXT versus placebo in healthy adults. Data were included for 24 fMRI studies that reported a total of 240 distinct foci. When contrasting intranasal OXT to placebo, the meta-analysis identified task-related increases in activity within a cluster within the left superior temporal gyrus and insula, and a decrease in activity in a cluster including the left amygdala and putamen. The findings are important as they implicate regions beyond the amygdala in the central effects of OXT. The outcomes from this meta-analysis can guide a priori predictions for future OXT research and provide an avenue for targeted treatment interventions.
Saturday 12.00 - 13.30 Grote zaal Poster #43 Connectivity/networks Megan E.J. Campbell
All authors & affiliationMegan Campbell, The Queensland Brain Institute, University of Queensland, Australia Vinh Nguyen, QIMR Berghofer Medical Research Institute, Australia Michael Breakspear, QIMR Berghofer Medical Research Institute, Brisbane, Australia Ross Cunnington, The Queensland Brain Institute and Department of Psychology, University of Queensland, Australia
Insula cortex gates parietal mirror response during counter-imitation.
Show/hide abstractSocial interactions essentially depend on two interrelated processes: perceiving another's actions and making actions in response. Human mirror system research has focused almost exclusively on imitation, defining mirroring as direct-matching of observed to executed actions (Ritzolatti & Fogassi, 2014). Yet, real-world interactions are variable, and mirror-matched representations are not always useful to prompting adaptive responses. The integration of cognitive-control processes with mirror processes has been neglected (for review: Campbell & Cunnington, 2017), here we investigated this by manipulating the task-relevance of mirroring of observed actions. Our fMRI task required participants (n=24) to observe and execute actions under two different response contexts: 1) predefined actions that then incidentally matched or mismatched stimuli; 2) stimulus-dependent responses with either intentional imitation or counter-imitation of observed actions. Task-dependent changes in BOLD-signal were used to inform 48 models of effective connectivity using dynamic causal modelling (DCM). We found evidence for differential engagement of frontal control networks for counter-imitation (bilateral insula and cingulate cortices), as distinct from incidental stimulus-response mismatch. Signal change within the action observation network (medial temporal, inferior parietal (IPL) and occipital regions) was related to imitation compared to counter-imitation, and was also distinctly greater for intentional, rather than incidental, imitation. Nonlinear DCMs included 4 nodes representing action perception (V5 area MT), action-mirroring (IPL), action planning (supplementary motor) and control (insula) regions. Bayesian model selection revealed a winning model with the IPL mirroring response gated by the insula during counter-imitation. When preparing to intentionally counter-imitate, connections of the SMA to IPL were downregulated by nonlinear influence of the insula. These results suggest that preparing to intentionally counter-imitate an observed action engages executive control processes which suppress more automatic mirroring processes. Hence the human mirror system is not so fixed as to be immune to domain-general cognitive control.
Saturday 12.00 - 13.30 Grote zaal Poster #44 Sensory processing Manuel Varlet
All authors & affiliationManuel Varlet, The MARCS Institute for Brain, Behaviour & Development, Western Sydney University, Australia. Sylvie Nozaradan, The MARCS Institute for Brain, Behaviour & Development, Western Sydney University, Australia. Peter E. Keller, The MARCS Institute for Brain, Behaviour & Development, Western Sydney University, Australia.
Neurophysiological bases of sensorimotor entrainment to auditory and visual rhythmic stimuli
Show/hide abstractBehavioural studies have shown that human movements are spontaneously entrained to auditory and visual environmental rhythms. Without intention, motor responses are thus automatically attracted towards synchrony with external rhythmic stimuli. However, the neural mechanisms underlying such sensorimotor entrainment remain largely unknown. Here we investigated with electroencephalography (EEG) and electromyography (EMG) the modulation of neural and muscular activities induced by the presentation of visual and auditory metronomes at either 1 or 2 Hz while participants were required to maintain constant pressure on a force sensor with their right index finger. Preliminary results suggest neural entrainment in the visual and auditory systems at the frequency of the stimulus presented (1 or 2 Hz) together with time-locked power modulations in the alpha (8-12 Hz) and beta (12-35 Hz) frequency bands, depending on the sensory modality involved. Furthermore, preliminary results suggest that although participants maintained constant finger force pressure and EMG activity when visual or auditory sequences were presented, some individuals exhibited time-locked modulations of EEG-EMG coherence in the beta (12-35 Hz) frequency band that could reflect the occurrence of motor entrainment. These preliminary findings provide new insights into the neurophysiological mechanisms underlying the occurrence and stability of sensorimotor entrainment.
Saturday 12.00 - 13.30 Grote zaal Poster #45 Developmental Didac Vidal-Pineiro
All authors & affiliationDidac Vidal-Pineiro, Center for Lifespan Changes in Brain and Cognition, Department of Psychology, University of Oslo, Oslo, Norway. Markus H Sneve, Center for Lifespan Changes in Brain and Cognition, Department of Psychology, University of Oslo, Oslo, Norway. Lars H Nyberg, Umeå Center for Functional Brain Imaging, Umeå, Sweden; Department of Integrative Medical Biology, Physiology Section and Department of Radiation Sciences, Diagnostic Radiology, Umeå University, Umeå, Sweden. Athanasia M Mowinckel, Center for Lifespan Changes in Brain and Cognition, Department of Psychology, University of Oslo, Oslo, Norway. Donatas Sederevicius, Center for Lifespan Changes in Brain and Cognition, Department of Psychology, University of Oslo, Oslo, Norway. Kristine B Walhovd, Center for Lifespan Changes in Brain and Cognition, Department of Psychology, University of Oslo, Oslo, Norway. Anders M Fjell, Center for Lifespan Changes in Brain and Cognition, Department of Psychology, University of Oslo, Oslo, Norway.
Maintenance and compensation in aging: Brain correlates of an episodic encoding task
Show/hide abstractMemory markedly declines with age. Nonetheless, substantial differences in memory function exist among older adults. It is proposed that such differences can be explained by underlying patterns of activity. Two patterns of functional activity often arise. Either older adults with preserved cognition are able to minimize age-related functional changes(1) or else, those with preserved memory recruit additional resources, thus showing compensatory patterns of activity(2). Brain maintenance relates to frontal(3) and posteromedial(4) function while compensatory patterns are generally associated with frontal overrecruitment(5). Evidence for both patterns were studied with 290 (age range=18-80) participants that were fMRI-scanned during an implicit source-encoding task(6). Memory for item-context associations was tested after 30' and the subsequent memory effects (SME) were estimated using activity for context versus item memory trials. SME were tested using age, memory performance, sex and age*performance interaction, all considered as factors. Three much-used analyses were carried: region-of-interest (ROI), vertexwise and conjunction analyses in addition to an activity gradient analysis -to study SME along an gradient of activity. An independent sample of young participants (n=55) was recruited to define ROIs and create the activity gradient. Both the ROI (p<0.05, FDR-corrected) and the vertexwise analyses (p<0.01; FWE) revealed that only in older participants, frontal SME were associated with memory performance. Low-performing older adults exhibited reduced frontal SME. Despite strong decrements with age, posteromedial SME were unrelated to memory in aging. The conjunction analysis revealed specific patterns of activity in older adults, though sparsely distributed. The activity gradient analysis confirmed that age and age*performance effects outside the SME network were negligible. Concluding, frontal cortex activity during source-encoding represents a primary determinant of memory maintenance in aging likely reflecting deficits/maintenance in strategic components. In contrast, the functional significance of the sparsely found compensatory patterns are of unclear functional significance.
Saturday 12.00 - 13.30 Grote zaal Poster #46 Developmental Jolanta Golan
All authors & affiliationJolanta Golan, University of East London* Kathryn Daubney, University of East London Grainne McLoughlin, Kings College London Elena Kushnerenko, University of East London
The role of theta frequency in managing cognitive load by five- to six-year-old children.
Show/hide abstractResearch suggests that younger children respond to cognitively demanding stimuli less efficiently and at a slower rate than older children and adults (Albrecht, et al., 2009; McDermott, et al., 2007). Although often disregarded, our interest was the cognitive conflict associated with error responses. Theta frequency associated with handling cognitive load was assessed with 128-channel EGI net. Eriksen flanker paradigm (Eriksen & Schultz, 1979) was used to assess performance of five- and six-year-old children. Our aim was to identify the relationship between behavioural responses and theta power thought to be present in conflicting situations in teenagers and adults (McLoughlin, et al., 2014). Twenty children participated in experiment one. The task was an adaptation of Eriksen flanker paradigm used by McLoughlin and colleagues (et al., 2014) with congruent and incongruent trials. It proved highly demanding for children of these ages, generating much fewer responses compared to older children, i.e., overall below 50% of all the trials. Nonetheless, when analysed, theta oscillations were present on trials with the incorrect responses. Therefore, to ensure the larger response rate in experiment two, duration of the the target stimuli was increased from 150 to 700 msec, based on a study by McDermott, et al. (2007). The sample included fifteen children. As a result, the behavioural performance on the task improved. However, as in the experiment one, frontal medial theta oscillations were present both when processing the stimulus and when locked to the response, specifically for the cognitive conflict, i.e., errors in incongruent trials. Overall, our findings indicate that theta frequency is important in handling high cognitive load and it develops in childhood before reaching efficiency in adulthood.
Saturday 12.00 - 13.30 Grote zaal Poster #47 Clinical Jin Young Park
All authors & affiliationJin Young Park, Department of Psychiatry, Gangnam Severance Hospital, Yonsei University Health System & Institute of Behavioral Science in Medicine, Yonsei University College of Medicine Jooyoung Oh, Department of Biomedical Science and Engineering (BMSE), Institute of Integrated Technology (IIT), Gwangju Institute of Science and Technology (GIST) Dongrae Cho, Department of Biomedical Science and Engineering (BMSE), Institute of Integrated Technology (IIT), Gwangju Institute of Science and Technology (GIST) Jaesub Park, Department of Psychiatry, Gangnam Severance Hospital, Yonsei University Health System & Institute of Behavioral Science in Medicine, Yonsei University College of Medicine Se Hee Na, Department of Anesthesiology and Pain Medicine, Gangnam Severance Hospital, Yonsei University Health System Jongin Kim, Department of Biomedical Science and Engineering (BMSE), Institute of Integrated Technology (IIT), Gwangju Institute of Science and Technology (GIST) Jaeseok Heo, Institute of Behavioral Science in Medicine, Yonsei University College of Medicine Cheung Soo Shin, Department of Anesthesiology and Pain Medicine, Gangnam Severance Hospital, Yonsei University Health System Jae-Jin Kim, Department of Psychiatry, Gangnam Severance Hospital, Yonsei University Health System & Institute of Behavioral Science in Medicine, Yonsei University College of Medicine Boreom Lee, Department of Biomedical Science and Engineering (BMSE), Institute of Integrated Technology (IIT), Gwangju Institute of Science and Technology (GIST)
Detection of delirium by analyzing heart rate variability using Machine Learning Technique
Show/hide abstractDelirium is an acute confusional state where an individual displays impaired cognitive functioning and limited attention. This brain failure status also effects the autonomic nervous system. We performed this study to distinguish the heart rate variability (HRV) data of delirium from those of a non-delirious state using the support vector machine (SVM). This study was a part of the ICU Distress and Delirium Management (IDDM) project. All the patients were admitted in ICU, and Electrocardiography (ECG) data was acquired for 7 minutes per every 3 hours out of 24 hours during their ICU stay. We performed their type of classifications (state-, patient-, both patient and state-based classification) by using SVM. 140 ICU patients were included, and 39 patients were delirious. A total of 4797 HRV data sets were analyzed using SVM. Delirium was successfully indicated in all three types of classifications. In both patient-state-based classification, the accuracy rate was highest. Important HRV index were LF/HF, alpha1/alpha2, TINN. In this study, acceptable accuracy were included in the classification. HRV index analyzed by the machine learning technique seems to be a marker to detection the delirium.
Saturday 12.00 - 13.30 Grote zaal Poster #49 Decision-making Dilushi Chandrakumar
All authors & affiliationDaniel Feuerriegel, The University of South Australia Tobias Loetscher, The University of South Australia Stefan Bode, The University of Melbourne Megan Grech, The University of South Australia Hannah Keage, The University of South Australia
Event-Related Potentials in Relation to Risk-Taking: A Systematic Review
Show/hide abstractEvent-related potentials (ERPs) have been used for over three decades to investigate neural mechanisms underlying risk-related decisions. To date, there has been no systematic evaluation of risk-related decisions and ERP components. Sambrook and Goslin (2015) carried out a meta-analysis on selected studies (i.e., not systematic) of an early negative component, the feedback-related negativity (FRN), with respect to reward prediction. The current study aimed to systematically evaluate the associations between risk-related decisions and commonly reported ERP components; P3, FRN, error-related negativity (ERN), and medial frontal negativity (MFN). We assessed associations between risk-taking and these ERP components following decisions and following feedback. A total of 411 articles were identified from PsychINFO and PubMed, with 71 meeting inclusion criteria. Selected articles assessed the early (FRN, ERN, and MFN) and the mid-latency P3 ERP components, all using gambling paradigms that involved selecting between choices of varying risk (e.g. Iowa Gambling Task, Balloon Analogue Risk Task and two-choice gambling tasks). The P3 component was consistently enhanced to the selection of risky options and when positive feedback (as compared to negative feedback) was provided. Although less consistently, early components were generally found to be larger for losses as compared to gains, both immediately after response selection (i.e., decision-making) and in response to feedback. Early and mid-latency ERP components appear to index risk-related decision processes, and these findings can be contextualised within existing cognitive theories. However, existing studies have been restricted to simple financial decisions in gambling tasks, which does not capture the diversity of risky decisions made in other areas of everyday experience, for example, social, health and recreational risk-related decisions.
Saturday 12.00 - 13.30 Grote zaal Poster #50 Attention Dragan Rangelov
All authors & affiliationDragan Rangelov, Queensland Brain Institute, The University of Queensland, Australia * Jason B. Mattingley, Queensland Brain Institute and School of Psychology, The University of Queensland, Australia
Maintaining multiple attentional sets decreases the specificity of cognitive control
Show/hide abstractThe information-processing capacity of human cognition has long been debated in cognitive neuroscience. In contrast to the classical view, which postulated severe capacity limitations, recent research has shown that humans can concurrently implement and maintain multiple attentional sets – at least as many as two – at no cost. The precise mechanism enabling such “super-capacity” remains unclear. One explanation is that maintaining multiple attentional sets comes at the expense of decreasing specificity. This hypothesis predicts weaker distractor suppression as more attentional sets are maintained. To test this prediction, we developed a paradigm in which two superimposed patches of moving grey dots were presented for several seconds. Three times per trial the colour saturation of the dots gradually increased and then returned back to grey. While highly saturated, the two patches had different colours, making it possible to discern the motion direction of each patch. Two times per trial, one of the patches transitioned to a target colour. The task was to report the average motion direction of the two target-coloured patches. In different blocks of trials, either a single colour or two different colours served as targets. Consistent with recent studies, analyses of response accuracy showed no differences between the single- and the two-target colour conditions. Analyses of concurrently recorded EEG showed that event-related potentials (ERPs) time-locked to the onset of the coloured-dots epochs were virtually identical for the single and the two-target colour conditions in epochs with a target colour. By contrast, the ERPs in epochs without a target colour were stronger for the two target-colours than the single target condition. These findings suggest that distractor suppression is weaker as more attentional sets are maintained, consistent with the idea that maintaining multiple attention sets comes at the expense of decreasing specificity of cognitive control.
Saturday 12.00 - 13.30 Grote zaal Poster #51 Decision-making Carmen Kohl
All authors & affiliationCarmen Kohl, City University London Laure Spieser, City University London Bettina Forster, City University London Sven Bestmann, University College London Kielan Yarrow, City University London
Exploring Neural Correlates of the Decision Variable
Show/hide abstractIn sequential sampling models, perceptual decisions are made by accumulating sensory evidence until a decision threshold is reached. Although these models are based on behavioural data, the build-up of evidence or Decision Variable (DV) has recently been proposed to be reflected in several neural signals in man. We explored these potential neural substrates of the DV in the context of the Speed Accuracy trade-off (SAT). Sequential sampling models commonly account for behavioural differences associated with the SAT by adjusting a single parameter, namely the decision threshold. To explore whether these model predictions are supported by neural signals, we conducted experiments during which participants completed perceptual decision making tasks, while we tracked their neural signals using either electroencephalographic ERPs or TMS-evoked MEPs. We manipulated the strength of sensory evidence, as well as the speed/accuracy focus, and compared the resulting neurometric patterns. Results indicate that, in line with previous findings, the amplitude build-up of both signals was steeper in easy compared to hard trials. Since sequential sampling models assume that accumulation is steeper for easier decisions, this finding supports the role of these signals as a neural substrate of the DV. However, contrary to the predictions made by the models, neither of the signals displayed a difference between speed and accuracy conditions. Sequential sampling models typically explain the behavioural SAT through an adjustment of the decision threshold. We therefore expected an amplitude difference in the neural signals to reflect this criterion adjustment. Neither signal displayed this difference. Since both the ERP and the MEP signals show the expected differences for the difficulty, but not the SAT, manipulation, we conclude that the way the SAT as commonly been thought to occur via sequential sampling models is not consistent with the most promising current human neural analogues of this process.
Saturday 12.00 - 13.30 Grote zaal Poster #52 Memory Flavia Schechtman Belham
All authors & affiliationFlávia Schechtman Belham, Institute of Cognitive Neuroscience, University College London (UCL) Stephanie Hatzifilalithis, Institute of Cognitive Neuroscience, University College London (UCL) Leun J. Otten, Institute of Cognitive Neuroscience, University College London (UCL)
When time is of the essence: does exposure duration affect encoding-related activity before negative events?
Show/hide abstractPast research has revealed that memory encoding is influenced by brain activity before an event. Such anticipatory brain activity (ABA) is mediated by emotion, being present for negative but not for neutral images. This result can be explained by an evolutionary account, where negative events are more crucial to survival of an organism, therefore requiring quick processing and prompt responses. The current study tested how removing the urgency of fast stimulus processing affects ABA for negative images. We predicted that ABA would primarily be present when limited time is available to take in and process a negative event, in which case the need to prepare appropriately and engage anticipatory activity may be especially pertinent. Healthy adults had their brain activity measured via scalp-recorded EEG while incidentally encoding emotional images in an indoor/outdoor judgment task. Each image was preceded by a schematic cue indicating the upcoming valence (neutral or negative) and exposure duration (1000 or 2000 ms). Memory performance, measured with a remember/know recognition task, was better for long-duration items irrespective of valence. As expected, ABA was evident only for negative images. Surprisingly, however, this effect did not interact with duration. Positive-going activity over posterior scalp sites started shortly after cue onset and preceded negative images that were later remembered. This was accompanied by smaller negative-going activity over frontal sites. These patterns resemble previous findings in the literature and may respectively reflect emotional and semantic preparatory processes. We suggest that, due to the salience and potential importance of negative stimuli, anticipatory processes are engaged to optimise the processing of the event when it arrives. The findings that such anticipation was necessary even in a more relaxed time frame, and that neutral items did not elicit the effect, are further evidence for the crucial modulatory role of emotion on cognitive processing.
Saturday 12.00 - 13.30 Grote zaal Poster #53 Social behavior Wioleta Walentowska
All authors & affiliationWioleta Walentowska, Psychophysiology Laboratory, Institute of Psychology, Jagiellonian University in Krakow, Poland & Cognitive & Affective Psychophysiology Laboratory, Ghent University, Belgium Katharina Paul, Cognitive & Affective Psychophysiology Laboratory, Ghent University, Belgium Mario Carlo Severo, Cognitive & Affective Psychophysiology Laboratory, Ghent University, Belgium Agnes Moors, Cognitive & Affective Psychophysiology Laboratory, Ghent University, Belgium & Research Group of Quantitative Psychology and Individual Differences, Center for Social and Cultural Psychology, KU Leuven, Belgium Gilles Pourtois, Cognitive & Affective Psychophysiology Laboratory, Ghent University, Belgium
Show/hide abstractThe Stimulus Preceding Negativity (SPN) component reflects the anticipatory phase of reward-related outcome processing. Its amplitude is usually larger for informative compared to uninformative upcoming stimuli, as well as for uncertain relative to predictable ones. In this study, we sought to assess whether these two effects, when combined together in the same experimental design, produced synergistic or instead independent effects on the SPN during performance monitoring. Thirty participants performed a speeded Go/NoGo task while 64-channel EEG was recorded concurrently. We focused on the SPN activity generated in anticipation of feedback, which was either positive (for correct and fast reactions) or negative (for correct but slow responses). Further, feedback's informativeness was alternated across blocks. In Experiment 1, when negative feedback was more frequent than positive one (hence uncertainty was low) due to the use of a strict response deadline, the SPN amplitude (measured at fronto-central electrodes) did not vary as a function of feedback's informativeness or valence. By comparison, in Experiment 2, when negative and positive feedback were equiprobable (uncertainty was high), the SPN was more pronounced for informative compared to uninformative feedback. Interestingly, in this condition, it was also larger at right fronto-central sites for positive than negative one. These ERP results suggest that both factors – informativeness and uncertainty – combine and influence reward anticipation at the SPN level.
Saturday 12.00 - 13.30 Grote zaal Poster #54 Electrophysiology methods Claudia Rodella
All authors & affiliationJesus Cespón, Cognitive Neuroscience Section, IRCCS Centro San Giovanni di Dio Fatebenefratelli, Brescia, Italy Carlo Miniussi, Center for Mind/Brain Sciences CIMeC University of Trento, Rovereto, Italy Paolo Maria Rossini, Institute of Neurology, Policlinico A. Gemelli, Catholic University of Sacred Heart, Rome, Italy Maria Concetta Pellicciari, Cognitive Neuroscience Section, IRCCS Centro San Giovanni di Dio Fatebenefratelli, Brescia, Italy
tDCS-induced changes in physiological and pathological brain aging: a multimodal study
Show/hide abstractPhysiological and pathological brain aging have been suggested to be on a continuum in terms of synaptic dysfunctions, with consequent neuroplasticity alterations in Alzheimer's disease (AD). A challenge in neuroscience is to identify neurophysiological markers of neuroplasticity changes on the cortical areas firstly impacted by aging and AD, i.e. prefrontal cortex. Our study aimed to investigate the age- and AD-dependent neuroplasticity mechanisms at cortical and behavioral levels. Transcranial direct current stimulation (tDCS) was used as neuro-modulation technique over prefrontal cortex, whereas TMS-EEG recordings and working memory task were performed to monitor the tDCS-induced modulations on cortical and behavioral measures. Twelve healthy young and elderly subjects, and nine AD patients received separately anodal, cathodal and sham tDCS over the left dorsolateral prefrontal cortex. Before and after tDCS, a verbal n-back task was performed, with different difficulty for each group. Combining TMS-EEG during resting state, cortical reactivity was assessed over prefrontal cortex. Although interesting tDCS polarity-dependent changes were observed in task performance, they did not reach the statistical significance. Nevertheless, group- and tDCS polarity-dependent cortical changes were observed. Specifically, a cortical reactivity increase after cathodal and a decrease after anodal tDCS in young subjects were observed. Opposite cortical patterns were found in elderly subjects, with a cortical reactivity increase after anodal and a decrease after cathodal tDCS. Finally, a decreased cortical reactivity after both anodal and cathodal tDCS was observed in AD patients. Our findings demonstrate that the cortical modulations induced by tDCS are not linear during physiological and pathological brain aging. Empirically measuring cortical plasticity in aging may provide critical predictors and early physiological markers of cognitive decline and might represent the starting point for the development of neuro-rehabilitation strategies.
Saturday 12.00 - 13.30 Grote zaal Poster #55 Decision-making Leendert van Maanen
All authors & affiliationLeendert van Maanen, University of Amsterdam Dimitris Katsimpokis, University of Amsterdam Guy E Hawkins, University of Amsterdam & University of Newcastle
Cue-induced vs deadline-induced speed-accuracy trade-off have different effects on the relationship between RT and accuracy
Show/hide abstractMany studies aimed at understanding how decision making is affected by time pressure employ a so-called speed-accuracy trade-off paradigm. Typically, this is induced by providing participants with a cue that states they should focus either on accurate responding, or fast responding. Another often used manipulation is providing participants with a response deadline. While the speed-accuracy trade-off paradigm has led to a better understanding of the cognitive and neural mechanisms of decision making under time pressure, in the current study we asked the question whether these two manipulations to induce a speed-accuracy trade-off are truly comparable. In a series of choice experiments, we applied both the cue and deadline manipulations in a full factorial design. The results show that (1) both manipulations increase error rate, and decrease response time, but that the presence of a deadline (2) reduces the positive skew of the response time distribution and (3) affects the distribution of errors throughout the RT distribution: Where the cue-based speed-accuracy trade-off manipulation does not induce a dependence of accuracy on RT, the deadline manipulation results in an increased proportion of slow errors. These results have theoretical implications for cognitive models of choice that implement decision making as an evidence accumulation process towards a threshold value. In particular, consistent with previous work, the cue manipulation seems to affect the overall level of the threshold. In contrast, the deadline manipulation seems to affect the speed with which participants adjust this threshold within a trial, sometimes referred to as decision urgency.
Saturday 12.00 - 13.30 Grote zaal Poster #56 Language Jona Sassenhagen
All authors & affiliationChristian J. Fiebach
Direct Encoding of Meaning in Brain Waves
Show/hide abstractCan the meaning of a word be decomposed into multiple 'atoms of meaning' (Churchland, 1989)? Or are words atomic, i.e., themselves the smallest units of meaning (Fodor & Leopore, 1999)? Brain patterns should be able to distinguish between these options. As of now, however, neural correlates of meaning processing usually reflect not the inherent semantics of the word, but how hard it is to process the word given its context. Thus, brain correlates of semantic processing have typically been interpreted either as reflecting the deviance from a context-dependent expectation (Lau et al., 2008) or the difficulty of integrating a new word into its sentence context (Hagoort et al., 2005). Studying the brain bases of meaning as such requires tracing the neural correlates of encoding the meaning of a word itself, not just how well this meaning matches a context. For this purpose, we employ a neural network (Mikolov, 2013) to learn high-dimensional (vector) representations of word meanings and construct a projection from 'meaning space' to 'brain space'. This allows us to simulate what the brain activity following any word should look like, given its (vector-space) meaning. We compare simulated to actual brain waves (64-channel EEG) from word list and sentence processing experiments . Given the stimulus' vectorized meaning representation, we can simulate brain patterns that strongly resemble (r ~.08, p<.001) actual brain activity, restricted to a narrow window between 350-450 msec. That is, in this time window, meaning itself is encoded in brain activity. We also show that the vector-based (context-invariant) meaning is a better predictor of brain activity than context-dependent factors (i.e., congruency with sentence context). We conclude that brains indeed reliably encode word meaning in a context-invariant fashion, and that the power of vectorized meaning representations to predict brain activity supports 'decompositionalist', not 'atomistic' models of meaning representation in the brain.
Saturday 12.00 - 13.30 Grote zaal Poster #57 Memory Sanne ten Oever
All authors & affiliationSanne ten Oever, Department of Cognitive Neuroscience, Faculty of Psychology and Neuroscience, Maastricht University Peter de Weerd, Department of Cognitive Neuroscience, Faculty of Psychology and Neuroscience, Maastricht University Alexander Sack, Department of Cognitive Neuroscience, Faculty of Psychology and Neuroscience, Maastricht University
Theta phase modulation of working memory maintenance and behavioral performance
Show/hide abstractIt has been proposed that working memory content is maintained through a cyclic repetition of to-be-remembered information in a 4-8 Hz (theta) frequency [1, 2]. In line with these theories, single unit recording studies in visual and prefrontal cortex of monkeys demonstrated that stimulus information is best represented at specific theta phases [3]. In human EEG and MEG, it has been proven difficult to measure the content of information maintained in working memory. Recently, however, distributed codes related to maintained information have been revealed by presenting a task-irrelevant, high-intensity impulse stimulus during the retention interval of a working memory task [4, 5]. We used this latter amplification method to investigate phase dependent modulation of maintained information in human EEG. A multivariate classifier was trained to determine the orientation of a to-be-remembered grating. Remarkably, memory amplification during retention revealed that decoding accuracy was modulated strongly by the theta and alpha phase at the onset of the impulse stimulus. Theta, but not alpha, phase also modulated the accurateness of memory performance. This result is consistent with the proposed function of theta in organizing maintained information, while alpha relates to inhibition of non-relevant stimuli and processes [2]. Importantly, the specific theta phase of maximal behavioral performance matched the phase at which memory decoding performance was highest. We here provide first concrete evidence that the amplification of memory content is retained within a restricted range of theta phases, and amplifying working memory at an optimal rather than a non-optimal phase enhances memory performance.
Saturday 12.00 - 13.30 Grote zaal Poster #58 Attention Alexander Luettich
All authors & affiliation(First author: Alexander Luettich; Oxford Centre for Human Brain Activity, University of Oxford) Ryszard Auksztulewicz; Oxford Centre for Human Brain Activity, University of Oxford; Wellcome Trust Centre for Neuroimaging, University College London Celine R. Gillebert; Oxford Centre for Human Brain Activity, University of Oxford; Department of Brain and Cognition, University of Leuven Kia Nobre; Oxford Centre for Human Brain Activity, University of Oxford
Neural mechanisms of spatial and temporal orienting in aging
Show/hide abstractRecent research has tried to delineate in how far the mechanisms of spatial and temporal orienting can be preserved with aging. Specifically, it has been claimed that older adults cannot use temporal information to improve performance, which was supported by neural evidence suggesting a lack of pre-target temporal orienting (Zanto 2011). In addition, while older adults were shown to exhibit most of the evoked markers of spatial orienting established in younger adults, evidence for cue-induced oscillatory lateralisation is still contradictory (Hong 2015, Leenders 2016). We developed a multimodal attention study, where both spatial and temporal information were manipulated. Specifically, we combined behavioural testing with electroencephalographic recording (EEG) and functional magnetic resonance imaging (fMRI) and tested 24 healthy elderly participants. We show that participants could benefit behaviourally from both spatial and temporal information. EEG analysis yielded significant effects on previously established cue (CNV, ADAN, LDAP) and target (N1 and P3 amplitude; P3 latency shift) evoked and induced (alpha lateralisation) modulations of attention. These results could be supported by regional evidence from fMRI, differentiating task positive and resting state networks. We confirm that healthy elderly participants can engage in preparatory spatial orienting and extend recent behavioural findings on spared temporal orienting with aging (Chauvin 2016) to the neural domain. The study serves as a baseline for a second branch of research, where we apply the paradigm to age-matched stroke survivors, assessing whether neural and behavioural signatures of spatial and temporal orienting are affected by focal subcortical and cortical lesions.
Saturday 12.00 - 13.30 Grote zaal Poster #59 Electrophysiology methods Shankha Sanyal
Show/hide abstractThe Indian Classical Music has long been known to evoke a wide range of emotional responses in the minds of the listeners, which can now be quantified with the help of biosensors like EEG and robust non-linear analysis techniques. The raga in Indian music is a sequence of musical notes and the play of sound which delights the hearts of people conveying a variety of emotional appraisals just by mere permutation/combination of notes. In this study we took a novel approach to study the brain response from two ragas (Chayanat and Darbari Kanada) which conventionally portray contrast emotions – joy and sorrow. EEG was taken from 20 respondents who were made to listen to 2 min clips of the two ragas with 2 min ‘resting period' in between and before the start. The obtained EEG data from 19 different electrodes were amplified to the same sampling as that of the sound signal data i.e. 44.1 kHz, so that they can be played as audio signal. On simple playing, distinct difference in tempo can be perceived in the EEG signals of the “rest state” and the “with music” state even by a layman i.e. there is significant difference in the amplitude of the two EEG signals. Next, MFDXA (Multifractal Detrended Cross Correlation Analysis) method was employed to study the degree of cross correlation between the EEG signals and the raw sound signal to assess the degree of correlation between the two. Interesting, considerable increase in the degree of cross-correlation was found in case of both the clips in comparison to the rest state. For, the first time a direct correlation between the source sound signal and the output EEG signal is reported which may have far reaching implications in the domain of neuro-cognitive processing of musical emotions by sonification of EEG data.
Saturday 12.00 - 13.30 Grote zaal Poster #60 Sensory processing Gustavo Garcia da Palma
All authors & affiliationGustavo Garcia da Palma, University of Sao Paulo
States of poetic presence mapped through Electroencephalography (EEG) and Oximeter (BPM) and artificial neural networks.
Show/hide abstractIn a correlation between procedures of acting and a review of the role of the emotions in actor's work using Affective Neuroscience based on Jaak Pankseep's theories, this study uses an Artificial Neural Network (ANN), to classify states of poetic presence from data collected during live performance using a low cost Electroencephalographic technology (EEG, Emotiv) and a heart rate sensor (ECG, Arduino). This paper will present an interface which implements the ANN as an example of how EEG and ECG can be used to study biometric patterns of actor's work offering a possible view to understand how works the performer's decision making process and how they create emotions on stage.
Saturday 12.00 - 13.30 Grote zaal Poster #61 Cognitive modeling Rico S. C. Lee
All authors & affiliationRico S. C. Lee1, Jeggan Tiego1, & Murat Yücel1 1Brain and Mental Health Laboratory, Monash University, Australia
Show/hide abstractA transdiagnostic, dimensional approach to mental health research is uniquely positioned to disentangle the underlying factors that contribute to illness and associated disability. Key to this research strategy is the interrogation of shared constructs with clear links to biology, which helps to develop unifying models that are explanatory in nature and inform novel therapeutics targeting specific neurocircuitry. To this end, the dimensions of impulsivity and compulsivity are commonly found across mental and substance use disorders (eg, obsessive-compulsive disorder, substance and behavioral addictions), and have been associated with clinical and functional outcomes across a spectrum of disorders. Accordingly, the current study sought to examine the heritability and comorbidity of constructs underpinning impulsivity and compulsivity (eg, tolerance of uncertainty, obsessional thinking) using a detailed assessment battery in a sample of Australian twins to estimate the influence of genetics, as well as shared and non-shared environmental factors on the prevalence of impulsive and compulsive behaviours in the general community. In total, 514 participants (68% female) aged 18 to 55 years old (M = 39.8; SD = 10.5) were recruited and included in the final analysis (134 MZ and 123 DZ twin pairs; 14.4% fraternal). Latent variable structural equation modeling was used to estimate the heritability and comorbidity of these cognitive-affective phenotypes and ascertain how these independently contribute to clinical symptoms (eg, compulsive gambling), functional impairment (eg, unemployment), and subjective quality of life (eg, satisfaction with social relationships). Moderator analyses were conducted to examine the impact of demographic factors (eg, age, sex). Findings from the present study will quantify the genetic and environmental basis of complex traits across common mental disorders and disentangle the unique contributions of these phenotypes to burden of disease.
Saturday 12.00 - 13.30 Grote zaal Poster #62 Clinical Judith Goris
All authors & affiliationJudith Goris, Senne Braem, Annabel Nijhof, Davide Rigoni, Eliane Deschrijver, Sander Van De Cruys, Roeljan Wiersema, & Marcel Brass
Early sensory prediction errors are less adjusted by global context in autism spectrum disorder
Show/hide abstractBackground: Recent theories on autism spectrum disorder (ASD) suggest that a key deficit in ASD concerns the inflexibility in modulating low-level sensory predictions as a function of global top-down expectations. As a direct test of this central hypothesis, we used electroencephalographic recordings to investigate whether processing of early sensory prediction error was less sensitive to global context in ASD. Methods: A group of 24 adults with ASD was compared to a gender- and age-matched group of 24 typically developing (TD) adults on a well-validated hierarchical predictive coding paradigm. In this auditory oddball task, participants listened to short sequences of either five identical sounds or four identical and a fifth deviant sound. The latter condition is known to generate the mismatch negativity (MMN) component, believed to reflect early sensory prediction error processing in a pre-attentive and non-conscious manner. Crucially, we manipulated the relative frequency of deviant sound sequences across blocks, as previous studies have shown that in blocks with frequent deviant sound sequences, top-down expectations seem to attenuate the MMN. We expected that this top-down modulation of the MMN would be smaller in the ASD group, in line with the hypothesis of inflexible context-dependent weighting of prediction error in ASD. Results: Both groups showed a MMN that was modulated by global context. However, this effect was significantly smaller in the ASD group as compared to the TD group. In contrast, the P3b, as an electroencephalographic marker of conscious expectation processes, was similar across groups, suggesting comparable top-down expectations. Conclusions: Our results demonstrate that people with ASD are less flexible in modulating their low-level sensory predictions, despite showing similar levels of global expectations, thereby confirming the central hypothesis of contemporary predictive coding accounts of ASD.
Saturday 12.00 - 13.30 Grote zaal Poster #63 Social behavior Mohammad Saleh Torkestani
All authors & affiliationMohammad Saleh Torkestani, Saeid Pourroostaei Ardakani, Zohreh Hashemi Allameh Tabataba’i University, Tehran, Iran
How tourism picture’s language and crowd could influence on effectiveness of visual advertisements (Case of Iran)
Show/hide abstractexamine the effect of language and the number of boards in some pictures on two dependent variables in order to find out what elements can significantly affect purchasing decision. We tried to figure out common purchasing patterns which led to seeking for a specific kind of information so as to buy a product. To discover the most eye-catching variables for Iranian people, we have used tourism pictures to find out whether Iranian people, who do not normally see advertisements in foreign language in their daily lives, would react to it or not. For accomplishing this research, we asked 28 participants to fill out a questionnaire after watching 22 photographs in a self-paced mood with an eye-tracking device named Tobii which can indicate data in form of time (average total visit duration) and a heat map. Findings confirm that the manipulated variables were relatively accepted. Although languages used in our pictures were equal in attracting the attention of the viewers (H1), they were partially successful in providing useful information for making travel decisions (H2) the various number of texts used in pictures differs in catching attention. It means that as the number of textual messages increases in each photograph, the amount of time being spent enhances significantly (H3). Also, multiple board pictures were more prosperous in stimulating the viewers’ interest to know more about the destination (H4). For Iranian, a foreign language would attract the same amount of attention as a native language. However, the extent they understand a language can provide them useful information about the place being advertised. Furthermore, pictures with more store signs can attract more attention as they provide more information about the destination.
Saturday 12.00 - 13.30 Grote zaal Poster #64 Electrophysiology methods Liqing Liu
All authors & affiliationLiqing Liu 1,2, Nils Rosjat1,2, Svitlana Popovych1,2, Azamat Yeldesbay1,2, Bin A. Wang2, Christian B. Grefkes2,3, Gereon R. Fink2,3, Silvia Daun1,2 1Heisenberg Research Group of Computational Neuroscience - Modeling Neural Network Function, Department of Animal Physiology, Institute of Zoology, University of Cologne 2Cognitive Neuroscience, Institute of Neuroscience and Medicine (INM-3), Research Centre Ju¨lich 3Department of Neurology, University Hospital Cologne
Age-related dynamic changes in human motor network connectivity
Show/hide abstractMotor actions are based upon complex interactions between several brain regions. These regions form distinct functional networks depending on the specific properties of the action. How normal ageing affects these functional motor networks and the complex assembly process to date remains poorly understood. In our previous study we found that older subjects compared to young ones exhibited a significant smaller event-related desynchronization (ERD) in the beta frequency band around movement onset in the regions of the motor cortex while the subjects performed a simple visually-guided motor task. In the current study, we re-analyzed electroencephalography (EEG) data from 18 young (11F, 22-35 years) and 24 elderly (12F, 60-78 years) right-handed healthy subjects as they were performing the same simple visually-guided motor task in order to unravel network effects which might be responsible for the decreased ERD observed in the older subjects. The task required participants to execute a left or right index finger button press triggered by a visual cue either directing to the left or right. We used Dynamic Causal Modeling (DCM) of induced responses to model dynamic changes in inter- and cross-frequency couplings between the regions of the motor, frontal and parietal cortex during the execution of the task, for both groups. The fixed-effects Bayesian model selection revealed the same winning model in both young and older subjects, which suggests that the main motor network connectivity pattern is preserved in the aging brain. However, the beta-beta frequency coupling strength significantly decreased in the older subjects in the following pathway: ipsilateral prefrontal cortex (PFC) - contralateral prefrontal cortex (RPFC) - supplementary motor area (SMA) - contralateral primary motor cortex (M1) - ipsilateral primary motor cortex (M1). We furthermore found decreased beta-beta coupling between the contralateral premotor cortex (PMC) and contralateral M1, as well as decreased beta-alpha coupling between SMA and ipsilateral M1. Our results add onto the current literature on event-related synchronization and desynchronization by revealing diminished or even partly vanished coupling pathways in the beta and alpha frequency bands as possible causes for the changes in ERD in older subjects.
Saturday 12.00 - 13.30 Grote zaal Poster #65 Sensory processing Daniel Gramm
All authors & affiliationDaniel Gramm 1, Martin Dietz1, Kristian Sandberg1, and Morten Overgaard1. Center of Functionally Integrative Neuroscience, Department of Clinical Medicine, Aarhus University.
Functional Localisation of the Ebbinghaus Illusion in Human Primary Visual Cortex
Show/hide abstractVisual illusions expose a discrepancy between the physical and perceived properties of a stimulus. This discrepancy can be exploited to reveal neuronal processes involved in visual perception. In this study, we replicate the finding that individual differences in central primary visual cortex (V1) surface area is correlated with visual illusion strength in the Ebbinghaus illusion. We used Population Receptive Field (pRF) mapping to measure the central V1 surface area (N = 24) and found a significant correlation with illusion strength induced by both a large surround and a small surround. We also found a significant correlation between central V1 surface area and mean pRF size which indicates that smaller V1s, in general, employ larger receptive fields to cover the visual field as efficiently as larger V1s. V1 functional Magnetic Resonance Imaging (fMRI) data was also acquired while the participants viewed the Ebbinghaus illusion and its constituent parts (central circle and surrounding circles). This allowed us to project the BOLD fMRI data into the previously created pRF model to investigate the between-subject activity in visual pRF-space. The resulting maps revealed an expected increase in activity corresponding to the location of the increased perceived size of the central circle from the behavioural results. Surprisingly, we also saw increased activity corresponding to the location of the surrounding circles. Our findings suggest that previous work attempting to model the Ebbinghaus illusion solely in terms of the effects of the central circle, might fail to capture important aspects of the illusion stemming from the surrounding circles.
Saturday 12.00 - 13.30 Grote zaal Poster #66 Attention Anderson Mora-Cortes
All authors & affiliationAnderson Mora-Cortes1, 2,3 K. Richard Ridderinkhof1,2, Michael X Cohen4 1 Department of Psychology, University of Amsterdam, Amsterdam, The Netherlands. 2 Amsterdam Brain & Cognition (ABC), University of Amsterdam, Amsterdam, The Netherlands. 3 Laboratory of Neuro- and Psychophysiology, Department of Neuroscience, KU Leuven, Leuven, Belgium. 4 Donders Center for Neuroscience, Radboud University Nijmegen Medical Centre, Radboud University, Nijmegen, Netherlands.
The steady-state visual evoked potential (SSVEP) and temporal attention
Show/hide abstractImprovements in perceptual performance can be obtained when events in the environment are temporally predictable. And temporal predictability improves attention and sensory processing. The amplitude of the steady-state visual evoked potential (SSVEP) has been shown to correlate with attention paid to a flickering stimulus even if the flickering stimulus is irrelevant for the task. However, to our knowledge the validity of the SSVEP to study temporal attention has not been established. Therefore, we designed an SSVEP- temporal attention task to evaluate whether the SSVEP and its temporal dynamics can be used to study temporal attention. We used a forced-choice perceptual detection task while presenting task-irrelevant visual flicker at alpha (10 Hz) and two surrounding frequencies (6 or 15 Hz). Temporal predictability was manipulated by having the inter-stimulus intervals (ISI) be constant or variable. Behavioral results replicated previous studies confirming the benefits of temporal expectations on performance for trials with constant ISI. EEG analyses were performed in combination with an optimal spatial filtering method and revealed robust SSVEP amplitudes for all flicker frequencies, even though a main effect of temporal expectations on SSVEP amplitude was not significant. Additional analyses revealed temporal predictability-related modulations of SSVEP amplitude at 10 Hz and its second harmonic (20 Hz). The effect of temporal predictability was also observed for the 6 Hz flicker and its second harmonic (12 Hz). Notably, the effect of temporal predictability was not observed for the 15 Hz flickering for any ISI condition. These results illustrate modest feasibility of the SSVEP technique to study temporal attention for stimuli with flicker frequencies around alpha.
Saturday 12.00 - 13.30 Grote zaal Poster #67 Attention Jason Baer
All authors & affiliationJason Baer Additional authors: Rajal G. Cohen Affiliations: Mind in Movement Lab, University of Idaho
Posture vs. Performance in a Dual-Task Paradigm
Show/hide abstractMaintaining upright sitting posture may help reduce or prevent neck pain. Biofeedback interventions, which draw attention to postural feedback cues, raise the question of a possible trade-off between attention to posture and attention to task performance, such that maintaining upright posture may negatively affect task performance. To investigate this possible trade-off, we assessed upright posture and score on a 10-minute computer game, played with and without biofeedback. To further investigate the possible influence of attention, we also collected data on cognitive factors known to affect attention: inhibitory control, mindfulness, and impulsivity.
Saturday 12.00 - 13.30 Grote zaal Poster #68 Sensory processing Anna Gaglianese
All authors & affiliationAnna Gaglianese1,2 Francisco Guerreiro Fernandez2,3 Alessio Fracasso1,4 Natalia Petridou1 1. Department of Radiology, Center for Image Sciences, University Medical Center Utrecht, Utrecht, The Netherlands 2. Department of Neurosurgery and Neurology, Brain Center Rudolf Magnus, University Medical Center Utrecht, Utrecht, The Netherlands 3. Instituto de Biofísica e Engenharia Biomédica, Faculdade de Ciências, Universidade de Lisboa, Portugal 4. Spinoza Center for Neuroimaging, Amsterdam Zuidoost, Netherlands
Spatial and temporal frequency selectivity in the human middle temporal cortex measured by 7T fmri
Show/hide abstractVisual motion processing has a key role in our daily life and it is processed by a core motion-sensitive area of the visual cortex: the human middle temporal complex hMT+. We recently showed, using electrocorticography, that some hMT+ neuronal populations exhibited separable selectivity for spatial and temporal frequencies to moving visual stimuli rather than speed tuning. It remains elusive whether this selectivity comprises a spatial organization within hMT+. Here, we used 7T fMRI to investigate the selectivity of hMT+ responses to different combinations of spatial and temporal frequency components of visual motion stimuli. fMRI data were acquired from four subjects on a Philips 7 T scanner using two 16-channel surface coils and a gradient echo (GE) EPI sequence (voxel size of 1.43 x 1.43 x 1.50 mm, TR/TE = 849/27 ms, SENSE factor = 2, 15 coronal slices covering hMT+). A functional localizer scan was acquired first, using a black and white dartboard moving at 5Hz alternating between moving and stationary period. The main experiment consisted on five runs of high-contrast black and white dartboards with different spatial and temporal frequency combinations (0.33 cycle/deg;1Hz, 0.33 cycle/deg; 3Hz, 0.33 cycle/deg; 5Hz, 0.2 cycle/deg; 3Hz, 1 cycle/deg; 3Hz). For each run the dartboard pattern expanded for 1s alternating with stationary periods of variable length. fMRI data were motion corrected in AFNI, without spatial smoothing. For each subject, hMT+ was functionally defined based on the contrast between moving and stationary dartboards. For each selected voxel and each spatial and temporal frequency condition (i.e. run) we computed the hemodynamic response function (HRF) using a deconvolution approach. Our results showed that hMT+ separates motion into its spatial and temporal components rather than decoding speed directly. Moreover, we detected clusters of activity for particular combinations of spatial and temporal frequencies suggesting a spatial organization within hMT+.
Saturday 12.00 - 13.30 Grote zaal Poster #69 Social behavior Dana Schneider
All authors & affiliationKathleen Kang, School of Psychology, the University of Nottingham, Nottingham, UK Stefan R. Schweinberger, Institute of Psychology, Friedrich Schiller University of Jena, Jena, Germany Peter Mitchell, School of Psychology, the University of Nottingham, Nottingham, UK
Dissociating neural signatures of mentalizing and classifying based on facial expressions
Show/hide abstractThe aim of this study was to identify neural correlates of mental state processing as participants viewed emotional facial expressions. Posed facial expressions have often been used as stimuli in mentalizing tasks, but there is no objective answer as to what the actor is actually experiencing (West and Kenny, 2011). Furthermore, even if perceivers are able to infer mental states, it is unsure if they are mentalizing or classifying their responses. Thus, this study aims to address these two issues. Here, we recruited target participants that viewed various emotional expressions (i.e., happiness, surprise, anger, disgust, neutral) as they were being recorded covertly. One group of perceivers was subsequently asked to choose the expression which best represents the target's expression out of a set of alternatives (classification), whereas another group of perceivers was asked to guess which expression the target was looking at out of a set of alternatives (mentalizing). Classification perceivers compared to mentalizing perceivers showed more prevalent activations at earlier ERP components N170, P200 and P300-600. Interestingly, at late components (N800-1000), an interaction between hemisphere and group for the frontal, fronto-temporal and temporal electrodes reflected more sustained activity over the right hemisphere than the left hemisphere for the mentalizing group. Hence, classifying of, and mentalizing about, facial expressions are accompanied by distinct patterns of brain activity.
Saturday 16.00 - 17.30 Grote zaal Poster #1 Developmental Moritz Koester
All authors & affiliationMoritz Koester, University of Osnabrueck Johanna Castel, University of Osnabrueck Thomas Gruber, University of Osnabrueck Joscha Kaertner, University of Muenster
Visual Cortical Networks Involved in Scene Perception Align with Visual Cognition in Early Childhood
Show/hide abstractThis study investigates how visual cortical networks align with inter-individual differences in visual cognition, namely the sensitivity for the object versus the background of a visual scene, in early childhood. To segregate object and background processing in the visual cortex in 5- and 7-year-old children, object and background were presented at different frequencies (12Hz or 15Hz), evoking disparate neuronal responses (steady state visually evoked potentials, ssVEPs) in the electroencephalogram (EEG). Furthermore, visual cognition was assessed in verbal picture descriptions and recognition memory tasks. In younger children compared to older children the background elicited higher ssVEPs. Visual cortical processing of object versus background were associated with visual cognition for older but not for younger children. This relation was strongest for verbal descriptions and generalized to the cortical processing of abstract stimuli and object and background presented alone. Thus, visual cortical networks restructure and align with acquired visual cognition in early childhood.
Saturday 16.00 - 17.30 Grote zaal Poster #2 Sensory processing Ruth Elijah
All authors & affiliationRuth Elijah, UNSW Sydney* Mike Le Pelley, UNSW Sydney Thomas Whitford, UNSW Sydney
Modifying the neural expectation that sensations follow immediately from self-initiated actions
Show/hide abstractAn important distinction that our perceptual system makes is between sensations that are self-generated versus those that are generated by others or the external world. In order to make this distinction, a prediction is made as to the anticipated sensory consequences of a self-initiated action, and the resultant sensation is attenuated. A fundamental prior assumption which is critical to making sensory predictions is that any resultant sensation should occur immediately after a self-initiated action. The current study aimed to determine whether this temporal expectation could be modified with repeated exposure to a self-initiated auditory sensation which was delayed relative to the initiating action. Participants underwent electroencephalographic (EEG) recordings while undergoing a task where they pressed a button to produce a tone. The onset of the tone occurred either immediately or 100ms after the button press. Training consisted of repeated exposure to either an immediate tone following each button-press, or to a delayed tone. Across two experiments it was shown that, prior to training, delayed tones evoked a larger N1 amplitude in the auditory-evoked potential compared to immediate tones, suggesting that the delayed tone was processed as more salient than the immediate tone. However, there was a significant decrease in N1 amplitude across training to a delayed tone, such that in a post-training test there was no longer a difference in N1 amplitude between the immediate and delayed tones. Furthermore, there was no significant change in N1 amplitude across training to an immediate tone, such that in this condition the significantly larger N1 amplitude for delayed tones compared to immediate tones remained post-training. These findings suggest that it is possible to modify the neural assumption that sensations follow immediately from actions. This opens the possibility that behavioural training might provide a viable avenue of treatment in people with sensory suppression deficits, such as has been consistently shown in schizophrenia.
Saturday 16.00 - 17.30 Grote zaal Poster #3 Connectivity/networks Marzieh Zare
All authors & affiliationZahra Rezvani, Institute for Cognitive and Brain Sciences, Shahid Beheshti University G.C., Tehran Marzieh Zare, School of Computer Science, Institute for Research in Fundamental Sciences (IPM), Tehran, Iran Melle Van der Molen, Institute of Psychology, Leiden University, The Netherlands, Leiden Institute for Brain and Cognition, Leiden University, The Netherlands Gojko Zaric, Department of Cognitive Neuroscience, Faculty of Psychology and Neuroscience University of Maastricht, The Netherlands Milene Bonte, Department of Cognitive Neuroscience, Faculty of Psychology and Neuroscience University of Maastricht, The Netherlands Jurgen Tijms, Rudolf Berlin Center, Amsterdam, The Netherlands Leo Blemort, Department of Cognitive Neuroscience, Faculty of Psychology and Neuroscience, University of Maastricht, The Netherlands Cornelis Jan Stam, Department of Clinical Neuropsychology and MEG Center, Neuroscience Campus Amsterdam, VU University Medical Center, Amsterdam, The Netherlands Maurits van der Molen, Department of Psychology, University of Amsterdam, The Netherlands Amsterdam Brain and Cognition, University of Amsterdam, The Netherlands Gorka Fraga Gonzalez, Amsterdam Brain and Cognition, University of Amsterdam, The Netherlands
Automatic classification of children with dyslexia using a support-vector machine and EEG resting-state connectivity
Show/hide abstractObjectives: Machine learning methods have been applied to a variety of high-dimensional data and could be used to assist clinical diagnosis. Deploying a machine-learning classifier, we aimed at increasing our accuracy in assigning children to either typical readers vs. children with dyslexia using EEG. Methods: We used EEG resting-state data of 29 dyslexics and 15 typical readers in grade 3. We calculated weighted connectivity matrices for multiple frequency bands using the phase lag index (PLI). From the connectivity matrices, we derived weighted connectivity graphs. A number of local network measures were computed from those graphs, and 27 False Discovery Rate (FDR) corrected features were selected as input to a Support Vector Machine (SVM) and a K Nearest Neighbors (KNN) classifier. In addition, cross validation was employed to assess the machine performance and random shuffling was performed to affirm that the performance of the classifier was appropriate. Results: Best performance was for the SVM using a polynomial kernel. Children were classified into the two groups with about 84% accuracy and 84% sensitivity. Conclusion: The automatic classification techniques were shown to be both robust and reliable and, importantly, to be applicable in distinguishing between typical and dyslexic readers. Significance: The ability to identify children with dyslexia using "automatic classification" strategies is a novel convergent approach that may facilitate early detection and improved characterization of similar developmental disorders. Keywords: EEG; Machine Learning; Support Vector Machine (SVM); KNN classifier; Local Network measures; Dyslexia; Children
Saturday 16.00 - 17.30 Grote zaal Poster #4 Consciousness Darinka Trubutschek
All authors & affiliationDarinka Truebutschek, [1] Ecole des Neurosciences de Paris Ile-de-France, 15 rue de l'ecole de medecine, 75006 Paris, France; [2] Ecole Doctorale Cerveau-Cognition-Comportement, Universite Pierre et Marie Curie, 4 Place Jussieu, 75005 Paris, France; [3] Cognitive Neuroimaging Unit, CEA DSV/I2BM, INSERM, Universite Paris-Sud, Universite Paris-Saclay, NeuroSpin center, 91191 Gif/Yvette, France Sebastien Marti, [1] Cognitive Neuroimaging Unit, CEA DSV/I2BM, INSERM, Universite Paris-Sud, Universite Paris-Saclay, NeuroSpin center, 91191 Gif/Yvette, France Stanislas Dehaene, [1] Cognitive Neuroimaging Unit, CEA DSV/I2BM, INSERM, Universite Paris-Sud, Universite Paris-Saclay, NeuroSpin center, 91191 Gif/Yvette, France; [2] College de France, 11 Place Marcelin Berthelot, 75005 Paris, France
The limits of non-conscious working memory
Show/hide abstractRecent reports have challenged the notion that working memory requires conscious awareness and sustained delay activity. However, these studies exclusively focused on the short-term (< 15s) non-conscious maintenance of information, without considering the hallmark of working memory: the ability to manipulate mental representations. We here combined visual masking with magnetoencephalography to interrogate this feature in the context of non-conscious working memory and dissect its neural mechanisms. In a spatial delayed-response task, participants reported the location of a subjectively unseen target much better than chance after a 3s-delay. Importantly, they continued to do so, even when asked to rotate the original location. Ongoing MEG analyses revealed that this mental rotation required consciousness: At the time of the presentation of the rotation cue halfway through the delay period, previously identified signatures of conscious working memory emerged on the unseen trials, including a sustained desynchronization in the alpha/beta band over frontal cortex. In addition, response location could only be decoded intermittently in posterior sensors during this period. Our findings are thus consistent with recent proposals of ‘activity-silent' working memory, during which slowly decaying synaptic changes allow cell assemblies to go dormant during the delay, yet be retrieved above chance-level after several seconds. However, they contradict the notion that all aspects of working memory can operate non-consciously and rather suggest that non-consciously maintained information may be willfully retrieved to perform complex conscious tasks. Future research should further delineate the boundary conditions of non-conscious working memory.
Saturday 16.00 - 17.30 Grote zaal Poster #5 Anatomy Kirsten Hoetting
All authors & affiliationKirsten Hoetting, Universitaet Hamburg, Biological Psychology and Neuropsychology Cordula Hoelig, Universitaet Hamburg, Biological Psychology and Neuropsychology Brigitte Roeder, Universitaet Hamburg, Biological Psychology and Neuropsychology
Physical exercise prevents hippocampal volume loss in middle-aged adults
Show/hide abstractPhysical exercise has beneficial effects on cognitive functioning and has been suggested as a means to prevent age-related decline. For example, one year of aerobic exercise in older adults has been found to counteract age-related volume loss in the hippocampus. However, studies in young or middle-aged humans are currently rare. In the present study, sedentary men and women between 40 and 55 years of age were randomly assigned to either an aerobic exercise group (cycling) or a control group (stretching and coordination). Both groups exercised twice a week for six months. Before and after the intervention, all participants underwent a structural magnetic resonance imaging (MRI) scan of the brain and a graded maximal ergospirometry to assess cardiorespiratory fitness. Thirty-four participants successfully completed the intervention and additionally took part in a pre- and post-training MRI assessment. The MRI data were processed with the longitudinal stream in FreeSurfer ( Gray matter volume of hippocampal subfields were extracted using the hippocampal segmentation tool implemented in FreeSurfer. Participants of the cycling group increased their cardiorespiratory fitness from pre- to posttest while there was no change in the stretching/coordination group. Aerobic exercise had a significant positive effect on hippocampal volume in the CA subfields and in the dentate gyrus: while gray matter decreased in the stretching/coordination group after six months, hippocampal volume remained stable in the cycling group. The effect was specific for hippocampal regions and was not seen for other subcortical structures such as the basal ganglia. These results suggest that aerobic exercise prevents age-related hippocampal volume loss not only in older adults but also in middle-aged adults. The highly selective effect in the CA subfields and in the dentate gyrus is reminiscent of animal findings which have reported exercise-induced neural proliferation in these hippocampal regions.
Saturday 16.00 - 17.30 Grote zaal Poster #6 Sensory processing Laetitia Grabot
All authors & affiliationVirginie van Wassenhove, CEA-Inserm, NeuroSpin, Unicog Laetitia Grabot, CEA-Inserm, NeuroSpin, Unicog
Time Order is a Psychological Bias
Show/hide abstractInter-individual variability in temporal order perception has rarely been tackled empirically although readily observed in the literature. Classically, temporal order perception has been studied using temporal order judgements (TOJ) in which participants report the order of two stimuli presented few milliseconds apart. The temporal delay, at which participants consider the order of stimuli to be at chance is called Point of Subjective Simultaneity (Pss) and is typically averaged across participants canceling out potential individual effects. Here, we asked whether an individual's Pss was stable over weeks, which would indicate an intrinsic or hardwire constant delay in perceptual systems. The present work aimed to test this stability and to distinguish this hypothesized intrinsic constant from an attentional effect, considering that temporal order perception is known to be modulated by attentional fluctuations. We designed a longitudinal psychophysics study using auditory, visual, and audiovisual TOJs. The experimental design also comprised unisensory (vision or audition attended) and divided (audition and vision attended) attentional conditions. A standard measure of Pss was extracted from the divided attentional condition, and a measure of Pss free of any attentional biases was computed from the unisensory attentional conditions. Our results show that individual standard Pss are stable over months, strengthening the hypothesis that the Pss is an individual marker of temporal perception and thus a defining feature of an individual's core biases. Attention could partially, but not fully, compensate for this bias. These results shed a new light in the debated issue of what is actually measured in a TOJ and stressed the importance to tackle individual differences.
Saturday 16.00 - 17.30 Grote zaal Poster #7 Memory Benchi Wang
All authors & affiliationJan Theeuwes, Vrije Universiteit Amsterdam Christian Olivers, Vrije Universiteit Amsterdam
When shorter delays lead to worse memories: Task disruption makes visual working memory temporarily vulnerable to test interference
Show/hide abstractEvidence shows that visual working memory (VWM) is strongly served by attentional mechanisms, whereas other evidence shows that VWM representations readily survive when attention is being taken away. To reconcile these findings, we tested the hypothesis that directing attention away makes a memory representation vulnerable to interference from the test pattern, but only temporarily so. When given sufficient time, the robustness of VWM can be restored so that it is protected against test interference. In four experiments, participants remembered a single grating for a later memory test. In the crucial conditions, participants also performed a letter change detection task in between, during the delay period. Experiments 1-3 demonstrate and replicate the predicted effect: The intervening task had an adverse effect on memory performance, but only when the test display appeared immediately after the secondary task. At long delays (of 3.5 seconds), memory performance was on a par with conditions in which there was no intervening task. By varying the similarity of the test pattern to the memorized pattern, Experiment 4 further showed that performance suffered at early test intervals, unless the test item was dissimilar to the memory item. In conclusion, VWM storage involves multiple types of representation, with unattended memories being more susceptible to test interference than others. This study is the first to provide positive evidence for test interference, and to demonstrate that the susceptibility to interference only occurs temporarily, as memory performance is restored when attention is allowed to return to the memorandum.
Saturday 16.00 - 17.30 Grote zaal Poster #8 Sensory processing Ulrike Horn
All authors & affiliationHorn Ulrike, University Greifswald Functional Imaging Unit Klepzig Kai, University Greifswald Psychology Department Hamm Alfons, University Greifswald Psychology Department Lotze Martin, University Greifswald Functional Imaging Unit
Brain activation underlying chill experiences during listening to unknown music
Show/hide abstractMusic is strongly linked to the experience of emotions(1). According to the component process model an emotional episode consists of changes in appraisal, motor expression, bodily symptoms, and subjective feeling experience(2). A chill experience during listening to music combines all of these components and is therefore often considered as a marker of peak emotion responses to music(3). Only few imaging studies investigated the underlying brain representation of this phenomenon(4,5). So far the stimuli were chosen by the participants. To establish a chill paradigm for emotion research such restriction should be avoided. We therefore tried to extend the knowledge about chill experiences to unfamiliar music. We used stimuli with a high potential to induce chills due to their musical structure6 but unknown to the participants. We investigated the chill phenomenon with fMRI in 16 healthy participants (22±3.4 years). Echo-planar-images were collected with a 3T Magnetom Verio using a 32-channel head coil (774 volumes, TR 2000ms). Chills were modelled at a 10s-time-window around the individual time point of subjective chill experience which was indicated by device pressures. The comparison to musical stimuli without chill events replicated the involvement of several emotion processing areas(4). Especially bihemispheric anterior insula, anterior cingulate, thalamus, and left putamen showed a profound activation. We could not replicate any findings regarding the activation in the nucleus accumbens. This lack of reward system involvement underlines the importance of the chill event as a physiological marker of intense arousal in general(5,7,3) rather than a necessarily pleasurable marker. As it has been shown that the recall of memories can be associated with enhanced activity in the striatum(8) the use of familiar music might have influenced the previous results. Further exploration of chill events during familiar and unfamiliar music is needed to establish suitable paradigms for emotion research in the auditory domain.
Saturday 16.00 - 17.30 Grote zaal Poster #9 Memory Fang-Wen Chen
All authors & affiliationFang-Wen Chen, Department of Psychology, National Taiwan University, Taiwan Bo-Cheng Kuo, Department of Psychology, National Taiwan University, Taiwan
Response Anticipation Activates Silent States in Working Memory
Show/hide abstractWorking memory (WM) allows us to maintain and manipulate information that is relevant to our goals over a short period of time. Previous neural evidence has shown persistent activity during retention interval of WM for delayed responses. However, a recent theoretical account suggests that the information can be maintained in an activity-silent state during WM maintenance. In this study, we investigated whether anticipation for responses can reactivate relevant WM representations from a silent state and elicit a ramp-up of item-specific neural activity prior to the test probe with electroencephalography (EEG). Participants (N =16) performed a delayed response WM task. They were instructed to view and remember the orientation of a sample grating. After a short delay, a probe grating appeared, and participants indicated which way it has been rotated (clockwise or counterclockwise) compared to the sample grating. We manipulated the delay variability with small (from 1300 to 1700 ms) and large variability (from 800 to 2200 ms). We showed better behavioural performance for small variability relative to large variability. Our EEG results revealed a significant modulation of response anticipation on alpha activity during WM maintenance. These results showed a decrease of alpha activity started 250 ms before the probe onset over the right frontal regions for the large variability relative to the small variability. Together, our findings provided oscillatory evidence for characterising activity-silent states in WM.
Saturday 16.00 - 17.30 Grote zaal Poster #10 Attention Mariam Kostandyan
All authors & affiliationHaeme Park, Dept. of Experimental Psychology, Ghent University, 9000 Ghent, Belgium Carsten Bundt, Dept. of Experimental Psychology, Ghent University, 9000 Ghent, Belgium Ruth Krebs, Dept. of Experimental Psychology, Ghent University, 9000 Ghent, Belgium C. Nicolas Boehler, Dept. of Experimental Psychology, Ghent University, 9000 Ghent, Belgium
Proactive and reactive cognitive control in a rewarded Stroop task: A simultaneous EEG-fMRI study
Show/hide abstractReward influence on cognitive control can manifest itself via proactive or reactive mechanisms of control. Typical experimental paradigms to study the influence of reward on cognitive control mechanisms are 1) monetary incentive delay paradigm (MID) where reward information is cue-locked on the trial level and 2) stimulus-reward association task (SRA) where reward information is stimulus-locked (and often overlaps with the task). SRA paradigms suggest that reward could enhance reactive control, and maybe specific item-/feature-level proactive control, whereas in MID tasks more global shifts from reactive to proactive control are observed in reward contexts. Yet, direct comparisons between these different reward manipulations are lacking. Moreover, possible reactive-control enhancements need to be further distinguished from alternative processes, e.g. voluntary top-down screening for the stimuli's feature. We conducted a simultaneous EEG-fMRI study with a mixed event-related/block design using a Stroop color naming task with four different block types: SRA, MID, Cued-SRA (C-SRA; cues were reward irrelevant and reflected preparatory attentional set), and neutral block. Behavioral results showed main effects of reward and congruency for all reward-related blocks. Preliminary fMRI data showed reward-related activations in inferior parietal cortex, orbitofrontal cortex, and anterior cingulate cortex for SRA, MID, and C-SRA conditions on target level. On the block level we observed a deactivation of the precuneus as part of the default mode in the MID block in contrast with the C-SRA and neutral conditions. ERPs analysis resulted in reward effects across all conditions reflecting a modulation of the P300 component and congruency-driven N450 component that was identified by contrasting incongruent and congruent trials. In addition, both the EEG data and the fMRI data showed a varied pattern of activity related to valuation and task implementation indicative of subtle differences across the different reward conditions.
Saturday 16.00 - 17.30 Grote zaal Poster #11 Attention Aleya Flechsenhar
All authors & affiliationOlivia Larson, Department of Psychology, McGill University, Montreal, Canada. Albert End, Department of Systems Neuroscience, University Medical Centre Hamburg-Eppendorf, Hamburg, Germany. Matthias Gamer, Department of Psychology, Julius Maximilians University of Wuerzburg, Wuerzburg, Germany.
Overt and Covert Shifts of Attention in the Presence of Social Features
Show/hide abstractVisual attention research has brought forth a multitude of studies investigating overt and covert attention. However, few have compared both viewing modalities using the same set of stimuli. In a study comprising two experimental setups, we investigated basic attentional mechanisms for overt and covert orienting in visual exploration of complex naturalistic scenes. A modified version of the dot-probe paradigm served as a measure of covert reflexive attention towards briefly presented (200ms) stimuli containing a social feature, i.e. a human being, on one side compared to a non-social element, i.e. objects or animals on the other. Furthermore, we controlled for physical saliency, which is known to heavily affect early visual orienting. Within the dot-probe experiment, we found a congruency effect displaying faster reaction times and fewer error rates for targets presented on the social side of the stimulus. These results prevailed even when we reduced the stimulus set to scenes for which the non-social side was physically more salient than the social one. The second experiment included eye tracking to assess overt attentional orienting for the same set of stimuli. The presentation time of the scenes was either 200ms, enabling measures of reflexive attention, or 5000ms allowing analysis of more sustained attentional processes, for which both bottom-up and top-down attentional mechanisms are involved. Results revealed a preference for social features (1) for the short presentation time analyzing the direction of the first saccade after stimulus onset and (2) for the long presentation time in which fixations were predominantly allocated on the heads of the displayed human beings. Therefore, these experiments indicate that social features dominate over physical saliency in automatic attentional allocation in covert and overt viewing conditions as well as for endogenously generated attentional shifts.
Saturday 16.00 - 17.30 Grote zaal Poster #12 Anatomy Anne Trutti
All authors & affiliationAnne C. Trutti * , Leiden University, Cognitive Psychology Unit & Leiden Institute for Brain and Cognition, Leiden, The Netherlands; University of Amsterdam, Integrative Model-based Cognitive Neuroscience Research Unit Group, Amsterdam, The Netherlands Laura Fontanesi, University of Basel, Department of Psychology, Basel, Switzerland Martijn Mulder, University of Amsterdam, Integrative Model-based Cognitive Neuroscience Research Unit Group, Amsterdam, The Netherlands; Netherlands Institute for Neuroscience, an Institute of the Royal Netherlands Academy of Arts and Sciences, Amsterdam, the Netherlands; Bernhard Hommel, Leiden University, Cognitive Psychology Unit & Leiden Institute for Brain and Cognition, Leiden, The Netherlands Birte Forstmann, Leiden University, Cognitive Psychology Unit & Leiden Institute for Brain and Cognition, Leiden, The Netherlands; University of Amsterdam, Integrative Model-based Cognitive Neuroscience Research Unit Group, Amsterdam, The Netherlands; Netherlands Institute for Neuroscience, an Institute of the Royal Netherlands Academy of Arts and Sciences, Amsterdam, the Netherlands
A probabilistic atlas of the VTA based on 7T high-resolution MR data
Show/hide abstractThe ventral tegmental area (VTA) represents, together with the substantia nigra pars compacta (SNc), the main source for dopamine (DA) in the brain. Recent developments high-resolution magnetic resonance imaging (MRI) allow for in-vivo investigations of an increasing amount of structures in the human subcortex thus MRI studies attempt to differentiate midbrain BOLD-activity as stemming from the VTA or the SNc, respectively. Yet, the resolution and signal-to-noise ratio even in high-resolution MRI make discrimination of the structures difficult. Especially in the case of the VTA, visualization is hard due to the VTA's heterogeneous anatomical characteristics and complex shape. Hence, this study attempts to overcome the challenges associated with mapping the VTA on the individual MR images by reconstructing the VTA on multiple 7T high-resolution MR data, in order to consequently gain a probabilistic atlas of the structure. The poster will provide information on the underlying 7T MR data, as well as the manual segmentation protocol the probabilistic VTA atlas is based on, and analyses and discusses the agreement of the two raters. The study emphasizes the issues associated with reports of VTA -- activity that we believe should be taken very cautiously, as there is no VTA atlas based on 7T MR data available yet. As a consequence, this study attempts to provide such an atlas of the VTA that can serve as an a priori defined anatomical region-of-interest and thus allows for the comparison of fMRI data.
Saturday 16.00 - 17.30 Grote zaal Poster #13 Attention Marius Rubo
All authors & affiliationRubo Marius, University of Wuerzburg* Gamer Matthias, University of Wuerzburg
Social attention in real life: Influence of social anxiety and visual saliency
Show/hide abstractPrevious research has shown a two-fold alteration of attentional mechanisms in socially anxious persons: they were found to show an initial attentional bias towards threatening cues like angry faces or threatening words, but then exhibit an avoidance of the detected stimulus compared to non-anxious controls (hypervigilance-avoidance hypothesis). In most studies in this field, participants are presented static, and typically impoverished images of persons. However, findings resulting from this research strategy are susceptible to differences between viewing images of persons and viewing real persons. The present study addresses attentional mechanisms in socially anxious persons by investigating attentional processes in a real situation using mobile eye tracking. We found that when walking in a populated train station, socially anxious participants fixated other persons more when they were far away, and less when they were close compared to non-anxious participants. Interestingly, no differences in viewing behavior between socially anxious and non-anxious persons were found in a matched group viewing the same scenes on a computer monitor. We argue that the results from the real-life group are in line with both the hypervigilance-avoidance-hypothesis and observations reported by clinicians. In a further analysis, we found visual saliency to predict gaze behavior better in a non-social situation than in the "social" train station. This was true both in the real-life and in the lab group, and did not depend on social anxiety. This finding demonstrates the predictive power of saliency algorithms even in everyday situations, and is in line with the idea of social information driving attention and thus reducing the effect of saliency on gaze behavior. Overall, this study highlights the relevance of ecological validity in the examination of attentional biases in clinical populations.
Saturday 16.00 - 17.30 Grote zaal Poster #14 Connectivity/networks Tirso Rene del Jesus Gonzalez-Alam
All authors & affiliationTirso RJ Gonzalez-Alam, University of York Department of Psychology Jonathan Smallwood, University of York Department of Psychology Elizabeth Jefferies, University of York Department of Psychology
Meaningful inhibition: Evidence for visual ventral stream involvement when behavioral inhibition depends on stimulus meaning
Show/hide abstractResponse inhibition plays an important role in behaviour, allowing us to act in a goal-oriented manner, organising our actions efficiently in response to changes in the world. Most research in response inhibition has used simple perceptual stimuli devoid of meaning, with low ecological validity. The aim of the present study was to describe the neural mechanisms that support inhibitory control when meaning is used to determine the behavioural response. We tested 90 participants in an fMRI Go / No - Go experiment (60 inside the scanner, 30 outside) in which we manipulated the modality of meaning, presenting the same stimuli as words and pictures (matched for difficulty with two control conditions where a perceptual feature determined the response). Our results show that regions of the Visual Ventral Stream (VVS) support response inhibition guided by images. This could not be attributed to a difficulty difference between pictures and words, because VVS activity was higher than in a perceptual condition that was matched for difficulty. Next we performed a resting state functional connectivity analysis in the offline sample to look for behavioural modulations related to functional connectivity. This illustrated that resting state connectivity from VVS regions to the pre-Supplementary Motor Area (pre-SMA) relates to variation in the efficiency of inhibition affected by the modality in which meaningful stimuli were presented. These results indicate a contribution of VVS regions to response inhibition when behaviour is guided by visual, high-level ecologically valid stimuli.
Saturday 16.00 - 17.30 Grote zaal Poster #15 Memory Kaifeng He
All authors & affiliationKaifeng He, Department of Psychology and Behavioral Sciences in Zhejiang University Fan Wu, Department of Psychology and Behavioral Sciences in Zhejiang University Hui Chen, Department of Psychology and Behavioral Sciences in Zhejiang University Mowei Shen, Department of Psychology and Behavioral Sciences in Zhejiang University Zaifeng Gao, Department of Psychology and Behavioral Sciences in Zhejiang University
Binding Recovery in Working Memory: Behavior and EEG Evidence
Show/hide abstractRecent studies have revealed that object-based attention plays key role in retaining bindings in working memory (WM). This finding was reached by interpolating a secondary task into the maintenance phase of WM, and showing that a secondary task consuming object-based attention led to significantly larger binding impairment than the constitute single features. In the current study, we examined whether the consumed resource by the secondary task would return back to the WM task after the completion of the secondary task (recovery time) and led to binding recovery. We manipulated the duration of the recovery time (500 ms vs. 1400 ms) and the type of the attention (object-based attention vs. space-based attention) consumed by the secondary task. We found a larger impairment for binding than for the constituent features with a short recovery-time after completing the secondary task. However, by giving 900 ms of additional time, a significant binding performance-recovery was demonstrated. Such selective binding impairment and recovery, however, were absent when a space-based visual-search task. Moreover, modeling studies shows that this recovery was not due to increasing the resolution of the representation, but was attributed to increased probability of storage. Finally, EEG studies showed that after the completion of the secondary task consuming object-based attention, a significant beta (13-18 Hz) activation was observed and lasted for at least 600 ms; while such enhanced beta was absent after the secondary task consuming space-based attention. These results together imply that WM and external object-based attention share one resource pool, and a trade-off exists between the two.
Saturday 16.00 - 17.30 Grote zaal Poster #16 Electrophysiology methods Mattia Federico Pagnotta
All authors & affiliationMattia F. Pagnotta, University of Fribourg Gijs Plomp, University of Fribourg
Dynamic Granger-causality: methods comparison in numerical simulations and benchmark EEG data
Show/hide abstractDynamic Granger-causality methods aim to quantify directed interaction strengths between brain areas with high temporal resolution, using simultaneously recorded electrophysiological signals. These methods are often based on time-varying multivariate autoregressive (tvMVAR) modeling, and while several such approaches have been proposed there currently is a lack of unbiased analyses and comparisons of their performance. Our aim was to compare the performance of commonly used tvMVAR methods using numerical simulations and real benchmark EEG data along fixed criteria. We compared classical Kalman filter (MVAAR), dual extended Kalman filter (DEKF), recursive least squares (RLS) and general linear Kalman filter (GLKF), and two ways of exploiting repeated observations: 1) single trial modeling followed by averaging, and 2) multi-trial modeling where one tvMVAR model is fitted across trials. Our results show that while most approaches can adequately model simulated and real data, the best performance was often achieved with GLKF and a multi-trial approach. This approach's accuracy, however, more strongly depended on model order choice and sampling rate. In fact, this algorithm produced highly variable estimates at high sampling rate when the imposed interaction delay exceeded the sampling interval by an order of magnitude, and a large model order was required. In this scenario downsampling successfully reduced the estimates' variability. Single-trial approaches using GLKF and MVAAR were more robust against setting model order too high and showed good performance at high sampling rates. For these algorithms downsampling degraded performance, because of slower adaptation speeds. Our findings help understand the strengths of various tvMVAR approaches and provide practical recommendations for their use in modeling dynamic directed interactions from electrophysiological signals.
Saturday 16.00 - 17.30 Grote zaal Poster #17 Connectivity/networks Thomy Nilsson
All authors & affiliationThomy Nilsson, University of Prince Edward Island
Spatial Multiplexing Lets Convergent-Divergent Networks Overcome Surface Bottlenecks to Information Transmission
Show/hide abstract The encompassing surface of any device limits its number of finite connections to other devices. Electronic devices overcome this bottleneck by minimizing size of the connecting pathways and temporally multiplexing information sent along those pathways. These are not options for biological systems using neurons and 2 millisecond action potentials. Consequently, small brains do not have sufficient surface area to receive enough sensory information to warrant a larger brain, nor can they send enough information to operate larger bodies needed to support larger brains. In 1955 Weddell, Taylor & Williams discovered that convergent-divergent organization of a tactile sensory system in rabbit overcame this bottleneck. While now recognized as a general principle in the organization of sensory and motor systems, how this organization works was elusive. This presentation demonstrates how spatial multiplexing enables convergent-divergent systems to transmit many inputs along a fractional number of pathways while maintaining the identity and strength of specific inputs, without coding or feedback. A linear array of n inputs connected to a linear array of n outputs demonstrates how a simple asymmetric, bifurcated branching network can transmit information using n/8 pathways. This is the limit for accurate convergence between linear arrays. Applying the Pythagorean equivalent of this branching pattern to information transmission between two surface arrays maintains accuracy using n/16 connecting pathways. This approaches the convergence found by Weddell, et al. However, these models have perfect resolution only with one input at a time. Two or more simultaneous inputs in close proximity are compromised in accuracy of location and strength. That is solvable by simultaneously scanning the transmissions over limited sets of pathways. The need for inherent scanning in convergent-divergent systems may be the basis of brain rhythms. Scanned spatial multiplexing may prove applicable to reduce connections in transmission networks where temporal multiplexing is infeasible.
Saturday 16.00 - 17.30 Grote zaal Poster #18 Clinical Louise Lavrencic
All authors & affiliationScott Coussens, Cognitive Ageing and Impairment Neurosciences Laboratory, School of Psychology, Social Work and Social Policy, University of South Australia Owen Churches, Brain and Cognition Laboratory, School of Psychology, Flinders University Hannah Keage, Cognitive Ageing and Impairment Neurosciences Laboratory, School of Psychology, Social Work and Social Policy, University of South Australia
Cognitive reserve is associated with neural processing during the n-back task in older adults
Show/hide abstractBackground Cognitive reserve can be used to explain individual differences in the effects of age- and dementia-related neuropathology on cognitive function. The underlying neural mechanisms are not understood. This study investigated how cognitive reserve associated with neural activity across tasks of varying difficulty, accounting for structural brain changes, to test the current theoretical model. Method Participants included thirty-four older adults (Mage=70.5, SD=6.1; 23 female). Cognitive reserve was indexed by the Lifetime of Experiences Questionnaire (LEQ). Electroencephalogram was recorded from 64 channels during three n-back tasks (0, 1, and 2-back). Mean amplitude of the P3 event-related potential component was calculated for each n-back condition, and for correct target and non-target trials. Diffusion Tensor Imaging fractional anisotropy values were used to index underlying white matter structural integrity. Linear mixed effects modelling was used to investigate predictors of P3 amplitude, using maximum likelihood estimation; LEQ scores, fractional anisotropy values, and n-back condition (repeated measures) were entered as fixed factors, and ID as a random intercept. Results Individuals with higher cognitive reserve had larger P3 amplitudes than those with low cognitive reserve (main effect of cognitive reserve). However, this effect differed across n-back conditions: as difficulty increased, the effect of cognitive reserve on P3 amplitude attenuated (interaction between cognitive reserve and n-back condition). There were no main effects of fractional anisotropy or n-back condition. Conclusion Results demonstrate that cognitive reserve associates with task-related neural activity, accounting for structural brain changes, an effect that is more prominent the lower the task difficulty. These findings contradict the dominant cognitive reserve theory, which proposes more pronounced differences in neural activity with increasing task difficulty (indicative of compensatory mechanisms). Understanding the relationships between cognitive reserve (as a modifiable factor), brain function, and brain structure is imperative for the development of interventions to promote brain health in late-life.
Saturday 16.00 - 17.30 Grote zaal Poster #19 Memory Lily Fu
All authors & affiliationLi Fu, Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, the Netherlands Joseph H.R. Maes, Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, the Netherlands Samarth Varma, Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, the Netherlands Roy P.C. Kessels, Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, the Netherlands; Department of Medical Psychology, Radboud University Medical Center, Nijmegen, the Netherlands Guillen Fernandez, Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, the Netherlands Sander M. Daselaar, Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, the Netherlands
The effects of semantic encoding effort on episodic memory in both young and older adults: A parametric fMRI study
Show/hide abstractIt is essential to develop effective interventions aimed at ameliorating age-related cognitive decline in our aging society. Our previous behavioral studies showed that effortful semantic encoding boosts episodic memory inolder adults (OAs) to bring them actually to the same level as young adults (YAs), particularly in OAs with high executive functioning. This finding seems critical for development of new tools to aid OAs' memory. However, it is still unclear what the neurophysiologic mechanisms are underlying these age-related advantages. Following these behavioral findings, we are currently using a subsequent memory fMRI encoding design, with the same task, including YAs (N=20), low-, and high-executive OAs (N=40). We focus on activation in executive and memory networks including the default mode network. We use 4 cognitive effort levels during incidental semantic encoding. Semantic effort is manipulated by varying decision-making demands based on Latent Semantic Analysis. EM performance, as indexed by d-prime, is later tested using a recognition task. Outlining the neurophysiological differences underlying the behavioral age effects of semantic effort on memory will further our knowledge needed for the development of effective intervention methods.
Saturday 16.00 - 17.30 Grote zaal Poster #20 Memory Daniel Labbe
All authors & affiliationDaniel Labbe, International Graduate School of Neuroscience. Department of Neuropsychology. Ruhr University Bochum. Germany Nikolai Axmacher, Department of Neuropsychology. Ruhr University Bochum. Germany
Visual Memory in Change Detection: What does the EEG signal represent?
Show/hide abstractThe "memory of the present" is a vital executive function. Short-term memory models typically postulate either a discrete or a continuous nature of items held in memory. While these models appear well-defined in cognitive terms, the underlying neurophysiology is ambiguous. How can specific EEG signatures inform a cognitive model of short-term memory? We administered a change detection task to capture discrete performance in healthy participants in an attempt to replicate previous ERP findings of a contralateral delay activity (CDA). Plans exist for a second task that captures "continuous resource"-based performance (i.e. accuracy) (cf. Ji Ma, Husain, & Bays, 2014, Nature Neuroscience). Our initial pilot data suggests that the CDA changes asymptotically towards the discrete capacity limit for simultaneously presented stimuli (i.e. 4+/-1), a finding that is in line with previous work (Vogel & Machizawa, 2004, Nature).
Saturday 16.00 - 17.30 Grote zaal Poster #21 Social behavior Saskia Heijnen
All authors & affiliationR. E. De Kleijn, Leiden University B. Hommel, Leiden University
The effect of human-robot synchronization on the tendency to anthropomorphize
Show/hide abstractIn order to elucidate the working mechanism behind anthropomorphism (1), this study aimed to investigate whether a robot would be anthropomorphized more if a human and a robot moved synchronously, compared to nonsynchronously. It was hypothesized that moving in synchrony would lead to anthropomorphization via increased perceived feature overlap (2), which in turn would activate features related to humans via pattern completion. To study this, we performed a study with synchronicity as within-subjects factor, and initiator (robot or human) as between-subjects factor. Participants rated the robot on two state and one trait anthropomorphism questionnaire, and performed a joint Simon task and one-shot Dictator game with the robot. Due to part of the data having non-normally distributed residuals, the intended ANOVA's could not be executed. Instead, the data was analyzed using clustured bootstrap analyses. The results overall did not confirm our hypotheses. The results from the questionnaires showed that, particularly for the group in which the human initiated the movements, the robot was anthropomorphized more after nonsynchronous, compared to synchronous interaction. For the Dictator Game, those in the human initiator group gave the robot more money than those in the robot initiator group. This effect did not depend on synchronicity. Finally, the joint Simon effect was subject to an interaction between initiator and synchronicity. Those in the robot initiator group showed a larger joint Simon effect in the nonsynchronous condition compared to the synchronous condition.
Saturday 16.00 - 17.30 Grote zaal Poster #22 Sensory processing Raquel London
All authors & affiliationRaquel London, Department of Experimental Psychology, Ghent University Christopher Benwell, Centre for Cognitive Neuroimaging, Institute of Neuroscience and Psychology, University of Glasgow Roberto Cecere, Centre for Cognitive Neuroimaging, Institute of Neuroscience and Psychology, University of Glasgow Michel Quak, Department of Experimental Psychology, Ghent University Gregor Thut, Centre for Cognitive Neuroimaging, Institute of Neuroscience and Psychology, University of Glasgow Durk Talsma, Department of Experimental Psychology, Ghent University
Trial-to-trial variability of pre-stimulus alpha oscillations predicts the temporal resolution of multisensory perception.
Show/hide abstractHere we present evidence from 40 participants that both power and instantaneous frequency of pre-stimulus alpha oscillations predict the temporal resolution of audiovisual perception in a temporal order judgement (TOJ) task. Alpha oscillations reflect fluctuations in attention and the excitability of the cortex. In detection tasks, higher alpha power predicts lower sensitivity to visual near-threshold stimuli and less susceptibility to multisensory illusions such as the sound-induced flash illusion. In turn, higher alpha frequency predicts better temporal resolution of both visual and audiovisual perception. In this study we went beyond detection of near-threshold or illusory stimuli. We measured the temporal resolution of audiovisual perception using a TOJ task with above-threshold auditory and visual stimuli. Additionally, we measured EEG and tested whether task performance could be predicted by pre-stimulus oscillations on a trial-by-trial basis. In TOJ tasks, the behavioral measure of temporal resolution (the "just noticeable difference" or JND) is typically not obtained at each trial, but over a large amount of trials. However, by applying a jack-knife procedure adapted for linking single-trial pre-stimulus activity to psychometric measures (Benwell et al., in prep.) we obtained a JND for each trial. Single-trial pre-stimulus power at each electrode from 1 to 45 Hz and from -1 to 0 sec was correlated with the JND across trials. We found that at alpha-range frequencies, higher posterior power predicted a higher JND, and thus a lower temporal resolution of audiovisual perception. Subsequently, we correlated the JND with single-trial instantaneous alpha frequency from -1 to 0 sec and found that higher posterior alpha frequency predicts a lower JND, and thus a higher temporal resolution of audiovisual perception. These results provide new insights into the role of moment-to-moment fluctuations of cortical excitability in the temporal aspects of multisensory perception.
Saturday 16.00 - 17.30 Grote zaal Poster #23 Attention Kelly Garner
All authors & affiliationKelly Garner, School of Psychology, University of Birmingham* Jane Raymond, School of Psychology, University of Birmingham
Outcome probability and value show distinct influences on visual selective attention
Show/hide abstractHow does prior experience alter which sensory information is prioritised for detailed analysis? Evidence suggests that learned associations between a signal and its outcome probability (van der Heijden 1989; Lanthier et al. 2015) or its outcome value (e.g. Raymond and O'Brien 2009; Chelazzi et al. 2014) can act to bias processing towards that signal. Controversy exists over how these two information sources (probability and value) combine to influence visual selection. Recent human behavioural data show an additive influence (Stankevich and Geng 2014; Munneke et al. 2015), whereas neural recordings in monkeys suggest an interactive influence (Platt and Glimcher 1999; Stănişor et al. 2013). However, previous investigations have only examined a limited range of outcome probabilities, and the experimental task structures have ensured that either operation (additive or interactive) is comparably optimal. The current study overcomes these limitations to provide the first direct test between additive and interactive models. Across two experiments, participants completed an endogenous cueing task. Probability of cue-validity varied across blocks, and targets appeared in coloured placeholders that were associated with high or low outcome values. In Experiment 1, a large range of cue-validity probabilities was assessed. In Experiment 2, potential reward value was manipulated so that an interactive operation sometimes had greater optimality. Across both experiments, the additive model outperformed the interactive model to account for observed performance, suggesting that outcome probability and value act on distinct processes controlling visual-selection. An exploratory analysis shows that the influence of outcome probability and value covaries positively across individuals, and that this in turn predicts total reward earnings. This suggests that although these associations act distinctly, they are both exploited to a greater degree by individuals who optimise task performance.
Saturday 16.00 - 17.30 Grote zaal Poster #24 Sensory processing Raphael Wallroth
All authors & affiliationRaphael Wallroth, German Institute of Human Nutrition Potsdam-Rehbruecke, NutriAct -- Competence Cluster Nutrition Research Berlin-Potsdam Kathrin Ohla, German Institute of Human Nutrition Potsdam-Rehbruecke
Taste categorical information is encoded in the delta band
Show/hide abstractThe ability to detect the presence of a chemical in the mouth plays a crucial role in nutrient sensing and identification of toxins. Little is known about the cortical dynamics of taste processing which represents the least studied sensory system in humans. We have recently shown that the earliest taste-evoked neural responses obtained from the human cortex code taste quality information (sweet, sour, salty, bitter; Crouzet, Busch, & Ohla, 2015). Yet, advances in the field have been greatly hampered due to poor signal-to-noise ratio of gustatory-evoked responses. We measured multi-channel head-surface electroencephalographic (EEG) in human participants while they were to detect tastants inducing salty, sweet, sour, or bitter sensations. Using multivariate pattern analysis of large-scale electrophysiological responses we show that taste quality information is largely contained in the lower (delta) frequency band and that isolating information in the delta band improves classification performance significantly. Given that delta oscillations have been implicated in the integration of cerebral activity with homeostatic processes such as hunger and also food reward processing (Knyazev, 2012) the current result suggests a previously unknown role of delta oscillations in food-related sensory processing.
Saturday 16.00 - 17.30 Grote zaal Poster #25 Social behavior Valtteri Wikstroem
All authors & affiliationKatri Saarikivi, Cognitive Brain Research Unit, University of Helsinki Valtteri Wikstroem, Cognitive Brain Research Unit, University of Helsinki Tommi Makkonen, Cognitive Brain Research Unit, University of Helsinki
Inter-brain synchronization during a joint learning task: preliminary findings from a pilot study
Show/hide abstractInteraction and cooperation between individuals can be investigated on many levels. Inter-individual synchronization of brain rhythms has recently been found in a variety of cooperative situations (Dumas et al., 2010; Dumas et al., 2012; Kawasaki et al., 2013; for review, see Konvalinka & Roepstorff, 2012). Inter-brain synchronization has been found emerge during collaboration but not competition (Cui et al., 2012). Similarly, Toppi and colleagues (2016) recently found increased connectivity in inter-brain networks during cooperation between pilots in a flight simulator. However, it is not yet known whether inter-brain synchronization is connected to the success of collaboration. In this study, we investigated the relation between inter-brain synchronization to the speed and success of joint learning in a dyad. We recruited pairs of subjects that considered each other friends. We recorded simultaneous EEG from both subjects during a multiplayer game that required maneuvering a race car along different courses. One of the subjects controlled the throttle and the other the steering of the car. The lap times, and number of mistakes (the vehicle went out of the course) were noted. Here, we present preliminary findings from 5 pilot measurements (10 subjects total) on 1) the success of the experimental set-up 2) the extent of synchronization during the joint learning task and 3) connections between the speed of joint learning, the lap times and the number of mistakes during the task and ratings of affiliation between the subjects.
Saturday 16.00 - 17.30 Grote zaal Poster #26 Memory Ya-Ping Chen
All authors & affiliationYa-Ping Chen, Department of Psychology, National Taiwan University, Taipei, Taiwan Bo-Cheng Kuo, Department of Psychology, National Taiwan University, Taipei, Taiwan
The influences of selection history on working memory: An EEG study
Show/hide abstractPrevious studies have revealed that context-driven selection history can influence the allocation of attention on target selection. Recent behavioural evidence has also shown that selection history can modulate the efficacy of attention allocation on working memory (WM) representations (Kuo, 2016). However, the neural correlates of the influences of selection history on WM remain largely unknown. Here we addressed this issue with electroencephalography (EEG). Participants performed a delayed response task. A display of one item (low load) or three items (high load) was shown for the participants to hold in WM. Selection history was defined as the number of items attended across trials in the task context within a block, manipulated by the stimulus set-size in the contexts with fewer possible stimuli (e.g. 4-item context) or more possible stimuli (e.g. 8-item context) from which the memorised content was selected. We showed that WM capacity measure (K score) was significantly reduced in the 8-item context relative to the 4-item context. This capacity reduction was more evident for high WM load relative to low WM load. Our EEG data showed that selection history resulted in greater regulation of alpha activity for 8-item context relative to 4-item context during WM maintenance. Importantly, we estimated the spatiotemporal pattern similarity in alpha activity for the influences of selection history on WM. The pattern similarity indexes were significantly higher for the 4-item context relative to 8-item context during the pre-trial stage over the posterior regions when WM load was high. In sum, our results highlight that the WM representations are highly flexible and susceptible to different task contexts.
Saturday 16.00 - 17.30 Grote zaal Poster #27 Memory Deniz Vatansever
All authors & affiliationDeniz Vatansever, Department of Psychology, University of York, York, United Kingdom Jonathan Smallwood, Department of Psychology, University of York, York, United Kingdom Elizabeth Jefferies, Department of Psychology, University of York, York, United Kingdom
Effortless Memory: Default Mode Contribution to Coherent Representations
Show/hide abstractBackground: Recent evidence suggests the default mode network's (DMN) central role in memory-based information processing. Easy semantic judgments, retrieval of autobiographical information, and naturally occurring stimulus-independent thoughts have all been associated with greater DMN activity. In contrast to the multiple demand network's (MDN) contribution to controlled processing, the DMN in this context may support cognition that depends on coherent representations stored in memory. In a functional magnetic resonance imaging (fMRI) based semantic association task, we assessed the differential activity of these two networks while manipulating the strength of stored semantic representations. Methods: A total of 30 participants underwent a functional MRI scan (3T GE HDx Excite, TR = 3000 ms, 217 volumes). The main objective of the task was to match the target word (e.g. bee) with either a strongly (e.g. honey) or a weakly (e.g. colony) related word out of 3 options. All fMRI data was preprocessed following a standard pipeline. Individual level condition-specific contrasts were set up, which were then carried forward on to group-level statistical t-tests. All results were cluster-level multiple-comparison corrected at Z > 2.6 (p < 0.05). Results: Behaviourally, participants were faster in responding to strongly as opposed to weakly associated words (t(29) = 16.180, P < 0.0001). The imaging results indicated that the main DMN hubs including the posterior cingulate, medial prefrontal cortices and bilateral angular gyri showed greater activity while participants matched strongly associated word pairs than weakly associated ones. The reverse contrast on the other hand, showed greater activity centred on the MDN regions as well as visual cortices. The poster will also include results for a related episodic memory task. Conclusions: The DMN may have an important contribution to cognition when task demands require that information from memory is retrieved in a manner that is coherent with prior expectations.
Saturday 16.00 - 17.30 Grote zaal Poster #28 Attention Martijn E. Wokke
All authors & affiliationTony Ro, City University of New York
Unconscious attentional control of alpha oscillations
Show/hide abstractTypically, events do not occur in isolation of their context. For instance, the context of being in Amsterdam draws your attention towards cyclists (trying to run you over, so it seems), after you've been walking the streets of Amsterdam for some time. Models of attention propose a dichotomy between endogenous and exogenous attentional control mechanisms, also referred to as top-down/voluntary and bottom-up/reflexive attention. In this framework, exogenous attentional control is determined by physical characteristics and salience of information from the external world, whereas endogenous attentional control is determined by current goals and can be drawn on volitionally. In the present study, we assessed whether endogenous attentional control occurs in the absence of explicit top-down goals. In two experiments, we used cues that provided contextual (implicit) information that was predictive of whether a target would appear or not. Participants were instructed to respond as quickly as possible when a slightly leftward or rightward tilted Gabor patch appeared while EEG signals were recorded. We first show that despite unawareness of the contextual information provided by the cues, responses were significantly faster when the cue indicated target presence. Using EEG, we are assessing whether the amplitude, latency, and phase of visual/parietal cortex activity can be differentially influenced by this unconscious endogenous attentional control. Preliminary results suggest that implicit contextual information affects both the phase and power of alpha oscillations, suggesting an unconscious increase in attentional control. Taken together, this study provides fundamental insights into essential operations of our attentional and visual sensory system. Specifically, we are able to elucidate whether and how implicit information employs endogenous attentional control via similar neural (oscillatory) mechanisms as conscious attentional control.
Saturday 16.00 - 17.30 Grote zaal Poster #29 Sensory processing Vinitha Rangarajan
All authors & affiliationVinitha Rangarajan, Department of Psychology, University of California, Berkeley, USA Kevin S. Weiner, Department of Psychology, Stanford University, California, USA Corentin Jacques, Psychological Sciences Research Institute (IPSY), Universite Catholique de Louvain, Belgium Josef Parvizi, Department of Neurology and Neurological Science, Stanford University, California, USA Robert T. Knight, Department of Psychology, University of California, Berkeley; Helen Wills Neuroscience Institute, University of California, Berkeley, USA Kalanit Grill-Spector, Department of Psychology, Stanford University, California, USA
Smaller, slower, and shorter-lived: Context-dependent temporal characteristics of visual adaptation in human ventral temporal cortex
Show/hide abstractRepeated presentations of visual stimuli produce decreased neural responses within human ventral temporal cortex (VTC) - a phenomenon known as repetition suppression or adaptation (refs 1-3). Several neural mechanisms, including scaling, sharpening, and facilitation, have been proposed to explain this phenomenon. Different than scaling and sharpening, facilitation proposes that neural responses are comparable in amplitude, faster to peak, and shorter in duration for repeated stimuli. We tested these three dynamics utilizing the high spatial and temporal resolution of electrocorticography (ECoG) from 6 subjects (2 female) implanted with intracranial electrodes over the right (n=4) and left (n=2) VTC. Subjects were presented with images of houses, bodies, cars and faces and we independently identified face-selective electrodes in each subject. We then assessed the facilitation hypothesis by quantifying the functional dynamics of repeated images presented after intervening stimuli (experiment 1) or immediately (experiment 2). We observed a dense concentration of face-selective electrodes (n=43) within the fusiform gyrus and inferior temporal gyrus. A majority of these electrodes exhibited significant high frequency broadband (HFB: 70-150 Hz) adaptation when presented with repeated exposures to faces. Contrary to predictions of the facilitation model, adaptation was slower (p<0.004), and both smaller, and shorter-lived (area under the curve for 1st vs 2nd presentation, experiment 2: p=1.8720e-09, KS-test=0.6744) for immediate repetitions. Interestingly, for repetitions with intervening stimuli, we found some smaller responses, but no significant changes to the onset (p>0.05) or duration (area under the curve, experiment 1: p=0.0569, KS-test=0.2791). Together, these findings provide strong evidence that mechanisms other than facilitation underlie processing of repeated stimuli and these mechanisms are context-dependent.
Saturday 16.00 - 17.30 Grote zaal Poster #30 Attention Ruth de Diego-Balaguer
All authors & affiliationTabinda Mohammad, Universite Caen Normandie Anna Martinez-Alvarez, Universitat de Barcelona Joan Orpella-Garcia, Universitat de Barcelona Clara Garcia-Gorro, Universitat de Barcelona Alicia Callejas, Universitat de Barcelona Estela Camara, Universitat de Barcelona Ruth de Diego-Balaguer, ICREA and University of Barcelona
Deficits in both explicit and implicit temporal orienting in Huntington's disease
Show/hide abstractTemporal processing involves a widespread network of brain regions overlapping with the motor system. In particular, the basal ganglia appear to be responsible for internally and externally triggered explicit timing. Huntington's disease (HD) entails an early degeneration of the basal ganglia. However, parietal and cerebellar brain structures necessary for implicit timing and temporal orienting of attention are also affected. Previous studies on HD have focused on explicit tasks. In this study, we were thus interested in assessing HD patients in tasks that require the allocation of attention in time explicitly and implicitly. In the explicit task, participants had to detect an auditory target preceded by an auditory cue. Cue pitch (high/low) was informative of the cue-to-target interval (SOA: 350/1350ms) with 75% validity. Catch trials consisted of a pink noise target to which participants were instructed not to respond. In the implicit task, participants were asked to discriminate the pitch of a syllable at the end of an alternating sequence of two syllables (co-te / pa-bi). The sequences could be rhythmic or non-rhythmic (rhythmicity factor). In half of them, the cue identity allowed for the prediction of the last syllable (predictiveness factor). In the explicit task, healthy matched controls (N = 22) showed validity effects for both short and long SOAs. In HD patients (N = 22) no validity effect at the long SOA was observed. In the implicit task, controls (N = 11) showed significant facilitation overall in rhythmic conditions and benefited the most from the convergence of both predictive and rhythmic cues. HD patients (N = 11) only showed facilitation in the rhythmic condition for non-predictive trials and in the predictive condition only in non-rhythmic trials. This pattern of results indicates that HD-related temporal orienting deficits are not limited to explicit but affect also implicit timing.
Saturday 16.00 - 17.30 Grote zaal Poster #31 Clinical Filipa Brito
All authors & affiliationFilipa Brito, Instituto de Saude Ambiental, Faculdade de Medicina, Universidade de Lisboa, Fundacao para a Ciencia e Tecnologia, Portugal Ana Virgolino, Instituto de Saude Ambiental, Faculdade de Medicina, Universidade de Lisboa and Instituto de Saude Ambiental, Faculdade de Medicina, Universidade de Lisboa Monica Fialho, Instituto de Saude Ambiental, Faculdade de Medicina, Universidade de Lisboa Ana Miranda, Instituto de Saude Ambiental, Faculdade de Medicina, Universidade de Lisboa and Instituto de Saude Ambiental, Faculdade de Medicina, Universidade de Lisboa Jose Peixoto, Instituto de Medicina Preventiva e Saude Publica, Faculdade de Medicina, Universidade de Lisboa Inês Neves, Instituto de Saude Ambiental, Faculdade de Medicina, Universidade de Lisboa and Instituto de Medicina Preventiva e Saude Publica, Faculdade de Medicina, Universidade de Lisboa Jose Camolas, Instituto de Saude Ambiental, Faculdade de Medicina, Universidade de Lisboa and Serviço de Endocrinologia, Hospital de Santa Maria, CHLN Maria de Fatima Reis, Instituto de Saude Ambiental, Faculdade de Medicina, Universidade de Lisboa and Instituto de Medicina Preventiva e Saude Publica, Faculdade de Medicina, Universidade de Lisboa Luis Carriço, Faculdade de Ciências, Universidade de Lisboa Osvaldo Santos, Instituto de Saude Ambiental, Faculdade de Medicina, Universidade de Lisboa and Instituto de Saude Ambiental, Faculdade de Medicina, Universidade de Lisboa
Do we know what really works? A systematic review about using videogames for cognitive training.
Show/hide abstractIntroduction: Videogames have been used as tools for training cognitive functions, with both clinical and non-clinical populations. However, the evidence about what makes videogames effective in promoting general or specific cognitive functions is still scarce. Objective: To identify, integrate and summarize knowledge about effective videogames and game elements in promoting attention, working memory and inhibitory control capacity. Methods: The systematic review of literature followed PRISMA guidelines. PubMed, SciELO and PsycARTICLES electronic databases were searched. Inclusion criteria were: peer-reviewed English-language papers published within last 10 years, assessing the impact of videogames on cognitive functions. Search-terms were chosen from two domains (MeSH terms whenever possible): cognitive outcomes (e.g., executive processes, attention, memory and inhibitory functions) and gamification/game-elements (e.g., game environment, game interaction, game tasks and game narratives). Articles selection and narrative syntheses were done independently by two (or, in case of disagreement) three team members. Results: From the 2744 initially identified articles, 155 were considered eligible for narrative synthesis. Preliminary findings suggest that most literature presents a poorly-detailed description of videogames and lack consensual definitions of game-elements (e.g., avatar vs. character) or technological borders (virtual environments vs. reality). Key-aspects of user-experience such as fun, enjoyment and usability only rarely are evaluated. Cognitive endpoints are usually considered as dependent variables of the whole gaming experience, but without adequate understanding of which specific game-elements have effect. Transference of cognitive competences is seldom assessed and effectiveness evaluation tends to be done for short-term follow-up periods. Conclusions: Both methodological limitations and lack of information about videogames' properties/characteristics hinder a proper assessment of the effectiveness of videogames in promoting cognitive functions. The trend on using videogames to promote cognitive function justifies and urges for a thorough evaluation of the impact of the different game-elements, enabling game's personalization to individuals' clinical condition.
Saturday 16.00 - 17.30 Grote zaal Poster #32 Decision-making Roland Nigbur
All authors & affiliationNigbur, Roland, Otto-von-Guericke-University Magdeburg Ullsperger, Markus, Otto-von-Guericke-University Magdeburg
Show/hide abstractRecent developments in cognitive neuroscience propose that affect and control are not opponent, but interconnected, reciprocal modulatory systems. It has been further suggested that neural implementations of performance monitoring functions, negative affect and pain perception overlap in anterior portions of medial frontal cortex. To investigate this matter, we combined a classic flanker task with interleaved blocks of electrical shocks. Before the experiment participants rated single electric stimuli to determine individual detection thresholds as well subjective stimulation maxima. Via this threshold range either painful (80% of previous pain maximum) or subtle (20% above detection threshold) trains of 20 stimulations were applied between flanker blocks. To quantify the physiological and psychological effects of the pain stimulation, we recorded galvanic skin responses (GSR) as well as heart rate (HR), corrugator activity (CA) and subjective mood ratings. Effects on performance monitoring functions were assessed via behavioral data and the error-related negativity (ERN). Direct pain stimulation led to significant changes in the affect-sensitive measurements, resulting in accelerated HR, increased GSR amplitudes as well as heightened CA. Mood ratings indicated that pain stimulation also enhanced subjective negative affect. Regarding the flanker task, main effects for congruency as well as a for pain stimulation were observed yielding generally faster reaction times after pain stimulation. In contrast, no pain effect was found for error rates. ERN amplitudes were decreased after pain blocks but differences did neither reach significance across group- nor subgroup-level based on pain intensity. Nevertheless, the results demonstrate that even performance-non-contingent pain leads to alterations in cognitive control and triggers behavioral adaptations. Whether these effects can be attributed solely to activity in medial frontal cortex remains debatable, but the present study might serve as a starting point for more refined experimental procedures aiming at examination of the interplay of negative affect, pain and cognitive control.
Saturday 16.00 - 17.30 Grote zaal Poster #33 Sensory processing Richard Hoechenberger
All authors & affiliationOhla, Kathrin, Psychophysiology of Food Perception, German Institute of Human Nutrition Potsdam-Rehbruecke, Nuthetal, Germany
Quick and reliable estimation of taste sensitivity using adaptive methods
Show/hide abstractMeasurements of perceptual sensitivity are a fundamental aspect of all psychophysiological investigations. In fact, the entire field of psychophysics was founded with the intention to unravel the relationship between physical and perceived stimulus intensity. While in the non-chemical senses, numerous methods for sensitivity threshold estimation have been developed during the past century, these methods have not yet been transferred to the chemical senses, especially the sense of taste, for several reasons. Firstly, stimulus delivery is rarely automated, which hampers repeated stimulation. Further, taste stimulation requires rather long inter-stimulus intervals of typically 20--30s, limiting the number of trials per experimental session. Lastly, liquid tastants must be readily prepared before testing; they cannot be created "on-line" during testing, unlike auditory or visual stimuli. Given these limitations, existing taste threshold estimation methods either suffer from low resolution due to few dilution steps, or they require multiple stimulations per trial, asking participants e.g. to identify an odd stimulus in a series. As participants are required to memorize and later retrieve previous perceptions, these procedures impose a high cognitive and memory load, and are therefore unsuitable for participants with limited cognitive resources, such as clinical populations. We set out to adapt two algorithms commonly used in the non-chemical senses, QUEST and SIAM, for simple, quick, and reliable estimation of taste thresholds. In both methods, participants are presented with only a single stimulus per trial, and must report whether the stimulus was perceived. The duration of both procedures was less than 10 minutes per taste quality. Both methods were found to be test-retest reliable, with higher reliability for QUEST. Further, thresholds estimated with both methods were highly correlated, indicating they were both measuring similar perceptual and sensory properties. We therefore recommend these methods for taste sensitivity estimation in populations with limited attentional or memory capacities.
Saturday 16.00 - 17.30 Grote zaal Poster #34 Sensory processing Jonas Vibell
All authors & affiliationJonas Vibell, University of Oxford, University of Hawaii Corinna Klinge, University of Oxford Massimiliano Zampini, University of Oxford Anna Christina Nobre, University of Oxford Charles Spence, University of Oxford
Endogenous Spatial Attention gives rise to Prior Entry: ERP Evidence
Show/hide abstractVibell, Klinge, Zampini, Spence, and Nobre (2007) demonstrated that endogenously attending to a sensory modality (vision or touch) modulated perceptual processing, in part, by the relative speeding--up of neural activation (i.e., the prior entry effect). It was unclear what was responsible for the shift in latencies, the fine temporal discrimination of the task of the Temporal Order Judgment (TOJ) task or the type of attentional modulation. The present study replicated Vibell et al.'s experimental design but crucially reversed the response and attentional dimensions to evaluate whether spatial attention would also yield similar prior entry latency effects in the early visual P1 potentials while maintaining an orthogonal design. The results show similar latency shifts by spatial attention, however, not in the sensory P1 potentials as in Vibell et al.'s study. The differences in onset of the latency shifts (prior entry effect) suggest that different neural mechanisms underlie attention to a specific sensory modality versus to a spatial location.
Saturday 16.00 - 17.30 Grote zaal Poster #35 Language Ece Bayram
All authors & affiliationEce Bayram, Ankara University Institute of Health Sciences Department of Interdisciplinary Neurosciences* Eda Aslanbaba, Ankara University Faculty of Medicine Department of Neurology* Muhittin Cenk Akbostanci, Ankara University Faculty of Medicine Department of Neurology*
Levodopa effects on spontaneous speech in Parkinson's disease
Show/hide abstractObjective:To assess levodopa effects on spontaneous speech in Parkinson's disease (PD). Background:Action-related language deficits are often reported in PD. These studies, however, include rather artificial language tasks such as verbal fluency, action-sentence compatibility effect and picture naming. Thus, in addition to patients not frequently reporting word deficits; verbal output during spontaneous speech in PD is reported to be similar to healthy controls. Methods: Ten healthy controls (HCs) and ten PD patients taking levodopa were included. Cognitive functions were assessed by Montreal Cognitive Assessment (MoCA), motor symptoms were evaluated using Unified Parkinson's Disease Rating Scale (UPDRS)- Part 3, and Cookie Theft Picture was used for spontaneous speech production. Patients were tested during "on" and "off" states. Five of the patients were first tested during "on" state, and the other five vice versa. Transcriptions and linguistic analyses were done by researchers blinded to "on" and "off" states. Total word count, mean length of utterance (MLU), verb and noun counts, number of different verb and nouns were analyzed. Results: Groups were matched fora ge, gender, years of education and handedness. MoCA scores were not different in between groups and states. UPDRS-Part 3 scores increased during "off". All aspects of the linguistic analyses were similar in between HCs and PD patients during "on". Whereas MLU, noun count and number of different nouns were similar during "on" and "off"; word count, verb count and number of different verbs were significantly lower during "off" compared to "on". Compared to HCs, PD patients produced significantly less verbs during "off". Conclusion: Levodopa appears to improve the verb deficit in PD during spontaneous speech in addition to motor improvement.
Saturday 16.00 - 17.30 Grote zaal Poster #36 Decision-making Trevor Chong
All authors & affiliationTrevor T-J Chong, Monash University, Australia Valerie Bonnelle, University of Oxford, UK Kai-Riin Veromann, University of Oxford, UK Julius Juurmaa, University of Tartu, Estonia Pille Taba, University of Tartu, Estonia Olivia Plant, University of Oxford, UK Masud Husain, University of Oxford, UK
Dissociation of Reward and Effort Sensitivity in Methcathinone-Induced Parkinsonism
Show/hide abstractMotivation requires us to weigh the perceived costs of an action against its potential benefits, with the basal ganglia playing a critical role in such cost-benefit valuations. Patients with basal ganglia pathology, such as those with idiopathic Parkinson's disease, can show significant deficits in motivation, but the mechanisms underlying these impairments are poorly understood. Methcathinone-induced Parkinsonism is a recently described extrapyramidal syndrome, characterised by globus pallidus and substantia nigra lesions, which provides a unique model of basal ganglia dysfunction. We assessed motivated behaviour in these patients to establish whether they exhibit altered sensitivity to either reward or effort. A novel cost-benefit decision-making task was used to evaluate effort and reward sensitivity in seven cases of methcathinone-induced Parkinsonism, and 18 age-matched, healthy controls. On each trial, participants decided whether they were willing to allocate varying levels of physical effort for different levels of reward. The reward required to motivate individuals to exert each level of effort on 50% of trials established their reward indifference points. Our main finding was that patients required greater rewards than controls to motivate them to exert a given amount of effort. However, the corresponding analysis on effort indifference points showed that patients and controls were no different in the amount of effort they were willing to invest for a given reward. Importantly, these results were not due to motor differences between groups. These results show a dissociation between reward and effort sensitivity in methcathinone-induced Parkinsonism. Pallidonigral complex dysfunction appears to bias cost-benefit decision-making, causing patients to become less sensitive to rewards, while maintaining normal sensitivity to effort costs. More broadly, this implies that diseases affecting different components of the basal ganglia may result in distinct amotivational syndromes, elements of which may be dissected using paradigms that are sufficiently sensitive to these differences.
Saturday 16.00 - 17.30 Grote zaal Poster #37 Language Sendy Caffarra
All authors & affiliationSendy Caffarra, Basque Center on Cognition Brain and Language Martha Mendoza, Basque Center on Cognition Brain and Language Doug Davidson, Basque Center on Cognition Brain and Language
Is the LAN effect an ERP artifact?
Show/hide abstractWords and morphemes are understood very quickly, but there are few techniques available to study the brain's response to them at the time-scale at which they occur. The left anterior negativity (LAN) has been proposed as the primary ERP candidate of morpho-syntactic processing, but recent findings have questioned whether the LAN effect, in fact, exists. Some authors hypothesize that this ERP effect arises from the overlap between two different components in different participants: the N400 (central-posterior negativity) and the P600 (i.e., posterior positivity, Osterhout & Mobley, 1995; Tanner, 2015). An alternative view describes the LAN as a functionally independent ERP component (Friederici, 2002; Molinaro et al., 2015), found in individual participants. The aim of the present project was to adjudicate between these two positions. EEG was recorded from eighty Spanish speakers who read 240 sentences containing determiner-noun gender violations or their associated controls. ERP data were analyzed in order to obtain ERP averages across subjects (grand average) and trials (subject average). Grand average ERP waveforms over participants, time-locked to the onset of the target noun, showed that gender violations elicited a clear LAN-P600 sequence (time windows: 300-500 ms; 500-800 ms) compared to controls. To test whether the observed LAN effect resulted from averaging together sub-populations of N400 and P600-effects, we examined the proportion of participants demonstrating different topographic patterns. Individual topographic maps of the gender violation effect between 300 ms and 500 ms showed a clear left-anterior negativity for 50% or more of the participants. A similar analysis of items showed the LAN effect associated with the majority of items. The present results support the idea that the LAN is a component at both the population as well as individual- levels of analysis. We argue that subject variability in component effects should be understood as random effects.
Saturday 16.00 - 17.30 Grote zaal Poster #38 Memory Lara Fine
All authors & affiliationMichael Weinborn, University of Western Australia Amanda Ng, University of Western Australia Erica Hodgson, University of Western Australia Denise Parker, University of Western Australia Romola Bucks, University of Western Australia
Aging and prospective memory: the importance of sleeping on it.
Show/hide abstractProspective memory -- the ability to remember to carry out intentions to be executed in the future -- is a cognitive skill that is vital for independent living, yet one that declines with age. Thus, identifying factors that influence this decline may lead to interventions to preserve older adults' prospective memory and ability to live independently. Disrupted sleep has recently come under scrutiny as a potential determinant of cognitive decline, with some researchers suggesting that age-related cognitive decline may be partially due to the poor sleep experienced by many older adults. The current research sought to evaluate whether sleep disruption mediates the relationship between older age and poorer prospective memory. Community dwelling older adults aged 55 to 93 years (N = 132) undertook objective assessment of sleep quality and quantity using actigraphy and participated in a laboratory-based prospective memory task. After controlling for education, sleep disruption, indeed, mediated the relationship between age and prospective memory. Specifically, more minutes awake (awakening length) explained the relationship between increasing age and poorer prospective memory (bootstrapped indirect path = -0.27, CI = -0.66 to -0.05) with 16% of the variance in prospective memory accuracy explained by the model, F (5, 126) = 5.10, p < .001. As well as being among the first to look at the association between sleep and prospective memory in older adults, this study is novel in demonstrating the potential importance of sleep for the relationship between age and prospective memory. Importantly, it suggests there may be opportunities to prevent prospective memory decline by addressing sleep disruption, helping older adults live better for longer.
Saturday 16.00 - 17.30 Grote zaal Poster #39 Sensory processing Malgorzata Wislowska
All authors & affiliationMalgorzata Wislowska, Centre for Cognitive Neuroscience, University of Salzburg, Salzburg, Austria Thomas Veale, School of Psychology, the University of Nottingham, Nottingham, United Kingdom Manuel Schabus, Centre for Cognitive Neuroscience, University of Salzburg, Salzburg, Austria Markus Bauer, School of Psychology, the University of Nottingham, Nottingham, United Kingdom
Prior expectations and sensory information interact in a Bayes-optimal way during perception
Show/hide abstractSurvival depends on our ability to efficiently process the available information. Contemporary neuroscience points out that the brain integrates information in a dynamic way, wherefore identical sensory information can impact the system differently. Top-down mechanisms like attention or expectation seem to have substantial influence on state of the brain, and ultimately on fate of incoming data [1]. Recent studies suggest that top-down and bottom-up information are fed back and forward though distinct frequency channels [2, 3], and integrated in a Bayes-optimal way [4]. In our study we challenged these theories through direct and independent manipulation of sensory and prior information, to examine their impact on behavior and underlying oscillatory processes. Twenty-four young and healthy participants performed a face/house discrimination task in an MEG scanner. Clarity of the pictures (sensory information) was manipulated by adding random noise to their phase components, titrated to performance using psychometric function fitting. Probability of occurrence of the two stimulus categories (prior information) was altered over the course of the experiment. Individual structural MR images were acquired for source-analysis. The results show that accuracy, criterion and reaction times were affected by sensory information. Importantly, expectancy seemed to have stronger impact on behavior for less certain stimuli, such that the shift in criterion was correlated with a Bayes-optimal observer. Finally, higher precision of sensory information led to enhanced post-stimulus gamma synchronization, whereas larger asymmetry of prior probability modulated predominantly pre-stimulus alpha/beta synchronization over occipital cortex. Our findings suggest that humans are capable to learn implicit priors and integrate them with stimulus information in a Bayesian way. The dynamics of the visual cortex revealed frequency specific separation of top-down and bottom-up information. Our findings are therefore convergent with the Bayesian perception hypothesis. Dynamic causal modelling will shed further light on the interaction between sensory and prior information.
Saturday 16.00 - 17.30 Grote zaal Poster #40 Electrophysiology methods Manuela Russo
All authors & affiliationManuela Francesca Russo, Queensland University of Technology Istvan Sulykos, Hungarian Academy of Science Patrick Johnston, Queensland University of Technology Jordy Kaufman, Swinburne University of Technology
Familiar cartographic contours as objects of visual expertise.
Show/hide abstractFacial inversion effects (FIE) (delayed reaction times to inverted faces, and delayed/increased amplitude N170 ERP) have been claimed to support the existence of face specific brain processes, because similar effects are not evident for other types of stimulus objects. However, Johnston et al. (2014) suggest that similar effects might exist for any class of familiar objects where particular conditions are met: that is, objects should 1) have a standard configuration of features; 2) a canonical view-point. A type of stimuli meeting these criteria are cartographic contours (maps), for which we propose, people may show culturally bound incidental expertise. As a proof of concept, we collected EEG data and RTs' data from Australians viewing a range of cartographic stimuli. During an ERP task, participants viewed upright versus upside down maps of Australia and Other countries. Results showed significantly greater N170 amplitudes to maps of Australia versus Other countries. There was an inversion effect on N170 latencies whereby peaks were later to inverted versus upright images of Australia. Such inversion effects were not evident for Other maps. Participants' RTs were also recorded for response times in identifying target images of Australia versus Other target maps, when stimuli were presented at eight different rotations. RTs were faster to Australia than to Other target maps. Furthermore, RTs to Australia were modulated by rotation (such that upright images promoted fastest responding, whereas fully inverted images lead to slowest responding), whereas RTs to unfamiliar images were slow regardless of image rotation. We conclude that, in general, people may show culturally bound incidental expertise for cartographic images of their home nation. We therefore propose that such stimuli offer unique opportunities for studying pre-existing (and non-motivated) visual expertise, since double dissociation effects may be evident in participants from different countries.
Saturday 16.00 - 17.30 Grote zaal Poster #41 Memory Samuel Fynes-Clinton
All authors & affiliationDr Lars Marstaller, University of Queensland, Centre for Advanced Imaging A/Prof. Hana Burianova, Swansea University, Department of Psychology; University of Queensland, Centre for Advanced Imaging
Children and adults show differences in brain activity during declarative memory retrieval
Show/hide abstractResearch demonstrates that healthy young and older adults engage a common functional network during autobiographical, episodic, and semantic memory retrieval. Behaviourally, research indicates that, compared to adults, prepubescent children show reduced declarative memory capabilities; however, to date, no studies have explored whether children engage the common functional network during declarative memory retrieval. Thus, the aim of this study was to investigate whether children activate similar neural correlates of declarative memory to adults. We used event-related functional magnetic resonance imaging (fMRI) to record brain activity in 21 children (8 females; Mean age = 10.90, SD = 0.81) and 21 adults, (11 females; Mean age = 26.71, SD = 3.79) while they performed a previously validated declarative memory task (Burianova & Grady, 2007). In our version of the task adapted for prepubescent children, participants were shown pictorial stimuli as cues for retrieval, and, by manipulating the retrieval demand, autobiographical, episodic, or semantic responses to the same stimuli were extracted. Results show that, compared to adults who, in line with previous research, recruited a large-scale bilateral temporo-parietal network, children engaged a left lateralised occipito-temporal network during all three types of declarative memory retrieval. This network showed early-onset sustained activity and was functionally connected to the left hippocampus. These findings suggest that prepubescent children might rely heavily on semantic processing during declarative memory, including autobiographical and episodic retrieval, whereas in adults this is associated with more distributed processing in the brain. Our findings also suggest that functional activity during declarative memory retrieval does not fully differentiate before pubescence and support the idea that autobiographical and episodic memory is related to the development of a sense of self.
Saturday 16.00 - 17.30 Grote zaal Poster #42 Memory Mathilde Bostelmann
All authors & affiliationMathilde Bostelmann, Laboratory of Brain and Cognitive Development, Institute of Psychology, University of Lausanne, Lausanne, CH Floriana Costanzo, Department of Neuroscience, Bambino Gesù Children's Hospital, Rome, Italy Deny Menghini, Department of Neuroscience, Bambino Gesù Children's Hospital, Rome, Italy Lorelay Martorana, Department of Neuroscience, Bambino Gesù Children's Hospital, Rome, Italy Paolo Ruggeri, Laboratory for Experimental Research on Behavior, Institute of Psychology, University of Lausanne, Lausanne, CH Stefano Vicari, Department of Neuroscience, Bambino Gesù Children's Hospital, Rome, Italy Pierre Lavenex, Laboratory of Brain and Cognitive Development, Institute of Psychology, University of Lausanne, Lausanne, CH Pamela Banta Lavenex, Laboratory of Brain and Cognitive Development, Institute of Psychology, University of Lausanne, Lausanne, CH
Competitive interaction between spatial memory systems in Down syndrome
Show/hide abstractWe have previously shown that individuals with Down syndrome (DS; a.k.a. Trisomy 21) are impaired on a high-resolution "hippocampus-dependent" place-learning task. To infer whether this impairment may be primarily due to the dysfunction of particular hippocampal circuits, we compared participants with DS to typically developing (TD) mental age-matched children and adults on a low-resolution place-learning task thought to rely specifically on the CA1 region of the hippocampus. Furthermore, because parallel spatial memory systems, supported by different neurobiological substrates, contribute different spatial representations necessary for real world navigation, we also assessed low-resolution spatial response learning. The spatial response learning system depends on the dorsal striatum and creates fixed stimulus-response representations. Our results showed that whereas TD children performed as well as adults at place learning, they performed less well than adults at response learning. Interestingly, DS individuals showed an intermediate profile: Although DS individuals exhibited relatively preserved performance on the place-learning task, their performance did not reach adult levels. Nevertheless, as many DS individuals as TD children performed above chance level at place learning. In contrast, more DS individuals than TD children performed above chance level at response learning. Our findings indicate that individuals with DS perform better on low-resolution than high-resolution place learning tasks, suggesting that spatial pattern separation abilities dependent on the dentate gyrus are more impaired than topological place learning abilities dependent on CA1. Moreover, based on the hypothesized competitive interaction between the hippocampal- and striatal-dependent memory systems subserving place and response learning, respectively, our results further suggest that in DS, impaired hippocampal function may enhance striatal response learning. Since most of the evidence suggesting a competitive interaction between spatial memory systems comes from rodent studies, our results provide rare evidence of the nature of this interaction in humans.
Saturday 16.00 - 17.30 Grote zaal Poster #43 Attention Corinne Bareham
All authors & affiliationTom Manly, MRC-Cognition and Brain Sciences Unit, Cambridge, UK Andres Canales-Johnson, MRC-Cognition and Brain Sciences Unit, Cambridge, UK, Department of Psychology, University of Cambridge, UK. Valdas Noreika, MRC-Cognition and Brain Sciences Unit, Cambridge, UK, Department of Psychology, University of Cambridge, UK Tristan A Bekinschtein, Department of Psychology, University of Cambridge, UK
Neural dynamics of spatial attention during drowsiness: is neglect-like hemispheric competition lost when falling asleep?
Show/hide abstract We recently showed using an auditory localisation task that when right-handers (N=26) become drowsy they showed a rightward bias in attention; judging more left tones as "right" than vice versa ("a la neglect"). We speculated that drowsiness might lead to a shift in hemispheric competition, leading to a rightward bias in attention. Here, the EEG dynamics showed that for the correct trials during alert periods, there is evidence of hemispheric competition with greater ERP amplitude for the left correct trials in the right parietal electrode cluster and for the right correct trials in the left parietal electrode cluster around 350 to 550 ms after the tone. For correct trials, this asymmetry dissipates with drowsiness. For errors however, there appears to be a different story. Left/right fronto-parietal asymmetry is evidenced but only in the drowsy trials and, in the opposite direction to the correct trials. Left errors are marked by decreased ERP amplitude in the right fronto-parietal clusters and increased amplitude in the left fronto-parietal clusters at 150-250 ms following the tone and vice versa for right errors. This asymmetry remains at 350-550 ms in frontal regions when errors in tone laterality judgments are made. Notably, the decreased amplitude of the left errors at 350-550 ms in the right frontal cluster in drowsiness is similar to the decreased amplitude of the right correct ERPs. This hemispheric asymmetry in the late ERPs during drowsy trials suggests that participants may activate a similar motor plan for left errors as right correct trials. Interestingly, this effect seems to be driven by earlier shifts in attention evidenced by a change in hemispheric competition in the early ERPs. Given the proposed competing models of spatial attention and neglect, our findings are consistent with the interhemispheric competition model with alertness as a modulator of hemispheric asymmetry.
Saturday 16.00 - 17.30 Grote zaal Poster #44 Language Mareike Floegel
All authors & affiliationFloegel Mareike, Cognitive Neuroscience Group, Brain Imaging Center and Department of Neurology, Goethe University, Frankfurt, Germany Fuchs Susanne, Leibniz Center for General Linguistics (ZAS), Berlin, Germany Kell Christian A., Cognitive Neuroscience Group, Brain Imaging Center and Department of Neurology, Goethe University, Frankfurt, Germany
Adaptation to dichotically presented spectral and temporal real-time perturbations of auditory speech feedback
Show/hide abstractWhen a speaker's auditory feedback is perturbed during speech production, the speaker adapts production to compensate for the disturbance [1,2] probably via processes in the right hemisphere [3]. Because, so far, only spectral feedback manipulations were investigated, it is unclear whether the observed right-lateralization reflects a right hemispheric specialization for feedback analyses in general [3] or a right hemispheric specialization for spectral processing [4]. We tested whether adaptation to spectral speech features lateralizes differently from adaptation to temporal speech features. German speakers' auditory feedback was altered spectrally (n=9) or temporally (n=9) during the production of CVC monosyllabic pseudowords. Spectral perturbations increased the vowel's F1, temporal manipulations decelerated the vowel (20% over 40 trials in steps of 0.05%) [5]. Auditory feedback was presented dichotically (feedback/stimulus manipulation only in one ear while the other ear perceived unperturbed feedback/stimulus) or diotically (perturbed/unperturbed feedback/stimulus in both ears). Participants decreased produced vowels' F1 over trials to compensate for spectral feedback perturbations when perturbed auditory feedback was presented to both ears, or only to the left ear. For temporal perturbations, participants shortened their vowel production over trials if altered auditory feedback was presented to both, or only the right ear. We further investigated spectral and temporal auditory discrimination on a purely perceptual level in a similar dichotic setting. Lateralization effects for spectral (right hemisphere) and temporal (left hemisphere) speech features were less pronounced compared to production and not significant. This identifies temporal vs spectral processing as an important determinant of functional lateralization and suggests production tasks amplify sensory-related lateralization.
Saturday 16.00 - 17.30 Grote zaal Poster #45 Social behavior Simmy Poonian
All authors & affiliationSimmy Poonian, Macquarie University Pierre Jacquet, Ecole Normale Superieure Florian Waszak, Universite Paris Descartes
Differences in neural oscillations for optimal movements and goal attainment when observing others actions
Show/hide abstractWhen we observe others' behaviour it is thought our brain makes predictions about their actions and intended goals. It currently remains unclear which aspects of the action sequence, i.e. the goals, grip type or optimal interactions between the two, are indexed by changes in alpha and beta oscillations. In this study, we investigated the temporal profile of the oscillatory response to others' actions, in relation to grip type, goals and optimality. In an EEG study participants were presented with videos of an actor interacting with an object (a box), where the type of grip (power/precision) was varied with the intended goal (open/light), to create observations where goal attainment was reached using optimal or sub-optimal actions. Cluster-based permutations revealed a temporal profile of alpha and beta oscillations in relation to grips, goals and optimality. Alpha differences in grip type were found in sensorimotor electrodes contralateral to the hand used in the video, during the time the actor was interacting with the box. Differences in beta between optimal and sub-optimal actions were revealed 200ms after the initiation of the goal-related action in fronto-central areas. We propose that this reflects activity related to whether the expectation about the goal-related action and the observed goal-related action matches or not. Finally, activity related to the intended goals, regardless of the type of action, was revealed through differences in beta activity. An initial early difference in beta activity over frontal areas was revealed between the two goal types, soon after the actor's hand had interacted with the box. A later difference in beta activity over right central electrodes occurred once the goal had been achieved. Our findings indicate a temporal profile when observing others actions, where movement and goal information is processed early on and is followed by processing of action-goal expectations and goal attainment.
Saturday 16.00 - 17.30 Grote zaal Poster #46 Attention Lisa Wittenhagen
All authors & affiliationLisa Wittenhagen, Queensland Brain Institute, The University of Queensland, Australia* Jason B. Mattingley, Queensland Brain Institute & School of Psychology, The University of Queensland, Australia
Role of spatial attention in modulation of neural representations of illusory shapes
Show/hide abstractThere has been a longstanding debate as to whether visual attention modulates the perception of illusory shapes (1). Separate lines of research have shown that both the presence of illusory shapes, and the allocation of spatial attention, are associated with enhanced neural responses (2,3,4,5). Here we asked whether spatial attention modulates neural responses to illusory shapes by having participants focus on stimuli presented in one visual field (left or right), while ignoring those on the other side. In two separate experiments, we used electroencephalography (EEG) to measure event-related potentials (ERPs) and steady-state visual evoked potentials (ssVEPs), elicited by a Kanizsa square (6) in one hemifield and a competing control stimulus, matched for all relevant physical features, in the other. In Experiment 1 (N=29), the inducers used to construct the Kanizsa and control stimuli flickered at unique frequencies (10 and 13.33 Hz) to elicit ssVEPs. In Experiment 2 (N=25), segments were removed briefly from static, circular placeholders within the left and right visual hemifields to induce ERPs. In both experiments, participants were instructed to attend covertly to the left- or right-sided stimulus to detect brief contrast changes of the inducers. ssVEP analyses revealed that illusory shapes that were ignored yielded a significantly larger neural response than those elicited by stimuli that did not form an illusory shape. By contrast, when the stimuli were attended, neural responses to the illusory and control stimuli were statistically indistinguishable. The results from Experiment 2 revealed that illusory shapes are processed ~30 ms earlier than control stimuli. Thus, it appears that illusory shape perception does not rely on visuospatial attention and arises at an early stage of visual processing, likely signaling the presence of an object, available for future goal-directed behavior (7,8).
Saturday 16.00 - 17.30 Grote zaal Poster #47 Clinical Julie Hall
All authors & affiliationJulie M Hall, Western Sydney University James M Shine, University of Sydney Kathryn M Broadhouse, University of Sydney Kaylena A Ehgoetz Martens, University of Sydney Moran Gilar, University of Sydney Ahmed A Moustafa, Western Sydney University Simon JG Lewis, University of Sydney
Similar organization of the structural connectome provides evidence that freezing of gait in Parkinson's disease is a functional disorder.
Show/hide abstractBackground: Freezing of gait (FOG) is a debilitating symptom of Parkinson's disease (PD) defined as a transient inability to initiate or maintain forward progression1. Although the pathophysiology is not yet fully understood, there is growing consensus that FOG is due to inefficient functional coupling between brainstem and fronto-parietal regions, linked by the basal ganglia (BG)2. The BG are a major subcortical hub involved in large-scale communication and integration of the human connectome3. As such, structural alterations within the BG might influence the organization of the connectome into either a more integrated or segregated state, deviating from the harmonious network organization. We hypothesized that FOG could be associated with a more segregated topology, impairing vital communication between key hubs. In contrast, a more integrated connectome might cause communicational conflict within competing yet complimentary pathways, which in turn may also lead to FOG. Methods: Forty PD patients with and twenty-seven PD patients without FOG were included in this study and were matched for gender, age, disease duration and levodopa dose equivalence. FOG patients had more severe disease severity as indicated by a higher Hoehn and Yahr score. Using diffusion weighted imaging, we investigated the structural organization of the whole brain connectome using the Lausanne 250 parcellation atlas. Data was analyzed using graph theoretical methods, including the module degree Z-score and the participation coefficient, the latter indicating the integrative or segregated state of the connectome. Results: Independent t-tests (FDR corrected) showed no significant differences on either of these measures between the groups. Conclusion: Our preliminary results indicate that structural white matter organization of the connectome does not significantly contribute to FOG. As such, it is likely that the underlying mechanisms responsible for FOG are functional in nature. Further analysis using resting state functional MRI can elucidate whether FOG is affected by a functional reorganization of the connectome.
Saturday 16.00 - 17.30 Grote zaal Poster #48 MRI methods Lukas Snoek
All authors & affiliationLukas Snoek, University of Amsterdam Steven Miletic, University of Amsterdam H. Steven Scholte, University of Amsterdam
Controlling for confounds in multivoxel pattern analyses: an alternative approach
Show/hide abstractOver the past decade, multivoxel pattern analyses (MVPA) have become a popular alternative to traditional mass-univariate analyses in fMRI research. One major advantage of MVPA is that it is more sensitive in detecting signal. Recently, several studies1,2 have shown that MVPA is also more sensitive to confounds. These studies additionally propose that confounds, that are expressed on a continuous scale, can be accounted for by "regressing out" the influence of confounds from the fMRI activity (or patterns) directly. We contend that this is an incorrect approach, as it removes confounding variance from the independent variable, but not the dependent variable (DV) Instead, we suggest that the influence of confounds on the dependent variable (i.e. the to-be-decoded factor) should be accounted for. For continuous DVs, we argue that the confounding variable should be regressed out of the DV itself. For categorical DVs, which is the most common scenario for MVPA studies, this is statistically not possible. Instead, we suggest a technique called inverse probability weighting3, which downweights samples in which the confounding variable is relatively more predictive of their true class. Using an existing large between-subject fMRI dataset, in which we aim to decode IQ both as a continuous and as a categorical DV from voxel-based morphometry (VBM) data, we empirically show that the proposed methods accurately controls for confounding factors (such as brain size and gender) while maintaining above-chance decoding performance. We contrast this to previously suggested methods1,2, which we show to erroneously lead to below-chance decoding accuracy. From our results, we conclude that confounds in MVPA analyses inflate effect sizes, but can be effectively controlled for using a set of straightforward techniques.
Saturday 16.00 - 17.30 Grote zaal Poster #49 Connectivity/networks Teresa Sousa
All authors & affiliationTeresa Sousa, Institute of Nuclear Sciences Applied to Health (ICNAS) - University of Coimbra, Institute for Biomedical Imaging and Life Sciences (CNC.IBILI) - Faculty of Medicine of the University of Coimbra, Faculty of Psychology and Neuroscience - University of Maastricht; Joao V. Duarte, Institute of Nuclear Sciences Applied to Health (ICNAS) - University of Coimbra, Institute for Biomedical Imaging and Life Sciences (CNC.IBILI) - Faculty of Medicine of the University of Coimbra; Gabriel N. Costa, Institute of Nuclear Sciences Applied to Health (ICNAS) - University of Coimbra, Institute for Biomedical Imaging and Life Sciences (CNC.IBILI) - Faculty of Medicine of the University of Coimbra; Valentin G. Kemper, Faculty of Psychology and Neuroscience - University of Maastricht; Ricardo Martins, Institute of Nuclear Sciences Applied to Health (ICNAS) - University of Coimbra, Institute for Biomedical Imaging and Life Sciences (CNC.IBILI) - Faculty of Medicine of the University of Coimbra; Rainer Goebel, Faculty of Psychology and Neuroscience - University of Maastricht, Department of Neuroimaging and Neuromodeling - Netherlands Institute for Neuroscience (KNAW); Miguel Castelo-Branco, Institute of Nuclear Sciences Applied to Health (ICNAS) - University of Coimbra, Institute for Biomedical Imaging and Life Sciences (CNC.IBILI) - Faculty of Medicine of the University of Coimbra
The Perceptual Integration of Visual Motion Revealed by hMT+ Interhemispheric Connectivity: a 7 Tesla study
Show/hide abstractThe mechanism by which human vision segments and binds stimulus' features to produce a coherent percept is still not well-understood (Treisman, 1996; Burwick, 2014). Perceptual interpretation of ambiguous moving stimuli, for which perception alternates between competing interpretations in conditions of identical sensory input, raises the question of how the visual system integrates global patterns of motion from its components (Leopold and Logothetis, 1999; Sterzer et al., 2009). The human motion complex (hMT+) is well known to be involved in motion perception (Tootell et al., 1995; Kolster et al., 2010) and has been shown to underlie the perceptual binding of overlapping moving surfaces (Castelo-Branco et al., 2002). However, its role in interhemispheric integration of visual motion information was only explored with apparent motion paradigms (Sterzer et al., 2003; Muckli et al., 2005; Rose and Buechel, 2005; Genç et al., 2011). Here we aimed to investigate how perceptual visual motion integration vs. segregation of interhemispheric non-overlapping 1D directional cues is modulated by interhemispheric functional connectivity. A previously described ambiguous moving stimulus (Wallach, 1935; Wuerger et al., 1996), which can be perceived as a coherent pattern comprehending both visual hemi-fields or as two separate non-overlapping component surfaces (one in each visual hemi-field), was used. Data from nine healthy participants were acquired using high-resolution functional magnetic resonance imaging (fMRI) at 7T and analyzed offline to estimate the variation of hMT+ interhemispheric correlation in time. We found that hMT+ interhemispheric correlation changes in time depending on whether participants integrate all motion features into the percept of a single coherent pattern or whether they segregate visual motion features and perceive two separate surfaces. We present the first fMRI-based evidence of a close relation between interhemispheric functional connectivity in hMT+ regions and the perceptual switches involving differential long-range integration of visual moving stimuli.
Saturday 16.00 - 17.30 Grote zaal Poster #50 Connectivity/networks David Pascucci
All authors & affiliationDavid Pascucci. Department of Psychology, University of Fribourg, Fribourg, Switzerland. Alexis Hervais-Adelman. Brain and Language Lab, Department of Clinical Neuroscience, University of Geneva, Geneva, Switzerland Christoph M. Michel. Functional Brain Mapping Lab, Department of Fundamental Neuroscience, University of Geneva, Geneva, Switzerland. Gijs Plomp. Department of Psychology, University of Fribourg, Fribourg, Switzerland.
Gating by induced asynchrony: The role of parietal cortex in selective attention
Show/hide abstractVoluntary selective attention operates through top-down mechanisms of signal enhancement and suppression, mediated by oscillations in the α-band. But how such top-down influences regulate processing in visual cortex remains poorly understood. In the present work, we combined dynamic Granger-causality analysis based on EEG source imaging, and phase-amplitude coupling (PAC) measures to characterize the pattern of large-scale directed interactions that orchestrates selective attention, and how these interactions affect stimulus processing in visual areas. Under changing task demands, subjects either attended to or ignored briefly presented gratings. Time-varying, directed connectivity analysis showed rapid increases of bottom-up γ-band interactions from visual areas in response to attended stimuli. Ignored stimuli, instead, evoked distributed and sustained top-down α-band interactions, originating from parietal cortex. These connectivity changes occurred together with increased α-γ PAC in visual areas for attended stimuli. Furthermore, multi-level modeling revealed that parietal α-band interactions disrupted the α-γ coupling in visual cortex, which in turn reduced the amount of γ-band outflow from visual areas. Our results are a first demonstration of how directed interactions affect cross-frequency coupling in downstream areas. These findings suggest that parietal cortex realizes selective attention by disrupting cross-frequency coupling at target regions in a way that prevents them from propagating task-irrelevant information.
Saturday 16.00 - 17.30 Grote zaal Poster #51 Attention Jana Tegelbeckers
All authors & affiliationTegelbeckers Jana, Child and Adolescent Psychiatry, Otto von Guericke University Magdeburg Brechmann Andre, Special Lab Non-Invasive Brain Imaging, Leibniz Institute for Neurobiology, Magdeburg Breitling Carolin, Child and Adolescent Psychiatry, Otto von Guericke University Magdeburg Bonath Bjoern, Child and Adolescent Psychiatry, Otto von Guericke University Magdeburg FLechtner Hans-Henning, Child and Adolescent Psychiatry, Otto von Guericke University Magdeburg Krauel Kerstin, Child and Adolescent Psychiatry, Otto von Guericke University Magdeburg
Neural mechanisms underlying the benefit of distraction: an fMRI study in adolescents
Show/hide abstractUnexpected events attract attention and disrupt ongoing behavior. This orienting response is of high evolutionary relevance but results in distraction in everyday cognitive tasks. On the contrary, recent studies repeatedly reported beneficial effects of short attentional distractions, particularly but not only in patients suffering from attention deficit hyperactivity disorder. To gain insight into the neural effects of potentially distracting unexpected sounds on attention, we paired a visual flanker task with novel/unrepeated and familiar/repeatedly presented sounds. 23 adolescents aged between 11 and 16 years underwent one passive as well as three active runs of the task during fMRI scanning (3T). We investigated the effects of novel sounds on behavior as well as on brain activity during task performance, without concurrent task and in terms of functional connectivity alterations. We found an improvement in error rates and reaction times following novel sounds compared to the baseline without sound. Furthermore, novel sounds elicited the expected activation pattern in the bilateral superior temporal gyri (STG) and the right inferior frontal gyrus (rIFG) but only during task execution. During passive listening, novel sounds solely activated primary auditory cortical areas. PPI analyses revealed increased functional connectivity between the rIFG as well as rSTG and the anterior cingulate cortex during task performance. Moreover, we found a significant correlation between STG activity and mean reaction times particularly when novel sounds were presented. Taken together, our results indicate that the orienting network consisting of the rIFG and bilateral STG is activated only when a disruption of behavior is necessary. In this case, the coupling between orienting network areas and cognitive control areas increases, leading to improved task execution. Thus, the role of the orienting response might be best described as intercepting and resetting current processes which might as well result in behavioral facilitation under certain conditions.
Saturday 16.00 - 17.30 Grote zaal Poster #52 Attention Hannah Hiebel
All authors & affiliationHannah Hiebel, University of Graz, Austria Joe Miller, University of Graz, Austria Clemens Brunner, University of Graz, Austria Andrey R. Nikolaev, University of Leuven, Belgium Margit Höfler, University of Graz, Austria Anja Ischebeck, University of Graz, Austria Christof Körner, University of Graz, Austria
Attention and Memory in Multiple Target Visual Search: A Combined EEG and Eye Tracking Study
Show/hide abstractWhen searching for two identical targets, one must continue the search after finding the first target and memorize its location. Thus, the multiple-target search paradigm allows us to study attention and memory at different stages of visual search. The aim of this study was twofold: First, we investigated the influence of set size on the neural response to a target. Second, we explored how memory encoding of the target is reflected in oscillatory brain activity. EEG and eye movements were co-registered while participants moved their eyes freely in a multiple-target visual search. They were asked to indicate the number of targets in displays containing either one or two targets among distractors. Set size of the displays was varied between 10, 22, and 30 items. In such a way, we manipulated the number of distractor fixations preceding the first target fixation, since the probability of an early target detection gradually decreases with set size. To investigate attentional processes, we analyzed fixation-related potentials for the first target. They revealed a late P300, most pronounced over parieto-occipital brain regions. The amplitude of this target-related P300 was larger for set sizes 22 and 30 than for set size 10. Since the P300 is a well-known indicator of deviant stimuli, this suggests that the number of inspected distractors defines the extent to which the target is perceived as a deviant. To investigate target memory, we analyzed fixation-related spectral perturbations time-locked to the first target fixation. We compared spectral power changes in a post-target period to a matched period without preceding target fixation. Our results help to understand how the brain responds to the detection of significant events such as a target, and how the brain maintains information about such events during continuing visual search.
Saturday 16.00 - 17.30 Grote zaal Poster #53 Decision-making Joanne Van Slooten
All authors & affiliationJoanne Van Slooten, Vrije Universiteit Amsterdam Sara Jahfari, Vrije Universiteit Amsterdam Tomas Knapen, Vrije Universiteit Amsterdam Jan Theeuwes, Vrije Universiteit Amsterdam
Pupil responses are indicators of value-based decision-making
Show/hide abstractCan we understand how people reach and evaluate value-based decisions just by looking at their pupil response? This would require a precise mapping between the cognitive processes underlying value-based decision-making and the variability in pupil responses (under constant illumination). To investigate this relationship, we analysed pupil responses during a probabilistic selection task (Frank, 2004), where variability in value expectation, violations of value expectation (signed reward prediction errors), and choice strategy (explore/exploit tendency) were formally estimated using a Bayesian hierarchical reinforcement learning model (Frank et al., 2007; Lee, 2011). Prior to choice, we observed that a dilatory pupil response predicted the estimated (per-trial) value difference between the two choice options. Furthermore, cross-subject differences in taking an exploitative or explorative choice strategy modulated the peak amplitude of this expected value-related pupil response. At the time of feedback (after choice), we observed a biphasic pupil response, where early dilation related inversely to trial-by-trial value-differences between choice options and late constriction closely tracked signed reward prediction errors. These findings show strong, and specific, relationships between per-trial pupil responses and reinforcement learning variables that describe value-based learning and decision-making. Our findings could benefit research on Parkinson's disease, as pupil responses can be used to assess therapeutic efficacy by providing non-invasive access to the affected reinforcement learning process.
Saturday 16.00 - 17.30 Grote zaal Poster #54 Attention Joe Miller
All authors & affiliationJoe Miller, Universty of Graz, Austria Hannah Hiebel, Universty of Graz, Austria Margit Höfler, Universty of Graz, Austria Ian Donald Gilchrist, University of Bristol, UK Christof Körner, Universty of Graz, Austria Anja Ischebeck, Universty of Graz, Austria
Fixation-Related fMRI Analysis of Target and Distractor Fixations in Visual Search
Show/hide abstractIn a serial visual search for multiple identical targets in the same display, participants must attend to items one after the other, discriminate targets from distractors, and store target locations in working memory. We combined fMRI and eye tracking to investigate which brain regions are involved in the detection and memorization of targets and whether task relevance of a target can be discriminated from its rarity. In a first experiment, participants were asked to indicate the number of targets (“T”s; 0-3 depending on search condition) among distractors (“L”s) in a display. Participants could move their eyes freely. To identify processes related to the target memorization, we contrasted trials from search conditions with 2 and 0 targets. This analysis revealed increases in bilateral activation in the precuneus, superior parietal lobules, and middle frontal gyri. Activation in these areas also increased from 0 to 3 targets. These regions are associated with visual working memory, suggesting that target locations were held in working memory. We next examined attentional processes related to target fixation. In this analysis each target fixation was matched to a corresponding distractor fixation of the same search condition. We found increased activation in prefrontal and parieto-occipital regions for target versus distractor fixations. Activations in these areas likely indicate an increase in attentional processes. To account for activation differences that were due to the rarity of the target relative to the distractors, we conducted a follow-up experiment. In this experiment, a task-irrelevant rare distractor (upside-down “L”) was present in 50% of displays. We expect to see stronger activation in frontal brain areas for target versus rare distractor fixations. This would indicate that frontal cortex activations are related to task relevance rather than item rarity. The results from these experiments also demonstrate the suitability of fixations as events in an fMRI analysis.
Saturday 16.00 - 17.30 Grote zaal Poster #55 Sensory processing Vittoria Spinosa
All authors & affiliationVittoria Spinosa, Brainlab-Cognitive Neuroscience Research Group, Institute of Neurosciences, University of Barcelona, Spain Iria SanMiguel, Brainlab-Cognitive Neuroscience Research Group, Institute of Neurosciences, University of Barcelona, Spain
Match and mismatch effects on auditory ERPs: separate mechanisms or a continuum?
Show/hide abstractGenerally, sensory event-related potentials (ERPs) are attenuated for expected stimuli (matching the predictions), while they are enhanced for unexpected stimuli (mismatching the predictions). Additionally, several specific error-related responses, such as the mismatch negativity (MMN) have been described. According to the predictive coding framework, sensory responses reflect the degree of match between input and prediction, thus match and mismatch effects are opposite ends of a continuum. However, in the ERP literature, match and mismatch effects have been mostly examined separately, and their relationship is unclear. Moreover, there is little understanding of the endogenous signals that cause these effects (i.e., the predictions). We recorded ERPs while participants performed button presses that were associated with either sounds or silences, and occasionally violated these associations. We examined ERPs reflecting the match and mismatch between predictions and inputs. Using silences as an input, we were able to uncover the endogenous signals underlying the match/mismatch effects on the stimulus-driven ERPs. Four event types (Expected sounds, unexpected sounds, expected silences and unexpected silences), all produced by a button press, were compared to the corresponding event recorded in an unpredictable control condition. Up to a latency of ca. 120 ms., brain signals were mostly driven by the expectations, being negatively displaced when a sound was expected, compared to when a silence was expected. Subsequently, match and mismatch effects followed similar time-courses modulating the sound-evoked potentials towards opposing polarities. The endogenous (silence-related) signals followed the same pattern, with a positive displacement for predicted silences and a negative displacement for unpredicted silences in the typical MMN time-window. The results provide partial support for the “degree of match” interpretation of sensory responses. This approach allowed to relate well-known prediction-related ERP effects to the predictive coding framework, and to understand the time-course of the endogenous signals underlying such effects.
Saturday 16.00 - 17.30 Grote zaal Poster #56 Attention Katarina Slama
All authors & affiliationS. J. Katarina Slama, Helen Wills Neuroscience Institute, University of California, Berkeley, USA.* Anne-Kristin Solbakk, Department of Psychology, University of Oslo, Norway. Tor Endestad, Department of Psychology, University of Oslo, Norway. Pål G. Larsson, Department of Neurosurgery, Division of Clinical Neuroscience, University of Oslo Hospital, Rikshospitalet, Norway. Jack J. Lin, Department of Neurology, University of California, Irvine, USA. David King-Stephens, Department of Neurology, California Pacific Medical Center, San Francisco, USA. Kenneth D. Laxer, Department of Neurology, California Pacific Medical Center, San Francisco, USA. Peter B. Weber, Department of Neurological Surgery, California Pacific Medical Center, San Francisco, USA. Robert T. Knight, Helen Wills Neuroscience Institute and Department of Psychology, University of California, Berkeley, USA.
Intracranial recordings define a cortico-hippocampal network in top-down and bottom-up visual attention
Show/hide abstractVisual attention enables humans to navigate a complex and cluttered world. It can be divided into top-down control and bottom-up capture. The contribution of prefrontal and parietal cortices to effortful top-down and automatic bottom-up attention has been extensively studied in humans and non-human primates using neuroimaging and electrophysiology. However, the precise spatio-temporal dynamics supporting these processes is unknown. We examined the spatiotemporal dynamics of top-down and bottom-up attention using neural signals recorded from subdural electrodes in patients with medically refractory epilepsy (n = 5). Subjects performed a visual search task designed to elicit top-down (tilted green triangle amidst green triangles) and bottom-up attention (red triangle amidst green triangles). Accuracy was higher and response times were faster in the bottom-up condition. Local field potential power in the broadband high gamma range (HG; 80-150 Hz) indexed local neural activity. Electrodes with significant HG increases over baseline (40% of all analyzed electrodes, FDR-corrected according to Benjamini-Hochberg procedure at p < 0.05) were considered task-active. Half (50%) of the task-active electrodes were found in the classical fronto-parietal attention network. Notably, 7% of task-active electrodes were found in parahippocampal areas, and 2% in anterior cingulate cortex. Within task-active electrodes, the majority (51%) had greater activity for top-down than bottom-up attention and 8% showed the reverse pattern with enhanced activity during bottom-up attention. Nearly half of the top-down electrodes (41%) were found in frontal cortex in contrast to only 8% in parietal cortex. Bottom-up electrodes were found in similar numbers in (43%) parietal cortex and frontal cortex (36%). We confirm a key role for the fronto-parietal attention network and observe a central role of prefrontal cortex during top-down attention. Further, we demonstrate the involvement of the anterior cingulate cortex and the parahippocampal regions in both top-down and bottom-up attention.
Saturday 16.00 - 17.30 Grote zaal Poster #57 Memory Roland Benoit
All authors & affiliationRoland G. Benoit, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany Daniel L. Schacter, Department of Psychology, Harvard University, Cambridge, MA, USA
Forming real-life attitudes via affective simulations: the contribution of the ventromedial prefrontal cortex
Show/hide abstractHumans are capable of imagining virtually any hypothetical episode. For example, we can picture what it would be like to meet with a friend at a particular place for the first time. Such episodic simulation is based on a core network of brain regions that include the ventromedial prefrontal cortex (vmPFC). Recent evidence indicates that this region supports simulations by co-activating affective representations of the individual elements (e.g., of the place and of the person) that make up the episode (Benoit, Szpunar, & Schacter, 2014). We hypothesized that such simulations, in turn, can also change our attitudes towards these very elements via a transfer of affective value between their respective representations within the vmPFC. To test this hypothesis, we scanned participants with fMRI while they imagined meeting liked or disliked people (UCS) at initially neutral, though familiar, places (CS). Following these simulations, participants' attitudes towards the places (CS) had changed with the affective value of the paired person (the respective UCS), demonstrating that mere imaginings can change real-life attitudes. The fMRI data further support the hypothesized contribution of the vmPFC in mediating this effect: First, using representational-similarity analysis, we observed that the activity pattern within this brain region coded for the identity of the individual elements and also reflected their respective affective value. The vmPFC thus indeed seems to code for affective representations of elements from our environment (such as familiar people and places). Secondly, activation in the vmPFC during the simulations not only reflected the affective value of the UCS (the person) but also predicted changes in attitude towards the CS (the place). The vmPFC may thus mediate simulation-induced attitude-change by transferring affective value between representations.
Saturday 16.00 - 17.30 Grote zaal Poster #58 Consciousness Shiri Makov
All authors & affiliationShiri Makov, The Gonda Center for Multidisciplinary Brain Research, Bar Ilan University, Israel; Sagol School of Neuroscience, Tel Aviv University, Israel. * Omer Sharon, Sagol School of Neuroscience, Tel Aviv University, Israel. Nai Ding, College of Biomedical Engineering and Instrument Sciences, Zhejiang University, China. Michal Ben-Shachar, The Gonda Center for Multidisciplinary Brain Research, Bar Ilan University, Israel; Department of English Literature and Linguistics, Bar Ilan University, Israel. Yuval Nir **, Sagol School of Neuroscience, Tel Aviv University, Israel; of Physiology and Pharmacology, Sackler School of Medicine, Tel Aviv University, Israel; Tel-Aviv Sourasky Medical Center, Tel-Aviv, Israel. Elana Zion-Golumbic **, The Gonda Center for Multidisciplinary Brain Research, Bar Ilan University, Israel. ** Equal senior authorship
Sleep Disrupts High-Level Speech Parsing Despite Preserving Basic Auditory Processing
Show/hide abstractThe extent to which the sleeping brain processes sensory information remains unclear. This is particularly true for continuous and complex stimuli, such as speech, that require hierarchical parsing and processing mediated by several brain regions. Previous studies of continuous speech processing in wakefulness provide metrics for assessing the depth of speech processing using non-invasive brain recordings. Such studies in wakefulness have demonstrated stronger and more extensive neuronal tracking when speech is intelligible and attended. Here we investigated, for the first time, the ability of the sleeping human brain to process continuous speech at different hierarchical levels, using a newly developed Concurrent Hierarchical Tracking (CHT) approach. Speech sequences were compiled with syllables, words, phrases and sentences occurring at fixed time intervals, such that parsing of different linguistic levels appear at distinct frequencies, allowing to distinguish their neural signatures in brain activity. We compared the neuronal tracking of intelligible vs. unintelligible (scrambled and foreign) speech across states of wakefulness and sleep, using high-density electroencephalogram (EEG). We found that neuronal tracking of stimulus acoustics was comparable across wakefulness and sleep, and similar across all conditions regardless of speech intelligibility. In contrast, neuronal tracking of higher-order linguistic parsing (words, phrases and sentences) was only observed for intelligible speech during wakefulness, and could not be detected at all during NREM or REM sleep. These results suggest that while low-level auditory processing is preserved during sleep, higher-level hierarchical linguistic parsing is severely disrupted, thereby revealing the capacity and limits of language processing during sleep. The current study sets the ground towards studying residual cognitive abilities and speech tracking in a wide array of unresponsive states and clinical populations.
Saturday 16.00 - 17.30 Grote zaal Poster #59 Memory Danying Wang
All authors & affiliationDanying Wang, School of Psychology, University of Birmingham, UK Andrew Clouter, School of Psychology, University of Birmingham, UK Qiaoyu Chen, School of Psychology, University of Birmingham, UK Kim L. Shapiro, School of Psychology, University of Birmingham, UK Simon Hanslmayr, School of Psychology, University of Birmingham, UK
Externally induced inter-regional theta phase synchronisation mediates human associative memory
Show/hide abstractHippocampal theta is thought to be crucial for binding sensory information from multiple cortical regions into coherent memory episodes. Studies in rodents showed that cortical inputs arriving at the appropriate hippocampal theta phase induce long-term potentiation, which is a possible mechanism for how theta phase modulates memory formation. Using a multisensory entrainment paradigm, we recently showed in humans that episodic memory performance is enhanced by theta (4 Hz) phase synchrony between visual and auditory cortices. Here, we investigate if such theta phase synchrony between visual and auditory brain regions varies as a function of successful episodic memory formation. We recorded EEG from 24 participants while they performed an associative memory task. Luminance of movies and amplitude of sounds were modulated from zero to full at 4 Hz with 0 degree, in-phase or 180 degrees, out-of-phase to induce a certain phase-relationship between visual or auditory cortices, respectively. Later, participants recalled the correct scene presented with a given sound clip. Associative memory performance was significantly better in the in-phase condition than the out-of-phase condition, replicating previous findings. EEG data shows that phase differences between the two sensory modalities varied with respect to the two entrainment conditions; the in-phase condition induced a 0 degree phase difference, whereas the out-of-phase condition induced a 180 degree phase difference between the visual and auditory cortices, showing that we succeeded in manipulating phase in the expected direction. Further analyses focus on phase differences of the theta activity in 4 Hz between the auditory source and the visual source, which is categorised depending on later associative memory performance. We found the distribution of phase difference between auditory and visual regions during encoding varied as a function of subsequent memory performance. The findings will further our understanding of theta phase synchrony for episodic memory encoding.
Saturday 16.00 - 17.30 Grote zaal Poster #60 Memory Casper Kerren
All authors & affiliationCasper Kerren, School of Psychology, University of Birmingham, Birmingham, United Kingdom Juan Linde-Domingo, School of Psychology, University of Birmingham, Birmingham, United Kingdom Simon Hanslmayr, School of Psychology, University of Birmingham, Birmingham, United Kingdom Maria Wimber, School of Psychology, University of Birmingham, Birmingham, United Kingdom School of Psychology, University of Birmingham, Birmingham, United Kingdom
Reactivation of memory patterns in the human EEG rhythmically fluctuates in the theta range
Show/hide abstractOur episodic memories allow us to mentally travel back and forth in time. We use this information to relive past experiences and to guide future behaviour. When successfully remembering, it is believed that a memory cue triggers the reactivation of the specific neural activity pattern that represents this particular past episode. However, little is known about the neural dynamics of this pattern completion process. Computational models of the hippocampus suggest that the brain constantly switches between a brain state that is optimal for remembering past events and a brain state optimal for encoding novel events, and that these switches are mediated by the hippocampal theta rhythm. These models have received some support from work in rodents, but have scarcely been tested in the human brain. Here, we let participants study novel word-object associations, where the words later, in the test phase, worked as a reminder for retrieving the associated object. In order to use multivariate pattern analysis to track the reactivation of target memories, the objects were categorised according to a semantic (animate and non-animate) or a perceptual (photograph or line-drawing) dimension. We used a linear discriminant analysis (LDA) to obtain measures of confidence by which the target memory was reactivated on each trial, as triggered by the reminder. We then specifically tested the hypothesis that target memories have a tendency to be reactivated during a specific phase of the theta rhythm. If so, we expected to find a rhythmic fluctuation in the classifier outputs within the theta band. Indeed, results from a frequency transformation of the classifier outputs show a peak in the theta frequency at 6 Hz. These results support recent computational models suggesting that slow oscillations drive pattern completion. These mechanisms seem to be shared between human and rodent brains.
Saturday 16.00 - 17.30 Grote zaal Poster #61 Electrophysiology methods Alice Tobin
All authors & affiliationAlice V. Tobin, The University of Western Australia Jason A. Bell, The University of Western Australia Patrick J. F. Clarke, Curtin University Allison M. Fox, The University of Western Australia
Electrode placement in transcranial direct current stimulation for modulation of working memory
Show/hide abstractBackground Working memory is a capacity-limited resource used for updating and manipulating information. When applied to the left dorsolateral prefrontal cortex (DLPFC), transcranial direct current stimulation (tDCS) can temporarily excite (anodal) or inhibit (cathodal) working memory performance. Previous research suggests that the position of the cathode is a critical variable influencing the tDCS ‘dose'. Placing the cathode over the right orbital region carries a risk of inducing inhibitory effects which may be counterproductive to achieving beneficial effects of tDCS on working memory. The current study evaluated the relationship between tDCS cathode placement and working memory by comparing two different cathode locations and monitoring behavioural and electrophysiological responses to a working memory task. Methods Fifty-two individuals (19 males; mean age = 20, SD= 0.49) participated in this study. Participants received either anodal or sham tDCS stimulation at rest with the cathode placed a) over the right orbital region or b) at the inion. Following 20 minutes of stimulation at 2 mA, all participants completed a parametric verbal n-back task (spanning 1-, 2-, and 3-back) with concurrent EEG recording at sites Fz, Cz, Pz. The amplitude of the P3 peak elicited in response to targets was extracted. Results As expected accuracy decreased and response time increased with increasing load. The amplitude of the P3 peak was negatively correlated with increasing n-back load. Neither performance nor P3 amplitude was significantly affected by cathode site. Performance was not affected by tDCS compared to sham. Discussion These findings suggest that the conditions under which tDCS improves working memory require careful consideration although there was no suggestion that cathode site modulated the lack of tDCS effect in our results.
Saturday 16.00 - 17.30 Grote zaal Poster #62 Language Fraga Gonzalez
All authors & affiliationFraga Gonzalez, G. *,1,2, Smit, D.J.A.3,4, Van der Molen, M.J.W.5,6, Tijms, J.2,7, Stam, C.J. 8, de Geus, E.J.C.3,4, Van der Molen, M.W. 1,9 1 Department of Psychology, University of Amsterdam, The Netherlands 2 Rudolf Berlin Center, Amsterdam, The Netherlands 3 Department of biological Psychology, VU University, Amsterdam, The Netherlands. 4 Neuroscience Campus Amsterdam, VU University, Amsterdam, The Netherlands. 5 Institute of Psychology, Leiden University, The Netherlands 6 Leiden Institute for Brain and Cognition, Leiden University, The Netherlands 7 IWAL Institute, Amsterdam, The Netherlands 8 Department of Clinical Neuropsychology and MEG Center, Neuroscience Campus Amsterdam, VU University Medical Center, Amsterdam, The Netherlands 9 Amsterdam Brain and Cognition, University of Amsterdam, The Netherlands
Characterization of EEG Resting State Functional Connectivity in Adult Dyslexics Using Phase Lag Index and Graph Analysis
Show/hide abstractDevelopment dyslexia may involve deficits in functional connectivity across broad cortical networks that enable fluent reading. We investigated the large-scale organization of EEG functional networks at rest in 28 dyslexics and 36 typically reading adults. For each frequency band, we assessed connectivity strength with the phase lag index (PLI). Network topology was examined using weighted graphs and minimum spanning tree (MST) graphs derived from the connectivity matrices. We found significant group differences in the alpha band (8-13 Hz). Dyslexic showed higher connectivity than typical readers over the left temporal, central and frontal sites. The graph metrics were significantly correlated with age in dyslexics but not in typical readers. The graph analysis indicated a more integrated network configuration but reduced functional specialization, i.e., modularity, in dyslexics compared to typical readers. The present findings support the sensitivity of graph metrics to characterize functional networks in adult dyslexia and extend our previous findings on children.
Saturday 16.00 - 17.30 Grote zaal Poster #63 Decision-making Matthias Liebrand
All authors & affiliationMatthias Liebrand a,b Jascha Kristeka, Elinor Tzvia, Ulrike M. Krämera,c Affiliations: a Dept. of Neurology, University of Lübeck, Ratzeburger Allee 160, 23538 Lübeck, Germany b Graduate School for Computing in Medicine and Life Sciences, University of Lübeck, Ratzeburger Allee 160, 23538 Lübeck,, Germany c Institute of Psychology II, University of Lübeck, Ratzeburger Allee 160, 23538 Lübeck, Germany
Ready for change: Oscillatory Mechanisms of Proactive Motor Control
Show/hide abstractProactive motor control is a preparatory mechanism facilitating upcoming action inhibition or adaptation. Previous studies investigating proactive motor control mostly focused on response inhibition, as in the classical go-nogo or stop-signal tasks. However, everyday life rarely calls for the complete suppression of actions without subsequent behavioral adjustment. Therefore, we conducted a modified cued go-nogo-change task, in which cues indicated whether participants might have to change to an alternative action or inhibit the response to an upcoming target. Based on the dual-mechanisms of control framework and using electroencephalography (EEG), we investigated the role of the sensorimotor cortex and of prefrontal regions in preparing to change and stop motor responses. We focused on mu and beta power over sensorimotor cortex ipsi- and contralateral to an automatic motor response and on prefrontal beta power. Over ipsilateral sensorimotor cortex, mu and beta power was relatively decreased when anticipating to change or stop the automatic motor behavior. Moreover, alpha phase coupling between ipsilateral motor cortex and prefrontal areas decreased when preparing to change, suggesting a decoupling of sensorimotor regions from prefrontal control. When the standard motor action actually had to be changed, prefrontal beta power increased, reflecting enhanced cognitive control. The present data highlight the role of the ipsilateral motor cortex in preparing to stop and change upcoming motor actions. Here, especially mu power and phase coupling seem to be critical to guide upcoming behavior.
Saturday 16.00 - 17.30 Grote zaal Poster #64 Developmental Liesa Ilg
All authors & affiliationLiesa Ilg1, Manousos Klados1, Nina Alexander2, Clemens Kirschbaum2, Shu-Chen Li1 1 Chair of Lifespan Developmental Neuroscience, Department of Psychology, TU Dresden, Dresden, Germany 2 Chair of Biopsychology, Department of Psychology, TU Dresden, Dresden, Germany
Long-term effects of prenatal glucocorticoid exposure on functional brain correlates of conflict monitoring in adolescence
Show/hide abstractPrenatal synthetic glucocorticoid (sGC) exposure, a routine treatment for women at risk of preterm delivery, was previously shown to have long-term consequences on structural brain development, especially in regions of the anterior cingulate cortex (ACC). A major role of the ACC is the monitoring of stimulus–response conflicts, which is fundamental for the regulation of cognitive control. To investigate long-term effects of prenatal sGC exposure on conflict monitoring at a neurofunctional level neurophysiological correlates of cognitive conflict monitoring were analyzed. A total of 52 term-born adolescents (now aged 14-18 yrs.) were re-recruited from a prior study (Alexander et al., 2012), with 28 participants been prenatally treated with sGC and a comparison group of 24 adolescents. Electroencephalograms (EEG) were recorded while the participants performed a cued continuous performance task that entail Go and response conflict eliciting NoGo trials (cf. Hämmerer et al., 2011) to investigate the long-term effects of sGC exposure on the N2 event-related-potential component, known to reflect conflict monitoring. Cortical source reconstruction was conducted using the sLORETA algorithm (Pascual-Marqui, 2007) that is integrated in the Brainstorm (Tadel et al., 2011; available for download under the GNU public license Adolescents prenatally exposed to sGC showed a significantly attenuated N2 component, which could be localized in regions of the ACC. Furthermore, neural activities source localized to the ACC region were significantly lower in these adolescents relative to the comparison group. Interestingly, both NoGo N2 amplitudes at the scalp level and the associated source activity in the ACC predict individual differences in performance. To our best knowledge, these results provide the first evidence for long-lasting impacts of prenatal sGC exposure on neurophysiological correlates of conflict monitoring in late adolescence, which is a period in life with perturbations in the neurocognitive development of the balance between cortical and subcortical networks.
Saturday 16.00 - 17.30 Grote zaal Poster #65 Developmental Alexandra Mogadam
All authors & affiliationAlexandra Mogadam*1,2, Paul D. Arnold3,4, Russell Schachar4,5, Margot J. Taylor1,2,6-8, Jason P. Lerch2,9,10, Evdokia Anagnostou1,2,8,11,12, Elizabeth W. Pang1,2,8 1Institute of Medical Science, Faculty of Medicine, University of Toronto 2Neurosciences and Mental Health, Sick Kids Research Institute, Toronto 3Mathison Centre for Mental Health Research and Education, University of Calgary 4Division of Psychiatry, Hospital for Sick Children, Toronto 5Department of Psychiatry, University of Toronto 6Department of Diagnostic Imaging, Hospital for Sick Children, Toronto 7Departments of Medical Imaging and Psychology, University of Toronto 8Division of Neurology, Hospital for Sick Children, Toronto 9Mouse Imaging Centre, Hospital for Sick Children, Toronto 10Department of Medical Biophysics, Faculty of Medicine, University of Toronto 11Holland Bloorview Kids Rehabilitation Hospital, Toronto 12Department of Paediatrics, Faculty of Medicine, University of Toronto
Comparing the Neural Correlates of Mental Flexibility in Children with Neurodevelopmental Disorders: an MEG Investigation
Show/hide abstractNeurodevelopmental disorders (NDs) are characterized by impairments and/or delays in the development of the central nervous system; they include autism spectrum disorders (ASD), attention deficit hyperactivity disorder (ADHD) and paediatric obsessive-compulsive disorder (OCD). Children with NDs experience set-backs in executive functioning, including mental flexibility. Mental flexibility (MF) represents the ability to alter mental processes continuously in response to environmental changes. Due to the heterogeneity of NDs, we hypothesized that the neurobiological substrates underlying MF would differ across ND subtypes. To investigate, we recruited 88 children with NDs (8-15yrs; 20 F; 38 ASD, 28 ADHD, 22 OCD) to complete an MF task while undergoing a magnetoencephalography (MEG) scan. MEG is a high-temporal and spatial resolution functional neuroimaging modality. The MF task required matching of stimuli, with matching criteria shifting every few trials. Groups performed comparably. MEG data were source reconstructed to create high resolution spatiotemporal activity maps of brain regions significantly involved in the task. MEG revealed similar patterns of activation for ASD and ADHD, with sustained early parietal activity and late recruitment of frontal regions. OCD displayed the expected activations in frontal regions, however their parietal activity was sustained also. This trend of greater similarities between ASD and ADHD, compared to OCD, is consistent with suggestions that the former two are more similar in etiology. Our results suggest that while the various groups of NDs show some differences in MF processing, they also share common neurobiological substrates. This is the first functional neuroimaging study to investigate MF across multiple NDs. This has implications for the future development of treatments and interventions for children with NDs.
Saturday 16.00 - 17.30 Grote zaal Poster #66 Social behavior Max Rollwage
All authors & affiliation Max Rollwage 1,2,3, Franziska Pannach3, Ulf Toelch 4, Igor Kagan5,6, Arezoo Pooresmaeili 3,6 1: Wellcome Trust Centre for Neuroimaging, University College London, London, United Kingdom 2: Max Planck University College London Centre for Computational Psychiatry and Ageing Research, London, United Kingdom 3: Perception and Cognition Group, European Neuroscience Institute, Göttingen, Germany 4: Biological Psychology and Cognitive Neuroscience, Freie Universität Berlin, Berlin, Germany 5: Decision and Awareness Group, German Primate Center (DPZ), Göttingen, Germany 6: Leibnitz Science Campus “Primate Cognition”, Göttingen, Germany
I judge you by your profit: self and other effort-judgements are affected by task contingent rewards
Show/hide abstractEstimating the effort someone spent on a task is a core dimension for evaluating own and other’s actions. Interestingly, it has been shown that self-judgements of effort are influenced by the magnitude of obtained rewards (Pooresmaeili, Wannig & Dolan, 2015). However, it is unclear whether the influence of reward on effort estimations is limited to self-judgements (i.e. a form of self-serving bias) or whether reward information is invariably incorporated when judging effort in other contexts. Here, we investigate whether people integrate reward magnitude when judging the effort exerted by another person. Pairs of participants (N=58) performed an effortful task in an interleaved manner, before rating the exerted effort. In half of the trials participants performed the task themselves and rated their own effort. In the other half, participants rated the other person’s effort after watching them performing the task (while hearing auditory feedback that co-varied with the amount of effort). After each trial, before the effort rating, both participants were informed of the gained reward for that trial. Higher rewards led to higher estimations of exerted effort. Importantly, this was true for self- as well as other-judgements. In line with the Bayesian cue integration framework, reward magnitude had a stronger effect when judging the effort of another person, compared to self-judgements, since people should have more reliable internal estimates of own effort. Supporting this claim, computational modelling revealed that reward information and objective cues of task difficulty were combined in a Bayes optimal manner to form effort estimates. Although optimal in terms of cue integration, reward-biased effort estimation, e.g. an inflated perception of expended effort for people with high income, could have a profound societal impact, potentially leading to stabilization of existing inequalities. Interestingly, the extent to which rewards influenced individuals was positively correlated with conservative world-views, suggesting interesting links between this effect and broader social attitudes.
Saturday 16.00 - 17.30 Grote zaal Poster #67 Connectivity/networks Alejandra Sel
All authors & affiliationAlejandra Sel*, Katharina Angerer, Miriam Klein-Flügge, Lennart Verhagen, Matthew Rushworth Department of Experimental Psychology, University of Oxford, 9 South Parks Road, OX1 3UD, UK
Associative corticocortical plasticity in action control mechanisms of the human brain
Show/hide abstractNoninvasive paired associative transcranial stimulation (PAS) selectively strengthens synaptic plasticity in corticocortical pathways in the human brain. Using PAS it is possible to augment physiological connectivity between two anatomically connected areas in the action control circuit; the ventral premotor cortex (PMv) and the primary motor cortex (M1). Converging evidence from neuroimaging, neurostimulation and clinical studies have shown that the pathway from PMv and adjacent inferior frontal cortex (IFC) to M1 has a crucial role when unwanted actions are inhibited. In the current study, we investigated whether it is possible to augment the influence that IFC has over M1 in an action control task. In two different task blocks, we presented participants with a go/no-go task while testing the influence of IFC over M1. We stimulated near PMv and M1 with a short 8-ms inter-pulse interval at 125ms after the cue onset in a selected number of trials. Between the two task blocks, we applied the same PAS protocol at 0.1Hz during a 15minute interval, which it is proven to evoke synchronous pre- and post-synaptic activity and which strengths interregional connectivity between pre-motor and motor cortex in a Hebbian-like manner. The results of the study showed selective changes in the motor-evoked potentials after the plasticity induction phase. The expression of the change in the pathway was state-dependent. This is, in the go trials the paired PMv-M1 stimulation led to an increased facilitatory influence of PMv over M1. Interestingly, the pattern of connectivity augmentation in no-go trials was highly dependent on the physiological excitability of M1 at rest. Additionally, we observed that the plasticity effect was dependent on stimulation order: repeated stimulation of PMv before M1 led to strengthening of the PMv–M1 pathway, while these effects were not present when repeated stimulation of M1 before PMv was applied.
Saturday 16.00 - 17.30 Grote zaal Poster #68 Neural modeling Alexandra Mogadam
All authors & affiliationAlexandra Mogadam*1,2, Paul D. Arnold3,4, Russell Schachar4,5, Margot J. Taylor1,2,6-8, Jason P. Lerch2,9,10, Evdokia Anagnostou1,2,8,11,12, Elizabeth W. Pang1,2,8 1Institute of Medical Science, Faculty of Medicine, University of Toronto 2Neurosciences and Mental Health, Sick Kids Research Institute, Toronto 3Mathison Centre for Mental Health Research and Education, University of Calgary 4Division of Psychiatry, Hospital for Sick Children, Toronto 5Department of Psychiatry, University of Toronto 6Department of Diagnostic Imaging, Hospital for Sick Children, Toronto 7Departments of Medical Imaging and Psychology, University of Toronto 8Division of Neurology, Hospital for Sick Children, Toronto 9Mouse Imaging Centre, Hospital for Sick Children, Toronto 10Department of Medical Biophysics, Faculty of Medicine, University of Toronto 11Holland Bloorview Kids Rehabilitation Hospital, Toronto 12Department of Paediatrics, Faculty of Medicine, University of Toronto
Comparing the Neural Correlates of Mental Flexibility in Children with Neurodevelopmental Disorders: an MEG Investigation
Show/hide abstractNeurodevelopmental disorders (NDs) are characterized by impairments and/or delays in the development of the central nervous system; they include autism spectrum disorders (ASD), attention deficit hyperactivity disorder (ADHD) and paediatric obsessive-compulsive disorder (OCD). Children with NDs experience set-backs in executive functioning, including mental flexibility. Mental flexibility (MF) represents the ability to alter mental processes continuously in response to environmental changes. Due to the heterogeneity of NDs, we hypothesized that the neurobiological substrates underlying MF would differ across ND subtypes. To investigate, we recruited 88 children with NDs (8-15yrs; 20 F; 38 ASD, 28 ADHD, 22 OCD) to complete an MF task while undergoing a magnetoencephalography (MEG) scan. MEG is a high-temporal and spatial resolution functional neuroimaging modality. The MF task required matching of stimuli, with matching criteria shifting every few trials. Groups performed comparably. MEG data were source reconstructed to create high resolution spatiotemporal activity maps of brain regions significantly involved in the task. MEG revealed similar patterns of activation for ASD and ADHD, with sustained early parietal activity and late recruitment of frontal regions. OCD displayed the expected activations in frontal regions, however their parietal activity was sustained also. This trend of greater similarities between ASD and ADHD, compared to OCD, is consistent with suggestions that the former two are more similar in etiology. Our results suggest that while the various groups of NDs show some differences in MF processing, they also share common neurobiological substrates. This is the first functional neuroimaging study to investigate MF across multiple NDs. This has implications for the future development of treatments and interventions for children with NDs.
Saturday 16.00 - 17.30 Grote zaal Poster #69 Cognitive modeling Daniel Fitousi
All authors & affiliationDaniel Fitousi Ariel University
Can we perceive two different colors at the same time? A direct test of the Boolean theory of selective attention
Show/hide abstractCan we access two spatially separated color patches (e.g., red and green) at the same time? Huang and Pashler (2007) have recently proposed a Boolean theory of attention (BTA) that provides a clear cut answer to this question, which is no. According to the BTA access to two features of the same dimension is limited and can therefore be held only in a serial manner, that is, one color after the other. The reason is that attention operates by constructing a Boolean map – a device that divides visual space into homogenous processing units adhering to conjunctive or disjunctive rules. The current study proposes a strong test of the BTA by dissociating serial, parallel and exhaustive architectures in the processing of a Boolean map. Here I have harnessed the System Factorial Technology (SFT, Townsend & Nozawa, 1995) – a powerful stochastic modeling tool – along with an allied set of technologies – the logical rule models (Fific, Little & Nosofsky, 2010) -- to examine the mental architectures and stopping rules involved in the processing of Boolean maps. The results showed that processing of a conjunctive map is held according to a serial architecture, providing considerable support in Huang and Pashler (2007) Boolean map theory.
Saturday 16.00 - 17.30 Grote zaal Poster #70 Developmental Lucia Amoruso
All authors & affiliationAmoruso, L. 1; Pinzino, M. 2; Narzisi, A2.; Finisguerra, A. 1; Fabbro1, F.; Muratori, F.2, Volzone, A.3; Urgesi, C1,3. 1 University of Udine, Laboratory of Cognitive Neuroscience, Udine, Italy. 2 Scientific Institute (IRCCS) Fondazione Stella Maris, Pisa, Italy. 3 Scientific Institute (IRRCS) Eugenio Medea, Polo Friuli Venezia Giulia, I-33078, San Vito al Tagliamento, Pordenone, Italy.
Contextual priors do not modulate action anticipation in children with autism
Show/hide abstractInfluential models on action comprehension postulate that optimal intention inference requires the combination of two types of information: the expectations about other’s likely behaviours based on previous experiences and the sensory evidence conveyed by perceptual movement kinematics. Recently, it has been suggested that individuals on the autism spectrum have difficulties in anticipating other people’s actions due to an attenuated influence of prior information. Here, we aimed to directly test this hypothesis in a group of 24 high-functioning children with autism and 24 ability-matched typically developing children. Based on the premise that the brain represents contextual information probabilistically, we developed a behavioural paradigm consisting of a probabilistic learning task (familiarization phase) followed by an action prediction task (testing phase). During the familiarization phase, children were presented with videos depicting a child actor grasping different objects to perform actions in diverse contexts and asked them to recognize actor’s intention. Importantly, within this phase, we implicitly biased action-context associations in terms of their probability of co-occurrence. During the testing phase, children observed the same videos but, in this case, the second-half remained occluded from view and participants were asked to predict action unfolding. We reasoned that, during this phase where movement kinematics became ambiguous, children’s responses would be biased to contextual priors acquired during the familiarization, thus compensating perceptual uncertainty. We found that, unlike ASD children, TD children exhibited a probabilistically modulated behaviour. More specifically, TD children were better at predicting those actions embedded in contexts with higher probability as compared to those with lower probability and this effect was absent in the ASD group. Collectively, our results suggest that ASD children have an impairment in building contextual priors and do not benefit from them when predicting other people’s actions.
Sunday 12.00 - 13.30 Grote zaal Poster #1 Memory David Wisniewski
All authors & affiliationDavid Wisniewski, Universiteit Gent, Belgium Senne Braem, Universiteit Gent, Belgium Jan De Houwer, Universiteit Gent, Belgium Marcel Brass, Universiteit Gent, Belgium
Using multivariate pattern analysis to reveal instruction-based and experience-dependent fear memories during fear reversal
Show/hide abstractIn human fear conditioning, mere verbal instructions have been shown to reliably elicit defensive reactions (Ugland et al. 2013). Yet, fear conditioning also relies in part on non-verbal processing of the contingency between the conditioned stimulus (CS) and an aversive unconditioned stimulus (US). For instance, it has been shown that experiencing CS-US pairings adds to self-reported fear, as compared to merely instructed CS-US pairings (Raes et al. 2014). Yet, the neural basis of instructed and experienced CS-US pairings remains poorly understood. Recent evidence suggests that brain regions differ in their susceptibility to verbal instructions in fear learning (Braem et al., submitted). Here, we employ an instructed fear reversal paradigm (Costa et al., 2015) in order to focus on the flexible adjustment of CS-US pairings. Will a reversal of CS-US contingencies affect instructed-only (CS+I) and instructed + experienced (CS+E) contingencies differently? Subjects first learnt both CS+I and CS+E pairings using a Pavlovian conditioning protocol. These associations were then reversed during the experiment, and we investigated reversal effects on US-expectancy ratings, skin conductance responses, as well as fMRI responses. Using multivariate pattern analysis methods, we assessed neural representations of both CS+I and CS+E in amygdala and other fear-related brain regions before and after fear associations were reversed. Our results highlight the role of insula and striatum in for human fear learning, as both regions closely track the instructed fear reversal irrespective of whether CS-US contingencies were merely instructed or also experienced. The results further highlight the role of the amygdala, finding some differences between instructed and experienced CS-US contingencies. Despite the reversal instructions, there seems to be an experience-dependent Pavlovian trace in this brain region. These findings illuminate and separate the roles of verbal and non-verbal factors contributing to fear learning and reversal learning in humans, promoting our understanding of emotion regulation.
Sunday 12.00 - 13.30 Grote zaal Poster #2 Decision-making Alexandra Sebastian
All authors & affiliationAlexandra Sebastian, University Medical Center of the Johannes Gutenberg University Mainz, Germany * Anne Maria Konken, University Medical Center of the Johannes Gutenberg University Mainz, Germany Oliver Tuescher, University Medical Center of the Johannes Gutenberg University Mainz, Germany Patrick Jung, University Medical Center of the Johannes Gutenberg University Mainz
A common fronto-basal-ganglia network for reactive action and reactive inhibition
Show/hide abstractIntroduction: Proactive and reactive control has usually been studied in the context of proactive inhibition (goal-directed action restraint) and reactive inhibition (stimulus-driven action cancelation). Recent studies indicate that proactive and reactive inhibition share some neural networks, whereas others are exclusively related to either proactive or reactive inhibition [1]. It is unclear, however, whether reactive inhibition and reactive action depend on common networks, unique networks, or a combination. This question was addressed in the current study. Methods: In total, 32 healthy participants (age: M=25.0 ±1.4 years) completed a cued go/nogo paradigm during functional magnetic resonance imaging (fMRI) comprising four conditions: proactive go (go cue -- go target, 37.5%), proactive nogo (nogo cue -- nogo target, 37.5%), reactive go (nogo cue -- go target, 12.5%), reactive nogo (go cue -- nogo target, 12.5%). A button press was required on go- but not on nogo-trials. Scanning was performed on a 3T TIM Trio (Siemens, Germany). FMRI data analysis was performed using standard SPM12 procedures. Results: Both forms of reactive control (i.e., reactive go>proactive go; reactive nogo>proactive nogo) significantly activated a bilateral fronto-striatal-parietal network including inferior frontal gyrus, anterior insula, caudate, pre-supplemental motor area and inferior parietal regions. Activity of the nucleus subthalamicus region was present at least at trend level. This was confirmed by a conjunction analysis. No regions were identified as being uniquely associated with either reactive action or reactive inhibition. Discussion: Reactive action and reactive inhibition activated the same neural network including fronto-striatal regions usually implicated in global stopping [2-4]. These results support the notion that unexpected stimuli interrupt action by recruiting a global suppressive fronto-basal-ganglia network for stopping [5]. The present findings extend this notion to reactive action where no action is ongoing or planned (i.e., nogo cue) but action is required following an unexpected event (i.e., go target).
Sunday 12.00 - 13.30 Grote zaal Poster #3 Attention Megan O'Neill
All authors & affiliationMegan H. O'Neill, Monash Institute for Cognitive and Clinical Neurosciences (MICCN), School of Psychological Sciences, Monash University, Melbourne, Australia. Gerard M. Loughnane, Trinity College Institute of Neuroscience, Trinity College Dublin, Dublin 2, Ireland. Renerus J. Stolwyk, Monash Institute for Cognitive and Clinical Neurosciences (MICCN), School of Psychological Sciences, Monash University, Melbourne, Australia Peter W. New, Rehabilitation and Aged Services, Medicine Program, Monash Health, Victoria, Australia; Department of Epidemiology and Preventive Medicine, Monash University, Victoria, Australia Redmond G. O'Connell,Trinity College Institute of Neuroscience, Trinity College Dublin, Dublin 2, Ireland; School of Psychology, Trinity College Dublin, Dublin 2, Ireland. Mark B. Bellgrove, Monash Institute for Cognitive and Clinical Neurosciences (MICCN), School of Psychological Sciences, Monash University, Melbourne, Australia.
Early target selection and evidence accumulation is impaired in the left hemifield after right hemisphere stroke
Show/hide abstractHemispatial neglect is a common and disabling neurological syndrome defined as the inability to detect, respond to, and orient towards stimuli on the side contralateral to cerebral damage. It is more common, severe and enduring following stroke affecting the right hemisphere, which results in inattention to left side. Despite decades of behavioral investigation, we know little about how the physiological processes supporting goal-directed behaviour are disrupted in neglect. Here we used a perceptual decision making paradigm and electroencephalography (EEG) to acquire reaction times (RTs) for left/right hemifield targets, and to isolate the distinct neural signals underpinning visuospatial target detection. These included the N2c/N2i -- event related potentials (ERPs) known to index early target selection; and the centro-parietal positivity (CPP) - a signal associated with evidence accumulation. Here we present data from three right hemisphere stroke patients with recovered neglect and 27 healthy control participants. Behavioural results confirmed the presence of neglect in stroke patients, with significantly slower RTs for left hemifield targets. Scalp topographies within the neglect group revealed intact N2c/N2i and CPP signals for right hemifield targets, with the timing comparable to healthy participants. However, for left hemifield targets, the amplitude of the N2c (recorded from right hemisphere) in the neglect patients was dramatically diminished compared to the N2c for right hemifield targets. Furthermore, the slope of the decision making process measured by the CPP was shallower for left compared with right hemifield targets, impacting evidence accumulation and leading to slowed perceptual decisions for the left hemifield. These data provide a model for the use of a perceptual decision making framework for understanding the physiological basis of attention deficits after right hemisphere stroke. Our method may in the longer term lead to the development of a set of neural metrics that may facilitate the objective assessment of spatial neglect.
Sunday 12.00 - 13.30 Grote zaal Poster #4 Anatomy Anneke Alkemade
All authors & affiliationMax C Keuken, Integrative Model-based Cognitive Neuroscience (IMCN) research Unit, University of Amsterdam, Amsterdam, The Netherlands Gilles De Hollander, Integrative Model-based Cognitive Neuroscience (IMCN) research Unit, University of Amsterdam, Amsterdam, The Netherlands Rawien Balesar, Integrative Model-based Cognitive Neuroscience (IMCN) research Unit, University of Amsterdam, Amsterdam, The Netherlands Birte U Forstmann, Integrative Model-based Cognitive Neuroscience (IMCN) research Unit, University of Amsterdam, Amsterdam, The Netherlands
Crossing the Styx: An integrative pipeline translating post mortem findings to MRI space
Show/hide abstractMagnetic Resonance Imaging (MRI) research represents a major focus of cognitive neuroscience research, and provides an excellent tool for studying how structure underlies function. However, the anatomical and chemical detail provided by MRI techniques is limited, and even today histological observations represent the gold standard in anatomical research. Intuitively, an integrative approach combining MRI techniques with microscopic anatomical detail obtained from post mortem studies has the potential to greatly improve our understanding of structure-function relationships in the human brain. However, postmortem observations validating MRI studies are scarce. For proof of concept we studied the human subthalamic nucleus, a small biconvex nucleus that functionally is part of the basal ganglia, and plays a role in motor, cognitive and limbic functions. Human post mortem brain specimens were collected and formalin fixed, embedded in paraffin and cut in 6 micrometer sections while performing blockface imaging with 300 micrometer intervals. Consecutive sections were sampled with 300 micrometer intervals and subjected to immunocytochemical staining using various markers including GABA-ergic, glutamatergic, dopaminergic, and serotonergic signaling. Sections were digitally imaged and the resulting images were analyzed using thresholding procedures available in the ImageJ software. 3D staining patterns were remodeled to individual MRI space using linear and non-linear registration. These studies allow us to compare MRI and histological results, thereby providing the tools to build a bridge between histological studies and functional neuroimaging studies.
Sunday 12.00 - 13.30 Grote zaal Poster #5 Electrophysiology methods Vicente Soto
All authors & affiliationVicente Soto, University of Liverpool John Tyson-Carr, University of Liverpool Katerina Kokmotou, University of Liverpool Hannah Roberts, University of Liverpool Stephanie Cook, University of Liverpool Nicholas Fallon, University of Liverpool Timo Giesbrecht, UNILEVER Research and Developement Andrej Stancak, University of Liverpool
Identification of neural responses to human faces using wireless multichannel EEG recordings.
Show/hide abstractPrevious neuroimaging studies have shown face-specific neural responses to faces compared to objects. The detection of face-specific brain activation in freely behaving and moving people has yet to be accomplished. The purpose of our research was to identify, using wireless multichannel EEG in freely moving participants, event-related potentials (ERPs) during viewing of human faces. Mobile EEG and eye tracking was recorded from 19 freely moving participants whilst they viewed a mock art gallery. Stimuli were presented on 20 panels displayed within corridors of University of Liverpool. Positive, negative and neutral valence images were viewed and later rated by the participants. Mobile EEG was recorded continuously using a 64-channel BrainProduct MOVE system. In absence of triggers indicating onsets of viewing of visual stimuli, a novel head-mounted wearable eye-tracker was used to capture real world video recordings and gaze locations. After synchronising the EEG and eye-tracking time-series recordings, BESA 6.1 program was used to process EEG data. Wireless EEG recordings allowed identification of a face-related ERP component in the latency interval ranging from 165 to 210 ms (N170 potential); this component was not seen whilst participants were viewing non-living objects (F(1,18) =23.9, P < 0.001). The face ERP component was sensitive to the emotional face expression; in particular, the amplitude of N185 potential was stronger during viewing disgusted compared to neutral faces (t(18) = -4.02, P < 0.001). Source dipole analysis revealed three equivalent current dipoles in the latency interval from 100 ms to 300 ms. Two source dipoles, located in the left extrastriate (BA19) and primary visual (BA17) cortex, modelled the visual P100 component, and one equivalent current dipole, fitted to the right fusiform gyrus (BA37), accounted for the face-related N170 potential. We demonstrate face-related ERP's in freely moving individuals in natural settings opening new possibilities in clinical, developmental, social or marketing research.
Sunday 12.00 - 13.30 Grote zaal Poster #6 Electrophysiology methods Annika Notbohm
All authors & affiliationAnnika Notbohm, Experimental Psychology Lab, Center for Excellence ‘Hearing4all', European Medical School, Carl von Ossietzky University, Oldenburg, Germany Johannes Vosskuhl, Experimental Psychology Lab, Center for Excellence ‘Hearing4all', European Medical School, Carl von Ossietzky University, Oldenburg, Germany Christoph S. Herrmann, Experimental Psychology Lab, Center for Excellence ‘Hearing4all', European Medical School, Carl von Ossietzky University, Oldenburg, Germany, Research Center Neurosensory Science, Carl von Ossietzky University, Oldenburg, Germany
The effects of sensory and electrical rhythmic stimulation on alpha oscillations and visual detection
Show/hide abstractIt has been shown in human electroencephalogram (EEG) that alpha oscillations (8-13 Hz) modulate perception via phase-dependent inhibition. If these alpha oscillations are coupled to an externally applied rhythmic oscillator (entrainment), inhibition maxima and minima of the intrinsic alpha rhythm should in theory appear more distinct in time and make potential phase-dependent perception predictable. There is, however, an ongoing debate on the efficacy of external rhythmic driving forces with regard to entrainment of intrinsic (alpha) oscillations. Here, we investigated the effect of visual stimulation on alpha oscillations in a first experiment. Secondly, we applied transcranial alternating current stimulation (tACS) to investigate the interaction of both types of rhythmic stimulation. The effect of pure sensory stimulation was investigated by probing the detection rates of brief visual light flashes, embedded in a 10Hz flicker stream. As predicted by the theory of entrainment, depending on the target position within the flicker stream (stimulation phase, either together with a flash or between two flashes), detection rates were found to be improved or diminished, respectively. Furthermore, following the physics concept of entrainment, a stronger driving force leads to stronger entrainment. This effect was found for the electrophysiological data (phase locking between stimulation train and alpha oscillation in the parietal cortex) as well as for the detection task. With brighter flicker stimulation, stronger phase-locking and increased detection modulation was observed. We further added tACS to the experimental design, either in-or out-of phase with the light flicker. A novel setup was implemented, that accounts for individual delays between visual flicker presentation and electrophysiological response of the entrained alpha oscillator in the parietal cortex. We were able to show that in general alpha oscillations can be entrained via external driving forces. This finding paves the way for future investigations on the causal relation between oscillations and their function.
Sunday 12.00 - 13.30 Grote zaal Poster #7 Social behavior Pasquale Cardellicchio
All authors & affiliationCardellicchio Pasquale, Center for Translational Neurophysiology of Speech and Communication, Istituto Italiano di Tecnologia, Via Fossato di Mortara 17/19, 44121, Ferrara, Italy.* Hilt Pauline, Center for Translational Neurophysiology of Speech and Communication, Istituto Italiano di Tecnologia, Via Fossato di Mortara 17/19, 44121, Ferrara, Italy. Olivier Etienne, Institute of Neuroscience, Universite catholique de Louvain, B-1200 Brussels, Belgium Fadiga Luciano, Center for Translational Neurophysiology of Speech and Communication, Istituto Italiano di Tecnologia, Via Fossato di Mortara 17/19, 44121, Ferrara, Italy, Section of Human Physiology, University of Ferrara, Via Fossato di Mortara 17/19, 44121, Ferrara, Italy. D'Ausilio Alessandro, Center for Translational Neurophysiology of Speech and Communication, Istituto Italiano di Tecnologia, Via Fossato di Mortara 17/19, 44121, Ferrara, Italy.
Early modulation of short intra-cortical Inhibition during the observation of action errors
Show/hide abstractEveryday while interacting with others, we continuously infer their intentions. In this process, a fast and effective discrimination of action errors is fundamental to promote flexible adaptation to other's behavior and consequently efficient social interaction. It's well known that action observation activates neural circuits that are partially overlapping with those recruited by the execution of actions [1--3]. However, the mechanisms specifically devoted to action error recognition remain poorly understood at the neural level. Some studies proposed that action error detection may rely on predictive mechanisms sensing subtle kinematic violations in the observed action via the recruitment of the Action Observation Network(AON)[4--6]. Alternatively, others studies have shown stronger facilitation in the AON when observing erroneous, impossible or uncommon actions[7--10]. This study focuses on the role of the motor system during the observation of action errors. Participants were asked to detect errors while observing someone tying a knot. We used two different type of errors, a procedural and a control one. In the first, a canonical tying error does not allow achievement of the action goal. In the second, the rope is cut in two segments, still impeding the achievement of the goal. We used a paired pulse Transcranial Magnetic Stimulation technique to measure intracortical inhibition (sICI), and facilitation (ICF) at three different time points (120, 180, 240 ms) after the errors. Our results show an early reduction of intracortical inhibition (120ms) in the procedural error. This release of inhibition could be explained by the greater mismatch with respect to the generated top-down predictions. Hence, sICI might be a potential candidate mechanism through which the AON could influence motor cortex output. In line with the predictive coding framework, the function carried out by the AON could be that of computing prediction errors based on the visual representation of actions.
Sunday 12.00 - 13.30 Grote zaal Poster #8 Attention Katya Olmos Solis
All authors & affiliationKatya Olmos Solis*, Anouk M. van Loon*,** Christian N. L. Olivers1*,** * Department of Experimental and Applied Psychology, Vrije Universiteit Amsterdam, The Netherlands ** Institute of Brain and Behavior Amsterdam, Vrije Universiteit Amsterdam, The Netherlands
Subtle eye movements reveal the temporal dynamics of preparing for visual search
Show/hide abstractThe search template is a visual working memory (VWM) representation that facilitates search by guiding attention towards matching features when they appear in view. Recent neurophysiological and behavioral data suggest that template activation occurs during the delay period prior to search, but the temporal dynamics of such activation remain unclear. In two experiments we tested if, in anticipation of a search task, the temporal dynamics of both general and template-specific preparatory effects can be inferred from subtle eye movements (mostly microsaccades). On each trial, participants memorized a target color (i.e. the template) for an upcoming search task. During the delay period prior to search, we presented an irrelevant Rapid Serial Visual Presentation (RSVP) of lateralized colored disks. The length of the delay was manipulated in blocks of long (4650 ms) and short delay (2790 ms) trials. We instructed participants to maintain fixation throughout the delay. Crucially, at different time points into the delay, the template color was inserted in the RSVP, allowing us to measure specific attentional biases towards the template location as a function of time. Results revealed a general suppression of saccade production in preparation to search: the closer in time participants were to the task, the fewer saccades they made. Strikingly, this suppression was stronger when, in a given time point, a template-matching disk was present compared to when absent. Moreover, although saccades became less frequent towards the search, they did become more informative about the current internal attentional goal. Specifically, in the time points where a template-matching disk was presented, relatively more and larger saccades went to the matching disk than to the irrelevant color, an effect that became stronger near the end of the trial, especially with long delays. Thus, we show that subtle eye movements track the preparation for selection prior to search.
Sunday 12.00 - 13.30 Grote zaal Poster #9 Memory Kate Ergo
All authors & affiliationEsther De Loof, Department of Experimental Psychology, Ghent University Kate Ergo, Department of Experimental Psychology, Ghent University Lien Naert, Department of Experimental Psychology, Ghent University Clio Janssens, Department of Experimental Psychology, Ghent University Filip Van Opstal, Center for Research in Cognition & Neurosciences, Universite Libre de Bruxelles Department of Psychology, University of Amsterdam Tom Verguts, Department of Experimental Psychology, Ghent University
Signed reward prediction errors drive declarative learning
Show/hide abstractFor humans to show optimal cognitive functioning, the brain must continuously make predictions. One of the features we make predictions about, is reward. Whenever there is a mismatch between the reward that was predicted and the reward actually obtained, we experience a reward prediction error (RPE). RPEs are thought to drive learning. This has been well established in procedural learning paradigms (e.g., classical and operant conditioning). However, empirical evidence on whether RPEs drive declarative learning -- a quintessentially human form of learning -- remains surprisingly absent. In this study, we used a declarative learning paradigm in which RPEs were coupled to the acquisition of Dutch-Swahili word pairs. The occurrence of signed RPEs (SRPEs; "better-than-expected" signals) during declarative learning improved recognition performance in a follow-up recognition test, with increasingly positive RPEs leading to better recognition. In addition, we demonstrate that classic declarative memory mechanisms such as time-on-task fail to explain recognition performance. The beneficial effect of SRPEs on recognition performance was subsequently affirmed in an EEG replication study. Moreover, we found oscillatory (high-beta and high-alpha) signatures for SRPEs during reward feedback, similar to SRPE signatures found in earlier procedural learning paradigms. Importantly, these results offer a powerful reinterpretation of the testing effect, with key implications for education.
Sunday 12.00 - 13.30 Grote zaal Poster #10 Attention Hesham ElShafei
All authors & affiliationRomain Bouet, (1) Olivier Bertrand, (1) Aurelie Bidet-Caulet, (1) (1) Brain Dynamics and Cognition Team, Lyon Neuroscience Research Center; CRNL, INSERM U1028, CNRS UMR5292, University of Lyon 1, Universite de Lyon, Lyon, France
Show/hide abstractAnticipation biases our attentional set towards upcoming stimuli. This attentional mechanism has been shown to operate by enhancing neural activity in cortical regions responsible for processing the upcoming stimulus, while suppressing activity in regions responsible for processing stimuli outside of the attentional focus. The role of alpha oscillations in sustaining these attentional modulations has been well demonstrated in the visual domain (e.g. Thut et al., 2006). However, only a handful of studies (e.g. Frey et al., 2014) have succeeded to establish this role in the auditory domain. MEG data were collected from fourteen healthy young participants performing an auditory pitch discrimination task. Anticipatory attention was manipulated by a visual cue that was either informative, i.e. indicating the target side (either left or right), or uninformative. Behaviorally, participants were significantly faster in the informative condition. Analysis of the post-cue alpha activity at the sensor, source and virtual electrode levels demonstrated that anticipatory attention modulated alpha activity differently in the auditory and visual cortices. Preceding the onset of the auditory target, alpha power decreased in the auditory cortex. This decrease was accompanied by an increase in alpha power in the visual cortex. Furthermore, these two modulation profiles were centered around different alpha frequencies (9Hz for the auditory cortex and 13Hz for the visual cortex). Moreover, alpha activity in both cortices were modulated differently according to the type of the visual cue and behavioral performances negatively correlated with the increase of alpha power in the visual cortex. Evidence of top-down modulation of auditory alpha has only been recently put forward. The present results corroborate this outgrowing literature and provide new insights into the dynamics of top-down modulation of alpha oscillations within the auditory and the visual cortices.
Sunday 12.00 - 13.30 Grote zaal Poster #11 Attention Anouk van Loon
All authors & affiliationAnouk M. van Loon, Department of Cognitive Psychology, Vrije Universiteit Amsterdam Johannes J. Fahrenfort, Department of Cognitive Psychology, Vrije Universiteit Amsterdam Christian N. L. Olivers, Department of Cognitive Psychology, Vrije Universiteit Amsterdam
Current and future goals warp object category space in opposite directions
Show/hide abstractPrevious studies have shown that only representations of currently task relevant goals can be decoded from brain activity. However, it is still unclear how and where representations of future goals are instantiated in the brain. Here, we measured fMRI of 24 human participants while on each trial we presented two real-world objects from different categories serving as targets for two consecutive visual search tasks. We manipulated the relevance of the objects with a cue that indicated which object to look for first (current), and which second (prospective). Before each search there was an eight second retention interval. We used multi-voxel pattern analysis to decode the dynamical changes in representational space of the object categories in object-selective cortex throughout the trial, as a function of current versus prospective task relevance. As predicted, we observed better category decoding for the currently relevant than for the prospectively relevant category right before the first search. In addition, when we trained the classifier on the currently relevant category and tested on the prospectively relevant category or vice versa, classification was below-chance during both searches. This indicates that current and future object categories are represented in opposite corners of the representational space. Indeed, representational similarity analyses confirmed that as a trial unfolds, object representations move from object category space (e.g. a cow) into relevance space (e.g. current target), where current and prospective targets of the same category are represented by opposite representational patterns. Taken together, our results demonstrate how the brain shields current from future targets and vice versa, by pushing the representational space into opposite directions.
Sunday 12.00 - 13.30 Grote zaal Poster #12 Memory Jane Herron
All authors & affiliationJane Herron, Cardiff University
Preparatory correlates of retrieval orientation are attenuated following working memory depletion.
Show/hide abstractPrevious research has shown that neural correlates of selective recollection are sensitive both to working memory capacity (WMC) and to working memory depletion, with participants showing a reduced ability to selectively recollect task-relevant information after completing a stroop task (Elward et al. 2010, 2013). It has been proposed that selective recollection is enabled by retrieval orientations that focuses retrieval efforts towards task-relevant content. An important question, therefore, is whether neural correlates of retrieval orientations are also sensitive to working memory depletion. Here, visually presented words were encoded in one of two tasks. One task required easy/difficult-to-draw judgments and the second required a function to be assigned. Participants then completed either a stroop or a control task for 6 minutes before two ‘exclusion' memory tests. Both tests required the endorsement of words from a designated encoding task (‘targets'), this being the Drawing task in one test and the Function task in the other. Preparatory correlates of retrieval orientation were obtained by time-locking ERPs to the pre-stimulus fixation asterisk and separating them according to target designation. The recording epoch spanned the pre-stimulus interval (1000ms). Preparatory ERPs diverged according to target designation in the control group only, for whom frontal ERPs were more positive-going for ‘Drawing' targets for the duration of the epoch. This preparatory orientation effect was eliminated following completion of the stroop task. The novelty of these findings are twofold; i) this is first time that preparatory measures of retrieval orientation have been obtained during consistent (as opposed to alternating) retrieval requirements, indicating that this effect is indexing the maintenance of retrieval orientations throughout the memory tests, ii) the attenuation of this effect following the stroop task indicates that the maintenance of retrieval orientations is sensitive to working memory capacity.
Sunday 12.00 - 13.30 Grote zaal Poster #13 Consciousness Daniel Bor
All authors & affiliationRogier A Kievit, Medical Research Council Cognition and Brain Sciences Unit, Cambridge. Darren Price, Medical Research Council Cognition and Brain Sciences Unit, Cambridge. Tristan A Bekinschtein, Psychology Department, University of Cambridge.
Awake Conscious Level Dynamics and Individual Differences in a Large MEG Sample
Show/hide abstractThere has been considerable recent progress in measuring conscious level (CL) using neuroimaging metrics. For instance, Casali and colleagues (2013) used the perturbational complexity index (PCI) to distinguish awake from sedated and sleeping subjects. However, this research almost exclusively compares awake with clearly non-conscious states, rather than the dynamics of CL during normal wakefulness. In addition, although cognitive capacity and CL are commonly studied separately, there is a clear between-species relationship between intelligence and ability to pass landmark conscious tasks, such as the mirror test, suggesting a putative link between CL and intelligence. Here we take advantage of a large Magnetoencephalography (MEG) dataset of healthy adults (n=623) across the lifespan (18-88 years), to examine within-subject CL dynamics during tasks, and whether between-subject differences in wakeful CL is associated with IQ. To measure CL, for each 4 s (arbitrary length) epoch, we calculated Lempel Ziv Complexity (LZC), a passive measure related to PCI, but suitable for MEG data alone, which assesses the spatiotemporal complexity (randomness) of the MEG signal (Schartner et al, 2015). Previous work suggests that higher signal complexity is related to higher CL. In addition, we calculated theta/alpha (T/A) frequency power ratio, to measure drowsiness. A reduction in alpha power and corresponding increase in theta power is related to increased drowsiness (Bareham et al 2014). Across an hour-long MEG session, both LZC and T/A measures exhibited marked fluctuations, during both passive and active tasks. Furthermore, lower CL (LZC) correlated with increased drowsiness (T/A) (r=0.68, p<0.0001), demonstrating convergence of these very different measures. When examining resting state, mean wakeful LZC correlated positively with IQ (r=0.36, p<0.00001), but negatively with age (r=-0.26, p<0.001), indicating that CL is maximal in younger, higher-IQ sections of the population.
Sunday 12.00 - 13.30 Grote zaal Poster #14 Connectivity/networks Irene de Caso
All authors & affiliationGiulia Poerio, University of York, Department of Psychology Elizabeth Jefferies, University of York, Department of Psychology Jonathan Smallwood, University of York, Department of Psychology
That's me in the spotlight -- Neural basis of individual differences in self-consciousness.
Show/hide abstractA long-standing literature implicates activity within the default mode network (DMN) to processes linked to the self (Northoff et al., 2006). However, contemporary work suggests that other large-scale networks networks might also be involved. For instance, goal-directed autobiographical planning requires positive functional connectivity (FC) between DMN and frontoparietal control (FPCN) networks (Spreng et al., 2010). The present study examined the inter-relationship between trait self-focus (measured via a self-consciousness scale; SCS), incidental memory in a self-reference paradigm, and resting state FC of large-scale networks. Behaviourally, we found that private SCS was linked to stronger incidental memory for self-relevant information. We also examined how patterns of FC differed according to levels of self-consciousness by using the SCS data to drive multiple regression analyses with seeds from the DMN, the FPCN and the limbic network. High levels of SCS was not linked to differences in the functional behaviour of the DMN, however, it was linked to stronger FC between FPCN and a cluster extending into the hippocampus, which meta analytic decoding using Neurosynth linked to episodic memory retrieval. Subsequent analysis demonstrated that trait variance in this pattern of FC was a moderator for the observed relationship between private SCS and enhanced memory for self-items. Together these findings suggest that interactions between the FPCN and hippocampus may support the memory advantage of self-relevant information associated with SCS and confirm theoretical positions that argue that that self-related processing does not simply depend upon the DMN, but instead relies on complex patterns of interactions between multiple large-scale networks.
Sunday 12.00 - 13.30 Grote zaal Poster #15 Sensory processing Christoph Braun
All authors & affiliationFioravanti Chiara, MEG Center, University of Tuebingen, Tuebingen, Germany; Institute of Medical Psychology and Behavioral Neurobiology, University of Tuebingen, Germany. Carboni Margherita, EEG an Epilepsy Unit, University Hospital, Geneva, Switzerland; MEG Center, University of Tuebingen, Tuebingen, Germany. Mazzetti Cecilia, Donders Center, Nijmegen, The Netherlands; 2 CIMeC, Center for Mind/Brain Sciences Cognitive Neuroscience, University of Trento, Trento, Italy; MEG Center, University of Tuebingen, Tuebingen, Germany. Kajal Singh Diljit, MEG Center, University of Tuebingen, Tuebingen, Germany; Institute of Medical Psychology and Behavioral Neurobiology, University of Tuebingen; Graduate School of Neural and Behavioral Sciences, International Max Planck Research School, Tuebingen, Germany. Ziemann Ulf, Faculty of Medicine, Department of Neurology and Stroke, University of Tuebingen; Hertie Institute for Clinical Brain Research, Tuebingen. Braun Christoph, MEG Center, University of Tuebingen, Tuebingen, Germany; CIMeC, Center for Mind/Brain Sciences Cognitive Neuroscience, University of Trento, Trento, Italy.
GABAergic Control in the Somatosensory System -- A Neuropharmacological MEG-Study
Show/hide abstractIntroduction: The interaction between sensory input of any modality and a tactile stimulus almost exclusively leads to a suppression of the somatosensory evoked response. Besides refractoriness, GABA mediated inhibition is hypothesized as underlying mechanism. Using pharmaco-MEG we investigated whether GABA modulates responses of primary (SI) and secondary somatosensory (SII) cortex and which type of GABA receptors are involved. Furthermore, we studied whether the response suppression in SII reflects the propagation of inhibited SI activity, or whether SII is specifically inhibited. Method: 16 male subjects were studied in 8 sessions receiving GABAA agonists alprazolam (Alp), ethanol (Eth) and GABAB agonist baclofen (Bac), or placebo (Pla) and tactile stimulation to index and middle finger of the left hand. Somatosensory evoked responses were recorded using a 275-channel MEG system. Sources in contralateral SI and in contra- and ipsilateral SII were localized and respective source activities were estimated. Drug dependent differences of peak amplitudes were evaluated. Result: Comparing effects of GABA-agonists to the placebo condition, a significant reduction in source activity was found for Alp in SI and for Alp and Eth in SII. Bac did not affect the activity neither in primary nor secondary somatosensory cortices. For Alp a differential effect was found for different latencies. The reduction of activity in SII for Alp could be explained by the propagation of inhibited SI activity. Discussion: Since Alp showed a differential effect on the somatosensory cortices, our data suggest the involvement of fast GABA receptors on the level of SI. Comparing SI and SII activations we conclude that the initial volley of SI activation is less affected by Alp than the second one and that SII activity is driven by the first volley of SI activity. Pharmaco-MEG appears to be a promising tool to shed light on the flow of sensory information
Sunday 12.00 - 13.30 Grote zaal Poster #16 Electrophysiology methods Andrew Corcoran
All authors & affiliationAndrew W. Corcoran, Cognition and Philosophy Lab, School of Philosophical, Historical and International Studies, Monash University, Clayton, Victoria 3800, Australia. Phillip M. Alday, Cognitive Neuroscience Lab, School of Psychology, Social Work and Social Policy, University of South Australia, Magill 5072, Australia. Matthias Schlesewsky, Cognitive Neuroscience Lab, School of Psychology, Social Work and Social Policy, University of South Australia, Magill 5072, Australia. Ina Bornkessel-Schlesewsky, Cognitive Neuroscience Lab, School of Psychology, Social Work and Social Policy, University of South Australia, Magill 5072, Australia.
restingIAF: A reliable, automated, open source method for quantifying individual alpha frequency
Show/hide abstractIndividual alpha frequency (IAF) is a promising electrophysiological marker of interindividual differences in cognitive function [1][2]. In particular, IAF has been shown to predict performance across a variety of psychophysical and cognitive tasks [3][4], and may underpin trait-like differences in information processing [5] and general intelligence [1]. IAF has also been cited as a useful anchor point for determining individually-tailored frequency band analyses [6]. Despite this large body of literature, however, there seems to be no clear consensus on the optimal method for estimating IAF. We therefore sought to develop a reliable, automated method for IAF estimation that could be easily integrated within existing analysis pipelines. We implemented a method of calculating two common IAF estimators (peak and gravity frequency) in MATLAB and Python (both available on GitHub; Python implentation also in the philistine package on PyPi). This restingIAF routine locates the bounds of the dominant alpha component according to the first and second derivatives of its Savitzky-Golay smoothed spectral density. We evaluated its performance characteristics in both empirical and simulated EEG datasets. restingIAF generated 61 IAF estimates from 63 healthy adults, thus yielding a higher proportion of estimates than both standard analysis (e.g. [7]) and Gaussian curve-fitting techniques (e.g, [8]). The distribution of IAF estimates was consistent with that derived using more complex algorithms within large-scale datasets (e.g., [9]). Preliminary analysis of simulated data revealed that restingIAF accurately extracts the peak frequency of underlying alpha-band components, even when signal-to-noise ratio is highly degraded. restingIAF is a novel method of extracting IAF estimates that can easily be applied to large datasets. It is fast, reliable, open source, and available in two popular programming languages, and thus easily integrated in the most popular M/EEG toolsets (EEGLAB, FieldTrip and MNE-Python). Widespread adoption might significantly improve the consistency and replicability of future IAF-related research.
Sunday 12.00 - 13.30 Grote zaal Poster #17 Sensory processing Kang Yong Eo
All authors & affiliationKang Yong Eo, Center for Cognition and Sociality, Institute for Basic Science (IBS) Yee-Joon Kim, Center for Cognition and Sociality, Institute for Basic Science (IBS)
Stimulus variability parametrically modulates theta oscillations during perceptual mean computation
Show/hide abstractThe visual system can efficiently extract statistical mean and variance from sets of similar objects. Although a number of behavioral studies have shown that the error of perceptual mean increases as stimulus variance increases, neural mechanisms underlying such parametric relation between stimulus variance and perceptual mean remain poorly understood. Inspired by the following facts: (1) frontal theta (4-7 Hz) activities parametrically increases as a function of memory load and (2) theta coupling between visual and frontal area is involved in encoding and maintaining stimulus identity, we investigated whether theta oscillations in the EEG parametrically vary with computational load caused by stimulus variance during mean orientation judgment task. We used an array of randomly oriented 36 Gabors whose orientations were drawn from a normal distribution with six possible mean orientations and five possible standard deviations of 0°, 7.5°, 15°, 22.5°, and 30°. An array of Gabors was presented for 150 ms, which was successively followed by an 800 ms delay and a randomly oriented probe Gabor at the center of the screen. Observers were instructed to adjust the orientation of a probe to match the mean orientation of the array of Gabors. We found that posterior and frontal theta amplitudes as well as phase-locking values parametrically increased as a function of array variance during encoding period from an array onset to 300 ms. Importantly, the sequential linear mixed-effects modeling showed that the posterior theta peak amplitude at around 200 ms correlated well with trial-to-trial variations in the amount of behavioral error. We also observed that phase-locking values in the posterior and frontal areas parametrically decreased as array variance increased after a probe onset. These results suggest that theta activity reflects stimulus variance in the encoding phase and the deviation from perceptual mean in the retrieval phase, respectively.
Sunday 12.00 - 13.30 Grote zaal Poster #18 Language Katarzyna Jednorog
All authors & affiliationKatarzyna Jednorog, Laboratory of Psychophysiology, Department of Neurophysiology, Nencki Institute of Experimental Biology, Warsaw, Poland* Magdalena luniewska, Laboratory of Psychophysiology, Department of Neurophysiology, Nencki Institute of Experimental Biology, Warsaw, Poland, Faculty of Psychology, Warsaw University, Poland Agnieszka Dębska, Laboratory of Psychophysiology, Department of Neurophysiology, Nencki Institute of Experimental Biology, Warsaw, Poland Katarzyna Chyl, Laboratory of Psychophysiology, Department of Neurophysiology, Nencki Institute of Experimental Biology. Anna Banaszkiewicz, Laboratory of Brain Imaging, Neurobiology Center, Nencki Institute of Experimental Biology, Warsaw, Poland. Agata zelechowska, Laboratory of Psychophysiology, Department of Neurophysiology, Nencki Institute of Experimental Biology, Warsaw, Poland Marcin Szwed, Department of Psychology, Jagiellonian University, Krakow, Poland Artur Marchewka, Laboratory of Brain Imaging, Neurobiology Center, Nencki Institute of Experimental Biology, Warsaw, Poland
The influence of familial risk and dyslexia on phonological processing -- a longitudinal fMRI study
Show/hide abstractPhonological awareness is a key factor in reading acquisition. Our aim was to establish the neural correlates of auditory rhyming in children at different stages of literacy acquisition and compare those who developed dyslexia with those who became typical readers and either had or did not have familial history of dyslexia (FHD). At time-point 1 (TP1) we tested phonological and reading-related skills in 120 Polish FHD+ and FHD- pre- and beginning readers (aged 5;6 -- 8;0). During fMRI children had to decide whether two aurally presented words rhymed (rhyme task) or whether they were spoken by speakers of the same gender (voice task). At TP2, 2 years later, tests were repeated and formal diagnosis of dyslexia was conducted. 3 groups were formed: FHD+ children who developed dyslexia (FHD+D, N=15), FHD+ (FHD+C, N=18) and FHD- (FHD-C, N=18) controls, matched for age, sex, SES, and IQ. Brain activity to rhyme>voice was compared between the groups at two TPs. FHD+D group underperformed in numerous reading and phonological awareness tasks compared to control groups with more pronounced differences at TP2. In the rhyme task FHD+D children scored lower than control groups only at TP1. At this stage both FHD+ groups (irrespective of later reading outcome) showed hypoactivation compared to FHD-C group, which overlapped in the left inferior frontal gyrus. This area was consistently underactivated in dyslexia and might reflect a dysfunction in efficient access to phonological representations. At TP2 FHD+D children displayed higher activation than both control groups, which overlapped in the left tempo-parietal junction (TPJ). Additionally in the left TPJ an interaction between group and TP was observed -- FHD+D had lower activity at TP1 and higher at TP2 than FHD-C group. Left TPJ activation might reflect an early, suboptimal, stage of grapheme-phoneme correspondence and that is why controls move away from it.
Sunday 12.00 - 13.30 Grote zaal Poster #19 Electrophysiology methods Marzieh Sorati
All authors & affiliationMarzieh Sorati, Norwegian University of Science and Technology(NTNU) Dawn M. Behne, Norwegian University of Science and Technology(NTNU)
How early in cortical processing does musical training facilitate perception and what does it facilitate?
Show/hide abstractMusical training facilitates perception, as has been researched through both electrical activity recordings and behavioral studies. While these studies have focused on relatively late cognitive components, some have indicated a possible early onset of facilitation without directly testing it. Furthermore, previous research has not directly tested whether facilitation is specific to auditory perception. The current study explicitly addresses whether N1 and P2 waveforms differ for trained musicians and non-musicians for auditory processing. ERP data were collected from six participants in each group using a high density EEG system during a detection task with audio and control conditions. The audio condition included the audio of a note played from a keyboard and a gray visual display. The control condition had the same visual display without the audio signal. The ERP waveform from the control condition was subtracted from the audio condition (difference wave) to isolate the auditory components. For each group onset latency and amplitude were calculated for N1 and P2 at Cz for the audio condition and for the difference wave. While for the difference wave no significant differences were observed between the two participant groups for N1 and P2 latency and amplitude, in the audio condition N1 onset latency was significantly earlier for musicians compared to non-musicians. This difference in N1 onset latency confirms implications from previous research by directly demonstrating that facilitation due to musical training occurs at N1. Furthermore, despite the difference in N1 onset latency between the two groups in the audio condition, that difference waves did not reliably differ between the two groups suggests that the perceptual facilitation observed for trained musicians is not auditory specific. These findings imply that factors other than audition alone underlie the early facilitation associated with musical training.
Sunday 12.00 - 13.30 Grote zaal Poster #20 Sensory processing Alexis Perez-Bellido
All authors & affiliationAlexis Perez-Bellido, Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen, Kapittelweg 29, 6500 HB Nijmegen, The Netherlands Valentin Wyart, Laboratoire de Neurosciences Cognitives, Inserm unit 960, Departement d'etudes Cognitives, ecole Normale Superieure, PSL Research University, 75005 Paris, France Floris P. de Lange, Donders Institute for Brain, Cognition and Behaviour, Radboud University Nijmegen, Kapittelweg 29, 6500 HB Nijmegen, The Netherlands
Sound enhances visual detection in the center but not in the periphery of the visual field
Show/hide abstractSounds can enhance visual detection, however the mechanisms that govern such multisensory interaction are largely unknown. In the present study, we tested whether the strength of the audiovisual interaction is dependent on visual eccentricity and audiovisual spatial congruency. Moreover, we examined how visual detection performance was modulated when a concurrent cue was provided in the visual modality, instead of the auditory modality. Participants had to detect noisy vertical Gabor patches at two different eccentricities (center and periphery) under visual-only (V), audiovisual (AV) and visually cued (VC) conditions. Using signal detection theory, we observed that sounds enhanced visual detection only in the central portion of the visual field. Interestingly, this auditory-induced visual enhancement only took place when the audiovisual signals were spatially congruent. In contrast, visual cues degraded, rather than improved, participant's sensitivity at the central and peripheral visual areas compared to the visual-only condition. Our results reveal that auditory-induced visual detection enhancement is selectively present in the central visual field and cannot be accounted by an explicit temporal cueing phenomenon (as previously suggested by Lippert et al., 2007). We argue that sounds may improve visual detection through a gain modulation of attention, which would be expected to be largest in the central portion of the visual field, where there is the largest density of neurons dedicated to the analysis of visual signals (cortical magnification).
Sunday 12.00 - 13.30 Grote zaal Poster #21 Memory Eugenia Marin-Garcia
All authors & affiliationPedro M. Paz-Alonso, Basque Center on Cognition, Brain and Language (BCBL)
Language use in true and false DRM memories: An fMRI study comparing monolinguals and bilinguals
Show/hide abstractConsequences of bilingualism in the memory domain remain not well understood. Research evidence suggests that second language use changes the ability to access semantic representations in the native language (Linck, Kroll & Sunderman, 2009). However, the consequences of this effect on relational semantic processing and on episodic memory are not clear. We used the Deese-Roediger-McDermott (DRM) paradigm (Roediger & McDermott, 1995) to investigate semantic relational processing leading to veridical and false memories in monolinguals and bilinguals. Our previous behavioral study (Marin-Garcia & Paz-Alonso, in prep.) showed that monolinguals exhibited more false memories, but similar true memories, relative to bilinguals; and the bilingual group showed more false memories in the more used language compared to the less used language. This suggests that language use modulates relational semantic processing probably by facilitating the access to semantic representations. The neural basis related to this modulation is unknown. In the present fMRI study, participants were divided in two groups based on their language profile: Spanish monolinguals and Spanish-Basque bilinguals. Language use/exposure, age of acquisition and proficiency were controlled. Participants were instructed to study word lists for a memory test. Materials consisted in Spanish DRM word lists that converge on a semantic theme captured in a critical word never presented in the list. After encoding, participants performed an old/new recognition test that included studied words, critical lures, and unrelated lures. Expected results of the recognition phase include no significant differences between groups for true greater than false memory in the Medial Temporal Lobe and Posterior Parietal Cortex. However, greater activation of the Inferior Frontal Gyrus for false greater than true memory is expected for monolinguals relative to bilinguals. Results will be discussed based on Spreading of Activation Theory (Roediger, Balota & Watson, 2001).
Sunday 12.00 - 13.30 Grote zaal Poster #22 Language Joost Rommers
All authors & affiliationJoost Rommers, Donders Institute for Brain, Cognition and Behaviour Kara D. Federmeier, University of Illinois
Lingering predictions: A pseudo-repetition effect for previously expected but not presented words
Show/hide abstractPredictions might support rapid language processing, as long as they are realized. But when a prediction is disconfirmed, is it suppressed, or does it linger? This study manipulated whether words were actually seen or were only expected and probed their fate in memory by presenting the words (again) a few sentences later. Thirty-six EEG participants read weakly constraining sentences ("The proofreader asked her to replace the word hot"; cloze probability 0.01). In the Previously Seen condition, the final word had been seen previously in a different weakly constraining sentence ("He was surprised when he found out that it was hot"; cloze 0.01). In the Expected But Not Seen condition, a likely prediction of the word had previously been disconfirmed ("Be careful, because the top of the stove is very dirty"; cloze "dirty" 0.01, cloze "hot" 0.86). In the Not Previously Seen condition, the word had not been presented before. Two sentences intervened between initial presentation/expectation and the experimental sentence. Fillers ensured that only 14% of the final words constituted a repetition. As expected, Previously Seen words elicited a strong N400 decrease relative to Not Previously Seen words (a repetition effect). Critically, a reliable N400 decrease was also observed in response to Expected But Not Seen words. In addition, late repetition effects were observed in the form of LPC enhancement and alpha power decreases, but only for Previously Seen words. The "pseudo-repetition effect" on the N400 suggests that disconfirmed predictions can linger at some stages of processing.
Sunday 12.00 - 13.30 Grote zaal Poster #23 Clinical Meadhbh Brosnan
All authors & affiliationMeadhbh Brosnan, Trinity College Dublin, Oxford University, Monash University Siobhan Harty, Oxford University, Jacob Levenstein, Oxford University, Redmond O'Connell, Trinity College Dublin, Paul Dockree, Trinity College Dublin, Ian Robertson, Trinity College Dublin, Nele Demeyere, Oxford University
The Temporal Dynamics of Attentional Engagement following Right Hemisphere Stroke
Show/hide abstractIt is well documented that the right hemisphere (RH) plays a privileged role in supporting sustained attention (cf Langner and Eickhoff 2013; Robertson 2014). However, the temporal dynamics of this relationship are less well understood. We assayed sustained attention capacity in stroke patients with unilateral RH damage (N=14), and neurologically-healthy older adults (N=23) using a task sensitive to attentional decrements over short periods of time (the Continuous Temporal Expectancy Task; (O'Connell et al. 2009). In this task, participants are required to continuously monitor a stream of of regularly alternating patterned stimuli in order to detect an infrequently occurring target stimulus, the duration of which was at least 330 ms longer than the standard stimulus. While both groups demonstrated linear decrements in accuracy across the task blocks, patients with RH damage exhibited relatively poorer overall target detection. Further, compared with neurologically healthy older adults, the RH stroke patients showed a significantly steeper decrease in accuracy within the task blocks indicating performance decrements occurring within very short temporal windows. Our data provide novel evidence that RH stroke patients exhibit a vigilance decrement within shorter temporal windows than previously noted in the literature. The break periods between blocks restored performance levels in the RH stroke patients, which provides valuable information for the development of neurorehabilitation protocols aimed at facilitate plasticity and recovery processes following acquired brain injury.
Sunday 12.00 - 13.30 Grote zaal Poster #24 Decision-making Yang Liu
All authors & affiliationYang Liu, University of Amsterdam; Gorka Fraga Gonzalez, University of Amsterdam; Wery van den Wildenberg, University of Amsterdam; Davide Rigoni, Ghent University; Marcel Brass, Ghent University; Reinout Wiers, University of Amsterdam; Richard Ridderinkhof, University of Amsterdam;
Free won't after two beers: an EEG study on intentional inhibition
Show/hide abstract It has been frequently reported that acute alcohol intake affects inhibition. The typically use stop signal task and go/no-go task, involve an external cue. However, in our daily life there isn't always an external stop signal, such as when we want to stop eating nuts (in favor of long-term dietary priorities). Intentional inhibition refers to the capacity to intentionally withhold an action that is already set in motion. In the current experiment we employed a novel motor chasing task to test cued and intentional inhibition in a single experiment. Electroencephalographic (EEG) activity was measured both under alcohol and placebo for each participant. A slow negative-going wave was observed, starting 1.5 seconds before participants intentionally initiate the action. A highly similar readiness potential was observed prior to intentional inhibition of the ongoing action. No such readiness potential was observed prior to externally triggered action engagement or inhibition. No main effect or interaction effect of alcohol was found. This finding suggests that intentional inhibition has neurophysiological precursors similar to intentional action initiation. We propose that this component reflects processes involved in the formation of an intention. Alcohol use seems to have a limited effect on intentional inhibition processes.
Sunday 12.00 - 13.30 Grote zaal Poster #25 Cognitive modeling Giovanni Vecchiato
All authors & affiliationGiovanni Vecchiato, Institute of Neuroscience, National Research Council, Parma, Italy Maria Del Vecchio, University of Modena and Reggio Emilia, Italy; Institute of Neuroscience, National Research Council, Parma, Italy Luca Ascari, Camlin Italy s.r.l. Fabio Deon, Camlin Italy s.r.l. Luca Kubin, Camlin Italy s.r.l. Jonas Ambeck-Madsen, Toyota Motor Europe Pietro Avanzini, Institute of Neuroscience, National Research Council, Parma, Italy Giacomo Rizzolatti, Institute of Neuroscience, National Research Council, Parma, Italy
Electroencephalographic correlates of braking and acceleration events during simulated car driving
Show/hide abstract
Sunday 12.00 - 13.30 Grote zaal Poster #26 Consciousness Charles VERDONK
All authors & affiliationVERDONK Charles, Neurosciences and operational constraints Department, French Armed Forces Biomedical Research Institute, Bretigny-sur-Orge, France DEL CUL Antoine, Psychiatry Department, AP-HP, Pitie-Salpetriere Hospital, Paris, France RAMDANI Celine, Neurosciences and operational constraints Department, French Armed Forces Biomedical Research Institute, Bretigny-sur-Orge, France
Does error monitoring system work when stimulus perception is subliminal?
Show/hide abstractAim. The purpose of our study was to evaluate whether error monitoring system works when the stimulus perception is subliminal. Partial Errors (PE) and Negativity Error (NE) as on-line markers, and post-error slowing as off-line marker, are available to explore this system. PE are early subthreshold electromyographic (EMG) bursts observed in some proportion of correct overt responses [1,2]. NE is an electroencephalographic (EEG) index of error processing [3,4]. And post-error slowing corresponds to an extension of Reaction Time (RT) in a correct trial following an error [5]. Method. We used a paradigm into which subjects had to perform a RT forced-choice task of comparing stimulus (digit 1, 3, 7 or 9) to the number 5. Stimulus is presented for 16ms, and after a variable Stimulus Onset Asynchrony (SOA) a mask was presented. With short SOA (33ms), in contrast with long SOA (66ms) and control SOA (100ms), stimuli were mostly unseen. EMG and EEG data were simultaneously recorded when subjects performed task. Results. Sixteen healthy subjects were included (mean age: 36 ± 8.34 years old). On the short SOA condition we highlighted presence of PE (12% of trials number)). It was significantly different from long SOA (16%) and control SOA (18%) conditions (p<0.05). EEG data showed, on long SOA and control SOA conditions, the classical modulation of NE amplitude according to trial correctness (errors > PE > correct). On the short SOA condition, NE amplitude was no longer different between correct, error and PE trials. A post-error slowing has been observed on long SOA and control SOA conditions, but also on short SOA condition. The post-error slowing was not significantly different between conditions. Conclusion. Taken together, our findings suggest that error monitoring system could be work even when the stimulus perception is subliminal. Nevertheless, mechanisms of this processing remain to be elucidated.
Sunday 12.00 - 13.30 Grote zaal Poster #27 Consciousness Hein van Schie
All authors & affiliationHein van Schie, Radboud University, Behavioural Science Institute, Nijmegen, Netherlands Willem Sleegers, Tilburg University, Department of Social Psychology, Tilburg, Netherlands Ruud van Loon, Radboud University, Behavioural Science Institute, Nijmegen, Netherlands Arjan ter Horst, Aeres University of Applied Sciences, Dronten, Netherlands
Being your avatar: Dysfunctional motor control and body awareness in a third person self perspective
Show/hide abstractIntroduction: Humans increasingly use of virtual environments in which they control the body of a virtual avatar. Recent developments in VR interfacing are using the human body as a controller for the virtual body. Little is known, about the functional mechanisms that allow humans to experience an external body as their own, about the mechanisms that serve motor control of a virtual body, and the possible consequences of virtual body ownership for the awareness of the real body. This study investigated impairments in motor control and body awareness accompanying third person body control (3PBC) in four experiments. Method: In all four experiments, participants were seated behind a table and performed a block-sorting task while wearing a head-mounted display (HMD). The HDM presented the live feed from a camera that was placed at an elevated position (left or right) behind the subject (experiments 1 and 2) or at a position directly above the workspace in (eight) different angles of view (experiments 3 and 4). Subjects used their left or right hand in different blocks. Results: Performance (speed, accuracy), self-report (agency, body ownership, sensation, self-body dissociation), and pain-threshold measures revealed problems of motor control and disembodiment to increase as a function of the angle of view (AOV) between the camera perspective and the orientation of the hand. This implies that within a particular camera orientation, markedly different levels of impairment can be found for the left and the right hand, depending on their angle of use. Conclusions: Performance, self-report, and pain-threshold revealed similar effects of AOV. This suggest that loss of motor control and disembodiment effects are closely related and reflect the multisensory conflict that accompanies 3PBC. These findings support current views on body awareness which suggest that body representations are constructed on the basis of multisensory integration and indicate that body awareness of individual body pars can be selectively disrupted in specific virtual configurations.
Sunday 12.00 - 13.30 Grote zaal Poster #28 Attention Tzvetan Popov
All authors & affiliationTzvetan Popov, University of Konstanz, Department of Psychology, Konstanz, Germany Paul Szyszka, University of Konstanz, Department of Biology, Konstanz, Germany
Show/hide abstractAlpha oscillatory activity dominating the human EEG, initially described by Hans Berger in 1929, is thought to mediate cognitive function in humans. However, the functional role and computational mechanisms of the Berger rhythm are matter of continuing debate. We have established the honey bee (Apis mellifera) as a simpler model system that displays key characteristics of alpha oscillatory activity known from human EEG: a prominent frequency peak in ongoing activity, amplitude reduction upon stimulus presentation, and phase biasing of high-frequency gamma (> 30Hz) activity. We recorded local field potentials (LFP) simultaneously from the left and right mushroom body (analogous to the left and right hemisphere). We found that the ongoing LFP was dominated by an alpha-like 18 Hz rhythm. Olfactory stimulation reduced the amplitude of this rhythm and gave rise to high-frequency gamma activity. The phase of the ongoing 18 Hz rhythm modulated the amplitude of the gamma, an analog to the alpha-gamma phase-amplitude coupling (PAC) that has been described during cognitive processing in humans. Crucially, in humans PAC is often interpreted as an index of altered neuronal spiking activity, which in turn affects cognitive processing. With the simpler model system of the honey bee, we were able to test this prediction and confirmed spike-field coherence in accordance with the 18 Hz oscillation in addition to a relationship between spikes and gamma activity during stimulus processing. These results suggest a common role of oscillatory neuronal activity across phyla and provide a promising new venue for causal studies on the function of ongoing and stimulus-induced oscillatory neuronal activity.
Sunday 12.00 - 13.30 Grote zaal Poster #29 Decision-making Rongjun Yu
All authors & affiliationMing Pu, South China Normal University
Who receives the pain? Receipt of aversive outcomes modulates neural representations of performance and responsibility monitoring
Show/hide abstractPerformance monitoring and responsibility assignment are crucial for adaptive behaviors in the ever-changing social situations. The neural dynamics of these processes in social interactions are still unclear. This study investigated how responsibility levels influence the processing of outcome evaluation for one's own pain and observed others' pain using (EEG). Participants played a cooperation game in which either the participant in the self-context session or the confederate in the other-context session would receive electric shock if one or both players responded incorrectly. At the performance feedback stage, we found that the FRN and P300 were sensitive to both own performance and others' performance, while the FRN and P300 were only sensitive to own performance in other-context. At the the pain shock (self context) delivery stage, the P2 in no responsibility was significantly smaller than P2 in full responsibility condition, even though participants' performance was identical. At the oberserved pain stage (other context), the P300 in response to pain videos was not sensitive to responsibility. Our results suggested that the receipts of aversive outcomes modulate how our brain encodes other's performance and the neural representations of own responsibility.
Sunday 12.00 - 13.30 Grote zaal Poster #30 Attention Iris Wiegand
All authors & affiliationWiegand, Iris. Max-Planck UCL Centre for Computational Psychiatry and Ageing Research, Max-Planck Institute for Human Development, Berlin, Germany Sander, Myriam C. Centre for Lifespan Psychology, Max-Planck Institute for Human Development, Berlin, Germany
Cue-evoked alpha phase dynamics influence subsequent stimulus-processing in younger and older adults
Show/hide abstractThe level of alertness refers to the system's state of general responsiveness, and can be temporarily increased by presenting a neutral warning cue shortly before an event occurs (Posner & Petersen, 1990). In younger but not in older adults, a phasic increase in alertness facilitates sensory and attentional processing of visual stimuli (Wiegand et al., 2017a,b). In a visual letter report task, in which half of the displays were preceded by an auditory warning tone, we measured report accuracy and EEG power and phase-locking in response to both the alerting cue and the subsequent stimulus in groups of younger (n=18) and older (n=17) adults. Alerting cues evoked a significant increase in power as well as in intertrial phase locking, with a maximum effect in the alpha frequency. In contrast to recent findings (Tran et al., 2016), phase-locking was stronger in older compared to younger adults (Sander et al., 2012), Furthermore, phasic alerting affected later stimulus processing in both age groups: We found reduced alpha phase-locking to the stimulus in both younger and older adults following a cue. Given that only younger adults showed a behavioral visual performance benefit by phasic alerting, we suggest that stronger re-setting of the system to external events that occur in close succession hinders older adults to effectively use the warning cue to foster subsequent stimulus processing (Lindenberger & Mayr, 2014).
Sunday 12.00 - 13.30 Grote zaal Poster #31 Attention cecilia mazzetti
All authors & affiliationCecilia Mazzetti, Donders Institute for Brain, Cognition and Behaviour, Radboud University, 6525 EN Nijmegen, The Netherlands Johanna M. Zumer, School of Psychology and Computational Neuroscience and Cognitive Robotics Centre, University of Birmingham, Birmingham, B15 2TT, United Kingdom Tobias Staudigl, Donders Institute for Brain, Cognition and Behaviour, Radboud University, 6525 EN Nijmegen, The Netherlands Sean James Fallon, Department of Experimental Psychology, Oxford, UK, OX1 3UD Ole Jensen, Centre for Human Brain Health, University of Birmingham, Birmingham, UK;
Hemispheric Asymmetry of Globus Pallidus Predicts Reward-related Posterior Alpha Lateralization
Show/hide abstract Being constantly exposed to a wide variety of stimuli, our brain implements mechanisms of selective gating of the sensory input in order to optimize the processing of relevant information and suppress distractors. At the neuronal level, this mechanism is reflected by fluctuations of neuronal activity in the alpha band (8-13Hz), in brain regions functionally associated with the sensory modality of the current task. Previous studies have shown that stimulus properties can mediate prioritization of objects associated with contextual gain/loss. Specifically, the presence of a salient item associated with a reward valence shapes the modulation of alpha oscillations during selective attention, by boosting or decreasing the hemispheric alpha power lateralization if serving as target or distractor, respectively. In the current study, we sought to shed light on the top-down mechanisms by which reward-value associations exert their effect on attentional processing. Using a GLM approach, we investigated the relationship between hemispheric asymmetry of basal ganglia structures, associated with incentive salience functions, and lateralized alpha modulation. Whilst controlling for the other independent variables, we found that the lateralization of the Globus Pallidus (GP) significantly predicted hemispheric modulability of alpha power. Specifically, subjects with larger pallidal volume in the left than in the right hemisphere were also better at modulating alpha oscillations in the left compared to the right hemisphere. This effect was significantly larger in trials where two reward contingencies were presented (both target and distractor characterized by salient stimuli), as compared to conditions with one or zero contingencies. These results support the more recent notion of a reinforcement learning network involving the pallidal complex as a core nucleus within the basal ganglia, which modulates cortical information flow and biases selective processing of reward related stimuli.
Sunday 12.00 - 13.30 Grote zaal Poster #32 Decision-making Nelleke van Wouwe
All authors & affiliationAuthor list 1.van Wouwe, Nelleke, Department of Neurology, Vanderbilt University Medical Center, Nashville, TN, USA 2. Neimat, Joseph, Department of Neurosurgery, University of Louisville, Louisville, KY, USA 3. van den Wildenberg, Wery, Department of Psychology, University of Amsterdam, Amsterdam, The Netherlands 4. Wylie, Scott, Department of Neurosurgery, University of Louisville, Louisville, KY, USA
Dissociable effects of focal stimulation of Subthalamic Nucleus subregions on conflict and stopping control
Show/hide abstractIndividuals with Parkinson's disease (PD) show specific deficits in inhibitory action control. Studies of PD patients treated with deep brain stimulation (DBS) offer a unique opportunity to establish causal links between inhibitory action control mechanisms and basal ganglia function. We previously studied the effects of subthalamic nucleus (STN) DBS on inhibitory control dynamics in a Simon conflict task and showed that bilateral STN DBS at clinical settings, in dopamine-medicated PD patients improved inhibitory control of incorrect action impulses (Wylie et al., 2010). The current study consisted of two experiments. In experiment one, we replicated the finding that bilateral STN DBS at clinical settings improves selective inhibition of interference from incorrect action impulses elicited in a Simon task even when patients were completely withdrawn from dopaminergic medication (van Wouwe et al., 2017). The second experiment aimed to modulate two distinct forms of inhibitory control with DBS STN, because the STN has recently become focal point of investigation of circuitries of both stopping and conflict control (Aron et al., 2016). We recruited a new group of patients with bilateral electrodes positioned within both dorsal and ventral subregions of the STN. In separate testing sessions, we applied spatially focused stimulation in bilateral dorsal and ventral STN subregions while PD patients performed the Simon conflict task (n=11) and the Stop task (n=9). Stimulating the dorsal, but not the ventral, STN subregion produced a drastic improvement in selective suppression of action impulses on the conflict task. In contrast, bilateral dorsal and ventral stimulation during performance of the Stop task showed that patients were faster at stopping with ventral versus dorsal stimulation. These findings further confirm the critical involvement of STN circuitry in inhibitory control and suggest that conflict control is linked to more dorsal and stopping control to relatively more ventral STN-cortical circuitry.
Sunday 12.00 - 13.30 Grote zaal Poster #33 Social behavior Sara Fernandez-Cabello
All authors & affiliationMatthias Tholen, Department of Psychology, University of Salzburg Mathias Schurz, Department of Psychology, University of Salzburg
Bottom-up connectivity to rTPJ during False Beliefs
Show/hide abstractTheory of Mind (ToM) is the ability to attribute mental states to other persons. One of the most widely used paradigm to assess ToM is the False Belief (FB) task 1, in which a character holds a false belief about reality. This task consistently elicits activation in a set of brain regions such as the medial prefrontal cortex (MPFC), the bilateral temporoparietal junction (TPJ) and precuneus 2,3. A key region of this network is right TPJ (rTPJ) 4, but its implication has mostly been pictured from fmri activation studies. In this study, we used DCM for fMRI 5 to test the reliability of the effective connectivity results in two different FB datasets. We tested directed connectivity between MPFC, rTPJ and the fusiform gyrus (FFG) within a group DCM framework 6. An open dataset 7 (n=28; OpenfMRI ds000109) and an in-house dataset (n=20) were included. Imaging data were preprocessed with SPM8 (re-alignment, coregistration, slice-timing, normalization, smoothing). A conjunction analysis (p=0.001 unc) was performed to obtain the overlapping in the MPFC (0,56,26) and rTPJ (56,-60,22) from the contrast FB > Control, and the input region FFG (38,-72,-12) from All > baseline. Individual timeseries were extracted at p=0.05 unc from 6mm radius spheres from the local maximas around the group peaks. A Parametric Empirical Bayes framework was used to explore group effects on the coupling parameters. Individual full-model DCMs were two-state, bilinear and deterministic. The FFG was the input region, there were fixed connections between all regions and modulations in all connections between and within regions. In both datasets, we found that bottom-up connectivity between FFG and MPF and FFg and rTPJ was increased under the modulation of False Beliefs (probability > 0.95). These results suggest that bottom-up connectivity is a reliable finding in False Belief DCM studies that include an lower level input region.
Sunday 12.00 - 13.30 Grote zaal Poster #34 Connectivity/networks Elizabeth Johnson
All authors & affiliationElizabeth L Johnson, University of California, Berkeley Robert T Knight, University of California, Berkeley
Dynamic and multiplexed networks for working memory in humans
Show/hide abstractWorking memory (WM) provides the neurobiological infrastructure for human cognition. Distributed network models posit that prefrontal cortex (PFC) supports WM by coordinating control over regions involved in sensory representation (e.g., Sreenivasan et al., 2014). In two studies, multimodal electrophysiology data reveal that PFC control over WM is rhythmic, fundamentally dynamic, and not necessary for all aspects of WM. In a single-trial visual WM task, human subjects encoded two colored shapes in specific spatiotemporal positions in preparation for a subsequent test on either the identity of each shape in the pair, or on the spatial or temporal relationship between the shapes in the pair. Direct brain recordings (n=10) demonstrate that PFC and medial temporal lobe (MTL) theta-band rhythms direct a complex system of higher-frequency neural activity across regions throughout the task, uncovering initial support for bidirectional PFC-MTL interactions underlying WM (Johnson et al. in review). Importantly, the direction of PFC-MTL interactions shifted dynamically as a function of whether subjects were encoding identity, spatial or temporal information (p<0.05, Bonferroni-corrected). Behavioral data from patients with unilateral PFC damage (n=14) demonstrates that PFC integrity is not necessary for all aspects of this WM task, with only 8% accuracy decrease in patients (Johnson et al., in review). In healthy controls (n=20), scalp electroencephalography data showed that delta-theta-band rhythms originating in PFC precess toward posterior cortex, concurrent with alpha-beta-band rhythms precessing in the opposite direction. All PFC effects are diminished bilaterally with unilateral damage (p<0.05, cluster-corrected), revealing an independent posterior neural mechanism supporting WM. These results are incompatible with simple models of PFC-guided network function -- and instead reveal that rapid and distributed parallel processing supports WM.
Sunday 12.00 - 13.30 Grote zaal Poster #35 Connectivity/networks Nadine Dijkstra
All authors & affiliationNadine Dijkstra, 1* Peter Zeidman, 2 Sasha Ondobaka, 2 Marcel A.J. van Gerven, 1 Karl Friston, 2 1. Radboud University, Donders Centre for Brain, Cognition and Behaviour, 6525 EN, Nijmegen, the Netherlands. 2. The Wellcome Trust Centre for Neuroimaging, UCL, 12 Queen Square, London, UK. * Corresponding author.
Directional connectivity during visual perception and imagery: a DCM study.
Show/hide abstractResearch suggests that perception and imagination engage neuronal representations in the same visual areas. However, the underlying mechanisms that differentiate sensory perception from imagination remain unclear. Here, we examined the directed coupling (effective connectivity) between fronto-parietal and visual areas during perception and imagery. We found an increase in bottom-up coupling during perception relative to baseline and an increase in top-down coupling during both perception and imagery, with a much stronger increase during imagery. Modulation of the coupling from frontal to early visual areas was common to both perception and imagery. Furthermore, we showed that the experienced vividness during imagery was selectively associated with increases in top-down connectivity to early visual cortex. These results highlight the importance of top-down processing in internally as well as externally driven visual experience and increase our understanding of the underlying neural mechanisms of perception and imagination.
Sunday 12.00 - 13.30 Grote zaal Poster #36 Clinical Tharaka Dassanayake
All authors & affiliationVajira Weerasinghe, Department of Physiology, Faculty of Medicine, University of Peradeniya, Sri Lanka Indika Gawarammana, Department of Medicine, Faculty of Medicine, University of Peradeniya, Sri Lanka Nicholas Buckley, School of Medical Sciences, The University of Sydney, Australia
Changes in P300 event related potential of organophosphate pesticide poisoning
Show/hide abstractOrganophosphate (OP) compounds inhibit the enzyme acetylcholinesterase, leading to widespread cholinergic hyperactivation in the brain. Studies in humans suggest that OP poisoning may cause neurocognitive deficits that outlast the period of cholinergic crisis [1]. In experimental animals exposed to large doses of OPs, acute effects are followed by cytotoxicity that damage different brain areas including medial temporal lobe which is thought to be among neural generators of P300(P3b) ERP component [2,3]. We hypothesised that acute large-dose OP pesticide poisoning in humans to have similar effects, which then should manifest as changes in P300. We compared the ERPs and behavioural data in an auditory oddball paradigm in 84 patients hospitalized with acute OP poisoning (OP group) with those of a Control Group of 27 with paracetamol overdose. Both groups were tested after being discharged from hospital (on average 12.7 days after exposure). The OP Group made significantly more misses (p=0.01), and had delayed reaction times (p=0.0001) even after adjusting for age, sex, years of education and presence of a major psychiatric illness. An ERP component x site x Group analysis-of-covariance (covariates: age, sex, years of education and presence of a major psychiatric illness) showed a steeper increase in successive ERP latencies (N1, P2, N2 and P300) in the OP Group [component x Group interaction: F(1.59, 315)=6.59, p=0.006]. Subsequent pairwise comparisons with adjustments for the covariates showed delayed mean P300 latencies in all recorded midline sites (FZ: p=0.005, CZ: p=0.027 and PZ: p=0.011) in the OP Group. Although there was a trend for the P300 amplitude to be smaller in the OP Group, the difference was not significant. Our findings indicate acute OP pesticide poisoning is associated with impaired stimulus classification mechanisms. A long-term follow up study that we presently conduct would elucidate the time-course of these deficits.
Sunday 12.00 - 13.30 Grote zaal Poster #37 Memory Ingmar de Vries
All authors & affiliationJoram van Driel, Department of Experimental and Applied Psychology, Faculty of Behavioural and Movement Sciences, Vrije Universiteit Amsterdam Christian N.L. Olivers, Department of Experimental and Applied Psychology, Faculty of Behavioural and Movement Sciences, Vrije Universiteit Amsterdam
EEG correlates of priority switches in working memory-driven visual search
Show/hide abstractVisual search is assumed to be guided by an active visual working memory representation of what we are currently looking for. This attentional template of the search target can be dissociated from accessory memory representations that are only needed prospectively, for a future task, and that until then should be prevented from guiding attention. Little is known about how the brain sequentially prioritizes memory representations for two consecutive tasks. We report evidence showing that such flexible prioritization is captured by posterior alpha-band (8-14 Hz) dynamics. We measured EEG while observers performed two consecutive working memory-guided visual search tasks. Prior to the first search task, a cue instructed observers which item to look for first (current template), and which second (prospective template). During the first delay, leading up to the first of the two searches, posterior alpha power was suppressed contralateral relative to ipsilateral to the memory item. Importantly, these lateralized alpha dynamics were stronger if the lateralized item was the imminent search target (the current template), than when the item was the subsequent search target (the prospective template). In a second study, we more closely investigated the electrophysiological mechanisms underlying the switch in priority, after the first search task and while preparing for the second search task. Importantly, in 40% of trials an auditory cue replaced the first search. Thus, at this point the first template could be dropped, in favor of now prioritizing the second, prospective template. Interestingly, dropping the first template resulted in clear alpha enhancement, while making the former prospective template the current search target resulted in alpha suppression, over regions contralateral to the respective memory item. This switching occurred around ~200 ms after the auditory cue. Together, these findings reveal a crucial role for alpha-band oscillations in the prioritizing and switching of template working memory representations.
Sunday 12.00 - 13.30 Grote zaal Poster #38 Connectivity/networks Xiuyi Wang
All authors & affiliationXiuyi Wang, Department of Psychology and York Neuroimaging Centre, Heslington, University of York, York, YO10 5DD, UK. Jonathan Smallwood, Department of Psychology and York Neuroimaging Centre, Heslington, University of York, York, YO10 5DD, UK. Elizabeth Jefferies, Department of Psychology and York Neuroimaging Centre, Heslington, University of York, York, YO10 5DD, UK.
The structural covariance network of semantic control
Show/hide abstractSemantic control shapes conceptual retrieval so that we are able to choose appropriate knowledge and behaviours for a given task or context. Although semantic control is an important ability in our daily lives, its neural basis is not fully elucidated and little is known about how individual differences in brain structure might relate to the efficiency of relatively automatic and more controlled semantic retrieval. In this study, we addressed this question by examining the structural covariance network of semantic control. 143 healthy participants performed two semantic tasks with high or low semantic control demands (tapping the retrieval of weak and strong associations respectively). First, we associated whole brain cortical thickness with the capacity for semantic control (weak associations controlling for strong associations, age and gender). We found cortical thickness in four regions --- bilateral posterior middle temporal gyrus (pMTG), right anterior temporal lobe (rATL) and left postcentral gyrus --- correlated with performance. We mapped structural covariance networks (SCNs) of each of these regions and found strikingly similar patterns. Bilateral dorsal medial prefrontal cortex, left supramarginal gyrus (SMG), left angular gyrus (AG), left dorsal inferior parietal sulcus (IPS), left posterior superior temporal gyrus (STG) and right pMTG co-varied strongly in their cortical thickness with bilateral pMTG, rATL and left postcentral gyrus. Our findings reveal the structural covariance network of semantic control.
Sunday 12.00 - 13.30 Grote zaal Poster #39 MRI methods ChienMing Lo
All authors & affiliationChienMing Lo, Brain and Consciousness Research Center, Taipei Medical University JianFeng Zhang, Center for Cognition and Brain Disorders, Hangzhou Normal University Niall W Duncan, Brain and Consciousness Research Center, Taipei Medical University
Using macromolecule suppressed MEGA-PREss magnetic resonance spectroscopy does not improve the correlation between GABA and BOLD responses in the occipital cortex
Show/hide abstractA number of studies have described a correlation between the BOLD response in the visual cortex in response to visual stimuli and MRS estimates of GABA concentrations in the same region [1]. More recent work, however, has suggested that such a relationship does not in fact exist [2]. In these studies, GABA concentrations are estimated using the MEGA-PREss protocol. Estimates from the standard version of this protocol include signal from both GABA and from non-specific macromolecules (GABA+MM). An adapted sequence seeks to reduce this macromolecule contamination through suppression of their signal [3]. Comparing GABA estimates made with the two sequences has suggested that there is, however, little correlation between them [4]. The current study thus had two aims. The first of these was to further investigate the relationship between GABA+MM and GABA- estimates in the same subjects. Following this, the second aim was to establish if there is a correlation between MRS estimates of GABA and BOLD responses in the visual cortex and which of the two MEGA-PREss sequences is better correlated, if either. To these ends, we recruited 31 healthy participants who underwent an fMRI visual task and MRS scanning in the occipital and posterior cingulate cortices. GABA+MM and GABA- concentrations were correlated in both regions but this correlation was weak. No relationship was found between either the GABA+MM or the GABA- concentrations and the parameters of the BOLD response (peak, latency, and FWHM) in the occipital cortex. These findings suggest that while GABA+MM and GABA- estimates of GABA concentrations are related to each other this connection is weak. They also support prior work pointing to there being no correlation between MRS measures of GABA and BOLD responses in the visual cortex and suggest that using a GABA- sequence will not improve this.
Sunday 12.00 - 13.30 Grote zaal Poster #40 MRI methods Hunar Abdulrahman
All authors & affiliationHunar Abdurlrahman, University of Cambridge, Cambridge, UK. Richard N. Henson, MRC- Cognition & Brain Science Unit, Cambridge, UK.
Decoding Novel from Repeated Faces using Multivariate Pattern Analyses
Show/hide abstractFMRI BOLD activity reduces after repeated exposure to stimuli (1). So far, the majority of the fMRI studies measured the change in the mean voxel activity across the repetitions (2). However, a non-uniform reduction in the BOLD activity is likely to change the spatial pattern across the voxels as well. Hence, it is important to investigate the decodability of this change in the voxel pattern. We can do that by removing the mean activity across the voxels for each trial and use a linear classifier to classify Novel (N) and Repeated (R) faces based on the difference between their pattern information alone. A successful classification in that case is indicative of a non-uniform reduction in the BOLD activity that extends beyond the simple assumptions of univariate repetition effects. In this study, we used a custom Multi-Voxel Searchlight Pattern Analyses to compare the classifiability of N vs R in a publicly available dataset (3) based on: 1) the overall mean activity difference between N & R and 2) the voxel pattern difference between N & R. We found that N & R are decodable above the chance level in the fusiform face area in both cases (p<0.001). This is true whether N is previously memorised (famous face) or a completely new face (unfamiliar face). This finding provides the evidence that a repeated exposure to a stimulus changes its subsequent distributed pattern activity in fMRI. This result is also important to consider in fMRI designs that have repeated trials of the same face class. We also stress that any model of repetition effects should be able to explain such drastic change in the distributed voxel pattern, which extends beyond the change in the overall mean activity.
Sunday 12.00 - 13.30 Grote zaal Poster #41 Memory Yanwei Shi
All authors & affiliationYanwei Shi, Department of Psychology and Behavioral Sciences, Zhejiang University, Hangzhou, China Lumin Feng, Department of Psychology and Behavioral Sciences, Zhejiang University, Hangzhou, China Hui Chen, Department of Psychology and Behavioral Sciences, Zhejiang University, Hangzhou, China Mowei Shen, Department of Psychology and Behavioral Sciences, Zhejiang University, Hangzhou, China Zaifeng Gao, Department of Psychology and Behavioral Sciences, Zhejiang University, Hangzhou, China*
The Working Memory Mechanism in Processing Biological Motion Retrieved from Long-term Memory
Show/hide abstractAlthough we have known much about working memory (WM) and long-term memory (LTM), the interaction between the two memory system remains largely unclear. In the current study, we investigated how WM addresses the information retrieved from LTM in a change-detection task. We took biological motion (BM) as the stimuli of interest, since it is the most common stimuli in our daily life and plays a critical role in our social cognition. Moreover, it has been revealed that WM recruits mirror neuron system (MNS) to process BM extracted from perception. We examined whether the MNS was activated during the encoding, maintenance, and comparison stages of WM, as well as whether the WM processing was modulated by memory load. Considering MNS can be manifested by 8-12 Hz Mu-suppression in the central cortex, and EEG is good at distinguishing WM stages, we recorded EEG during the experiment. The participants were first trained to form long-term links between specific BM scenarios (containing 1 or 3 stimuli) and digits (e.g., a walking BM was always linked to digit 5), then they finished a change-detection task: A digit was presented and participants had to recall the corresponding BM set, after 2000 ms a visual BM was probed. We found that Mu was not suppressed at the encoding or maintenance phase of WM, but was suppressed at the comparison stage. Moreover, the memory load did not modulate the amplitude of occipital N1 or P1 in the encoding stage, or the frontal late positive component in the maintenance stage. However, the amplitude of P3 in the comparison stage was significantly reduced under high BM load. These results suggest that WM treats the retrieved BM from LTM as a non-biological unit in the encoding and maintenance stages, which unfolds in the comparison stage when necessary.
Sunday 12.00 - 13.30 Grote zaal Poster #42 Memory Paul Rignanese
All authors & affiliationPaul Rignanese, VU University Amsterdam, the Netherlands Martijn Meeter, Department of Educational Sciences, VU University Amsterdam, the Netherlands Marlieke van Kesteren, Department of Educational Neuroscience, VU University Amsterdam, the Netherlands
Integrating memories: How congruency affects reinstatement of old and integration with new memories
Show/hide abstractTo successfully construct knowledge, we integrate separately learned instances by inferring their association. Neural reactivation - or reinstatement - of prior knowledge during new learning is suggested to aid this integration of old and new memories, but it is unknown exactly how this reinstatement best enables effective knowledge building. Here, we evaluated whether congruency between two separately learned, but overlapping associations enhances memory integration through reinstatement. We performed an AB-AC inference paradigm where student participants learned overlapping associations (A=pseudoword, B=scene, C=object) in the MRI-scanner. Half of the BC-combinations were congruent with prior knowledge (i.e. bathroom-rubber duck) and half were not (i.e. bathroom-hammer), and this was counterbalanced over participants. After learning in the scanner and a week later, participants were tested on C item recognition and BC associative memory. We expect enhancement of B-reinstatement in the Parahippocampal Place Area (PPA) and the hippocampus during AC-learning through congruency with prior knowledge. We also expect the extent of this enhancement to relate to subsequent associative memory. This will be examined using a MultiVoxel Pattern Analysis (MVPA) classifier, trained on independent localizer data to detect distributed patterns related to visual processing of scenes. Furthermore, integration success is predicted to relate to interactions between the hippocampus and the medial prefrontal cortex (mPFC), as these regions are implicated to be involved in integrating new with old information. Additionally, we will ask participants to indicate the subjective strength of their reactivation of B after learning AC. Based on a previous pilot, we expect this subjective reactivation score to positively relate to subsequent memory and B-reinstatement as well. These results will further understanding of how old and new memories are integrated in the brain and which factors can modulate this process. Findings will additionally be of interest for educational settings where successfully integrating knowledge is imperative for success.
Sunday 12.00 - 13.30 Grote zaal Poster #43 Memory Gerd Waldhauser
All authors & affiliationGerd T. Waldhauser, Ruhr-University Bochum Malte Kobelt, Ruhr-University Bochum Nikolai Axmacher, Ruhr-University Bochum
The neural signature of involuntary memory intrusions
Show/hide abstractMemory intrusions are a hallmark symptom of various psychiatric conditions, most notably posttraumatic stress disorder. Employing electroencephalography (EEG), the present study set out to investigate what distinguishes involuntary intrusions from intentional retrieval of desired memory representations. We contrasted neural activity between two subsequent experimental phases, an intentional retrieval and a visual discrimination task. Both tasks utilized the same stimulus material consisting of pairs of centrally presented abstract cues and pictures of everyday objects presented in the left or right visual hemifield as targets. During intentional retrieval, participants engaged in item recognition on the cues and were instructed to remember the targets and their respective hemifield of presentation. During visual discrimination, cues were presented in an identical manner, but participants now completed a speeded visual discrimination task, subsequently indicating involuntary intrusion of the target. Voluntary retrieval was characterized by an increase of brain oscillatory power at theta (5-9 Hz) frequency at parietal and frontal electrodes in a time window from 700-1200 ms, consistent with recollection and post-retrieval monitoring. In contrast, involuntary intrusions led to prolonged reaction times in the discrimination task and an early (200-700 ms) frontocentral theta power increase, supposedly signaling interference of unwanted memory representations. These results suggest that involuntary memory intrusions trigger an early interference signal but are not followed by indices of recollection and post-retrieval processing observed in voluntary retrieval. Potentially, these insights could lead to the development of new therapeutic techniques such as the modulation of theta power by means of brain stimulation in service of controlling involuntary memories.
Sunday 12.00 - 13.30 Grote zaal Poster #44 Attention Pauldy Otermans
All authors & affiliationPauldy C.J. Otermans, Brunel University London Andre J. Szameitat, Brunel University London
The central attentional bottleneck demands executive functions of working memory
Show/hide abstractThis functional magnetic resonance imaging (fMRI) research studies the effect of multitasking on working memory (WM). A processing bottleneck causes interference when performing dual-tasks due to its capability of processing only one task at a time. Executive functions are needed to resolve this interference and it has been suggested that these executive functions are used by both dual-task performance and WM tasks. While previous neuroscientific research indeed showed that dual-tasks as well as WM tasks rely on lateral-prefrontal cortices (LPFC), it remains unknown whether both tasks activate the same areas or different sub-areas of the LPFC. Therefore, this study investigated how the neuroanatomical correlates of both dual-task and WM compare to each other. To study this, a complex WM span task was created by combining the paradigm of the psychological refractory period (PRP) with a WM task. Participants were presented with a sequence of letters to remember, followed by a task block in which they had to perform either a single task (visual or auditory) or a dual-task, and finally were asked to recall the letters. Behavioural results showed that WM performance, i.e. the amount of letters recalled in the correct order, decreases when performing a dual-task as compared to a single task. This indicates that dual-task performance relies at least partially on the same cognitive processes as WM. The brain activation for the PRP and WM tasks show considerable overlap as well as some differentiation. Both tasks activated, among other areas, the inferior frontal junction. With respect to differences, the PRP task activated more the inferior middle frontal gyrus (MFG) whilst the WM component activated more the superior MFG. Thus, results support the assumption that PRP dual-tasks demand the executive functions of working memory.
Sunday 12.00 - 13.30 Grote zaal Poster #45 Attention Angela Renton
All authors & affiliationDavid R. Painter, School of Psychology, The University of Queensland, St Lucia 4072, Australia Jason B. Mattingley, Queensland Brain Institute & School of Psychology, The University of Queensland, St Lucia 4072, Australia
Differential mobilization of visual selective attention during dynamic approach and avoidance behavior
Show/hide abstractActive visual exploration in a complex world requires selective attention. Attention acts to prioritise a subset of stimulus inputs and to filter out or suppress others. In typical attention experiments, stimuli are designated as task-relevant (targets) or task-irrelevant (distractors). Findings from a range of brain imaging studies have consistently revealed enhancement of neural responses to task-relevant targets, and suppression of neural responses to task-irrelevant distractors. In naturalistic settings, however, different visual stimuli might be equally task-relevant, but might be prioritized in terms of the distinct behaviors they elicit. Here we examined how attentional selection is influenced by prioritization of moving subsets of visual stimuli cued to elicit the distinct behavioral goals of approach and avoidance. In two experiments (N=24 participants each) electroencephalography was used to measure neural activity while participants maneuvered a cursor amongst large sets of moving objects. Participants were rewarded for approaching objects in one color and punished for failing to avoid objects in a different color, with a third set of colored objects serving as task-irrelevant distractors. In Experiment 1, participants undertook the approach and avoidance behavior concurrently, while in Experiment 2 approach and avoidance behavior was undertaken separately in randomly interleaved blocks of trials. Approached, avoided and task-irrelevant objects were tagged at unique flicker frequencies, producing steady-state visual evoked potentials (ssVEPs). In both experiments, frequency-based analyses revealed that ssVEP amplitudes were higher for approached than task-irrelevant objects, and lower for avoided than task-irrelevant objects. In Experiment 2, the total ssVEP amplitude across all frequency-tagged stimuli was equivalent across the approach and avoidance behaviors. However, the allocation of attention across stimuli, indexed by ssVEP amplitude, differed between tasks. These results suggest that visual attention not only depends on task-relevance, as predicted by numerous models of attention, but also on the specific behavioral goals associated with task-relevant objects.
Sunday 12.00 - 13.30 Grote zaal Poster #46 Developmental Janelle Cleary
All authors & affiliationJanelle Cleary, University of New England (Australia) Graham Jamieson, Ph.D., University of New England (Australia) Ian Evans, University of New England (Australia)
Infant EEG discriminates mothers' versus strangers' fearful faces at 10-12Hz in left BA10 and BA11
Show/hide abstractResearch investigating the impact of extinction based infant sleep interventions on the development of affective self-regulation is yet to consider cortical activity in terms of either electroencephalographic (EEG) spectral data or event-related potentials (ERPs). This study measured baseline responses to affective face stimuli (happy and fearful) of mothers and a stranger in infants at three months of age. Mobile EEG testing took place in the homes of 53 infants who were shown four different faces (mother happy, mother fearful, stranger happy and stranger fearful) across 80 trials (4 x 20 each). Trials consisted of a brief two second video (no human features) to gain attention, followed by a fixation cross then one of the four face stimuli with jittered time-lags between all stimuli. So far spectral data has been processed using FASTER ICA and source activity estimated using eLORETA for the EEG recordings from 15 of these infants. EEG in the frequency band 10 -- 12Hz was stronger in response to the mothers' fearful face compared to the stranger's fearful face with peak (p = .008) in left BA10 and BA11. This effect may index the impact of negative affect in attachment figures on the mobilisation of executive control at the earliest stages of infant self-regulation development. Further interpretation will be closely tied to the functional significance of frequency bands within infant EEG. Longitudinal data from these infants (which also includes the Still Face Paradigm) will also be analysed at six and twelve months of age to track the development and efficiency of neural networks of affective self-regulation.
Sunday 12.00 - 13.30 Grote zaal Poster #47 Language Basil Preisig
All authors & affiliationEggenberger Noëmi, Perception and Eye Movement Laboratory, Departments of Neurology and Clinical Research, Inselspital, Freiburgstrasse 46, CH-3010 Bern, Switzerland Cazzoli Dario, Gerontechnology and Rehabilitation Group, ARTORG Center for Biomedical Engineering Research, University of Bern, Murtenstrasse 50, CH-3008 Bern, Switzerland Nyffeler Thomas, Center of Neurology and Neurorehabilitation, Luzerner Kantonsspital, CH-6000 Luzern, Switzerland Gutbrod Klemens, University Neurorehabilitation Clinics, Department of Neurology, Inselspital, and University of Bern, Anna-Seiler-Haus, Freiburgstrasse 41c, CH-3010 Bern, Switzerland Annoni Jean-Marie, Neurology Unit, Laboratory for Cognitive and Neurological Sciences, Department of Medicine, Faculty of Science, University of Fribourg, Chemin du Musee 5, CH-1700 Fribourg, Switzerland Meichtry Jurka, University Neurorehabilitation Clinics, Department of Neurology, Inselspital, and University of Bern, Anna-Seiler-Haus, Freiburgstrasse 41c, CH-3010 Bern, Switzerland Nef Tobias, Gerontechnology and Rehabilitation Group, ARTORG Center for Biomedical Engineering Research, University of Bern, Murtenstrasse 50, CH-3008 Bern, Switzerland Mueri Rene, Perception and Eye Movement Laboratory, Departments of Neurology and Clinical Research, Inselspital, Freiburgstrasse 46, CH-3010 Bern, Switzerland
Nonverbal communication in aphasia: Gesture perception and production during face-to-face interaction
Show/hide abstractThe role of nonverbal communication in patients with aphasia is not yet fully understood. This study investigated how aphasic patients perceive and produce co-speech gestures during face-to-face interaction, and whether distinct brain lesions predict the frequency of spontaneous gesturing. For this purpose, we recorded samples of conversation in patients with aphasia and healthy participants. Gesture perception was assessed by means of a head-mounted eye-tracking system and the produced co-speech gestures were coded according to a linguistic classification system. We found that gestures, which convey meaning (i.e., communicative gestures), attract more visual attention in patients than meaningless hand movements. Patients also showed a higher probability to fixate gestures made by their interlocutor than healthy participants. This implies that patients with aphasia benefit from the multimodal information provided by co-speech gestures by fixating communicative gestures more frequently. On the level of gesture production, we found that patients with damage to the anterior part of the arcuate fasciculus showed a higher frequency of communicative gestures. This area lies in close vicinity to the premotor cortex and is considered to be important for speech production. This suggests that the use of communicative gestures depends on the integrity of patients' speech production abilities. Our findings provide novel evidence for the notion of a compensatory use of co-speech gestures in aphasia, contradicting the assumption of a parallel breakdown of speaking and gesturing.
Sunday 12.00 - 13.30 Grote zaal Poster #48 Attention Tzu-Yu Hsu
All authors & affiliationTzu-Yu Hsu, Research Center for Brain and Consciousness, Taipei Medical University, Taiwan; Research Center for Brain and Consciousness, Shuang-Ho Hospital, Taiwan; Graduate institution of Health and Biotechnology Law, Taipei Medical University, Taiwan
Sustained or transient control process in congruency sequence effect: An EEG study
Show/hide abstractReduced reaction times to conflict situations are observed when the preceding event was also an incongruent compared to a congruent situation. Such a reduction, known as congruency sequence effect (CSE), is considered to reflect a control process that monitors conflict situations and effectively adjusts current behavior to avoid interference from distractors. Some evidence has suggested that effects from this process can persist for some time after a conflict trial has been presented, influencing the response to subsequent trials. This remains to be tested in humans, however. To approach this question, we adopted a prime-probe arrow task to induce response compatibility. Meanwhile, EEG was recorded to measure the mechanism underlying the CSE. Short (133 ms) or long (500 ms) inter-stimulus intervals (ISI) between prime and probe were manipulated to investigate whether the duration of the ISI affects CSE. Our behavioral results showed a typical compatibility effect in the current trials and a significant CSE irrespective of whether the ISIs were short or long. In line with the behavioral results, the amplitude of the N2 reflected the conflict level on current trials. Analyzing trials according to the content of the previous one, we saw an increased probe N2 where the prior trial was incongruent and where the ISI was long. Interestingly, we did not, however, see any difference during the ISI period itself. The preliminary results suggest that whilst there may be a persistent CSE, it may require a quiescence period before it affects subsequent stimulus responses.
Sunday 12.00 - 13.30 Grote zaal Poster #49 Cognitive modeling Dilene van Campen
All authors & affiliationA.D. van Campen, Department of Psychology, University of Amsterdam, Amsterdam, The Netherlands L. van Maanen, Department of Psychology, University of Amsterdam, Amsterdam, The Netherlands B.U. Forstmann, Department of Psychology, University of Amsterdam, Amsterdam, The Netherlands K.R. Ridderinkhof, Department of Psychology, University of Amsterdam, Amsterdam, The Netherlands R. Ratcliff, Department of Psychology, The Ohio State University, Columbus, Ohio, USA
Evidence Accumulation Modeling of Benchmark Phenomena in the Simon Task
Show/hide abstractOver the last half-century, enormous effort has been put forward to explain behavioral interference effects emerging from conflict paradigms such as the Simon task. Computational modeling has proven to be useful in providing insights into the mechanisms underlying conflict effects. However, quantitative modeling in the Simon task domain has been sparse. Therefore, the aim of the present study is to use formal quantitative modeling to capture three well described representative behavioral benchmark phenomena of the Simon task. Conflict in the Simon task arises when the stimulus occurs on one side and the required response is on the other side (the stimulus location and the response are incongruent). This congruency effect is referred to as the Simon effect. In addition to this general effect, the Simon task is renowned for producing relatively pronounced and robust benchmark phenomena: congruence sequence effects (CSE), response capture (fast incongruent errors), and fade-out (reduction of interference effects for slower responses). Based on experimental evidence, we examine which parameter variations of two quantitative models (the Diffusion model and the Linear Ballistic Accumulator model) account for these benchmark phenomena. Response capture and CSE are well accommodated by bias toward congruent responses or toward repetition of congruence conditions, respectively; a combination of bias toward congruent responses and a compensatory increase in drift rate for incongruent responses jointly account for fade-out. The revealed model outcomes provide new insights and increase the understanding of the underlying mechanisms involved the Simon task.
Sunday 12.00 - 13.30 Grote zaal Poster #50 Cognitive modeling Benjamin Gagl
All authors & affiliationBenjamin Gagl, Department of Psychology, Goethe University Frankfurt and Center for Individual Development and Adaptive Education of Children at Risk (IDeA), Frankfurt am Main, Germany Jona Sassenhagen, Department of Psychology, Goethe University Frankfurt Sophia Haan, Department of Psychology, Goethe University Frankfurt Fabio Richlan, Centre for Cognitive Neuroscience, University of Salzburg, Hellbrunnerstrasse 34, 5020 Salzburg, Austria. Christian J. Fiebach, Department of Psychology, Goethe University Frankfurt and Center for Individual Development and Adaptive Education of Children at Risk (IDeA), Frankfurt am Main, Germany
Predictive processing is key for reading: Evaluating a visual information-optimization hypothesis in visual word recognition.
Show/hide abstractHow do we process visual information in reading? Here we propose and implement a visual optimization algorithm that "explains away" redundant visual information as described by predictive coding. The result of this computation is a representation based on unpredicted visual features (prediction error; PE), which allows efficient higher-level processing since the PE-representation includes only word specific visual information. This hypothesis was evaluated against current visual word recognition concepts and by multiple behavioral (lexical decision tasks; N=78/54) and neuronal datasets (fMRI/EEG data from silent reading; N=39/31). Interestingly, the PE reflects orthographic familiarity, in a pure form, since PE is associated to orthographic word-characteristics (e.g. Orthographic neighborhood) without a relation to higher-level concepts (e.g. Word frequency). In the behavioral data we found a PE effect that was reflected in an increase of reaction times with PE for words and a decrease for non-words. We propose that this interaction reflects an information accumulation process related to lexical access. In the neuronal data we found that bilateral-occipital brain regions (fMRI) and the N170 component (EEG) were related to the PE. Furthermore, left/medial-frontal regions and the N400 component showed an interaction of PE with word lexicality as also observed in behavior. Finally, we investigated the PE-based visual optimization algorithm in noisy environments. Behavioral data showed the PE by lexicality interaction only for correct responses (N=54) and the PE effect on the N170 component of the EEG could be replicated (N=30). Visual noise increased PE, leading to error prone behavior. Also, the PE by lexicality interaction could not be found in the N400 component under noise conditions. On the basis of these results, we conclude that in reading, visual information is optimized by a prediction based on redundant visual information from words providing an optimal representation for efficient meaning extraction in reading.
Sunday 12.00 - 13.30 Grote zaal Poster #51 Decision-making Emmet McNickle
All authors & affiliationEmmet McNickle, Trinity College Dublin Ger Loughnane, Trinity College Dublin David McGovern, Trinity College Dublin Simon Kelly, University College Dublin Redmond O'Connell, Trinity College Dublin
Motor and evidence accumulation signals trace stimulus-response incongruity in the Simon Effect
Show/hide abstractThe ‘Simon Effect' is characterised by slower reaction times when a stimulus is presented on the opposite side of the visual field to the responding effector. This behavioural effect is commonly assumed to arise from conflict at the motor level but the possibility that other processing levels in the sensorimotor hierarchy may also play a role has not been examined. To this end, we used EEG to simultaneously track indices of evidence accumulation and motor preparation during performance of a motion discrimination task in which participants indicated the direction of coherent motion in one of two bilateral patches. Participants detected coherent motion at near ceiling accuracy; however ‘congruent' trials (where coherent motion side matched correct response side) were reported significantly earlier than ‘incongruent' trials. The lateralised readiness potential (LRP) exhibited an early lateralisation reflecting the location of coherent motion, followed by a lateralisation that reflected the motion direction. In addition, the build-up of the Centro-Parietal Positivity evidence accumulation signal was faster on congruent than incongruent trials; an effect that occurred significantly later than the initial LRP deviation. These results indicate that the Simon Effect does not solely arise from conflict at the motor level but also from subtler effects on the perceptual decision making process itself.
Sunday 12.00 - 13.30 Grote zaal Poster #52 Sensory processing Genevieve Quek
All authors & affiliationGenevieve L. Quek, University of Louvain Bruno Rossion, University of Louvain
Category-selective brain processes elicited in dynamic visual streams are immune to temporal predictability
Show/hide abstractHuman brains give meaning to the extremely rich and dynamic sensory world surrounding them by grouping perceived entities into classes, a process called perceptual categorisation. The combination of visual periodicity and neurophysiological recordings offers an objective and efficient way to quantify perceptual categorisation in the visual domain. When human observers view category exemplars (e.g., faces) at strict periodic intervals within a rapid stream of items that do not belong to that category (e.g., objects), neural responses reflecting visual categorisation of these exemplars are tagged with their specified frequency, measurable both on the scalp (Rossion et al., 2015) and inside the brain (Jonas et al., 2016). However, since periodicity entails predictability, an outstanding issue is whether category-selective signals yielded by periodicity are generated partly by temporal expectation. Here we present a stringent test of this hypothesis. In Experiment 1, we demonstrate that EEG responses to highly variable face images embedded in a fast (12 Hz) visual stream of non-face objects are immune to temporal predictability. In Experiment 2, we assess potential responses to rare (10%) omissions of periodic faces in the same fast visual stream, and show there is no identifiable category-selective neural response to violations of temporal predictability even after a long period of entrainment. We conclude that temporal predictability does not generate or modulate long duration face-selective processes elicited in dynamic visual streams. These observations undermine a predictive coding framework interpretation of category-change detection in the human brain and have important implications for understanding human perceptual categorisation in dynamic visual environments.
Sunday 12.00 - 13.30 Grote zaal Poster #53 Memory Gizem Göktepe
All authors & affiliationIgor Mapelli, METU, Department of Health Informatics Tolga Esat Özkurt, METU, Department of Health Informatics
Would somatosensory stimuli increase LTM performance? A preliminary EEG study
Show/hide abstractThe psychotherapy treatment named “eye movement desensitization and reprocessing” (EMDR)[1] suggests that bilateral stimuli may enhance eliciting episodic memories. In this preliminary study, our aim was to investigate brain oscillatory signatures of bilateral somatosensory stimuli during a long term memory task. We designed and implemented a face recognition paradigm presenting 240 human face images in the encoding session: (i) 1 s of fixation cross, (ii) 4 s of face image presentation, (iii) 4 s of cue question, (iv) 4 s of math quiz to reduce recency effect and increase cognitive load. After an hour break, the recognition session followed. Bilateral tactile (slow and fast) stimuli to hands were sent, while subjects were questioned on whether they had seen the face images (336 faces, 126 new faces) before. A control condition without any somatosensory stimuli was also accomplished. EEG data of two subjects were collected solely during the recognition session via a 32-channel BrainAmp system with sampling frequency of 1000 Hz. EEG data were bandpass filtered (0.2-100 Hz, 4th order Butterworth filter) and demeaned. Independent component analysis (fast ICA) was applied to identify and remove eye blink artifacts. Time-frequency analysis was conducted using Fourier basis with a fixed Hanning window length of 0.5 s (shifts of 50 ms) in the frequency range of 4-30 Hz with 2 Hz increments. Baseline correction was realized by subtracting power in [-1 -0.5] s. Preliminary results indicated that there was increase in face recognition performance with tactile stimulus compared to none. Alpha power activity before and after the face image onset modified conspicuously in frontal and occipital channels. Tactile stimulus induced a stronger alpha synchronization before the image onset [-0.5 0] s. Moreover, stronger desynchronization with the stimulus was observed during the face image presentation [0s–4s], most prominently on the right occipital regions.
Sunday 12.00 - 13.30 Grote zaal Poster #54 Memory Frank van Schalkwijk
All authors & affiliationFrank J. van Schalkwijk, Laboratory for Sleep, Cognition and Consciousness Research, Centre for Cognitive Neuroscience [CCNS], University of Salzburg, Salzburg, Austria. Cornelia Sauter, Department of Neurology, Medical University Vienna, A-1090 Vienna, Austria; Competence Center of Sleep Medicine, Charité – University Medicine, Berlin, Germany. Kerstin Hoedlmoser, Laboratory for Sleep, Cognition and Consciousness Research, Centre for Cognitive Neuroscience [CCNS], University of Salzburg, Salzburg, Austria. Dominik P.J. Heib, Laboratory for Sleep, Cognition and Consciousness Research, Centre for Cognitive Neuroscience [CCNS], University of Salzburg, Salzburg, Austria. Theresa Hauser, Laboratory for Sleep, Cognition and Consciousness Research, Centre for Cognitive Neuroscience [CCNS], University of Salzburg, Salzburg, Austria. Gerhard Klösch, Department of Neurology, Medical University Vienna, A-1090 Vienna, Austria. Doris Moser, Department of Neurology, Medical University Vienna, A-1090 Vienna, Austria. Georg Gruber, Dept. of Psychiatry and Psychotherapy, Medical University of Vienna, Vienna, Austria. Peter Anderer, Dept. of Psychiatry and Psychotherapy, Medical University of Vienna, Vienna, Austria. Bernd Saletu, Department of Psychiatry, Medical University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria. Silvia Parapatics , Department of Psychiatry, Medical University of Vienna, Waehringer Guertel 18-20, A-1090 Vienna, Austria. Birgit Weber-Bischof, Institute of Psychology, University of Vienna, A-1010 Vienna, Austria. Herbert Bauer, Institute of Psychology, University of Vienna, A-1010 Vienna, Austria. Florian Fischmeister, Institute of Psychology, University of Vienna, A-1010 Vienna, Austria. Josef Zeitlhofer, Department of Neurology, Medical University Vienna, A-1090 Vienna, Austria. Manuel Schabus, Laboratory for Sleep, Cognition and Consciousness Research, Centre for Cognitive Neuroscience [CCNS], University of Salzburg, Salzburg, Austria.
Short and long-term effects of full-night sleep on procedural memory performance.
Show/hide abstractStudies on sleep and memory primarily focus on overnight performance changes; leaving potential long-term effects out of consideration. Furthermore, associations between performance changes and sleep architecture benefit from more in-depth analyses of sleep characteristics such as sleep spindles and slow oscillations. The present study investigated the effects of full-night sleep on procedural memory performance after a full-night of sleep and a one-week follow-up. Additionally, the functional significance of sleep spindles and slow oscillations for the consolidation of procedural memory will be investigated. Participants (N = 23, Mage = 23.17 ± 2.66 years, range = 20-30 years; 48% male) were trained on a procedural mirror-tracing task during counterbalanced learning (mirrored vision) and control (normal vision) conditions. Participants traced 12 figures as quickly and accurately as possible within a 90s timeframe during the evening training session. Performance was evaluated for speed and accuracy directly following training, as well as after a full-night of sleep and a one-week follow-up. Twenty-one channel electroencephalography was recorded during task execution, with polysomnography recorded during the night. Performance changes, sleep architecture, as well as sleep spindle and slow oscillation activity were analyzed for both conditions. Preliminary findings indicate no performance changes for the control condition. For the learning condition, speed showed no overnight improvement but rather a significant improvement for the subsequent follow-up. By contrast, accuracy significantly decreased following full-night sleep and stabilized for the subsequent follow-up. Further associations between performance changes, sleep spindles, and slow oscillations will be investigated. The present study shows that performance on a procedural memory task can be improved by full-night sleep, but that additional performance changes can occur after a longer time period.
Sunday 12.00 - 13.30 Grote zaal Poster #55 Language Charlotte Vogt
All authors & affiliationCharlotte Vogt, Brain Imaging Center and Department of Neurology, Goethe University Frankfurt Christiane Arnold, Brain Imaging Center and Department of Neurology, Goethe University Frankfurt Suzana Gispert-Sánchez, Molecular Neurogenetics, Goethe University Frankfurt Christian A. Kell, Brain Imaging Center and Department of Neurology, Goethe University Frankfurt
Oxytocin modulates brain activity associated with overt speaking
Show/hide abstractSpeech is a means of social communication. It has been proposed that many socio-affective behaviors, such as speech, are modulated by the neuropeptide oxytocin (Tops, van IJzendoorn et al., 2011; Heinrichs, von Dawans et al., 2009). Yet it remains unclear to what extent polymorphisms of the oxytocin receptor gene (OXTR) modulate brain functions underlying the production of speech. To determine the physiological contribution of the well-characterized rs53576 OXTR polymorphism (Lucht, Barnow et al., 2009) during speech production, we performed a cue-target fMRI reading paradigm in 69 healthy participants. The paradigm consisted of a covert reading condition, as well as two overt reading conditions during which participants read a sentence with either neutral or happy intonation. For group analyses, participants were divided into two groups, AA/AG and GG, based on the genotype of their rs53576 OXTR polymorphism (Riem, Madelon et al., 2011). Previous findings have indicated that carriers of the GG genotype show increased efficiency in processing social auditory cues (Tops, van IJzendoorn et al., 2011). The AA/AG group activated the left dorsal premotor cortex and dorsal inferior frontal gyrus more strongly than the GG group (p=0,05; FWE corrected). The group difference in the left dorsal speech processing stream was especially pronounced during the generation of happy intonation. Processing in the dorsal speech stream relates to auditory-motor mapping (Hickok and Poeppe,l 2007) and usually increases when speech processing is challenging (Murakami, Kell et al., 2015). However, there were no behavioral differences during speech production between the two groups. We concluded that participants with more efficient oxytocin signaling process speech in the left dorsal speech stream more efficiently than participants with less modulation of auditory processing by oxytocin. Our results thus indicate a modulation of auditory-motor mapping during speech production by oxytocin, primarily when social information is conveyed through speech melody.
Sunday 12.00 - 13.30 Grote zaal Poster #56 Language Tatiana Goregliad Fjaellingsdal
All authors & affiliationTatiana Goregliad Fjaellingsdal, Department of Psychology, University of Oldenburg, Oldenburg Esther Ruigendijk, Cluster of Excellence Hearing4all, University of Oldenburg, Oldenburg, Germany & Department of Dutch, University of Oldenburg, Oldenburg, Germany Stefan Scherbaum, Department of Psychology, Technische Universität Dresden, Dresden, Germany Martin G. Bleichner, Department of Psychology, University of Oldenburg, Oldenburg, Germany & Cluster of Excellence Hearing4all, University of Oldenburg, Oldenburg, Germany
Measuring the N400 effect under conversation-like circumstances using wireless EEG
Show/hide abstractStudying the neurophysiological underpinnings of conversations may provide a better understanding of language processes. However, it remains a challenge to create experiments that retain sufficient constraints to measure neurophysiological data while preserving conversational elements to a large extend [1]. The question is further whether established event-related potentials (ERPs), for example the N400 [2], can be recorded in paradigms with a higher ecological validity, which are closer to natural conversations. Employing wireless electroencephalography (EEG), more naturalistic paradigms may be used to study language under greater realistic circumstances and with fewer constraints than posed by classical lab based EEG. Our focus here lies on the effects of the person's involvement in the conversation (active speaker or passive listener) and the influence of speaker switches (turn take) on ERPs. In the present experiment, we therefore studied the N400 effect in a conversation-like scenario using wireless EEG [e.g., 3]. We presented sentences with a congruent (75%) or incongruent (25%) sentence ending (final word) in two conditions. Participants had to either read aloud (active condition) or to listen (passive condition) to the first part of the sentence. In both conditions, the final word was presented auditorily by another speaker, simulating a turn take. Results show, as expected, a significant main N400 effect, i.e., a larger N400 for semantically incongruent than congruent final words. This effect was most pronounced over centro-parietal electrodes. The N400 effect was present independent of whether the participants actively read or passively listened to the sentence. There was no statistically significant difference between conditions on the N400. Our results are a first indication that language-related ERPs can be studied in a conversation-like paradigm using wireless EEG. In the follow-up studies, we loosen further constraints of this still controlled paradigm with the aim to study language processes in freely interacting partners.
Sunday 12.00 - 13.30 Grote zaal Poster #57 Sensory processing Sara Jahfari
All authors & affiliationSara Jahfari, Department of Experimental and Applied Psychology, Vrije Universiteit van Amsterdam, The Netherlands Tomas Knapen, Department of Experimental and Applied Psychology, Vrije Universiteit van Amsterdam, The Netherlands Jan Theeuwes, Department of Experimental and Applied Psychology, Vrije Universiteit van Amsterdam, The Netherlands
Perceptual activity relates to the value of stimuli at the time of choice, and (signed) prediction errors at the time of feedback.
Show/hide abstractReinforcement learning can bias decision-making towards the option with the highest expected outcome. This effect is well described by cognitive learning theories, and associated with the constant tracking of stimulus-values in prefrontal cortex, as well as the careful evaluation of choice outcomes within the striatum. Nevertheless, at the onset of each decision we often begin with the processing of incoming sensory information, and to date, far less is known about the interplay between learning and perception. In this study, fMRI was recorded from a probabilistic learning experiment that used facial stimuli to examine how the activity pattern in perceptual regions relates to the computation of stimulus values, or the evaluation of choice outcomes. A hierarchical Bayesian reinforcement model was used to estimate trial-by-trial stimulus-beliefs, and the evaluation of choice outcomes (reward prediction error; RPE). In addition, the fusiform face area (FFA) was segmented for each participant using a separately recorded perceptual mapper. When participants were presented with a choice between two faces (the decision phase), the BOLD response of the FFA was closely related to the estimated trial-by-trial values of the two faces presented. However, with the presentation of feedback, results showed a close relationship between the FFA BOLD response and the estimated trial-by-trial reward prediction errors (RPE). Notably, both the signed and unsigned RPE regressors correlated with the estimated FFA BOLD response, but the correlation occurred earlier in time for the unsigned RPE regressor. These results provide novel insight into the relationship between perception and learning, and demonstrate how the tracking of trial-by-trial value beliefs (at the moment of choice) and the evaluation of feedback (during the presentation of outcomes) can already be detected in the activity pattern of perceptual regions.
Sunday 12.00 - 13.30 Grote zaal Poster #58 Memory Sebastian Michelmann
All authors & affiliationSebastian Michelmann, School of Psychology, University of Birmingham Benjamin Griffiths, School of Psychology, University of Birmingham Bernhard Staresina, School of Psychology, University of Birmingham Maria Wimber, School of Psychology, University of Birmingham Ramesh Chelvarajah, Neurosciences, Queen Elizabeth Hospital Birmingham, UK David Rollings, Neurosciences, Queen Elizabeth Hospital Birmingham, UK Vijay Sawlani, Neurosciences, Queen Elizabeth Hospital Birmingham, UK Hajo Hamer, Epilepsy Center, Department of Neurology, University Hospital Erlangen, Germany Stephanie Gollwitzer, Epilepsy Center, Department of Neurology, University Hospital Erlangen, Germany Gernot Kreiselmeyer, Epilepsy Center, Department of Neurology, University Hospital Erlangen, Germany Howard Bowman, School of Psychology, University of Birmingham & Centre for Cognitive Neuroscience and Cognitive Systems and the School of Computing, University of Kent at Canterbury Simon Hanslmayr, School of Psychology, University of Birmingham
A mental chronometry of dynamic memory replay in humans
Show/hide abstractWhen we mentally travel through episodes from our past we can let fluent representations unravel in front of our mental eye. Memories from our last summer vacation could involve walking out the hotel lobby, driving to the beach and finally stepping into the cool water of the sea. This phenomenon implies that our brains have the remarkable ability to temporally guide us through traces in our memory which may have been laid out years ago. Until now it remains largely unexplored which mechanisms orchestrate the replay of dynamic memories in humans, in particular what the mental chronometry of such dynamic replay is. Recent evidence suggests that oscillatory activity in the alpha rhythm plays an important role in this process and that decreases in power are involved in the temporal organisation of neural representations (1-3). We therefore conducted a Magnetencephalography (MEG) study in which participants learned associations between words and specific scenes within short videos. Using a simplified version of this task we also recorded intracranial electrophysiological activity from epileptic patients undergoing pre-surgical monitoring. During memory retrieval, we asked patients and healthy volunteers in which scene (first, second or third) of the video they learned a certain word. This way we prompted the reinstatement of the temporal structure of the videos. Crucially, in both experiments we presented (and subjects remembered) the same videos several times but associated with different words. This enabled us to use representational similarity analysis (RSA) to track the replay of individual scenes. We found sustained power decreases in the alpha frequency range to be associated with successful memory. Studying the time course of replay for different scenes provided new indications of how dynamic memories are replayed and how their neural representations unfold over time.
Sunday 12.00 - 13.30 Grote zaal Poster #59 Memory Marilina Mastrogiuseppe
All authors & affiliationMarilina Mastrogiuseppe, University of Trento, CIMeC Natasha Bertelsen, University of Trento, CIMeC Sang Ah Lee, University of Trento, CIMeC
Spatio-temporal subcomponents of Episodic Memory in Williams Syndrome
Show/hide abstractWilliams Syndrome (WS) is a rare neurodevelopmental disorder characterised by a number of medical problems as well as mental delay, hypersociability, and visuospatial impairments. The condition is caused by gene deletion on chromosome 7q11.23, which in turn prevents the synthesis of specific proteins needed for the normal functional development of the hippocampal formation (Landau & Ferrara, 2013). Recent neuroimaging studies have produced evidence that the hippocampus is involved in the retrieval of associations between events and their spatial contexts (Deuker et al., 2016). Given the hippocampal functional impairment that affects WS individuals, the present study set out to investigate spatial binding abilities in this population by evaluating their performance on an episodic memory hiding task and comparing it to typically developing chronological and mental age matched controls. Overall, episodic memory retrieval was delayed in WS participants when performance was compared with mental and chronological age controls (F (1,17) = 5,15 p = ,037). More importantly, WS participants where found to be selectively impaired in their ability to retrieve the spatio-temporal information associated with the distinct pattern of hiding they had previously encoded, as compared to the two control groups (F (2,34) = 5,57 p = ,008). The resulting impaired memory profile is consistent with hippocampal anomalies documented in Williams Syndrome (Meyer-Lindenberg et al., 2005), and further supports a specific role of the hippocampus in the binding together and subsequent retrieval of the spatio-temporal subcomponents of episodic memory.
Sunday 12.00 - 13.30 Grote zaal Poster #60 Decision-making Diego Urgeles
All authors & affiliationDiego Urgeles, Facultad de Informatica, Universidad Complutense de Madrid Guadalupe Miñana, Facultad de Informática, Universidad Complutense de Madrid Yolanda García, Facultad de Informática, Universidad Complutense de Madrid
Big Data Integration on Bipolar Disorder and Drugs Dependency
Show/hide abstractThe use of Big and Open Data in social and health sciences is increasing every day due to the improvements in prediction and pattern recognition processes, among others. Machine learning and new technologies provide the opportunity to use new sources of information which add an important value on our information systems and facilitate decision making. However, data sources are diverse and datasets are non-structured, therefore, in most cases, adequate preprocessing of data is necessary: collecting, cleaning, integration are key processes that we cannot ignore if we want the results of traditional statistical analysis to be reliable. In this work we show two examples in which several sources of information must be processed: medical history data, automated real-time monitoring data (from accelerometers) and data provided by the patient through a mobile application (with and without supervision). Both cases, bipolar disorder and drugs dependency studies, data collected includes geo-localization of the patient, voice recording, image recognition and many other qualitative and qualitative data. Data from accelerometers are collected every 5 seconds, while geographical position on the map is collected several times per day. A great amount of data from monitoring via smartphones is collected at least once per day as well. With all these data, the system will be able to build a profile of the patient in real time and compare it with patient euthymic stage. This information facilitates the crisis prediction. Anyway, a correct collecting and integration of the data requires the execution of integration processes of NoSQL databases and other processes for cleaning data as detection of outliers or principal component analysis for reducing dimensionality of the problems before data analytics. In this work we use MongoDB for data integration and Python and R as programming languages.
Sunday 12.00 - 13.30 Grote zaal Poster #61 Sensory processing Eva Landova
All authors & affiliationEva Landova, Silvie Radlova, Jakub Polak, Kristyna Sedlackova, Jaroslav Tintera, Daniel Frynta; all authors are from the National Institute of Mental Health, Klecany, Czech Republic
Not all snakes are alike: fMRI study of the affective response to different snake species
Show/hide abstractThere is about 3500 extant snake species, which together cover a wide morphological variability. The list includes small, underground worm-like species, colorful milk snakes, vipers and rattlesnakes with prominent scales, huge boid snakes, and many more. We asked respondents to rank a set of snake pictures and, based on the results, we selected species that elicited exclusively fear or disgust the most. Using a block fMRI design, we measured responses of 60 participants with differing susceptibility to fear and disgust (measured using the SNAQ and DS-R questionnaires) to the pictures of these fear-eliciting and disgust-eliciting snakes. Pictures of leaves were used as control (neutral) stimuli. The results showed no differences between participants with different SNAQ and DS-R scores. Insula and amygdala activation, often reported in studies dealing with strongly affective fearful and disgusting stimuli, showed only in a few single participants that reported exceptionally high fear and/or disgust. However, the results showed a clear difference between the fear-eliciting and disgust-eliciting snake stimuli across all groups. The activation of the visual cortex bilaterally was significantly higher (condition: fear > disgust, p<0.001) when watching the fear-eliciting snakes. The occipital fusiform gyrus, lingual gyrus and inferior occipital gyrus were bilaterally activated only during the fear-eliciting condition. These results are in agreement with findings that more extensive activation of the visual system facilitates the perceptual processing of survival-relevant stimuli. It suggests that human respondents are able to differentiate between various snake morphotypes. The fear-eliciting snakes (mainly vipers and rattlesnakes) present life-threatening stimuli and thus selectively attract motivated attention, while disgust-eliciting snakes, although also strongly affective, do not possess an immediate threat and are thus perceived with less concern.
Sunday 12.00 - 13.30 Grote zaal Poster #62 Open science and replication Roel van Dooren
All authors & affiliationRoy de Kleijn, Leiden University, Institute of Psychology, Cognitive Psychology Unit & Leiden Institute for Brain and Cognition, Leiden, the Netherlands Vera Mekern, Leiden University, Institute of Psychology, Cognitive Psychology Unit & Leiden Institute for Brain and Cognition, Leiden, the Netherlands Roberta Sellaro,Leiden University, Institute of Psychology, Cognitive Psychology Unit & Leiden Institute for Brain and Cognition, Leiden, the Netherlands Bernhard Hommel, Leiden University, Institute of Psychology, Cognitive Psychology Unit & Leiden Institute for Brain and Cognition, Leiden, the Netherlands
No evidence for cross-domain priming from visual to cognitive search
Show/hide abstractResearch on cross-domain priming reveals an interesting relationship between the exploration vs. exploitation trade-off in visual- and cognitive search (Hills, Todd, & Goldstone, 2008, 2010). One cognitive control mechanism may underlie the modulation of this trade-off in different domains. Previous work in our lab aimed to identify support for this shared mechanism in cross-domain priming by, albeit indirectly, studying the role of striatal dopamine in the exploration vs. exploitation trade-off. As the hypothesized cross-domain priming effects were not obtained, we decided to run a pure replication of Study 1 by Hills, Todd and Goldstone (2010) to acquire a better understanding of the absence of these effects. Participants performed a visual foraging task in which they searched spatial environments containing resources (i.e., ‘food') which were either clumped together or were more spatially distributed (i.e., clumpy vs. diffuse) in a visual scene. Prior to and following this task, participants performed a Scrabble task. We expected participants foraging diffuse environments (i.e., priming exploration) to switch more frequently between lettersets as compared to participants who foraged clumpy environments (i.e., priming exploitation). We found no priming effects of visual search on cognitive search, thereby not replicating the results reported by Hills, Todd and Goldstone (2010). Additional exploratory analyses supported the lack of cross-domain priming effects. As we found no evidence of cross-domain priming, we argue that the relationship between the exploration vs. exploitation trade-off in visual- and cognitive search is not yet fully understood. Specifically, it is unclear whether the current paradigm was ineffective in capturing cross-domain priming, or whether priming occurs at all.
Sunday 12.00 - 13.30 Grote zaal Poster #63 Language Antonia Tovar
All authors & affiliationAntonia Tovar (1)(2); Jesús Ruiz-Idiago (3); Celia Mareca Viladrich (3); Edith Pomarol-Clotet (5)(6); Aina Garí Soler (7); Vitor C. Zimmerer (8); Pedro Roy-Millan (3); Joana Rossello (1); Wolfram Hinzen (4)(5) 1) Universitat de Barcelona, Barcelona, Spain. (2) Universitat Pompeu Fabra, Barcelona, Spain. (3) Huntington Disease Programme. Hospital Mare de Déu de la Mercè. Hermanas Hospitalarias, Barcelona, Spain. (4) ICREA (Catalan Institution for Research and Advanced Studies), Universitat Pompeu Fabra, Departament de Traducció i Ciències del Llenguatge, Barcelona, Spain. (5) FIDMAG, Germanes Hospitalàries Research Foundation. Barcelona, Spain. (6) CIBERSAM (G15), (7) University of Malta, Msida, Malta. (8) University College London, Department of Language and Cognition, London, United Kingdom.
Language and cognitive impairments in Huntington's disease population
Show/hide abstractHuntington's disease (HD) is an autosomal dominant neurodegenerative disorder. While motor and neuropsychiatric symptoms are its core features, cognitive impairments are also widely recognized, including deficits in ‘theory of mind' (ToM). A recent study (Hinzen et al., 2016) has systematically profiled linguistic decline in this disease based on an analysis of spontaneous speech, demonstrating that HD affects core aspects of linguistic organization and that it does so even before the onset of motor symptoms. Methodology: firstly, we replicate these results in a different sample of patients, and secondly, we probe specific aspects of language experimentally, in linguistic comprehension. The sample consisted of 8 pre-symptomatic gene-carriers, 8 early symptomatic, and 16 controls. All participants were administered a neuropsychological test battery as part of a standard cognitive assessment. Spontaneous speech was obtained from an open interview. Based on the result, a novel sentence-picture-matching test was designed to specifically test the comprehension of sentences with both factive subordinated complements, e.g. The man knows it is warm outside and non-factive ones, e.g. The man thinks it is warm outside. Results: Spontaneous speech analysis indicated systematic structural changes in language confirming our earlier result, including in pre-symptomatic gene-carriers whose neuropsychological scores were normal. Stage 1-2 patients performed worse on the SPM task than controls, and significantly worse on non-factives than on factives. This result predicts difficulties in this population in reasoning about mental states whose contents need not correspond to reality, and so in understanding people's actions. To explore this prediction, we further asked participants to watch two short videos, one verbal and one non-verbal, and to narrate their contents. Results showed that all patients except one were at chance in understanding the nonverbal video. Overall, these results suggest that cognitive decline in HD not only affects language but more specifically linguistic resources required for representing mental states.
Sunday 12.00 - 13.30 Grote zaal Poster #64 Memory Samarth Varma
All authors & affiliationSamarth Varma, Donders Institute of Brain, Cognition and Behaviour Atsuko Takashima, Donders Institute of Brain, Cognition and Behaviour, Max Planck Institute of Psycholinguistics Sander Krewinkel, Donders Institute of Brain, Cognition and Behaviour Maaike van Kooten, Donders Institute of Brain, Cognition and Behaviour Lily Fu, Donders Institute of Brain, Cognition and Behaviour W. Pieter Medendorp, Donders Institute of Brain, Cognition and Behaviour Roy P.C. Kessels, Donders Institute of Brain, Cognition and Behaviour Sander M. Daselaar, Donders Institute of Brain, Cognition and Behaviour
Non-Interfering Effects of Active Post-Encoding Tasks on Memory Consolidation
Show/hide abstractStudies show that when learning is followed by or split across periods of quiet wakeful rest, memory is better consolidated. Tasks involving only complex and meaningful information such as picture-naming, psychometric tests and autobiographical thinking cause forgetting due to reallocation of resources away from consolidation-relevant activities. In contrast to this, we investigated whether interference can be elicited using a task that has minimal semantic and hippocampal processing needs, and reduces task-unrelated mindwandering by having high attentional demands. To this end, we ran seven experiments (N= 212) that compared subsequent memory performance of items learnt prior to a 9-minute long N-Back task, an equivalent period of quiet wakeful rest, and a cued-autobiographical thinking task. We manipulated the design of the N-Back task to target only general/executive resources at first (N-Back with numbers), and then resources overlapping with ongoing consolidation as well (N-Back with encoding-like stimuli). Challenging the assumptions of standard interference theory, we did not find any differences between N-Back and rest conditions on memory retention, measured using both recall and recognition tests. Critically, we also show that subsequent memory performance is superior when learning is followed by the N-Back task as compared to a period of autobiographical thinking. Our results indicate that: (1) quiet, wakeful rest is not necessary for episodic memory consolidation, (2) post-encoding cognitive engagement does not interfere with memory consolidation when task-performance has minimal semantic and hippocampally-based episodic memory processing demands, and (3) mindwandering may not be crucial for successful consolidation. From an educational point of view, this study has implication for the understanding of the Distributed Learning effect, and the effect of prevocational or skill-learning tasks in a classroom environment.
Sunday 12.00 - 13.30 Grote zaal Poster #65 Memory Andrew Clouter
All authors & affiliationAndrew Clouter, Kimron L. Shapiro, Simon Hanslmayr
Gluing memories via oscillations: Theta phase synchrony drives associative memory formation in humans
Show/hide abstractThe objective of our experiments was to investigate the causal role of neural synchrony between visual and auditory processing regions on associative memory formation for multisensory events. Multisensory episodic memories rely on successfully binding elements that are processed in separate, specialised brain regions. The formation of episodic memories is thought to rely on the synchronization between distant brain regions in the theta frequency band. However, causal evidence for this idea from humans is missing. To provide such evidence we developed a novel multisensory memory paradigm where participants encode sound-movie associations. Modulating the luminance and amplitude of the videos and sounds independently allowed us to control the degree of phase synchrony between the auditory and visual cortex. We then show in two experiments that memory for the sound-movie associations differs drastically depending on the degree of inter-sensory phase synchrony. In the first experiment, in the encoding phase, all participants were shown short (3-second) videos that were luminance modified with a 4 Hz sine wave, with an accompanying audio clip that had been amplitude modulated with a 4 Hz sine wave. The phase offset (i.e., synchrony) between the audio clip and the video was 0, 90, 180, or 270 degrees. In a second experiment, the videos and sounds were modulated at 4 Hz, 1.7 Hz (delta), and 10.5 Hz (alpha). On each trial, participants rated how well the audio clip suited the contents of the video clip. Each of six blocks contained 16 audio-video pairings (four at each phase angle), and was followed by a brief distractor task and an associative recognition test. Associations were better remembered in the synchronous compared to the asynchronous condition. This effect was specific to theta (i.e. 4Hz) and did not occur in a faster (10.5 Hz) or slower frequency (1.7 Hz). These findings suggest that episodic memory formation in humans relies on a theta specific synchronization mechanism.
Sunday 12.00 - 13.30 Grote zaal Poster #66 Consciousness Lucie Berkovitch
All authors & affiliationLucie Berkovitch
Impaired conscious access in schizophrenia is associated with abnormal attentional amplification
Show/hide abstractPrevious research suggested that the conscious perception of masked stimuli is impaired in schizophrenia, while unconscious bottom-up processing of the same stimuli, as assessed by subliminal priming, can be preserved. Here, we test this postulated dissociation between intact bottom-up and impaired top-down processing and evaluate its brain mechanisms using high-density recordings of event-related-potentials. Sixteen patients with schizophrenia and 16 normal controls were exposed to peripheral digits at various degrees of masking, under conditions of either focused attention or distraction by another task. In the distraction condition, the brain activity evoked by masked digits was drastically reduced in both groups, but early visual activation could still be detected and did not differ between patients and controls. By contrast, under the focused attention instructions, a major impairment was observed: in patients, contrary to controls, the N1 component was insufficiently amplified by attention, and the late non-linear ignition component associated with the P3 component was drastically reduced. Interestingly, the patients showed an essentially normal attentional amplification of the P1 and N2 components. These results suggest that some but not all top-down attentional amplification processes are impaired in schizophrenia, while bottom-up processing seems to be preserved.
Sunday 12.00 - 13.30 Grote zaal Poster #67 Sensory processing Ehsan Darestani Farahani
All authors & affiliationEhsan Darestani Farahani affiliation: Research Group Experimental ORL, Department of Neurosciences, KU Leuven - University of Leuven, Belgium - Additional authors: Jan Wouters, Astrid van Wieringen affiliation: Research Group Experimental ORL, Department of Neurosciences, KU Leuven - University of Leuven, Belgium
Non-primary cortical sources of auditory temporal processing
Show/hide abstractAuditory information is transmitted to the higher brain centers through the primary and the non-primary auditory pathways. The primary pathway goes from the brainstem, to the midbrain, and then to the thalamus before terminating at the primary auditory cortex. In a parallel pathway, the non-primary pathway initiates at the cochlear nuclei and connects to the reticular formation, a region of the brainstem with interconnected nuclei. These fibers project through reticular formation into the thalamus and then to the association cortices. Despite several studies on the sub-cortical centers of the non-primary pathway, detailed information on the non-primary cortical centers that are responsive to auditory stimuli is lacking. The main aim of this study is to gain insight into cortical sources underlying auditory temporal processing, more specifically, non-primary cortical sources. To accomplish this, the brain sources of auditory steady-state responses (ASSRs) to amplitude modulated acoustic stimuli are reconstructed. This study provides a new insight into the regions of association cortex that are electrophysiologically responsive to acoustic inputs. In order to reconstruct ASSR sources, we applied independent component analysis (adaptive mixture ICA) with subsequent equivalent dipole modeling to the concatenated EEG data of all subjects (young adults, 20-30 years of age). These data were based on responses to white noise stimuli, amplitude modulated at 4, 20, 40, or 80 Hz. The independent components that exhibited a significant response at the respective modulation frequencies were recognized as ASSR sources. Our results suggest that the cortical sources underlying auditory temporal processing are not restricted to the primary auditory cortex and other sources in the association cortex are also involved in auditory temporal processing. For the four modulation frequencies, the identified sources, either primary or non-primary, were located in the physiologically and anatomically plausible locations.
Sunday 12.00 - 13.30 Grote zaal Poster #68 Memory Moritz Koester
All authors & affiliationMoritz Koester, Holger Finger, & Thomas Gruber
Neuronal oscillations indicate sleep-dependent changes in the cortical memory trace
Show/hide abstractSleep promotes the consolidation of newly acquired associative memories. Here we used neuronal oscillations in the human EEG to investigate sleep-dependent changes in the cortical memory trace. The retrieval activity for object–color associations was assessed immediately after encoding and after 3 hr of sleep or wakefulness. Sleep had beneficial effects on memory performance and led to reduced event-related theta and gamma power during the retrieval of associative memories. Furthermore, event related alpha suppression was attenuated in the wake group for memorized and novel stimuli. There were no sleep-dependent changes in retrieval activity for missed items or items retrieved without color. Thus, the sleep-dependent reduction in theta and gamma oscillations was specific for the retrieval of associative memories. In line with theoretical accounts on sleep-dependent memory consolidation, decreased theta may indicate reduced mediotemporal activity because of a transfer of information into neocortical networks during sleep, whereas reduced parietal gamma may reflect effects of synaptic downscaling. Changes in alpha suppression in the wake group possibly index reduced attentional resources that may also contribute to a lower memory performance in this group. These findings indicate that the consolidation of associative memories during sleep is associated with profound changes in the cortical memory trace and relies on multiple neuronal processes working in concert.
Sunday 12.00 - 13.30 Grote zaal Poster #69 Memory Karsten Rauss
All authors & affiliationKarsten Rauss1*, Nicolas D. Lutz1,2, Ines Wolf1†, Stefanie Huebner1†, Jan Born1,3 1Institute of Medical Psychology and Behavioral Neurobiology, University of Tübingen, Otfried-Mueller-Strasse 25, 72076 Tuebingen, Germany; 2Graduate Training Centre of Neuroscience/International Max Planck Research School for Cognitive & Systems Neuroscience, University of Tuebingen, Oesterbergstrasse 3, 72074 Tuebingen, Germany; 3Werner Reichardt Centre for Integrative Neuroscience, University of Tuebingen, Otfried-Mueller-Strasse 25, 72076 Tuebingen, Germany; †equal contributions
Sleep improves and impairs distinct aspects of visual sequence learning
Show/hide abstractSleep is known to support memory consolidation, but it remains unclear whether implicit sequence learning benefits from sleep. Here, we tested whether sleep may exert differential effects on different behavioral indicators of sequence learning. If sleep literally consolidates sequence information, this should make performance more vulnerable to unpredictable deviant stimuli or changes in stimulus timing. On the other hand, recent studies suggest that sleep supports abstraction of knowledge, which should render sequence learning immune to such changes. We addressed these questions using a modified serial reaction-time task, where human participants responded to a deterministic sequence of 12 stimuli arranged in two dimensions. Individual deviant stimuli were interspersed with regular standard stimuli just before and right after an 11-hour retention interval. Four groups (n=32 each, overall N=128) were trained and tested using different combinations of short (200 ± 50 ms) and long (2000 ± 500 ms) response-to-stimulus intervals (RSIs). Different subgroups (n=16 each) underwent training before daytime wakefulness and nighttime sleep, respectively. Our results show that sequence knowledge can be extracted from and applied to different temporal contexts, independent of whether subjects slept during the retention interval. Sleep effects were only observed under standard conditions, with short RSIs during both training and retrieval. This demonstrates that sleep affects sequence learning when it is probed using a restricted number of deviant stimuli within a deterministic sequence. Crucially, sleep-dependent consolidation of sequence learning both diminished and improved distinct aspects of performance: errors in response to deviants became more frequent after sleep, an effect partly mediated by a higher number of prediction-informed errors. At the same time, sleep supported accurate performance for stimuli immediately following deviants, thus putting the system back on track after unpredictable disturbances.
Sunday 12.00 - 13.30 Grote zaal Poster #70 Attention Kyungun Jhung
All authors & affiliationKyungun Jhung, MD, Department of Psychiatry & Behavioral Neuroscience, International St. Mary's Hospital, Catholic Kwandong University, Incheon, South Korea Jin Young Park, MD, Department of Psychiatry and Institution of Behavioral Science in Medicine, Yonsei University College Of Medicine, Seoul, South Korea Yu-jin Jeong, Graduate Program in Cognitive Science, Yonsei University, Seoul, South Korea
Neurophysiological correlates of executive function in children and adolescents with attention-deficit/hyperactivity disorder
Show/hide abstractObjectives: Attention-deficit hyperactivity disorder (ADHD) is characterized by significant impairments in various executive functions, with a prevalence of approximately 5-10% of all children worldwide. Developing reliable neural markers of executive function will improve understanding of ADHD and aid in making a precise diagnosis. The goal of this study was to examine the relationship between various executive functions and electrophysiological activities in children and adolescents with ADHD. Methods: In 31(19 male, age 11.5+-3.6 years) patients with ADHD, resting-state EEG during eyes open and eyes closed was recorded, and Comprehensive Attention Test (CAT), Stroop Color-Word Inference Test (Stroop CWIT), Trail Making Test (TMT), and Wisconsin Card Sorting Test (WCST) were administered. Korean version of the ADHD Rating Scale (K-ARS) and Korean Child Behavior Checklist 6-18 (K-CBCL) were assessed. Results: Alpha and beta power positively correlated with the Attention Quotient (AQ), while delta power negatively correlated with AQ from CAT. In the Stroop CWIT, decreased delta power and increased beta power were related to higher performance. Power of the alpha band increased with higher TMT performance. Moreover, delta power negatively correlated with good performance on the WCST, while alpha and high gamma band showed a positive correlation. Correlation with the parent-rating of ADHD symptoms showed a negative correlation between alpha power and higher scores on the K-ARS. Conclusions: These findings indicate that relative power in higher frequency bands of EEG is related to the higher executive function in children and adolescents with ADHD, while the association with the relative power in lower frequency bands of EEG seem to be vice versa. Furthermore, the findings suggest that qEEG may be a useful adjunctive tool in assessing patients with ADHD.
Sunday 16.00 - 17.30 Grote zaal Poster #1 Sensory processing Yu-Fang Yang
All authors & affiliationYu-Fang Yang,CIAMS, Univ. Paris-Sud, Universite Paris-Saclay,CIAMS, Univ. Paris-Sud, Universite Paris-Saclay* Brunet-Gouet,Centre Hospitalier de Versailles, HandiResp, EA4047, Universite Versailles Saint-Quentin Burca,Centre Hospitalier de Versailles, HandiResp, EA4047, Universite Versailles Saint-Quentin Kalunga,LISV, UVSQ, Universite Paris-Saclay Amorim,CIAMS, Univ. Paris-Sud, Universite Paris-Saclay,CIAMS, Univ. Paris-Sud, Universite Paris-Saclay
Does the brain process similarly photos and sketches when categorizing an emotional facial expression?
Show/hide abstractAn upside-down face leads to a disrupted holistic processing in face perception increasing the N170 event-related potential (ERP) component. The inversion effect on N170 is similar between neutral photo and sketch faces during passive viewing [1]. Yet, the effect of stimulus type and orientation remains unclear when comes to recognise facial emotion. Hence, we created a subset of emotional sketch faces (neutral, happy, fear, angry, sad) from the Radboud faces database [2], by extracting the most diagnostic high-spatial frequency facial features (e.g. eye, nose, and mouth). Methods: P100 ERP component (O1 and O2 electrodes), and N170 (P7 and P8 electrodes), were recorded among twenty-five participants (M=26.4 years ± 6.5, 17 males). ANOVAs were conducted with stimulus type (sketch vs. photo) and orientation (upright vs. inverted) as within-subjects factors. Results: The behavioral results (unbiased hit rate and RTs) showed similar high level performance when stimuli were upright, and a detrimental inversion effect, greater for sketch than photo faces. The increased P100 peak amplitude on inverted stimuli (p<.003) suggested the difficulty in structural encoding, whereas a reduced P100 amplitude was found for sketch faces (p<.0001) reflecting impoverished visuosensory input [3]. P100 peak latency increased only for inverted photo faces (p<.05). N170 peak amplitude showed greater values for sketch faces only (p<.003). We found an additive effect of stimulus type (6ms increase for sketches, p=.006) and stimulus orientation (5ms increase for inverted stimuli, p<.04) on N170 peak latency; with a greater inversion effect (6ms) at the right hemisphere. Conclusions: The results of P100 latency suggested that photo faces elicited an early full face processing that required additional time to reorient inverted stimulus, contrary to the sketch faces. N170 presented an intermediate pattern of modulation showing that faces need extra-processing time but don't exhibit a dramatic latency slow-down such as in behaviour.
Sunday 16.00 - 17.30 Grote zaal Poster #2 Connectivity/networks Laura O'Halloran
All authors & affiliationLaura O'Halloran 1, Zhipeng Cao 1, Kathy Ruddy1, Lee Jollans, Matthew D. Albaugh, Ph.D. 1, Alexi Potter, PhD., Tobias Banaschewski M.D., Ph.D.2; Ruediger Bruehl Ph.D.3; Arun L.W. Bokde Ph.D.4; Uli Bromberg M.D.5; Christian Buechel M.D.5; Anna Cattrell Ph.D.6; Patricia J. Conrod Ph.D.7,8; Sylvane Desrivieres Ph.D.6; Herta Flor Ph.D.9; Vincent Frouin Ph.D.10; Juergen Gallinat M.D.11; Robert Goodman, Ph.D.12; Penny Gowland Ph.D.13; Yvonne Grimmer, M.D.9; Andreas Heinz M.D., Ph.D.14; Viola Kappel, Dipl.-Psych.15; Jean-Luc Martinot M.D., Ph.D.16; Marie-Laure Paillere Martinot M.D., Ph.D.17; Frauke Nees Ph.D.9; Dimitri Papadopoulos Orfanos Ph.D.10; a i e ttil , .D.18; Luise Poustka M.D.2; Tomaš Paus M.D., Ph.D.19; Michael N. Smolka M.D.20; Maren Struve, Ph.D.9; Henrik Walter M.D., Ph.D.14; Gunter Schumann M.D.6 ; Hugh Garavan, Ph.D. 1; 1 Clare M. Kelly1 Robert Whelan Ph.D.1
Functional correlates of reaction time variability in a large sample of adolescents and ADHD symptom adolescents
Show/hide abstractReaction time variability, often quantified by the intra-individual coefficient of variation (ICV), is an index of consistent attention and executive control processes, and is often disrupted in attention deficit and hyperactivity disorder (ADHD). Here, for the first time, we investigate the brain regions and the functional connectivity underlying ICV in a large population-based sample of adolescents (n=758). Activity in the precentral gyrus, posterior cingulate and thalamus during a speeded choice task correlated with ICV. Increased functional connectivity within motor-related regions was positively correlated with ICV. In contrast, functional connectivity between prefrontal regions and the cerebellar network, and also between the default mode network, cerebellar and limbic regions were both negatively correlated with ICV. Testing these findings in a separate sample of adolescents with ADHD symptoms (n=30) in comparison to a matched asymptomatic control group (n=30), the ADHD symptom group had significantly increased behavioral ICV, as well as increased connectivity in ICV-related regions, including within the motor network, and between the prefrontal and cerebellum, and DMN and cerebellar networks. We propose that attentional abilities are underpinned by ICV and encompass a large scale of brain networks, and that functional connectivity between the right dorsolateral prefrontal gyrus and cerebellum in particular provides an objective and robust index of sustained attention.
Sunday 16.00 - 17.30 Grote zaal Poster #3 Memory Saima Mohan
All authors & affiliationSaima Mohan, Hitachi India Private Limited Stephanie Sutoko, Hitachi Limited Atsushi Maki, Hitachi Limited
A Cross-cultural Study: Effects of Mood State Analysis on Working Memory for Development of Biomarker
Show/hide abstractAs per mental health study, day-to-day anxiety is bending towards depression; affecting a person's thinking, mood and function on daily basis. In previous near infrared spectroscopy (NIRS) experiments conducted on native language, the influence of negative mood state on verbal working memory (WM) were analyzed and observed that left pre-frontal cortex (PFC) showed high significance. It remains a challenge to demonstrate if such interactions of mood on cognition are reproducible with subcontinent group comprising of varied melanin chromophores. To investigate this, a non-invasive, compact, wearable optical topography, measuring frontal cortex activities alone in real-time is used. We acquired anonymized NIRS samples of 20 healthy controls, subjected with different memory-load conditions and multiple trials of spatial and verbal WM in non-native language. With group analysis study it was observed, the performance measures such as response time(RT) for verbal WM was low with higher accuracy level, which requires to be investigated further with native language. With high-memory load, significant channels found from observations of oxy-hemoglobin (oxy-Hb) after one-sample t-test exhibited dominance on dorsal PFC. To reduce Type-I error, False discovery rate(FDR) correction was applied only on set of significant channels and results indicated dorsal lateralization during verbal WM task. To evaluate different facets of mood, participants were asked to take Profile of Mood States (POMS) questionnaire. The mood scores of group was correlated with activation values of spatial and verbal WM separately; and in comparison, verbal neural activity indicated stronger negative correlation (p<0.05, FDR corrected). Hence, the results suggest that the neural activity response of verbal WM in PFC are associated to individual's natural mood state. These findings are supported with confounding factors such as accuracy, RT and gender. We also found with little memory load conditions no significant differences in PFC activation were observed.
Sunday 16.00 - 17.30 Grote zaal Poster #4 Decision-making Yun-Hsuan Chang
All authors & affiliation1. Yun-Hsuan Chang, Department of Psychology, Asia University; Department of Psychology, College of Medical and Health Science, Asia University, Taichung, Taiwan; Department of Medical Research, China Medical University Hospital, China Medical University, Taichung, Taiwan; Department of Psychiatry, College of Medicine, National Cheng Kung University, Taiwan 2. Shulan Hsieh, Department of Psychology; Institute of Allied Health Sciences; Public Health, National Cheng Kung University, Tainan, Taiwan
Association among online-game playing, resting fMRI state and multitasking in a virtual environment
Show/hide abstractEvidence has suggested potential cognitive skills training of extensive video-game playing, especially in eye-hand coordination and reaction time, spatial attention, divided visual attention, selective attention and attentional capacity, but the effective in multitasking ability is unclear. As the dramatic development of computer and internet, online-game playing has become a common entertainment in the current generation. Online-game players often do several tasks at the same time; often tasks interfere with each other and might increase mistakes. This study proposed to explore the relationships between online-game playing types and multitasking ability in a virtual environment, using Edinburgh Virtual Errands Test (EVET). In this study, 79 participants with mean aged from 20-30 years old were recruited. Chen's Internet Addiction Scale (CIAS), internet use questionnaire, EVET, working memory tests and resting fMRI were given to each participant. The results showed a positively association between multitasking ability and working memory (WM). No significant difference of spent time between MOBA and OGPs, showing both groups could be expertise in online-game playing. Three groups with varied online-game playing, multiplayer online battle arena (MOBA), other online-game players (OGPs) and no-players (NGPs) were compared with their EVET performance, WM and time spent of online-game use. The MOBA players had better performance on EVET plan follow and total score, and spatial memory than did other two groups, indicating a benefit of MOBA on planning following and increase the efficiency of multitasking. Using Cohe-ReHo analysis of resting fMRI showed significant association between both sides of temporal lobe, caudate, occipital lobe, anterior cingulate cortex and parahippocampus and EVET. In conclusions, online-game playing may provide a role in training multitasking and helps to maintain the planning. Multiplayers online game genres may benefit in participants' spatial memory. Several brain function activities were found, indicating collaboration among multiple brain areas involving in multitasking performance.
Sunday 16.00 - 17.30 Grote zaal Poster #5 Sensory processing Robin Fondberg
All authors & affiliationRobin Fondberg, Karolinska Institutet Johan N. Lundstroem, Karolinska Institutet Mats J. Olsson, Karolinska Institutet Janina Seubert, Karolinska Institutet
Flavor formation and evaluation: The relationship between congruency, pleasantness and odor referral to the mouth
Show/hide abstractDuring food consumption, odor and taste combine to create a holistic flavor object; an inherently multisensory experience that extends beyond the mere addition of its components. While semantic congruency is generally known to be important for multisensory binding, little is known about its specific role in flavor processing. The aim of the present study was therefore to delineate the effect of congruency between odor and taste on two key features of flavor perception: the mislocalization of food odors to the mouth, and flavor pleasantness. Moreover, we assessed the interrelationship between these indicators to investigate whether the shift in localization might modulate pleasantness evaluations. Aqueous solutions with different combinations of odors and tastes were prepared to create bimodal mixtures with varying degrees of semantic overlap, ranging from maximally incongruent to maximally congruent in equidistant steps. Participants reported where they perceived the odors, and how much they liked the solutions. Congruency had a positive, linear effect both on the tendency to perceive odors in the oral cavity, and on the pleasantness ratings. However, pleasantness ratings did not differ depending on whether the odor had been localized to the mouth or not. This indicates that as an olfactory-gustatory mixture approximates a familiar food, the two become increasingly merged into one perceptual entity that is perceived to be arising from the mouth. Also, the mixture is evaluated as increasingly pleasant, which likely serves as a survival mechanism by promoting consumption of common foods. These findings clearly demonstrate the modulatory role of congruency in both features of flavor processing. They further highlight that the two processes do not, however, mutually rely on each other, suggesting that they are driven by distinct neural processes.
Sunday 16.00 - 17.30 Grote zaal Poster #6 Attention Freek van Ede
All authors & affiliationFreek van Ede, Sammi Chekroud, Mark Stokes, Kia Nobre, Oxford Centre for Human Brain Activity (University of Oxford)
Decoding the Influence of Anticipatory States on Visual Perception in the Presence of Temporal Distractors
Show/hide abstractWhile it has long been recognised that anticipatory states amplify early EEG responses to visual targets in humans, it remains unclear how such modulations relate to the actual content of the neural representation. Using multivariate orientation decoding of high temporal-resolution EEG recordings, we demonstrate that anticipation also increases the amount of stimulus-identity information contained in these early brain responses. Furthermore, this approach enabled us to characterize the influence of temporally adjacent distractors on target identity decoding, revealing that anticipation did not just attenuate distractor interference on target representations but, instead, delayed it. Enhanced target decoding and distractor resilience were further predicted by the attenuation of posterior 8-14 Hz alpha oscillations. The findings offer several novel insights into how anticipatory states prioritise relevant sensory input in time, and they highlight the potential of non-invasive multivariate electrophysiology to track cognitive influences on perception in tasks with rapidly changing displays.
Sunday 16.00 - 17.30 Grote zaal Poster #7 Attention Judith Nicolas
All authors & affiliationJudith Nicolas, Lyon Neuroscience Research Center, INSERM, Unit 1028, Lyon, France Aline Bompas, School of Psychology, Cardiff University, Cardiff, Wales, UK Romain Bouet, Lyon Neuroscience Research Center, INSERM, Unit 1028, Lyon, France Olivier Sillan, Lyon Neuroscience Research Center, INSERM, Unit 1028, Lyon, France Eric Koun, Lyon Neuroscience Research Center, INSERM, Unit 1028, Lyon, France Christian Urquizar, Lyon Neuroscience Research Center, INSERM, Unit 1028, Lyon, France Alessandro Farne, Lyon Neuroscience Research Center, INSERM, Unit 1028, Lyon, France Aurelie Bidet-Caulet, Lyon Neuroscience Research Center, INSERM, Unit 1028, Lyon, France Denis Pelisson, Lyon Neuroscience Research Center, INSERM, Unit 1028, Lyon, France
Saccadic Adaptation Increases Sustained Gamma Band Activity
Show/hide abstractAttention and saccadic adaptation are critical components of visual perception, the former enhancing sensory processing of objects of interest, the latter maintaining the accuracy of saccadic eye movements toward these objects. Recent studies propelled the hypothesis of a tight functional coupling between these two mechanisms. Indeed adaptation of reactive saccades towards the left hemifield increases the processing speed of unpredictable stimuli (Habchi et al., 2015), conversely attentional load boosts saccadic adaptation (Gerardin et al. 2015). This hypothesis is also consistent with the partial overlap between the cortical substrates of saccadic adaptation (Gerardin et al. 2012) and visual attention (Corbetta and Shulman 2002). Here, we used magnetoencephalography to better understand the neurophysiological bases of this coupling. We compared visual discrimination performance of 12 healthy subjects before and after an adaptation or control task involving reactive saccades. A backward adaptation procedure (saccade amplitude decrease) was used. Eye movements and magnetic signals were recorded continuously. The neurophysiological analysis focused on gamma band power during the pre-target period of the saccadic adaptation and the discrimination tasks. Although attentional modulations by saccadic adaptation failed to impact behavioral performance in our paradigm, we demonstrated its impact at the electrophysiological level as an increase of gamma band power within an extended brain network during and after the saccadic adaptation. The beamformer source reconstruction showed a predominance of this effect in the left hemisphere (ipsilateral to the direction of adapted saccades) and an extension of this effect to the right hemisphere in the following discrimination task. These results reveal that gamma oscillations are involved in the coupling between saccadic adaptation and attention, suggesting that the sustained effect of this oculomotor adaptation procedure could generalize to other cognitive tasks relying on gamma activity.
Sunday 16.00 - 17.30 Grote zaal Poster #8 Attention Tobias Katus
All authors & affiliationMartin Eimer, Birkbeck, University of London
Independent capacities of spatial working memory in vision and touch
Show/hide abstractCowan's theory of working memory (WM) posits that information is maintained by a central attention mechanism whose capacity is shared across different modalities. In the event-related potential, the capacity of visual WM is mirrored by contralateral delay activity (CDA) amplitudes that increase in a load-dependent fashion until WM capacity is reached. We recently demonstrated the existence of a tactile CDA (tCDA component) which is sensitive to tactile WM load. In two studies, we here separately varied tactile and visual load, and unpredictably tested memory for either modality after each trial. Converging with behavioral performance, the tCDA/CDA components elicited in our EEG experiment reflected memory load effects that were strictly modality-specific. The CDA increased for visual load increments from 1 to 2 and 3 items. The tCDA was enhanced when tactile load increased from 1 to 2 items, but showed no further enhancement for 3 tactile items. Critically, neither did tactile load influence the visual CDA, nor did visual load modulate the tCDA, suggesting an independence of attentional maintenance processes for visual and tactile information. In a behavioral study, modality-specific set size effects were accompanied by crossmodal effects, as tactile/visual task performance decreased with load increments in both the tested and untested modalities. However, even for 12 multisensory items, this dual-task interference was considerably smaller than predicted by a shared capacity account. No correlation between individuals' capacities for tactile and visual information was found in baseline trials with 6 unimodal items. We propose that dual-task interference effects are not caused by a competition between multisensory items for a shared WM representation. Instead, they reflect competitive interactions between modality-specific maintenance processes with independent capacities that are sustained by signals from a shared top-down control system, which is not directly implicated in the storage of information.
Sunday 16.00 - 17.30 Grote zaal Poster #9 Memory Yi-Jhong Han
All authors & affiliationYi-Jhong Han, Institute of Cognitive Neuroscience, University College London (UCL), London, UK Nicholas D. Cole, Institute of Cognitive Neuroscience, University College London (UCL), London, UK Daisy V. Yip, Institute of Cognitive Neuroscience, University College London (UCL), London, UK Leun J. Otten, Institute of Cognitive Neuroscience, University College London (UCL), London, UK
The effect of closeness to others on the self-reference effect
Show/hide abstractMemory tends to be better for information that pertains to the self rather than another person. In the literature, the precise relationship between the self and another has not typically been considered. In three experiments, we addressed the role of closeness to another in the self-reference effect and the underlying neural mechanisms. Participants were asked to judge how well trait adjectives described themselves or people they were familiar with, but who varied in level of emotional attachment. Participants saw a series of trait adjectives, each preceded by a cue that indicated the person about whom the upcoming judgment should be made. Memory for the adjectives was probed with a remember/know task shortly afterwards. In the first experiment, memory performance was considered for the self, a close other and a distant other. In the second and third experiments, performance was considered alongside electrical brain activity to compare the self with a close and distant other respectively. The ability to recollect adjectives decreased gradually the more distant another was. Regarding brain activity, the frontally-distributed positive subsequent memory ERP effect typically associated with deep semantic encoding was elicited by adjectives judged in relation to the self and a close other. Successful recollection of adjectives related to the self was associated with a widespread positive old/new effect early on. In stark contrast, adjectives encoded in relation to a close other elicited a later negative-going effect that was largest over posterior scalp sites. Adjectives judged in relation to a distant other did not show significant effects during encoding or retrieval. These findings suggest that information pertaining to the self and a close other are encoded in qualitatively similar ways, but recollected via separate neural mechanisms. The self may have a distinct role in memory that cannot exclusively be explained by closeness to others.
Sunday 16.00 - 17.30 Grote zaal Poster #10 Memory Lukaš Hejtmanek
All authors & affiliationMartin Tomašek, Department of Neurology, Charles University in Prague, 2nd Faculty of Medicine, Czech Republic Petr Marusic, Department of Neurology, Charles University in Prague, 2nd Faculty of Medicine, Czech Republic Kamil Vlcek, Department of Neurophysiology of Memory, Academy of Sciences, Czech Republic
Theta power modulation during egocentric and allocentric navigation in virtual and real environment
Show/hide abstractNavigation processes can use two basic reference frames; egocentric reference frame is using self position and rotation in determining direction and distance of objects and is localised to parietal lobe, whereas allocentric navigation uses layouts and landmarks irrespective of self position and is dependent on the function of medial temporal lobe and hippocampus. There is still continuous debate to what extent these two processes interlay, where such interfacing takes place and how they differ during spatial representation acquisition and wayfinding. It has been demonstrated that increase in theta power in hippocampal area is correlated with movement onsets and modulated by speed. But it has not been addressed what type of spatial reference frame was used in such cases. Our research aims to clearly differentiate between egocentric and allocentric navigation. Using intracranial electroencephalography recordings from patients with refractory epilepsy we are able to obtain highly spatially accurate data as well as uncover the dynamics of navigation processes with sub-second precision. We modelled a Morris maze analogy in virtual and real environment and collected data from 5 patients during both virtual and real world navigation while they solved tasks requiring them to find a hidden goal using either egocentric or allocentric reference frames. We then used continuous wavelet transform to uncover dynamics of theta power modulation in medial temporal lobe, parietal lobe and hippocampus in relation to movement onsets, speed and type of conducted task in both real and virtual settings. Our research is the first monitoring iEEG activity in humans during free movement as well as effectively differentiating egocentric and allocentric navigation and their possible interfacing. Acknowledgments: Supported by GAcR 16-07690S and RVO: 67985823
Sunday 16.00 - 17.30 Grote zaal Poster #11 Attention Lara Roesler
All authors & affiliationLara Roesler, Department of Psychology, Julius Maximilians University of Wurzburg, Wurzburg, Germany Albert End, Department of Systems Neuroscience, University Medical Center Hamburg-Eppendorf, Hamburg, Germany Matthias Gamer, Department of Psychology, Julius Maximilians University of Wurzburg, Wurzburg, Germany, Department of Systems Neuroscience, University Medical Center Hamburg-Eppendorf, Hamburg, Germany
Orienting towards social features in naturalistic scenes is reflexive
Show/hide abstractPrevious eye-tracking studies have shown that observers prioritize social information (i.e. human heads or bodies) over low-level physically salient elements when freely viewing an image. These findings are in contrast to saliency-based models of visual attention which suggest that attention is automatically oriented to image regions with high physical saliency. What remains unknown is whether the observed prioritization of human features is a voluntary mechanism following an initial reaction to highly salient scene elements or takes place reflexively. To investigate early stages of social attention in more detail, we presented naturalistic scenes with or without social features for 200 ms while participants' eye movements were recorded. Analyses revealed significantly more first eye movements to areas containing social features than a chance level distribution of saccades would suggest. Furthermore, using a generalized linear mixed model, we showed that the social content of an image region is a better predictor of saccade direction than its physical saliency suggesting that social features partially outweigh the influence of low-level saliency on gaze patterns. Considering the brief image presentation time, our findings provide persuasive evidence for a reflexive component in social attention. Moreover, the current study accentuates the importance of taking social influences into consideration for a more coherent understanding of human attentional selection.
Sunday 16.00 - 17.30 Grote zaal Poster #12 Language Carmen Kung
All authors & affiliationElaine Schmidt, Cambridge University Peter de Lissa, University of Fribourg Anne Castles, Macquarie University Sachiko Kinoshita, Macquarie University Blake Johnson, Macquarie University Katherine Demuth, Macquarie University
Children don't use subvocal prosody to process comma: Evidence from concurrent eye-tracking and EEG
Show/hide abstractAccumulating ERP evidence suggests that readers rely on subvocal prosody (an ‘internal voice') to parse commas during on-line processing (Drury, Baum, Valeriote, & Steinhauer, 2016). However, it is not known whether children process commas like adults. This is primarily due to the unsuitability of the experimental paradigm known as RSVP (rapid serial visual presentation) a computer-controlled word-by-word paradigm typically used in adult studies. Compared to natural reading, RSVP imposes a heavier working memory load, and thus more processing difficulties since readers cannot regress to an earlier point in the text (Schotter, Tran, & Rayner, 2014). As an alternative to RSVP, the current ERP study used a natural-reading paradigm to examine the processing of commas by 24 9-11-year-old children: We presented participants with 100 sentences containing temporarily ambiguous transitive/intransitive verbs (half containing comma violations) while co-registering their eye movements and brain responses. If children process commas like adults, we expect similar ERP effects reported by previous adult studies: a closure positive shift (CPS) to commas indicating a perception of commas as prosodic boundaries, and an N400-P600 effect reflecting processing difficulty elicited by comma violations. However, the results showed only an N400 effect to the intransitive verb and an N400 to the following noun, reflecting children's difficulties in processing intransitive structures per se. The absence of the expected ERP effects raised the question of whether our participants were aware of the comma violations. An examination of their eye-tracking data showed that they did: Compared to correct sentences, children showed longer dwelling time at the violations and more regressions from violations to earlier parts of sentences. Taken together, the results demonstrate that children—unlike adults—do not rely on subvocal prosody to parse commas, even though they notice comma violations.
Sunday 16.00 - 17.30 Grote zaal Poster #13 Sensory processing Sasskia Bruers
All authors & affiliationSasskia Brueers, CerCo UMR 5549 CNRS Universite de Toulouse 3 Rufin VanRullen, CerCo UMR 5549 CNRS Universite de Toulouse 3
Does alpha power modulate perception independently of endogenous factors?
Show/hide abstractThe amplitude of ongoing parieto-occipital EEG oscillations in the alpha band predicts visual perception. Moreover, it also co-varies with other endogenous factors such as attention, vigilance or alertness. In turn, these endogenous factors are known to influence visual perception. Therefore, it remains unclear how much of the relation between alpha and perception is indirectly mediated by endogenous factors, and how much reflects a direct causal influence of alpha rhythms on sensory neural processing. Here, we aim to disentangle the two causal routes (direct/indirect) by introducing modulations of alpha amplitude, independently of any fluctuations in endogenous factors. To this end, we use white-noise sequences to constrain the brain activity of 20 participants. In a first step, we apply a cross-correlation between white-noise luminance sequences and the concurrently recorded EEG to extract impulse response functions (a model of the systematic relationship between stimulation and brain response). These are then used in a second step to reconstruct (rather than record) the brain activity linked with new sequences (by convolution). Interestingly, this reconstructed EEG only contains information about oscillations directly linked to the white-noise stimulation; fluctuations in attention and other endogenous factors may still modulate EEG alpha rhythms during the task, but our reconstructed EEG signal is impervious to these factors. We found that the detection of near-perceptual threshold targets embedded within the white-noise sequences depended on the ~10Hz amplitude of the reconstructed EEG over parieto-occipital channels (p=0.00018, cluster correction): around target onset, higher amplitude led to poorer performance. Thus, fluctuations in alpha power (induced here by random luminance sequences) can directly influence perception. Though it is possible that these fluctuations would also causally modulate attention, we can nonetheless rule out a causal "primacy" of endogenous factors: the relation between alpha power and perception is not a mere consequence of fluctuations in endogenous factors.
Sunday 16.00 - 17.30 Grote zaal Poster #14 Decision-making Johnny King Lau
All authors & affiliationJohnny King, Lau, School of PCLS & CINN, University of Reading, UK; Anthony Haffey, School of PCLS, University of Reading, UK; Kei Kuratomi, Research Institute, Kochi University of Technology, Japan; Hiroki, Ozono, Faculty of Law, Economics, and Humanities, Kagoshima University, Japan; Asuka, Komiya, Graduate School of Integrated Arts and Science, Hiroshima University, Japan; Kou, Murayama, School of PCLS & CINN, University of Reading, UK
When the seductive power of curiosity overrides prospective risk -- the underlying neural mechanism
Show/hide abstractWhy would Pandora open the box, even after she was warned? This study examined how curiosity biases decision-making, even in the face of physical risk (e.g. expecting electric stimulation), and evaluated the relevant underlying neural mechanisms using fMRI. To induce curiosity, we pilot-tested a set of videos of magic tricks performed by professional magicians. A magic trick shows an event that appears impossible, thus an ideal material to create strong knowledge gap as a source of curiosity [1]. Thirty-one healthy, right-handed participants (5 male; age: 18-30yos) were presented with magic trick videos (n=36) and images of food in a 3T Siemens scanner. In every trial, after the presentation of a fixation and a video, participants were asked to rate how curious they were to know the solution. Then, they were shown a wheel of fortune representing a lottery which visualized the probability of winning (and losing), and were asked to decide whether to gamble. If they gambled and won, they were provided with a ticket to see the solution. They were instructed that if they lost, they would receive a mild electric shock after the experiment. Participants could also opt to skip the lottery. Prior to the fMRI experiment, we evaluated the paradigm with a separate participant group. Behavioural data were fitted into a generalised linear mixed-effects model, which showed that increased probability of expecting a reward, as well as curiosity, increased individual's tendency to take risk (corroborating past findings e.g. [2]). Neuroimaging analysis (using SPM12) compared the BOLD signals of the ‘accepted' and ‘rejected' trials at the time of decision making. In particular, the accepted (versus rejected) magic trials were associated with greater striatal activity, commonly regarded as part of the brain's reward system[3].
Sunday 16.00 - 17.30 Grote zaal Poster #15 Memory Elisabeth Friedrich
All authors & affiliationElisabeth V. C. Friedrich, Research Unit Biological Psychology, Department of Psychology, LMU Munich, Leopoldstr. 13, 80802 Munich, Germany* Anna Lena Biel, Research Unit Biological Psychology, Department of Psychology, LMU Munich, Leopoldstr. 13, 80802 Munich, Germany Paul Sauseng, Research Unit Biological Psychology, Department of Psychology, LMU Munich, Leopoldstr. 13, 80802 Munich, Germany
Fronto-Medial tACS but not tDCS has an Effect on Working Memory Processes
Show/hide abstractCortical slow oscillations in the theta frequency range (4-8 Hz) in the medial prefrontal cortex have been associated with executive control of working memory processes. Moreover, it was shown that this frontal midline theta activity can be transiently modulated by transcranial Direct Current Stimulation (tDCS) placing an electrode over a fronto-medial stimulation site (Fz) and the return electrode on the chin. The goal of this study was to investigate whether this fronto-medial stimulation montage also has an effect on working memory performance using (1) tDCS and (2) transcranial Alternating Current Stimulation (tACS). First, 84 adults performed a 2-back task before and after tDCS was applied. Participants were divided in three groups in which either anodal, cathodal or sham stimulation was applied. It was predicted that tDCS with the anode over Fz (and cathode on the chin) should lead to a better performance compared to when the cathode is placed over Fz (and anode on the chin). The difference in d' between the baseline and the post-tDCS block was calculated. No significant differences between the three groups in working memory performance were found. Second, tACS was applied at theta frequency (6 Hz) to 59 adults with the same montage. Participants performed a 2-back task before and during they received either real or sham tACS. The hypothesis was that real stimulation leads to better performance than sham stimulation. The difference in d' (delta d') between the baseline and the stimulation block showed a trend for a significant group difference in the expected direction. Delta d' was significantly different from zero in the real stimulation, but not so in the sham condition. To conclude, tDCS with this fronto-medial stimulation had no effect on working memory performance, whereas tACS at theta frequency could improve performance.
Sunday 16.00 - 17.30 Grote zaal Poster #16 Attention Bronagh McCoy
All authors & affiliationBronagh McCoy, Vrije Universiteit Amsterdam Jan Theeuwes, Vrije Universiteit Amsterdam
Overt and covert attention to location-based reward
Show/hide abstractRecent research has examined the impact of location-based reward on attentional orienting, claiming that reward factors play an important role in spatial priority maps (Chelazzi et al., 2014). The current study investigated whether and how reward associations based on spatial location translate from overt eye movements to covert attention. We hypothesized that if reward associations can be tied to different locations in space, and if overt and covert attention rely on similar overlapping neuronal populations, then both overt and covert attentional measures should display similar spatial-based reward learning. Our results demonstrate that location- and reward-based improvements in one domain do not lead to similar improvements in the other. Specifically, although we found the greatest improvement at a high reward location during overt attentional learning, this translated to the greatest impairment at that location during the subsequent covert task. We interpret this as resulting from an increased link between the high reward location and the learned eye movement response, leading to an increase in manual response time during covert attention when the eyes had to remain fixated and the learned eye movement response was suppressed. We conclude that the learning of location-based reward is intimately linked to the learned response type, leading to alternative outcomes in different task contexts.
Sunday 16.00 - 17.30 Grote zaal Poster #17 Electrophysiology methods Sandon Griffin
All authors & affiliationArjen Stolk, Helen Wills Neuroscience Institute, University of California, Berkeley, Berkeley, CA 94720, USA Sandon M. Griffin, Helen Wills Neuroscience Institute, University of California, Berkeley, Berkeley, CA 94720, USA Roemer van der Meij, Department of Cognitive Science, University of California, San Diego, La Jolla, CA 92093, USA Callum Dewar, Helen Wills Neuroscience Institute, University of California, Berkeley, Berkeley, CA 94720, USA Ignacio Saez, Helen Wills Neuroscience Institute, University of California, Berkeley, Berkeley, CA 94720, USA Jack J. Lin, Department of Neurology, University of California, Irvine, Irvine, CA 92697, USA Giovanni Piantoni, Massachusetts General Hospital and Harvard Medical School, Boston, MA 02114, USA Jan-Mathijs Schoffelen, Donders Institute for Brain, Cognition, and Behaviour, Radboud University, 6500 HB Nijmegen, The Netherlands Robert T. Knight, Helen Wills Neuroscience Institute, University of California, Berkeley, Berkeley, CA 94720, USA Robert Oostenveld, Donders Institute for Brain, Cognition, and Behaviour, Radboud University, 6500 HB Nijmegen, The Netherlands
An integrated platform for human intracranial data analysis
Show/hide abstractThe exquisite spatiotemporal precision of human intracranial EEG recordings (iEEG) permits characterizing neural processing with a level of detail inaccessible to scalp-EEG, MEG, or fMRI. However, the same qualities that make iEEG an exceptionally powerful tool for investigating the brain also present unique complexities for researchers [1,2]. Until now, the fusion of anatomical data (MR and CT images) with electrophysiological data and its subsequent analysis has relied on technologically challenging and error-prone combinations of software. Here, we present an integrated platform to address the series of challenges associated with human iEEG analysis, in order to guide a rapidly expanding scientific audience from the multitude of raw data files to integrated observations, fast and efficiently. The platform is fully embedded within an open-source toolbox for non-invasive electrophysiological data analysis (FieldTrip [3]). This allows iEEG researchers to build on a continuously growing set of analysis methods that, over the past decade, have been employed by a large research community. The platform consists of two parallel but interrelated workflows. The first workflow entails the processing of anatomical data, including image preprocessing, fusion of the CT with the MRI, electrode placement, cortical surface rendering, brain-shift compensation [4], anatomical labeling, and spatial normalization. The second workflow focuses on improving the signal-to-noise ratio of the functional data while optimally preparing it for follow-up analyses, and encompasses data segmentation, digital filtering, identification of non-physiologiccal artifacts and epileptiform activity, and remontaging. The integration of the two workflows culminates in an interactive and anatomically informed data exploration tool and facilitates the subsequent representation of neural activity overlaid on cortical and subcortical surface models, in figure or video format. Overall, our aim for the platform is to further the state-of-the-art in the field of human iEEG while serving as a resource for both newcomers and experts in the field.
Sunday 16.00 - 17.30 Grote zaal Poster #18 Sensory processing Bertille Somon
All authors & affiliationBertille Somon, ONERA The French Aerospace Lab, Information Processing and Systems Department, Salon Cedex Air, France Aurelie Campagne, Univ. Grenoble Alpes, CNRS, LPNC UMR 5105, F-38000, Grenoble, France Arnaud Delorme, Centre de recherche Cerveau & Cognition, Pavillon Baudot, Hopital Purpan, BP-25202, Toulouse, France Bruno Berberian, ONERA The French Aerospace Lab, Information Processing and Systems Department, Salon Cedex Air, France
ERP study of performance monitoring during execution and supervision
Show/hide abstractIncrease of automation in our everyday-life drastically modified our activities. People who were previously performing tasks, now supervise them. Performance or error-monitoring is characterized by several Event-Related Potentials (ERPs): ERN, CRN, Pe and FRN. They have been identified and replicated both in error execution and in other agent supervision tasks. In our ERP study we propose to assess whether the neurophysiological correlates of performance monitoring during system supervision are similar to those observed for other agent supervision, and error execution. Participants performed a modified flanker task with feedback. They had to identify the orientation of a vertical arrow going upwards or downwards in the presence or absence of flanking arrows (difficult or easy condition respectively). For each task difficulty three types of task were performed -- execution, human agent supervision and system supervision -- on 3 separate days. Electroencephalographic data (EEG) was recorded and mean ERPs were identified across participants for each task type and each difficulty. Preliminary results show an ERN and FRN at FCz location in the execution task, which peak around 50ms after an error and 250ms after the feedback, respectively. Except for a Pe component observed at Pz location between 200 and 400ms and between 400 and 600ms after error execution and error detection for both supervision types respectively, no error-related ERP previously mentioned is observed. However, we observe another positive wave at FCz peaking at 400ms after error detection for both supervision types. The amplitude of all components is reduced with increasing task difficulty, except for the FRN going oppositely. There is no significant difference in ERP amplitude between supervision of a system and human agent for both difficulties. These results are a first step for the evaluation of the out-of-the-loop (OOL) performance problem, and the degradation of error-monitoring, in more or less complex situations.
Sunday 16.00 - 17.30 Grote zaal Poster #19 Language Yingying Tan
All authors & affiliationYingying Tan, Max Planck Institute for Psycholinguistics Randi Martin, Department of Psychology, Rice University
Interference and Executive Control in Sentence Comprehension: An ERP Study of Sentence Comprehension in Chinese
Show/hide abstractInterference during sentence comprehension occurs when readers retrieve earlier information to integrate with later information and intervening material matches target information on semantic or syntactic properties, resulting in slower processing and less accurate comprehension. However, the interaction of semantic and syntactic interference, and the underlying executive control mechanisms supporting each process remained unclear. To address these two issues, the present study recorded EEG signal during the reading of Chinese sentences containing semantic and/or syntactic interference, and related the interference effects to subjects' performance on nine cognitive ability measures, including working memory, executive control, and vocabulary tasks. Forty native Chinese speakers read sentences while the EEG was recorded. Interference when retrieving the subject of the main verb was manipulated in a 2 (high-low semantic) x 2 (high-low syntactic) design. English translations of example sentences are shown below. The semantic manipulation varied the plausibility of the distracting noun as the subject of the verb (as shown in brackets) and the syntactic manipulation varied whether the distracting noun was a subject or object. Low-Syntactic interference: The man saw that the scholar who rejected the [conference/visitor] today waited... High-Syntactic interference: The man saw that the scholar who the [conference/visitor] rejected today waited... The ERP results at the main verb showed a LAN-like effect for both syntactic and semantic interference. In addition, syntactic interference was evident in a P600 effect, while semantic interference was evident in a late left anterior negativity (600 -- 800 ms). We suggested that LAN effect reflected conflict detection in the high interference conditions, while the late effects most likely reflected syntactic and semantic revision. In addition, for syntactic but not semantic interference effect, subjects with better inhibition ability showed a smaller late effect (P600), suggesting a role of inhibition underlying sentence comprehension as obtained in some previous studies.
Sunday 16.00 - 17.30 Grote zaal Poster #20 Memory Michel Quak
All authors & affiliationMichel Quak, Ghent University Zachary D. Langford, Ghent University Raquel E. London, Ghent University Durk Talsma, Ghent University
Bilateral but not lateralized posterior slow wave activity reflects feature load in visual working memory
Show/hide abstractA major ongoing debate in visual working memory research concerns the question whether visual working memory capacity is determined only by the number of objects that have to be memorized, or by the number of relevant features contained within these memorized objects. Here, we examine the effect of feature load on visual working memory capacity, change detection sensitivity, and posterior slow wave event-related brain potential (ERP) activity during memory retention using a change detection task with multi-feature objects. Working memory capacity and sensitivity decreased significantly as a function of both the number of objects and the number of features memorized per object. Contralateral Delay Activity (CDA) was strongly sensitive to the number of objects, but not to the number of features. Additional analyses of both ipsilateral and contralateral brain activity, however, revealed a pattern that also reflected feature processing. We conclude that objects as well as features contribute to limitations in visual working memory capacity and that bilateral and lateralized slow wave activity might reflect two separate systems that underlie feature and object processing.
Sunday 16.00 - 17.30 Grote zaal Poster #21 Language Kaisa Lohvansuu
All authors & affiliationKaisa Lohvansuu, Department of Psychology & Jyvaeskylae Centre for Interdisciplinary Brain Research, University of Jyvaeskylae* Jarmo A. Haemaelaeinen, Department of Psychology & Jyvaeskylae Centre for Interdisciplinary Brain Research, University of Jyvaeskylae Paavo H. T. Leppaenen, Department of Psychology & Jyvaeskylae Centre for Interdisciplinary Brain Research, University of Jyvaeskylae
Infant ERPs reveal speech perception as a predictor of reading skills in adolescence
Show/hide abstractDyslexia, a specific reading disorder, is highly heritable. By studying brain responses of infants of dyslexic parents, it is possible to reach infants who will encounter reading difficulties later in school-age. Previous studies show that brain responses (event-related potentials, ERPs) to speech in infants at risk for dyslexia are atypical already at birth, and predictive of pre-reading and emerging reading skills. There are no studies on associations between speech perception deficits at infancy and reading fluency in adolescence, when reading skills have stabilized. However, during early school years reading skills develop at different rates in children, and therefore examining the relationship between early brain responses and reading level in adolescence is important. ERPs to syllables differing in phonemic length presented in an oddball experiment (ka/, probability 12%; /ka:/, probability 88%), were measured at the age of 6 months. The sample consisted of infants at-risk for dyslexia (N = 8, diagnosed as dyslexics at-school age, and N = 8, who became typical readers), and control infants (N = 18). Same participants were studied at 14 years when their reading fluency was assessed. We found that processing of the frequently presented syllable /ka:/ at the right parietal site was significantly different in infants diagnosed as dyslexics at school-age as compared to the typically reading groups. In the dyslexic group, a high correlation between infant brain responses and reading speed at 14 years was found (r = -.893, p < .01). Brain activation at the right parietal site explained 59% of reading speed score at 14 years. Our results show that speech perception is deficient in at-risk infants who later became dyslexics, and this deficit is predictive to their reading speed in adolescence.
Sunday 16.00 - 17.30 Grote zaal Poster #22 Attention Hadj Boumediene MEZIANE
All authors & affiliationDr. Paolo Ruggeri, Laboratory of Experimental Research on Behavior (LERB), Institute of Psychology, Lausanne University, Switzerland Prof. Catherine Brandner, Laboratory of Experimental Research on Behavior (LERB), Institute of Psychology, Lausanne University, Switzerland
EEG dynamics of interference and selective attention: effects of sex and anxiety
Show/hide abstractInterference effect has been highlighted by Stroop tasks where naming the font colour of a printed word slows down reaction time and increases errors when the meaning of the word and the font colour are incongruent. Although sex and anxiety have gained increasing interest over the last decade, their impact upon brain functioning remains scarcely explored. Here, we examine the differences between high and low anxious men and women subgroups in the spectral content of electroencephalography (EEG) recordings during congruent and incongruent trials of a Stroop task. First, a slower response time was observed in women by comparison with men. Event-related synchronization/desynchronization (ERS/ERD) analysis was performed on EEG data for the whole epoch. During the preparation period (1.5 s before the target stimulus onset), the alpha ERS was significantly higher in the right as compared to the left parietal-occipital hemisphere. Moreover, high anxious individuals revealed higher parietal-occipital alpha ERS as compared to lower anxious individuals. During the execution period (between 400-600 ms after target stimulus onset), alpha ERD was found significantly higher in low anxious than in high anxious individuals. In addition, low anxious women had higher alpha ERD than men. In the medial-frontal cortex (FCz electrode), frontal-midline (FM) theta ERS was significantly higher in the incongruent as compared to congruent trials and in Color-Stroop than Word-Stroop task. Higher FM theta ERS was observed in men by comparison with women. These results highlight the possible roles of the anterior cingulate cortex (ACC) in maintaining attentional set and in processing conflict. In addition, the ACC might be a generator of the FM theta rhythm characteristically observed during intense concentration suggesting for both tasks different engagement of cognitive control mechanisms adopted by men and women during task preparation and execution periods depending on the level of anxiety.
Sunday 16.00 - 17.30 Grote zaal Poster #23 Decision-making Jessica Dully
All authors & affiliationJessica Dully, Trinity College Institute of Neuroscience Redmond O' Connell, Trinity College Institute of Neuroscience
Efficiency of Strategic Adaptation in Older Adults in a Perceptual Decision Making Task
Show/hide abstractA well-replicated finding in the domain of cognitive aging is that older adults exhibit slowed response times in choice reaction tasks, including tasks requiring participants to switch between an emphasis on responding as quickly as possible versus responding as accurately as possible. It remains unclear whether this is due to poorer strategic adaptation, some other age-related factor (e.g. delays in sensory encoding or motor execution), or both. In the present experiment, healthy older adults (65 -- 80 years) and younger adults (18 -- 35 years) engaged in a two-alternative contrast discrimination task (n = 30), consisting of two superimposed leftward/rightward gratings which gradually changed in relative contrast. Stimuli were presented under two different speed pressure conditions, imposed via verbal instruction to participants, as well as feedback in the form of points. Instructions to switch between tasks were given on a block-by-block basis, allowing for analysis of task switching performance across relatively long time windows. Young participants showed a greater behavioural effect of speed emphasis, while old participants were less flexible at adapting their speed-accuracy trade-off. Continuous 128-channel EEG data was recorded, to allow for probing of the distinct stages of the sensorimotor hierarchy (sensory encoding, evidence accumulation, and motor preparation). Electrophysiological analyses will explore the neural origins of reduced flexibility on the part of the elderly participants. This will add insight to performance differences between younger and older adults, and will establish an idea of the limitations faced by older adults in this domain.
Sunday 16.00 - 17.30 Grote zaal Poster #24 Sensory processing Bradley Jack
All authors & affiliationBradley N. Jack, UNSW Sydney Mike E. Le Pelley, UNSW Sydney Oren Griffiths, UNSW Sydney David Luque, UNSW Sydney Thomas J. Whitford, UNSW Sydney
Fractionating the electrophysiological brain signatures of action- and context-based prediction-errors
Show/hide abstractThe human brain makes predictions about upcoming sensory input in at least two ways. First, it predicts sensations that result from actions producing them: Self-generated sounds elicit smaller neural responses than externally-generated sounds. Second, it predicts sensations based on past experience and context: Relatively frequent sounds—controls, elicit smaller neural responses than infrequent sounds—deviants. The theory is that predictions prepare sensory and associative cortices to receive and ‘explain away' sensory input, meaning that less cortical activation is required to process predicted sounds than unpredicted sounds; that is, sounds that elicit prediction-error. We set out to determine when, where, and how action- and context-based prediction-errors interact in the brain by comparing electrophysiological responses to sounds that orthogonally varied in terms of the action that produced them (i.e., whether they were self- or externally-generated) and the context in which they were presented (i.e., whether they were controls or deviants). Consistent with previous research, we found that the N1 indexed action-based prediction-errors and that the mismatch negativity (MMN) indexed context-based prediction-errors. We also found that action- and context-based prediction-errors interacted at the P3: Self-generated deviants elicited larger amplitudes than externally-generated deviants—an action-based prediction-error, and self-generated deviants elicited larger amplitudes than self-generated controls—a context-based prediction-error. Using standardized low-resolution brain electromagnetic tomography (sLORETA), we found that the action- and context-based prediction-errors at the P3 shared similar location of neural sources in the postcentral gyrus of the parietal lobe. These results support the view that prediction-error signals are the brain's lingua franca: the common, computational principles underlying interrelated processes of perception and action.
Sunday 16.00 - 17.30 Grote zaal Poster #25 Memory Ann-Kathrin Zaiser
All authors & affiliationAnn-Kathrin Zaiser, Experimental Neuropsychology Unit, Department of Psychology, Saarland University, Saarbruecken, Germany Patric Meyer, Department of Cognitive and Clinical Neuroscience, Central Institute of Mental Health, Heidelberg University, Medical Faculty Mannheim, Mannheim, Germany Department of Psychology, SRH University of Applied Sciences, Heidelberg, Germany Axel Mecklinger, Experimental Neuropsychology Unit, Department of Psychology, Saarland University, Saarbruecken, Germany Regine Bader, Experimental Neuropsychology Unit, Department of Psychology, Saarland University, Saarbruecken, Germany
Rapid Integration of Novel Associations by Means of Fast Mapping - The Role of Feature Overlap
Show/hide abstractTraditional theories of declarative memory suggest that the acquisition of novel associations depends on hippocampal structures. However, an incidental learning procedure called Fast Mapping (FM) might enable rapid incorporation of associations into semantic networks without hippocampal involvement. Yet the underlying mechanisms of learning through FM are poorly understood. In typical FM experiments, a picture of a previously unknown target (e.g., the animal numbat) and a previously known distractor (e.g., a zebra) are presented together with a question referring to an unfamiliar label ("Is the numbat´s tail pointed up?"). By recognizing the distractor, it can be inferred that the label belongs to the target. In a recent study, patients suffering from severe hippocampal damage performed as well as healthy controls in a recognition test if associations were acquired within an FM paradigm but not through explicit encoding (EE). However, patients with lesions extending to extra-hippocampal structures including the perirhinal cortex (PrC) did not benefit from FM (Sharon, Moscovitch, & Gilboa, 2011). The PrC is ascribed a major role in object discrimination when objects share many features (Clarke & Tyler, 2015). As target and distractor in FM paradigms are typically similar, FM learning success might be mediated by PrC activity. If so, associations encoded together with similar distractors (FM high overlap, FMHO) are expected to be integrated better than associations acquired together with dissimilar distractors (FM low overlap, FMLO) and EE. To measure integration, we conducted a semantic priming task with the newly acquired labels serving as primes for semantically related versus unrelated familiar nouns. A priming effect was observed in the FMHO condition but not after FMLO encoding and EE. The marginally significant difference between FMHO and FMLO encoding provides further support for the assumption that successful integration of associations through FM only occurs if target and distractor share many features.
Sunday 16.00 - 17.30 Grote zaal Poster #26 Consciousness Hamed Haque
All authors & affiliationHamed Haque, Helsinki Institute of Life Sciences, Neuroscience Center, University of Helsinki, Finland Muriel Lobier, Helsinki Institute of Life Sciences, Neuroscience Center, University of Helsinki, Finland Matias Palva, Helsinki Institute of Life Sciences, Neuroscience Center, University of Helsinki, Finland Satu Palva, Helsinki Institute of Life Sciences, Neuroscience Center, University of Helsinki, Finland
Effect of detection and discrimination on the neural ignition of conscious access
Show/hide abstractSubjective conscious perception is associated with strong amplification, global ignition, of cortical activity in a distributed network [1, 2] and with gamma-band oscillations [3, 4]. Stimuli may nevertheless induce only partial consciousness, an intermediate between null and full consciousness, where the presence but not the identity of an object can be reported [5]. We hypothesized that partial consciousness is correlated with evoked responses (ERs) in early visual areas and full consciousness with gamma band oscillations and ERs in later visual areas. We recorded magnetoencephalography (MEG) during a visual discrimination task. Participants were shown one of two objects at either low or high contrast (low: 50% detection, high: 77% discrimination rate) and asked to report object identity when the object was perceived. For the low contrast, object identification was at chance level. For the high contrast, object detection was at ceiling. We filtered the source reconstructed data with a broadband 1-45 Hz to estimate ERs and with Morlet filters to estimate oscillations amplitudes. We then contrasted detected with undetected trials for the low contrast and correctly identified with incorrectly identified trials for the high contrast. We estimated statistical significance with t-tests and corrected for multiple comparisons. ERs were stronger for detected compared to undetected trials not only in primary visual areas but also in higher order cortical areas, in line with the global ignition hypothesis. Identification was associated with a stronger ER ~400 -- 500 ms from onset in IT and prefrontal cortex. Both low contrast detection and high contrast discrimination were associated with a decrease in oscillation amplitude in beta and gamma frequencies. In summary, these data suggest that consciousness is graded and can evoke either full or partial consciousness depending on the strength of evoked activity.
Sunday 16.00 - 17.30 Grote zaal Poster #27 Attention Agnieszka Fudali-Czyz
All authors & affiliationRafal Lewkowicz, The Military Institute of Aviation Medicine, Poland Bibianna Balaj, The Department of Psychology, The Faculty of Humanities, Nicolaus Copernicus University in Torun, Poland Piotr Francuz, The John Paul II Catholic University of Lublin, the Department of Experimental Psychology and Perception & Cognition Lab, Poland Pawel Augustynowicz, The John Paul II Catholic University of Lublin, the Department of Experimental Psychology and Perception & Cognition Lab, Poland Pawel Strozak, The John Paul II Catholic University of Lublin, the Department of Experimental Psychology and Perception & Cognition Lab, Poland
The impact of spatially disorienting cues on attentive blank stares in pilots and non-pilots
Show/hide abstractSpatial disorientation (SD) is the loss of the ability to determine position of the airplane and the pilot relative to the ground or to some other aircraft. SD leads to cognitive impairment, possibly by diverting attention resources to regaining orientation. The cognitive task which is important during piloting is change detection. Attentive blank stares (ABS) mean a failure to notice changes in a visual scene, despite looking at the area of change. In our previous study by means of eye fixation-related potentials (EFRP) method we showed that ABS cases are really attentive blank because these cases in comparison to successful change detection were accompanied by lower amplitude of the lambda response. The goal of our present studies was to test the effect of different SD cues on attentive blank stares in pilots and non-pilots. Two experiments were carried out: 1. in the flight simulator GYRO-IPT on a sample of 30 pilots, with head-mounted oculograph, 2. under laboratory conditions with 30 non-pilots, with the EFRP method. In the first experiment pilots get a double task with flight control and change detection. There were six flight profiles with 3 visual and 3 vestibular SD cues and six of their counterparts serving as a baseline. In the second experiment the only task was to detect changes. In half of trials we introduced visual SD cue by means of the optokinetic stimulation. In the first study we showed more ABS cases in profiles with SD cues than control conditions. In the second study we showed decrement of the visual awareness negativity in SD cues conditions than control one. Our studies are a part of project supported by a NCN grant (UMO-2013/09/B0HS6/03266).
Sunday 16.00 - 17.30 Grote zaal Poster #28 Memory Karen Campbell
All authors & affiliationKaren L. Campbell, Department of Psychology, Brock University, Canada Kevin P. Madore, Department of Psychology, Harvard University, USA Roland G. Benoit, Max Planck Institute for Human Cognitive & Brain Sciences, Leipzig, Germany Preston P. Thakral, Department of Psychology, Harvard University, USA Daniel L. Schacter, Department of Psychology, Harvard University, USA
Increased hippocampus to ventromedial prefrontal connectivity during episodic future simulation
Show/hide abstractBoth the hippocampus and ventromedial prefrontal cortex (vmPFC) appear to be critical for imagining future experiences (episodic simulation). Damage to either structure can affect one's ability to imagine the future, and both structures are commonly activated as part of a wider core network during episodic simulation (Benoit & Schacter, 2015). However, the precise role played by each of these structures, and the direction of information flow between them during future imagining, is not well understood. To address these issues, we scanned participants using fMRI while they imagined future events in response to object cues or performed a non-episodic control task (i.e., generating a sentence using the cue word). We identified peak regions in the hippocampus and vmPFC (imagine > object control) and used dynamic causal modeling (DCM) to examine effective connectivity between these nodes (Friston et al., 2003). Bayesian model selection was used to compare model families that differed in the location of modulation by episodic simulation (hippocampus to vmPFC, vmPFC to hippocampus, bidirectional, and no modulation). Our results show that while there was strong bidirectional intrinsic connectivity between these regions throughout scanning, only the hippocampus to vmPFC connection increased during episodic simulation. These findings are in line with recent work showing an initial flow of information from the hippocampus to the PFC during context-guided memory (Place et al., 2016), and suggests that the hippocampus may be driving the integration of event details in the vmPFC.
Sunday 16.00 - 17.30 Grote zaal Poster #29 Attention Maria Viktoria Stuckenberg
All authors & affiliationAndreas Widmann, Institute of Psychology, University of Leipzig, Leipzig, Germany Erich Schroeger, Institute of Psychology, University of Leipzig, Leipzig, Germany
Stimulus probability modulates visually induced auditory expectations whereas task requirements have only minor impact
Show/hide abstractThe human auditory system establishes sensory representations of expected auditory events based on predictive visual information. The amplitude of the N1 auditory evoked potential, observed around 100 ms after sound onset, has been shown to be enhanced in response to unexpected compared to expected but otherwise identical sounds. This amplitude difference was interpreted as enhanced prediction error signal for unexpected sounds relative to expected sounds in the predictive coding framework and termed Incongruency Response (IR). Here we examined the impact of sound probability and task requirements on the elicitation of the IR. Each sound of either high or low pitch was preceded (stimulus-asynchrony 600 ms) by a note symbol either presented above or below a fixation cross. In 90% of the trials the visual symbol correctly predicted the pitch of the upcoming sound (congruent sound), whereas in 10% of the trials the sound was predicted incorrectly (incongruent sound). High and low pitch sounds were presented with equal (50% each) probability in one condition and with different probabilities (83.3 vs. 16.7%) in another condition (balanced across participants). Participants had to discriminate high vs. low pitch sounds in half of the blocks and congruent vs. incongruent sounds in the other half. Results reveal that the task had only minor impact on the IR. Importantly, a significant IR for unexpected (rare visual cue before frequent tone) compared to expected (frequent visual cue before frequent tone) but otherwise identical sounds was confined to the condition with unequal stimulus probabilities. This suggests that it is easier to induce an auditory expectation by a visual cue when an established auditory representation can be re-activated. We assume that the increase of the relative frequency of the stimulation in a trial-by-trial design supports, similar to contextual information or automatization, a strong relationship between visual and auditory stimulation that is required to form a visual based auditory sensory prediction.
Sunday 16.00 - 17.30 Grote zaal Poster #30 Sensory processing Jennifer K. Bertrand
All authors & affiliationJennifer K. Bertrand, Faculty of Physical Education, University of Alberta Nathan J. Wispinski, Department of Psychology, University of Alberta Kyle E. Mathewson, Department of Psychology, University of Alberta Craig S. Chapman, Faculty of Physical Education, University of Alberta
Are neural oscillations responsible for the misperception of flicker brightness?
Show/hide abstractConsider an optical illusion where lines on a printed page appear to be moving. Though you perceive motion, you know the ink is stationary, suggesting that you have some awareness that your current perception is inaccurate. However, instances of misperception also exist that go undetected by the conscious brain. Here, we explore a subconscious misperception of brightness induced by the flicker rate of a visual stimulus. We explore both a profound behavioural effect and the underlying neural activity (via electroencephalography; EEG) with the hypothesis that the neural oscillations induced by different frequencies of flicker are at the root of the misperception. Over 100 years ago, Brucke (1867) showed that the flickering frequency of a light significantly impacted its perceived brightness relative to a non-flickering baseline [1,2]. In two experiments we replicated this study and extended it to judgements between two flickering stimuli. In experiment 1, we presented pairs of stimuli with the same physical luminance, but differing across a range of frequencies (0Hz, 4Hz, 9Hz, 13Hz and 17Hz). We replicate and extend Brucke's behavioural finding, showing that the 13Hz and 17Hz stimuli are perceived as dimmest, the 0Hz and 9Hz stimuli are perceived as having average brightness and, remarkably, despite having the identical luminance, the 4Hz stimulus is perceived as the brightest nearly 80% of the time. In experiment 2 we replicated experiment 1, but targeted a specific range of frequencies (4Hz, 9Hz, 13Hz) and collected EEG data. Preliminary EEG analysis suggests that the amount of phase locking induced by the flicking stimuli is correlated to its perceived brightness -- the more the brain locks to a target's frequency, the brighter it appears. This study therefore presents a potential neural mechanism responsible for brightness misperception that we hope will extend to other biased perceptual decisions, or misperceived stimuli.
Sunday 16.00 - 17.30 Grote zaal Poster #31 Connectivity/networks Johann Philipp Zoellner
All authors & affiliationZoellner Johann Philipp, Cognitive Neuroscience Group, Department of Neurology, University Hospital, Goethe University Frankfurt, Germany Hansmeyer Laura, Cognitive Neuroscience Group, Department of Neurology, University Hospital, Goethe University Frankfurt, Germany Forster Marie-Therese, Department of Neurosurgery, University Hospital, Goethe University Frankfurt, Germany Kropff Ines, Cognitive Neuroscience Group, Department of Neurology, University Hospital, Goethe University Frankfurt, Germany Fuhrmann Silke, Cognitive Neuroscience Group, Department of Neurology, University Hospital, Goethe University Frankfurt, Germany Hok Pavel, Department of Neurology, University Hospital, Palacky University of Olomouc, Czech Republic Kell Christan A., Cognitive Neuroscience Group, Department of Neurology, University Hospital, Goethe University Frankfurt, Germany
Assessing transient disruptions in human functional speech networks following tumor surgery of the dominant hemisphere
Show/hide abstractAphasia is the clinical manifestation of a dysfunctional language network in the speech dominant half of the brain. Aphasia resulting from left hemispheric stroke results in over-recruitment of homologous brain regions in the right hemisphere1. This suggests increased interhemispheric connectivity2 as a compensation attempt. Yet, stroke research does not allow analyzing pre-lesional language networks. In contrast, studying the frequent transient aphasia following tumor surgery3 allows comparing pre- to post-surgery functional networks. Analyzing resting state measurements identifies largely state-independent connectivity changes in neural networks4. We thus performed repetitive resting state fMRI measurements in 7 well-characterized patients undergoing awake tumor surgery of their speech dominant left temporal lobe. None had lasting speech or language symptoms prior to surgery and postoperative aphasia was present in all patients. fMRI datasets were analyzed using a seed-based whole brain functional connectivity analysis, using 6 seeds in a priori defined language regions and their contralateral homologues (pars triangularis, pars opercularis, articulatory motor cortex, premotor cortex, superior parietal-temporal area, superior temporal sulcus). Paired t-tests were used to compare functional connectivity between the pre-surgery scan and the time point of maximal aphasia. Significance was assumed at p<0.05 (FDR-corrected at cluster level, cluster-forming threshold of p<0.001, uncorrected). During postoperative aphasia, intrahemispheric connectivity between frontal and temporal language regions was indeed reduced in the left hemisphere, with the exception of an increased connectivity from left pars triangularis to the left supramarginal gyrus. At the same time, interhemispheric connectivity increased from bilateral temporal lobe seeds to contralateral motor regions. Our results demonstrate that aphasia is indeed associated with a hypoconnection of the left language network. At the same time, our results show a less pronounced hyperconnection in the left dominant hemisphere as well as in interhemispheric connections between temporal and frontal regions during postoperative aphasia, potentially hinting at compensation efforts.
Sunday 16.00 - 17.30 Grote zaal Poster #32 Language Hyojin Park
All authors & affiliationHyojin Park, Institute of Neuroscience and Psychology, University of Glasgow Robin A. A. Ince, Institute of Neuroscience and Psychology, University of Glasgow Gregor Thut, Institute of Neuroscience and Psychology, University of Glasgow Joachim Gross, Institute of Neuroscience and Psychology, University of Glasgow
Neural decomposition of synergy and redundancy in interaction between audiovisual speech rhythms and brain oscillations
Show/hide abstractIn audiovisual speech processing, auditory and visual information interact and are integrated to lead to a unified percept of speech. Previously, we have shown that low-frequency brain oscillations separately track auditory and visual speech signals to facilitate speech comprehension. However, it is still unclear to what extent auditory and visual information is represented in brain areas, either individually or jointly. Here, we applied a recently developed tool from the Information Theory (Partial Information Decomposition; PID) to decompose multivariate mutual information between auditory, visual and brain signals. This method allows quantification of the unique information the brain signals carries of each modality (auditory, visual). Furthermore, we can now address the question if activity in a certain brain area carries a synergistic or redundant representation of both sensory signals. We used low-frequency theta phase of auditory and visual speech signals and brain signals at each voxel measured by MEG. In adverse audiovisual speech condition in which attention to visual speech is critical for speech comprehension, we found redundant information in auditory/temporal region including posterior superior temporal gyrus (pSTG) while synergistic information in left motor and inferior temporal cortex. Importantly, redundant information in the left pSTG and synergistic information in the left motor cortex predict speech comprehension. By means of this novel information theoretic tools, we here show for the first time evidence for neural decomposition of information of entrained audiovisual speech rhythms interacting with brain oscillations for facilitating speech comprehension. Our finding demonstrates how the brain processes audiovisual inputs efficiently-taking advantage of common information as well as making greater information from multisensory inputs that enable remarkable ability in human communication.
Sunday 16.00 - 17.30 Grote zaal Poster #33 Decision-making Tessa Rusch
All authors & affiliationMichael Spezio, Department of Psychology, Scripps College Claremont Prashant Doshi, Department of Computer Science, University of Georgia Athens Jan Glaescher, Institute of Systems Neuroscience, University Medical Center Hamburg
Characterizing Mentalizing-Processes during Cooperative Decisions
Show/hide abstractMentalizing capacities are classically examined with the so called "False Belief Task" (Wimmer & Perner, 1983). We transferred this task's basic concept into the realm of social decision making to investigate how people predict other agents' behavior during value guided choices: Two individuals engage in a simple choice task, in which probabilistic outcomes that also depend on the partner's choice have to be learned. Cooperative choices are highly rewarded. After periods of successful cooperation one player's (Learner's) outcome distribution is reversed, but only the other player (Teacher) is informed about the reversal. The Learner thus has a false belief about the state of the world. To maximize the reward, the Teacher must track how the Learner's false belief evolves over time and make choices to "communicate" the contingency switch to the Learner. During each trial both players make explicit predictions of their partner's choices before making their own. In line with the Learners' false belief, the Teacher initially predicts unchanged behavior for the Learner. However, she switches her own choice. One the one hand, the Teacher's choice switch assures a maximal possible outcome for both players given the Learner's false belief, on the other hand, it informs the Learner about the contingency switch. The Learner detects the Teacher-signal and after accumulating enough evidence reacts to the new reward distribution and adjust his/her choice. The interactive nature of the task and the necessity to track and incorporate the partner's beliefs into the own decision making process allows to examine mentalizing processes in an ecologically valid setting. On a model free level increased mentalizing processes in both players show in vast increases in RTs after reversals and the players' adapted choice behavior. Furthermore, we use an interactive partially observable Markov decision model (I-POMDP, Gmytrasiewicz & Doshi, 2005) to examine value guided mentalizing processes in a fine grained manner.
Sunday 16.00 - 17.30 Grote zaal Poster #34 Memory Tobias Staudigl
All authors & affiliationTobias Staudigl, Donders Institute for Brain, Cognition and Behaviour, Radboud University, Nijmegen, The Netherlands Ole Jensen, Centre for Human Brain Health, School of Psychology, University of Birmingham, Birmingham, UK Christian F. Doeller, Kavli Institute for Systems Neuroscience, Centre for Neural Computation, Egil and Pauline Braathen and Fred Kavli Centre for Cortical Microcircuits, NTNU & St. Olavs Hospital, Trondheim, Norway
Hexadirectional signals during exploration of visual space in human MEG data
Show/hide abstractGrid cells are one of the core building blocks of navigation and play a pivotal role in memory processes. While single cell recordings in rodents have shown 60° periodic representations of space during spatial navigation (Hafting et al., 2005), studies in non-human primates revealed similar hexagonal firing patterns of cells during image viewing (Killian et al., 2012). Human fMRI studies provide evidence that grid-like signals are also accessible on a macroscopic level (Doeller et al., 2010). The present study set out to investigate grid-like signals during visual exploration in human neuromagnetic activity. We simultaneously recorded MEG and eye-tracking data from 36 healthy participants during a free viewing encoding task of natural pictures. MEG data were aligned to saccade onsets and source-level gamma power was extracted during saccadic eye movements. One half of the data was used to estimate the potential grid orientation, applying a quadrature filter approach. In the remaining data, trials aligned to the main grid axis were compared to misaligned trials. We found significantly higher 60° periodic gamma power for aligned versus misaligned trials in the anterior part of the medial temporal lobe. Importantly, control analyses (45° and 90° periodicities) did not show this effect, and the 60° periodic gamma power modulation could also not be explained by oculomotor activity. Our findings show, for the first time, grid-like activity during visual exploration in human neuromagnetic signals. Converging with findings in non-human primates, this speaks to a generalization of hexadirectional signals during spatial navigation to grid-like representations of visual space.
Sunday 16.00 - 17.30 Grote zaal Poster #35 Electrophysiology methods Jessica Loke
All authors & affiliationLoke, University of Amsterdam
Dog or Poodle: Recurrent processing in object categorizatio
Show/hide abstractHumans recognize objects at a remarkable speed. Previous studies have shown that objects can be recognized at less than 30ms. This speed has led theorists to attribute a rapid feed-forward processing. Then, what is the role of recurrent processing and what determines its occurrence? We propose two hypothesis. First, we hypothesize that recurrent processing is necessary for complex stimuli. In our experiment, this hypothesis is explored using occluded and segmented stimuli, with occluded and non-segmented stimuli being more complex stimuli. Second, we hypothesize that recurrent processing is necessary for complex task. In our experiment, this hypothesis is explored using ordinate-level and subordinate-level categorization, with subordinate-level categorization being the more complex task. All stimuli are presented masked and unmasked - masking has been shown to interfere with recurrent processing. Seventy-nine participants from the University of Amsterdam completed our object categorization task while also providing EEG measurements. We found significant effects of masking, stimuli complexity and task complexity. This finding expands our current understanding of object recognition and the role of recurrent processing.
Sunday 16.00 - 17.30 Grote zaal Poster #36 Open science and replication Caspar Goeke
All authors & affiliation1. Holger Finger, University of Osnabrueck, Institute of Cognitive Science, Osnabrueck, Germany 2. Dorena Diekamp, University of Osnabrueck, Institute of Cognitive Science, Osnabrueck, Germany 3. Prof. Dr. Peter Koenig, University of Osnabrueck, Institute of Cognitive Science, Osnabrueck, Germany
Proposing, Testing, and Evaluating a Unified JavaScript Framework for Professional Online Studies
Show/hide abstractRecently online-based research has gained increasing attention from various fields of research in the cognitive sciences. Technological advances in the form of online crowdsourcing (Amazon Mechanical Turk), open data repositories (Open Science Framework), and online analysis (Ipython notebook) offer rich possibilities to improve, validate, and speed up research. However, until today there is no cross-platform integration of these subsystems. Furthermore, online studies still suffers from the complex implementation including server infrastructure, database programming, security considerations. Thus, online studies have not yet lived up to their promise. Here we present LabVanced, a JavaScript framework that constitutes methodological innovation by combining three essential aspects for online research. With our framework studies can be implemented in an intuitive graphical user interface without programming. Second, the framework takes care about participant recruitment. Third, it outlines options for data visualizations and statistical analysis. Additionally, the framework can be used for sharing not only the recorded data, but also the study design and the analysis. This will support the idea of open data repositories, but also increase the validity and transparency of experimental paradigms and offers an opportunity see how the data was analyzed. To demonstrate the functionality of the framework we present the results of a pilot study in the field of spatial navigation that was conducted using the framework. Specifically, we recruited over 2000 subjects with various cultural backgrounds and consequently analyzed performance difference in dependence on the factors culture, gender and age. Overall, our results demonstrate a strong influence of cultural factors in spatial cognition. Such an influence has not yet been reported before and would not have been possible to show without the massive amount of data collected via our framework. In summary, we conclude that our new framework constitutes a wide range of advantages improving and accelerating online research.
Sunday 16.00 - 17.30 Grote zaal Poster #37 Decision-making Friederike Behrens
All authors & affiliationFrançois Quesque Dr, University of Lille Mariska Kret Dr, University of Leiden
The Pupil - The Window to Judging Social Intentions
Show/hide abstractHumans are well adapted to reading other's emotions and intentions from various signals and based on these, decide to cooperate or compete. Studies have shown that humans' personal or social intentions are reflected in their kinematics and that observers can classify intentions purely based on that. Additionally, these signals resonate within the observer's own body. For instance, the pupil has been reported to react to another's facial or bodily expressions of emotion and can reflect a person's decision. The current study investigated whether people could recognize another's personal or social intentions from their kinematics by i) asking about the observed intentions explicitly and ii) measuring observer's pupil size. Specifically, participants classified videos from an interaction game, where one player placed an object between him and another player (preparatory action) and subsequently, the object was placed aside by either the same (personal intention) or the other player (social intention). The classification videos showed the arm movements of the preparatory action and participants classified the movements as showing a personal or social intention while their pupil size was measured. Replicating previous studies, participants could correctly recognize whether another's intentions were personal or social based on subtle changes in kinematics only. Intriguingly, these recognitions were manifested in the participants' pupil size. Independent of luminance, pupils dilated more and earlier in response to personal compared to social intention videos, especially when participants responded intuitively. Interestingly, an observer's decision was already reflected in the pupil size before the explicit response was made, increasing the probability of seeing a social intention with larger pupils. This research shows an alternative route towards people's interpretation of a social situation without having to explicitly ask to categorize that situation. This can have potential applications in clinical research or research in non-/pre-verbal populations (i.e. autistic individuals, infants or other species).
Sunday 16.00 - 17.30 Grote zaal Poster #38 Decision-making Laetitia Grabot
All authors & affiliationLaetitia Grabot, CEA-Inserm, NeuroSpin, Unicog Tadeusz Kononowicz, CEA-Inserm, NeuroSpin, Unicog Tom Dupre La Tour, LTCI, Telecom ParisTech, Universite Paris-Saclay Alexandre Gramfort, LTCI, Telecom ParisTech, Universite Paris-Saclay Valerie Doyere, Paris-Saclay Intitute of Neuroscience (Neuro-PSI), Cognition & Behavior, Paris Sud University, CNRS, Paris Saclay University Virginie van Wassenhove, CEA-Inserm, NeuroSpin, Unicog
Alpha/beta nesting during time production and meta-cognition.
Show/hide abstractHumans can both estimate a duration and evaluate their self-performance. The exact neurocognitive mechanisms underlying these abilities remain elusive. One the one hand, recent evidence has involved beta power during time estimation (Kononowicz et al., 2015; Kulashekhar et al, 2016 et al, 2016) and on the other, temporal attention is indexed by the phase of alpha oscillations (Rohenkohl et al., 2011) and temporal expectation by the phase-amplitude-coupling (PAC) of theta/beta (Cravo et al., 2011). Additionally, recent studies have shown that alpha/gamma PAC plays a role in the allocation of attention in visual tasks (Voyteck et al., 2010; Roux & Uhlhaas, 2013), and theta/gamma PAC has been proposed as a mechanism for temporal expectation (Henry & Herrmann, 2014). Here, we explored the hypothesis that alpha/beta PAC may be implicated in a task requiring time production and temporal self-estimation (i.e. attention to one's time performance). We exhaustively tested all possible functional oscillatory nesting in a time production task in which participants produced 1.45 s time intervals by pressing a button twice followed by self-evaluation on their temporal performances. PAC (Tort et al., 2009) was computed during time production, i.e. between the two button presses. PAC was also tested as control before the first button press, to test for motor preparation. A robust and significant alpha/beta PAC was readily observed for each individual (n=12). The alpha/beta coupling was specific to the timing production task but the coupling strength did not predict performance or the self-estimated duration. We will discuss the role of PAC in time estimation in the context of current proposals for temporal multiplexing (Gu et al., 2015) and the more general idea of the maintenance of current sensorimotor state (Engel & Fries, 2010).
Sunday 16.00 - 17.30 Grote zaal Poster #39 Consciousness Camile Corrêa
All authors & affiliationCamile M.C. Corrêa, Department of Psychology, University of Amsterdam, Amsterdam, The Netherlands. Jan Willem de Gee, 1. Department of Psychology, University of Amsterdam, Amsterdam, The Netherlands. 2. Department of Neurophysiology and Pathophysiology, University Medical Center Hamburg-Eppendorf, Hamburg, Germany Matthew Weaver, Department of Psychology, University of Amsterdam, Amsterdam, The Netherlands. Tobias H. Donner, 1. Department of Psychology, University of Amsterdam, Amsterdam, The Netherlands. Department of Neurophysiology and Pathophysiology, 2. University Medical Center Hamburg-Eppendorf, Hamburg, Germany. 3. Amsterdam Brain & Cognition, University of Amsterdam, Amsterdam, The Netherlands. Simon van Gaal, 1. Department of Psychology, University of Amsterdam, Amsterdam, The Netherlands. 2. Amsterdam Brain & Cognition, University of Amsterdam, Amsterdam, The Netherlands
Feedback-related pupil dilation and P3 waveforms encode prediction error responses only when feedback is conscious
Show/hide abstractThe extent to which subjective awareness is necessary to process performance feedback, and how the confidence in our decisions modulates the impact of this feedback is largely unknown. To test this issue, 32 participants performed a challenging perceptual task in which participants had to judge the direction of a noisy Gabor stimulus while performance was staircased at 70% correct. On each trial, participants had to indicate the confidence in their decision, after which veridical feedback was presented (using the Dutch words for "wrong" and "correct"). Crucially, on half of the trials, the feedback word was masked from awareness. During task performance EEG and pupil size were recorded. After conscious feedback, both pupil size and the P3 mirrored a prediction error response, as reflected in an interaction between confidence (high/low) and feedback valence (error/correct). Further, a single-trial regression analysis of pupil diameter to electrophysiological brain activity at the time of feedback presentation revealed that pupil size scaled with P3 amplitude on a trial-by-trial basis. Interestingly, after masked feedback, the data pattern looked rather differently, and showed no sign of a prediction error response, although there were stronger modulations on error versus correct trials, as well as for low versus high confident trials. These results suggest that confidence modulations on outcome processing depend on feedback awareness, which may reflect the necessity of consciousness in integrating confidence and performance outcomes to update and optimize future decisions.
Sunday 16.00 - 17.30 Grote zaal Poster #40 Attention Stephen Whitmarsh
All authors & affiliationStephen Whitmarsh, ecole normale superieure Jerome Sackur, ecole normale superieure Catherine Tallon-Baudry, ecole normale superieure
Stomach-Brain coupling co-varies with spontaneous fluctuations in attention
Show/hide abstractMechanisms by which the stomach exerts an influence on the brain are numerous. Here, we focus on direct coupling between the brain and the slow gastric rhythm, intrinsically generated in the stomach at a very low frequencies (approx. 0.05Hz), potentially within the time-scale of cognitive processes. Evidence for direct stomach-brain coupling comes from Richter et al. (2017), who used electrogastography to measure gastric pace-maker cells and brain activity using MEG during resting-state. The amplitude of cortical oscillations in the alpha range (10-11Hz) was shown to be coupled to the phase of the gastric rhythm due to an ascending influence from the stomach. In the current study we set out to test for the first time whether differences in stomach-brain coupling occur in tandem with changes in cognitive performance. Given the known role of alpha oscillations in attention, we designed a task with long (10-30s) trials in which subjects reported fluctuations of attention during tactile detection, and recorded both brain and stomach activity. The ability of subjects to accurately report on tactile attention (Whitmarsh et al. 2013, 2016) was confirmed by objective neural correlates: both contralateral somatosensory alpha and evoked response correlated with subjective reports. Stomach-brain coupling, measured as the strength of the coupling between the phase of the gastric rhythm and occipito-parietal alpha at 10-11Hz, was increased when subjects reported to be distracted from the task. These findings provide the first evidence that the stomach-brain coupling has a direct influence on cognition.
Sunday 16.00 - 17.30 Grote zaal Poster #41 Sensory processing David Richter
All authors & affiliationDavid Richter, Radboud University Nijmegen, Donders Institute for Brain, Cognition and Behaviour. Matthias Ekman, Radboud University Nijmegen, Donders Institute for Brain, Cognition and Behaviour. Floris P. de Lange, Radboud University Nijmegen, Donders Institute for Brain, Cognition and Behaviour.
Dampened sensory response to predictable stimuli throughout the ventral visual stream
Show/hide abstractPrediction plays a crucial role in perception, as prominently suggested by predictive coding theories (Friston, 2005). Previous neurophysiological research in macaques has shown expectation suppression for predicted object stimuli in object-selective inferior temporal (IT) cortex (Meyer & Olson, 2011). However, the generalizability of this to humans remains uncertain, with one study showing an increased response in object-selective lateral occipital cortex (LOC, the putative homologue of monkey IT) to predictable sequences of objects (Turk-Browne et al., 2009). In the present study we set out to investigate the neural basis of expectation suppression for object stimuli throughout the visual hierarchy using a rapid event-related fMRI design. Participants (n=24) were exposed to pairs of naturalistic object images in a statistical learning paradigm, in which the first object correctly predicted the identity of the second object on most trials. Image transitions were not task relevant; thus all learning of statistical regularities was incidental. We found that neural responses to expected compared to unexpected stimuli are suppressed throughout the ventral visual stream, from the primary visual cortex (V1), object-selective LOC, to temporal occipital fusiform cortex and the posterior parahippocampal gyrus. Additionally, we present tentative evidence that expectation suppression may be modulated by the strength of conditional probability (i.e., the prediction strength). Taken together, our results suggest that expectation suppression is a widespread neural mechanism that is visible throughout the visual hierarchy. Since tuning for specific object stimuli only arises later in the visual cortical hierarchy, this widespread expectation suppression suggests that high-level object expectations feed back to modulate feature processing at lower levels in the visual cortical hierarchy, in line with a predictive coding framework of perception.
Sunday 16.00 - 17.30 Grote zaal Poster #42 Sensory processing Jade Frost
All authors & affiliationJade D. Frost, School of Psychology, University of Newcastle, Australia. Kelly McDonnell, School of Psychology, University of Newcastle, Australia. Istvan Winkler, Institute of Cognitive Neuroscience and Psychology, Research Centre for Natural Sciences, MTA, Budapest, Hungary Juanita Todd, School of Psychology, University of Newcastle, Australia.
Cognitive and attentional demand abolishes first impressions during early auditory relevance filtering
Show/hide abstractThe predictive coding framework is proving to be a powerful theoretical platform for testing hypotheses about the mechanisms underlying sensory inference and elucidating how the brain uses experience to predict future states whilst remaining sensitive to change. The mismatch negativity (MMN) component of the event-related potential (ERP) is used to measure the brains response to varying sound probability across time and can be used to infer the mechanism by which we filter sound relevance. The presence of MMN indicates that predictions about future states based on experience have been violated by unexpected input, whereas its amplitude can index the strength of underlying predictions. Though many models integrate a framework for interpreting the MMN under conditions in which its elicitation mirrors the statistical properties of the environment, less is understood about how contextual cues modulate the MMN signal. In our lab, participants typically hear the multi-timescale sequence comprising both local probabilistic and contextual information whilst watching a film. In several published studies, we show that MMN succumbs to a first impression bias during relevance filtering that gives rise to powerful distortions in later learning that do not reflect probabilistic and transitional statistics only. In the present study, we test two hypotheses about the hierarchical network resources presumably required for bias in learning to emerge and provide evidence that distortions in learning are no longer observed if participants hear the multi-timescale sequence whilst concurrently engaged in either: 1) a cognitively-demanding working memory task, or 2) a simple perceptual task that requires attentional resources. Successfully abolishing patterns of bias shows that learning over longer timescales is sensitive to manipulation of working memory resources that impact attentional capacity. Our data support the notion that auditory sensory inferences is hierarchical and engages a network that can be disrupted by the concurrent demands on attention resources.
Sunday 16.00 - 17.30 Grote zaal Poster #43 Sensory processing Tamar Regev
All authors & affiliationTamar Regev, Edmond and Lily Safra Center for Brain Sciences, The Hebrew University of Jerusalem * Israel Nelken, Department of Neurobiology, and Edmond and Lily Safra Center for Brain Sciences, The Hebrew University of Jerusalem Leon Y. Deouell, Department of Psychology, and Edmond and Lily Safra Center for Brain Sciences, The Hebrew University of Jerusalem
Automatic representation of pitch in human auditory cortex is linear and not helical
Show/hide abstractThe perceptual organization of pitch is frequently described as helical, with a constantly increasing dimension of height, and a circular dimension of pitch class (chroma) (e.g. 1, 2, 3). While height representation has clear manifestations in neural activity, the way in which chroma representation is manifested in the brain is currently debated. We tested whether chromatic regularity - the same musical note distributed across octaves - is detected automatically by the human brain. We used the mismatch-negativity (MMN), an ERP component indexing automatic detection of deviations from auditory regularity. In two experiments, musicians first trained to classify pure or complex tones across 4 octaves, based on chroma -- C versus G (27 subjects, Experiment 1), or C vs. F# (30 subjects, Experiment 2). Next, they were passively exposed to two MMN protocols, while watching a silent video: a classic 2-tone protocol of 80% standard tones and 20% pitch-height deviants, and a novel 5-tone protocol, designed for testing chroma deviation -- the standard was equally divided between Cs from 4 octaves (20% each), and the middle G or F# (20%) served as deviant. Finally, in an 'attend-chroma' block the 5-tone sequence was presented and subjects had to press a button each time the chroma deviant appeared. Subjects accurately detected the deviant tones in the 'attend-chroma' block. However, in the passive blocks, a significant MMN was found only to pitch-height deviation in the 2-tone sequences but not to chroma deviation, even for perfect performers in the active task. Additionally, early (~100 ms) N1 responses followed the distribution of pitch height rather than chromaticity. These results indicate that at the level of pre-attentive, automatic processing, the brain represents pitch height and not chroma even in trained musicians. Processing the musical dimension of chroma might require more elaborate cognitive processes, such as attention and working memory.
Sunday 16.00 - 17.30 Grote zaal Poster #44 Language Xueqiao Li
All authors & affiliationXueqiao Li, Department of Psychology, University of Jyvaeskylae Elina Haimi, Department of Psychology, University of Jyvaeskylae Janne Pesonen, Department of Psychology, University of Jyvaeskylae Iris Juvenius, Department of Psychology, University of Jyvaeskylae Emilia Tuhkanen, Department of Psychology, University of Jyvaeskylae Piia Astikainen, Department of Psychology, University of Jyvaeskylae
Preserved Processing of Irony in Depression: An ERP Study
Show/hide abstractUnderstanding the cognitive bias of depression and its neural mechanisms is expected to stimulate potential new treatment strategies that are needed for patients suffering from this disabling condition. However, it is surprising that few studies have focused on depression-related dysfunction in processing of humorous contents, which is an important high-level cognitive activity that plays a crucial role in social life and is meaningful for emotional wellbeing. Ironic humor that is recognized as the negative style of humor was applied in the present study. Participants were presented with a spoken dyadic conversation consisted of an introductory sentence and a punchline paired with a picture. The punchline was either congruent or incongruent with the picture, the latter providing an ironic meaning. Each sentence was paired with two different pictures enabling the comparison of responses to identical sentences in two conditions, i.e. neutral and ironic conditions. Event-related potentials (ERPs) and behavioral responses were measured from 20 depressed and 22 non-depressed participants. On the first day, participants were not required to respond to the stimuli, but on the second day they were instructed to evaluate the funniness