Poster Session


Foyer MPI CBS, Leipzig
Stephanstr. 1A, 04103 Leipzig

Poster Session I

Tuesday, 26 June , 17:30-18:45

Poster Number


Title with Abstract

Neuroimaging Physics and Signal Processing


Brammerloh, M.

Biophysical modelling of iron-induced MRI contrast in the brain

Pathologic iron accumulation in the human brain is a biomarker and a potential cause of several neurodegenerative diseases. In Parkinson’s disease (PD), iron overload in dopaminergic preceeds neuron loss in substantia nigra. Also in Alzheimer’s disease (AD) and Multiple Sclerosis (MS), iron is connected to the pathogenesis. Magnetic Resonance Imaging (MRI) provides a promising method for non-invasive brain iron quantification in vivo. Accumulation of paramagnetic iron is revealed by changes in the transverse (R2) and effective transverse relaxation rates (R2*). A quantitative understanding of iron’s effect on these MRI parameters could be used to develop early stage biomarkers for AD, PD and MS. However, thus far only simplified models have been used for iron quantification in vivo, disregarding the chemical form of iron and its microscopical distribution. We developed a model of iron-induced R2* contrast in substantia nigra as a step toward diagnosing PD. This model predicts contributions of different chemical forms of iron and different cellular populations to R2 and R2*. It was informed by quantitative iron maps obtained with ion-beam microscopy and validated with MRI on post mortem samples of human brain tissue. It could be shown that the linear exponential part of static dephasing relaxation is a possible biomarker for iron in dopaminergic neurons. Furthermore, iron in different chemical forms contributed in different patterns to R2*, suggesting that prior models are an oversimplification. A generalization of the developed model to further areas and pathologies as well as its translation to in vivo will be the next steps.



Respiratory and cardiac noise removal from fMRI data: Evaluation of retrospective methods

Physiological signals of non-neuronal origin have been seen to hamper the identification of cognition related activity in blood oxygenation level dependent (BOLD)-contrast functional magnetic resonance imaging (fMRI) data. A gold standard method for the removal of such nuisance signals however doesn’t yet exist; its identification mainly hindered by indirect nature of BOLD contrast fMRI. In the present work, an attempt has been made to identify such a standard by comparing the performance of three commonly used retrospective physiological noise correction methods: RETROICOR (Retrospective image based correction), aCOMPCOR (anatomical image- Component based noise correction) and PICA (Probabilistic independent component analysis) on simulated noisy data sets, degraded by the same amount of physiological noise (tCNR = 0.5). The t-statistic values of peak activated voxels and the number of significant voxels per cluster (p < 0.001) were used as the parameters for comparison, with the parameters obtained from the statistical analyses of noiseless datasets serving as the standard. Noise correction methods of RETROICOR and aCOMPCOR, which involves modeling of physiological noise were found to perform better at the selected tCNR than PICA based noise classification, with RETROICOR performing the best in terms of the number of significant voxels detected per cluster.


Gast, R.

PyRates - A Python framework for rate-based neural simulations

In neuroscience, computational modeling has become an important source of insight into brain states and dynamics complementary to classic experiments. This is due to the potential to observe and manipulate variables in models that are difficult to assess in the living brain. Neural mass models (NMMs) are computationally efficient models for simulating large-scale brain dynamics as observable with neuroimaging techniques such as EEG/MEG or fMRI [1]. They model dynamic interactions between large, lumped populations of different cell types at the meso- and macroscopic scale. In this work, we present PyRates, a Python framework for building NMMs and simulating their dynamic behavior within a well defined mathematical structure. To ensure a highly customizable neural mass design, we provide both an differentio-differential approach for maximal computational efficiency and an integro-differential approach that allows for a flexible implementation of various neurobiological features. We show via numerical simulations how PyRates can replicate and extend established NMMs like the popular Jansen-Rit model [2], and build novel NMMs based on the same formal skeleton. Finally, we discuss the framework’s interface to other neuroimaging and network analysis tools and conclude, that PyRates makes a substantial contribution to open and reproducible neuroscience by providing a unified, fully tested and well documented neural simulation framework. References [1] M. Breakspear: Dynamic models of large-scale brain activity. Nature Neuroscience, 20: 340-352, 2017. [2] B. H. Jansen and V. G. Rit: Electroencephalogram and visual evoked potential generation in a mathematical model of coupled columns. Biological Cybernetics, 73: 357-366, 1995. o visual pathways .


Haenelt, D.

Mapping colour-selective columns in V2 across cortical depth using GE-EPI and SE-EPI

Using ultra-high field fMRI makes in vivo examinations of columnar and laminar structures in the human cerebral cortex feasible. However, it is still under debate which acquisition technique is optimal for high-resolution fMRI. Typical acquisition protocols at lower field strengths use GE-EPI , which possesses high BOLD sensitivity but lacks specificity due to large draining veins. Because of the anisotropic macrovascular distribution in the cortex, the needed specificity towards microvascular contributions does also depend on whether columnar or laminar structures are of interest. Here, we compare single-shot 2D GE-EPI and SE-EPI at 7 T in their ability and specificity to delineate colour-selective stripes in extrastriate cortex V2. Participants performed multiple sessions at different days. In single sessions, stimuli consisted either of chromatic or achromatic sinusoidal moving gratings shown in different blocks. GE-EPI and SE-EPI were acquired on different days. V2 was defined based on retinotopy and layers were estimated according to the equi-volume approach. With both sequences, stripes can be reliably identified. The column width across cortical depth is used as a measure for tangential spread. The analysis shall shed light on the issue if GE-EPI is significantly affected in tangential blurring towards the pial surface by delineating cortical columns.

Brain Computer Interfaces, Big Data, and Statistics


Hohmann, M.

A large platform for Big Data neurophysiology

Neurophysiological research has come under scrutiny for being underpowered due to inadequate sample sizes. This limitation is particularly profound in research on rare neurological diseases like Amyotrophic Lateral Sclerosis, which only affects approx. 2 in 100.000 people in Europe. One reason for this limitation is the necessity to conduct studies in a controlled lab environment as the setup is complicated, expensive, and prone to environmental influences. Our goal is to provide a platform that breaks with this limitation. By developing a smartphone application with a state-of-the-art user-interface and pairing it with a low-cost Electroencephalogram (EEG), we aim to enable patients to participate in controlled, neuroscientific experiments, from wherever they are, whenever they want. In a pilot study, we evaluated the first prototype of our smartphone application with 30 subjects. We could show that subjects were able to self-administer the full experimental session by only following the instructions that the smartphone application provided. Subjects were able to repeatedly fit the EEG headset onto their heads by following feedback from a life signal-quality display. Lastly, data quality of the acquired EEG signals was sufficient to distinguish between the performed cognitive tasks in an offline analysis. Our results indicate that that careful application design with a focus on psychological factors that drive acceptance of new technology has the potential to make lab-technology accessible to everyone, and it may help to break with the existing limitations of neuroscientific research.


Jayaram, V.

MOABB: A platform for brain-computer interface meta-analyses

BCI algorithm development has long been ham-pered by two major issues: Small sample sets and a lack of reproducibility. We offer a solution to both of these problems via a software suite that streamlines both the issues of finding and preprocessing data in a reliable manner, as well as that of using a consistent interface for machine learning methods. By building on recent advances in software for signal analysis implemented in the MNE toolkit, and the unified framework for machine learning offered by the scikit-learn project, we offer a system that can improve BCI algorithm development. To validate this system, we analyze of a set of state-of-the-art decoding algorithms across 12 open access datasets, with over 250 subjects. Our analysis confirms that different datasets can result in very different results for identical processing pipelines, highlighting the need for trustworthy algorithm benchmarking in the field of BCIs, and further that many previously validated methods do not hold up when applied across different datasets, which has wide-reaching implications for practical BCIs.


Scharf, F.

Principles behind variance missallocation in temporal exploratory factor analysis for ERP data: Insights from an inter-factor covariance decomposition

Temporal exploratory factor analysis (EFA) is commonly applied to ERP data sets to reduce their dimensionality and the ambiguity with respect to the underlying components. However, the risk of variance missallocation (i.e., the incorrect allocation of condition effects) has raised concerns with regard to EFA usage. Here, we show that variance misallocation occurs because of biased factor covariance estimates and the temporal overlap between the underlying components. For example, a direct consequence of our expositions is that researchers should use oblique rather than orthogonal rotations. Furthermore, researchers should be aware of the implicit assumptions about factor time courses inherent in any factor rotation method. A Monte Carlo simulation confirms our results by showing, for instance, that characteristic biases occur only for orthogonal Varimax rotation but not for oblique rotation methods such as Geomin or Promax. We outline the practical implications of our results.

Language and Communication


Adamson, H.

Structural brain changes associated with second language learning

Second language (L2) acquisition necessitates a reorganisation of the brain, however, unlike motor skill and visuospatial training, it remains unclear precisely how these plastic changes occur as an individual becomes increasingly bilingual. Research on multilingualism has primarily focused on comparing monolinguals to bilinguals, with few longitudinal studies following the same learners. Moreover, most have not tried to tease apart the structural changes associated with different aspects of learning. Evidence suggests specific elements of language (e.g. syntax, semantic, phonology, oral production) may have localized structural correlates in both connections and nodes within the language network. We hypothesized that structural adaptations in these areas should co-occur and correlate with changes in specific aptitude measurements in these different domains. We therefore collected longitudinal scans over the course of 1.5 years (3 months intervals, 6 time-points total) during L2 acquisition to investigate the dynamic plastic changes associated with language learning. Fifty-six Arabic native speakers were taught German using two different teaching methods, one focused on syntax, one on semantics. To assess structural network changes, high resolution diffusion MRI scans were acquired and fractional anisotropy (FA) was computed as a measure of white and grey matter microstructural organisation. To assess behavioural changes, a battery of language aptitude tests were administered. Voxel-based statistics were used to examine whole brain changes over time, differences between teaching groups, and the correlation of FA change with specific language aptitude measures. Our preliminary data demonstrate widespread and non-linear structural adaptations of brain regions involved in attention, memory, learning, in addition to the primary language processing areas. Subcortical and medial temporal lobe regions show an initial change during the first 6 months, while more frontal cortical regions only display plastic changes over the last 6 months. The correlation analysis revealed associations of behaviour with changes in the arcuate fascicle, internal capsule, medial temporal lobe, and superior longitudinal fascicle. These early results provide the first steps in profiling the dynamic recruitment and maintenance of plastic changes across the brain during L2 learning.


Chien, P.J.

Neural networks for lexical tone and intonation in Mandarin Chinese

In Mandarin Chinese (henceforth Mandarin), pitch information can both indicate lexical meanings (Chao, 1968) and intonational modulations (Yuan et al., 2002). This dual function of pitch leads to an intriguing question on how Mandarin speakers process the two types of pitch information. Previous research suggests similar cognitive mechanisms in the two processes by showing behavioural interferences (Liu et al., 2016) and overlapping neural activations (Gandour et al., 2004). Nevertheless, the underlying neural correlates specific to tone and intonation remain unclear, and methodological limitations of the previous work also preclude conclusions. The current study aims to identify and compare the neural networks for tone and intonation in Mandarin speakers. We first plan a behavioural experiment examining pitch processing at the psychophysical level and then an fMRI experiment investigating the neural correlates. Our pilot tests adopted monosyllabic word stimuli morphed across tone (tone 2, tone 4) and intonation (statement, question) and had participants judge one of these dimensions in separate blocks. The results consistently revealed participants’ categorical perception in tone but not in intonation, suggesting that the long-term representation of intonation in Mandarin speakers might be weaker than that in non-tonal language speakers (Sammler et al., 2015). Meanwhile, with the stimuli well-controlled in acoustic cues, the results suggest that pitch information might not be the dominant cue in Mandarin intonation. While the previous studies might have taken the assumption that pitch information essentially signals statement-question in Mandarin, this finding importantly calls the attention to cross-cultural diversity in research (Jack et al., 2018).


Just, A.C.

Paralinguistic use of prosody in Mandarin Chinese and German: Is intention transmission affected by tone?

Speech prosody – our vocal tone – is a crucial channel in conversations as it conveys social information about the speaker’s mind. A recent study showed robust evidence that German speakers use conventionalized prosodic forms to communicate their interpersonal intentions such as criticism, suggestion or wish, irrespective of the semantic meaning of the utterance (Hellbernd & Sammler, 2016). But how does this prosodic channel of communication work across cultures? Are there culture-specific aspects of prosody that may lead to in-group advantages for listeners of the same native language, but may cause cross-cultural misunderstandings as proposed by the dialect theory of emotion recognition (Elfenbein & Ambady, 2002)? We addressed this question by comparing the ability of 20 German and 21 Mandarin Chinese native listeners to recognize communicative intentions in the prosody of their native and the respective other culture’s language. Mandarin Chinese is an interesting case because it is a tonal language where pitch variations carry lexical meaning which might interfere with the recognition of communicative intentions, particularly in German listeners. Listeners performed a six-alternatives forced-choice classification on disyllabic words uttered by four German and four Mandarin Chinese native speakers in six different intentions. As expected, results show an in-group advantage in German listeners who performed significantly better on German than Mandarin Chinese prosody. Surprisingly, performance of Chinese listeners was equally low for both German and Mandarin Chinese prosody, although significantly higher than chance levels, suggesting that prosody may be a less important cue for communicative intentions for Chinese listeners.


Lutz, C.

Wor(l)ds apart – does the N400 reflect bilingual language distance and meaning in translation? An ERP study of the effects of L1-L2 distance and translation direction in German-English and Cantonese-English bilinguals

Globalization. Interconnection. Bilingualism. More and more people grow up bi- or multilingual. Research on differential language processing in bilinguals suggests that similar languages cause more interference and consequently activate more control brain structures. We investigate the process of overt translation and differences between two contrasting language combinations and translation directions. 23 Cantonese-English and 23 German-English bilinguals translated 140 single words from L1 to L2 and vice versa, while reaction time (RT) and EEG were recorded. Mixed effects models (MEM) was used as statistical methodology. Accuracy was lower in forward translation for Chinese bilinguals. RT was slower in forward than backward translation, and was responsive to translation direction, stimulus valence, target familiarity, and direction-group-proficiency interactions. We report modulation of N400 by main effects of group, electrode location, concreteness and several interactions. Notably, N400 amplitude overall was larger in German-English than Cantonese-English speakers and greater for forward translation in Cantonese. Use of MEM changed our RT interpretations, correcting for influencing factors. The findings of N400 are in accordance with a hypothesis of greater influence and thus greater cognitive control for overlapping languages. However, RT and accuracy do not reflect this, since forward appears more effortful than backward translation in Cantonese natives. Upon closer examination, Cantonese-English feature higher negativity in frontal areas in forward translation. These difficulties might arise due to script properties or due to difficulty in L2 word retrieval, but many answers remain elusive. Thus, it seems that with (language) distance does not necessarily come clarity, at least not in translation.

Auditory Perception and Music


Chien, S.C.

Deviance detection: On and off responses, omission response, and mismatch negativity

The omission response, elicited by an expected but omitted stimulus, has been interpreted by existing computational models as different from the typical mismatch negativity (MMN). From the viewpoint of prediction-based models, the omission response purely reflects prediction signals rather than prediction error. For adaptation-based models, the omission response can be treated as the rebound response of neural oscillators. However, both interpretations cannot explain the experimental observation that the omission response shows stronger amplitude than the responses evoked by the standard stimuli. In this simulation study, we propose to treat MMN as a mixture of the On response to the deviant and the Off response to the omitted standard. In this case, the omission response is regarded as the Off response that is time-locked to the omitted stimulus. We created an auditory cortex model consisting of connected excitatory and inhibitory neural populations. The model has a two-layer structure, where only the first layer receives direct thalamic inputs. In the demonstration, the model reproduces various types of cortical On/Off responses and the cortical omitted-stimulus responses, where the response latencies are proportional to the stimulus onset asynchrony. The simulation results suggest that the Off response produced by the two-layer structure along with the sustained resonance in the first layer representing the temporal regularity can account for the omission response. To conclude, with a goal of a unifying framework of MMN generation, we provided a new solution to interpret the omission response as a same neural mechanism as MMN generation.


Gugnowska, K.

Neural bases of interpersonal coordinated behaviour during music performance

Coordinated interactions like speaking, dancing, or playing music in ensemble are important elements of human social behaviour. Those behaviours are increasingly associated with brain-to-brain coupling, i.e. patterns of similar neural oscillations unfolding in the brains of two collaborating individuals. The present poster will outline a dual-EEG experiment aimed to investigate the psychological and neurophysiological mechanisms that underlie coordinated interpersonal behaviour. The project particularly strives to pinpoint mechanisms of interbrain synchronization during coordinated joint action planning in piano duos. Pairs of pianists will jointly perform simple music excerpts while planning a (cued) tempo change that will be played with disabled audio (i.e. mute) feedback. The resulting tempi will be congruent or incongruent between players. This manipulation will allow the examination of coordinated motor planning between the pianists (i.e. planning of synchronous / asynchronous tempi) and its effect on the quality of behavioural coordination. Notably, the analysis of coupling between players’ EEG signals recorded during a silent planning phase will allow us to eliminate the confounding influence of (shared) auditory and sensorimotor input on brain-to-brain coupling, a limitation of previous research investigating this topic. It is expected that brain-to-brain coupling and behavioural synchrony will be stronger during the planning of congruent, as compared to incongruent, tempo changes. Additional analyses will also allow us to explore which neural oscillations play the key role in interpersonal coordination.

Memory, Sleep, and Executive Function


Kohler, N.

Joint action in music: An fMRI study

To date, research on joint action has mainly followed two seemingly mutually exclusive theoretical frameworks: one with focus on interpersonal synchronisation, the other with focus on internal co-representations of others’ actions. Recent efforts have started to integrate both approaches into a coherent framework (Colling & Williamson, 2014) by showing that joint actions are characterized by a fine balance between synchronisation and co-representation depending on situational demands (Novembre, Sammler, & Keller, 2016). In the present project we adopt this view and investigate the underlying neural networks using fMRI. Pianists perform previously rehearsed Bach chorales together. Thereby, one pianist plays only the (right hand) melodies in the MRI, while the other pianist plays only the (left hand) basslines outside of the MRI. To manipulate pianists’ co-representation of their partner’s actions, they are either motorically familiar with their partner’s part (i.e. they trained it before) or unfamiliar. Simultaneously, the pianists’ task is to adapt to either the same or a different tempo in the second half of the pieces, which influences interpersonal synchronisation as previously shown (Novembre et al. 2016). In this way, we expect to shed new light on the neural mechanisms underlying joint action.


Roscher, K.

Hierarchy of musical action planning - revisited

Playing music requires complex action planning processes. These processes were proposed to be hierarchical in that the sequence-level planning regards the musical structure (i.e., harmonic relationships) and guides the planning of single-act movement kinematics (i.e., fingering). So far, this action hierarchy has been shown in production tasks in which pianists imitated hands playing chord sequences depicted in pictures or videos (Bianco et al., 2016; Sammler et al., 2013). What remains unclear, however, is whether this hierarchical planning generalizes to production regardless of the source of musical input (e.g., whether it is hands imitation or scores reading). We addressed this question by asking 22 expert pianists to play chord sequences on a muted keyboard by reading scores with fingering notations. Sequence-level planning of the final chords was either strong or weak depending on whether they were embedded in long (5 chords) or short sequences (2 chords), respectively. Final chords were structurally expected or unexpected and had to be played with conventional or unconventional fingering, lending insights into planning of musical structure and movement kinematics, respectively. Preliminary results confirm sequence-level planning of musical structure, but also reveal strong sequence-level planning of movement kinematics, indicating that the hierarchy of musical action planning might be more complex than previously thought. Possibly, music performance based on scores speeds up structural planning allowing to also plan movement kinematics ahead. In order to investigate the time course of the reaction cycle more thoroughly, EEG should be performed in the future.


Tu, H-F.

An investigation of entrainment and visual processing in toddlers using eye tracking

Music is a multi-dimensional, strong temporal form of art. Temporal structures, such as beats and meters, play an important role to human brains in sensory, motor, and cognitive processes. Entrained neuronal oscillations are observed in adults when attention is allocated to an ongoing periodic event stream. In line with Jones’ Dynamic Attending Theory, there is a strong link between entrainment and the enhanced performance of visual attention tasks. This experimental study aims to employ a wearable eye tracker to investigate whether i) visual processing is enhanced by music; whether ii) unattended beats in or out-of synchronization with visual stimuli, and iii) fast or slow tempi, affect the visual search performance of toddlers with different background of exposure to music. The study features two experiments that employ methods modified from extant research. Two groups of toddlers will be recruited from conventional and music-based day care centers. Three hypotheses are (a) compared to silent phase, visual processing is enhanced by periodic auditory stimuli, (b) when the visual stimuli are synchronized with unattended auditory beats, it will increase the efficiency of visual search, and (c) compared to slow tempi, fast tempi will enhance the quality of visual search skills.

Social Cognition and Emotion


Baccolo, E.

The development of sensitivity to social traits of faces

The ability to discriminate social signals from faces is a fundamental component of human social interactions. So far, only a few studies have investigated the developmental origins of the sensitivity to these fine-grained characteristics of faces, focusing only on explicit trustworthiness judgements. In this study, a group of 5-year-old and 7-year-old children and a group of adults performed two tasks aimed to measure their implicit perceptual sensitivity to physical cues to trustworthiness (oddmanout task) and their ability to make explicit trustworthiness judgments (pairwise preference). In the oddmanout task, participants observed three simultaneously presented faces and selected the one they judged to be more different from the others. In the pairwise preference task, participants selected the face they trusted more among two simultaneously presented faces. For both tasks, the stimuli consisted of 7 variations of the same female face identity varying along a continuum of expressed trustworthiness. Preliminary results show that 5-year-old children perform significantly worse than both 7-year-olds and adults in both tasks. Nevertheless, multidimensional scaling (MDS) analyses performed on dissimilarity scores derived from the oddmanout task show that already at the age of 5 children represent faces in memory as a function of the level of trustworthiness they express, and provide explicit judgments favouring the face that display more intense physical cues to trustworthiness. This overlap between implicit and explicit judgements for younger as well as older children suggests that sensitivity to social signals from faces is already present early in the development and specializes in time.


Cai, Q.

The processing of laughter in people with autism

Background: Laughter is a universal non-verbal emotional vocalisation in human beings. As a social vocal emotion, the engagement of mentalising ability is crucial in the processing of laughter’s social functions and using it as a social signal in daily interaction, resulting in a long-term benefit in people’s social life well-being. Aims: To investigate laughter processing in people with autism would extend our knowledge of the role laughter plays in establishing and maintaining social bonds. Moreover, it is essential for the characterisation of the relationship between cognitive and affective mentalising and social interactions in autism. Methods: Behavioural studies are proposed to investigate the implicit and explicit processing of laughter in people with autism and matched groups of neuro-typical controls. Future research will investigate the role the mentalising system and orofacial mirror system plays in the processing of laughter in people with autism using neuroimaging techniques (e.g. fMRI). Results: In the process of collecting data. However, in my previous MSc project, we found that relative to TD adults, ASD adults performed less well in distinguishing the spontaneous and social laughter. In addition, the perceived contagion of laughter by ASD adults is less influenced by the type of laughter. Conclusion: We predicate that autistic individuals have difficulties in processing spontaneous and social laughter relative to neuro-typical people, suggesting that a high level of cognitive skill such as mentalising ability is crucial in social emotion perception.


Hopkins, A. K.

Individuals with high fear of negative evaluation show greater learning for self-negative evaluation

Fear of negative evaluation (FNE) is a core feature of social anxiety disorder. People who are highly fearful of negative evaluation display biased processing of social-evaluative information when related to the self. They select fewer positive words when asked to predict how another agent would evaluate them, but display no bias when making predictions about an unknown other agent. However, it is unclear whether this effect arises from a reduced ability to learn from positive information or from an enhanced learning from negative information, a distinction that has important therapeutical implications. We aimed to investigate the mechanism underlying the negative self-bias in high FNE individuals using computational modelling. Data from a probabilistic social learning task (n=100) (Button et al., 2015), completed by participants that varied along a continuum of low to high FNE, was modelled using adapted Rescorla-Wagner reinforcement learning models. The winning model contained separate learning rates for self-negative and self-positive information but a single learning rate for other information, indicating that the updating of information for the self is valence specific, whereas for others it is the same across valences. Learning rates were higher overall for self-positive information, indicative of an optimism bias. Crucially, high and low FNE individuals show the same ability to learn from self-positive information, but high FNE individuals have higher learning rates for self-negative information. Therapeutical targets for social anxiety might therefore focus on reducing this bias towards negative information.


Keshmirian, A.

Punishment of a collective behavior: Contribution of intention and outcome in third-party punishment for group moral transgressions

Animals form groups to avoid being eaten by the predators, which is known as “safety in numbers” rule. Humans also benefit from collective acts in many ways. We know, for instance, that people attribute less blame, hence less regret to themselves when they cause a collective harm (i.e. avoiding first-party punishment). But we don't know if, by collective acts, individuals also avoid less punishment from an impartial third-party observer. To the best of our knowledge, no study so far investigated if someone who is not directly involved in a collective harm also attributes less punishment to them (i.e. third-party punishment) which is analogous to the so-called “safety in numbers” rule but in humans and for punishment. Psychological and brain imaging studies have shown that attribution of punishment is a dual process, composed of two distinct neuro-cognitive components: intention inference (did they mean to harm?) and outcome monitoring (what was the consequence of the harm?). It was expected that the amount of punishment would be reduced in a collective harm in comparison to the individual harm, in general. Since the intention inference component of the dual-process theory is more cognitively demanding, we also expected that in the collective harm, the punishment would be more reduced in case of negative outcome rather than the negative intention (i.e. failed attempted harm). To test these hypotheses, we asked the participants to attribute punishment to “X” in two different conditions: i) “X” belongs to a group of 3 and ii) “X” does the harm individually.

Clinical Neuroscience


Albrecht, F.

Predicting alien/anarchic limb syndrome in corticobasal syndrome by structural magnetic resonance imaging

Alien limb phenomenon is a rare syndrome associated with a feeling of non-belonging and disowning toward one’s limb. In contrast, anarchic limb phenomenon leads to involuntary but well-executed movements of the limb. Alien/anarchic limb phenomena are frequent in corticobasal syndrome (CBS), an atypical parkinsonian syndrome characterized by rigidity, akinesia, dystonia, cortical sensory deficit, and apraxia. The structure-function relationship of alien/anarchic limb was investigated in multi-centric structural MRI data. Whole-group and single-subject comparisons were made in 25 CBS and eight CBS-alien/anarchic limb patients vs. controls. Furthermore, support vector machine learning was used to see if CBS patients with and without alien/anarchic limb syndrome could be distinguished based on underlying structural MRI patterns. Whole-group comparison of CBS vs. controls revealed asymmetric frontotemporal gray matter density differences. CBS with alien/anarchic limb syndrome vs. controls showed differences in frontoparietal gray matter density including the supplementary motor area contralateral to the side of the affected limb. Classification of CBS patients yielded accuracies of 79%. CBS with alien/anarchic limb syndrome was differentiated from patients without alien/anarchic limb syndrome with an accuracy of 81%. Predictive differences were found in the cingulate gyrus spreading to frontomedian cortex, postcentral gyrus, and temporoparietoocipital regions. We present the first analysis on alien/anarchic limb in corticobasal syndrome identifying syndrome-specific atrophy based on structural MRI with support vector machine classification. Results pave the way for individual clinical syndrome prediction and allow understanding the underlying neurocognitive architecture.


Ballarini, T.

Predicting individual dopaminergic treatment response in Parkinson’s disease with structural MRI

Objective: We aimed at testing the potential of biomarkers in predicting the individual patient response to dopaminergic therapy (DT), the most common symptomatic treatment for Parkinson´s disease (PD). Methods: Treatment efficacy was assessed in 30 PD patients as the motor symptoms improvement between the unmedicated and the medicated state as assessed by the Unified Parkinson’s Disease Rating Scale score III. A multiple regression was implemented to test the prediction accuracy of age, disease duration and treatment dose and length. Patients were stratified into weak and strong responders according to the individual treatment response. Univariate voxel-based morphometry was applied to investigate differences between the two groups on age-corrected T1-weighted magnetic resonance images. Finally, multivariate support vector machine classification was used to predict individual treatment response based on structural neuroimaging data. Results: Increasing age, but neither disease duration nor treatment dose and duration, predicted a weaker DT response. In addition, weak responders showed greater brain atrophy in left supplementary motor area, superior frontal gyrus, and left temporoparietal operculum. Support vector machine classification revealed that gray matter volume in these brain regions, including additionally the cerebellum and primary motor cortex, was able to differentiate weak and strong DT responders with 74% accuracy. Conclusions: Results suggest that both increasing age and reduced gray matter volume are valid and independent predictors of DT response in PD. By predicting individual treatment response with clinical data/biomarkers our results will pave the road to future applications in the framework of personalized medicine in clinical routine. red for accurate group studies.


Blöchl, M.

“Post-stroke depression”? Depressive symptoms before and after first-incidence stroke

Observations of high levels of depressive symptoms in stroke survivors have fuelled the assumption that stroke can lead to depression – a notion reflected in the term “post-stroke depression”. Indeed, DSM-V lists stroke as one of the few direct causes of depression. However, there is some evidence to suggest that vice versa, depression increases the risk of stroke. Thus, stroke patients may suffer from increased levels of depressive symptoms even before an acute insult. The extent to which stroke induces changes in depressiveness therefore remains unclear. Here, we used longitudinal data from a large cohort study in the United States (MIDUS) to test the hypothesis that stroke accelerates an increase in depressive symptoms. During the course of MIDUS, N = 89 participants reported a first-time incidence of stroke. Using propensity matching, we generated a control group of non-stroke participants with similar demographic and cardio-vascular risk factors. Multi-level models will be applied to compare changes in depression between the stroke and non-stroke group. We will also examine whether changes in depressiveness are specific to stroke or also occur in patients with other cardio-vascular diseases (e.g. myocardial infarction) or different life-threatening illnesses (e.g. cancer), respectively. Lastly, we will explore the potential role of physical disability as a driving factor for depressive symptoms post-stroke and other diseases. The results will contribute to a better understanding of the concept of “post-stroke depression” and may also provide insights into the aetiology of mood disturbances following stroke


Dermody, N.

Relating MRI to histopathological surrogate biomarkers in frontotemporal dementia

Background: Frontotemporal dementia (FTD) comprises a spectrum of clinical syndromes of which four main subtypes are recognised: behavioural-variant FTD (bvFTD), semantic dementia (SD), progressive non-fluent aphasia (PNFA), and logopenic progressive aphasia (LPA). Clinical diagnosis is hampered by disease heterogeneity and reliable methods to diagnose and monitor FTD are lacking. Both changes in the concentration of proteins from cerebrospinal fluid (CSF) or blood and regional atrophy patterns could serve as disease biomarkers. Objective: To relate disease-specific atrophy patterns, as measured with magnetic resonance imaging (MRI), in FTD subtypes to commonly studied proteins from CSF/blood. Methods: We compared CSF concentrations of total-tau, phosphorylated-tau, β-amyloid 1-42, progranulin, ubiquitin, neurofilament light chain (NfL) and phosphorylated neurofilament heavy chain (pNfH) in 152 FTD patients (75 with bvFTD, 22 with SD, 36 with PNFA and 19 with LPA), and 40 Alzheimer's disease (AD) patients. We also compared NfL serum concentrations in an additional 29 FTD patients, 5 AD patients and 23 healthy controls. For each disease, we correlated protein concentrations with volumes of disease-affected brain regions. Results: AD and LPA exhibited similar protein concentration profiles. Serum NfL was higher in SD compared with all other groups, except PNFA. Significant correlations between protein concentrations and brain volume loss were seen for CSF NfL in AD and bvFTD, and CSF pNfH in bvFTD and SD. Conclusion: The data indicate that proteins from CSF and blood have potential to differentiate between FTD subtypes and AD, and elevated levels of neurofilament proteins may reflect greater atrophy.

Memory and Attention


Chiou, S.-C.,

Observation of whole-body movement sequences: Memory encoding, retrieval, and the influence from higher cognitive functions

Working memory (WM) for human actions plays an important role in not only social interactions but also the observation-imitation process of motor skill learning due to an unavoidable time delay between visual perception and motor output. However, how the spatial and temporal information of a complex whole-body movement sequence are encoded, retained and retrieved from the WM remains unclear. In Experiment 1, to differentiate between the perception and memory processes during action observation, we manipulated the retention interval (RI) of a delayed discrimination task, in which participants watched two sequentially displayed movement sequences (different in spatial and/or temporal domains) and made a same/different judgment. The results showed robust spatial but weak temporal representations of movement sequences in both short RI (perception) and long RI (memory) conditions. In Experiment 2, we manipulated both the movement length and the RI to further investigate how the effects of memory load and maintenance delay influence memory performance. The results showed a non-decaying, near-perfect recognition performance of movement trajectory, while the recognition of rhythm (i.e., pattern of movement durations) was jointly influenced by memory load and maintenance delay. When the memory load was low, performance remained high if RI was short, but deteriorated if RI was lengthened, illustrating a time-based forgetting. On the contrary, when the memory load was high, performance remained low across all RIs, indicating that the memory load might have surpassed the limited capacity of WM. Future studies will focus on the effect of attention and potential interactions between different information streams.


Souza, C.

The modulation of episodic recollection by conceptual knowledge in Alzheimer's Disease and Autism Spectrum Disorder: Examining the role of the hippocampus.

Memory is central to many cognitive processes, impacting in future decisions for a better social adaptation. For example, every day when we wake up, we plan our day based on long-term memories, as our previous experiences (episodic memory) and also as our general knowledge (semantic memory). Long-term memory theories suggest that the hippocampus sustains memories formation, but disagree upon the participation of this structure in all types of declarative memories. Also, the idea that hippocampus areas supports the development of an abstract (semanticized) version of episodic memory traits in cortical areas remains in debate. In healthy-individuals, the conceptual knowledge (i.e., schemas and item-typicality) enhances memory formation (episodic traits). However, in Autism-Spectrum-Disorder (ASD) and Alzheimer's Disease (AD) there is no benefit from conceptual knowledge, specially item-typicality, when forming episodic memories. The possible explanation is because both clinical groups have hippocampus compromised. This project examines the role of encoding schema (abstract mental representation) and item-typicality (goodness of an exemplar to represent it category) on declarative memories systems. We will contrast typical developmental (TD) adults with ASD-group as well as comparing AD-group with TD-elderly group in a recognition memory task with remember-know paradigm. The innovativeness of this proposal rests on indirectly comparing the neurocognitive profile of two clinical groups with robust behavioral and EEG source localization and connectivity estimates. The project is likely to clarify the semantic-episodic system relationship, particularly the role of the hippocampus-cortical network, and to uncover neurocognitive signatures in these clinical populations for earlier diagnosis and rehabilitation purposes.


Maier, M.J.

Forgiveness and cognitive control - provoking revenge via theta-burst-dtimulation of the dLPFC

Forgiveness is a highly relevant process in the daily social life. Based on theoretical models and correlational evidence, it has been suggested that forgiveness relies on inhibition as a sub-function of cognitive control. We combined an ultimatum game (UG) and a dictator game (DG) with inhibitory, continuous theta-burst stimulation (cTBS; verum vs. placebo, within-subjects design) of the right dorsolateral prefrontal cortex (rDLPFC) to examine the effect of reduced cognitive control on forgiveness behavior. After an inhibitory theta-burst stimulation we expect lower rates of forgiveness behavior. Participants first played an UG against fair and unfair opponents, where they had to accept or refuse (fair and unfair) monetary offers, and then received a cTBS before playing a DG against the same opponents with reversed roles. The participants now had the opportunity to forgive the unfair opponents (distribution of a fair amount of money) or to take revenge. Following verum cTBS, participants allocated significantly less money to unfair opponents than in the placebo cTBS condition. Additionally, reaction times (RTs) as well as neural activation in the right DLPFC differed significantly between verum and placebo cTBS for unfair opponents (higher RTs and lower DLPFC activation following verum stimulation). For fair opponents, no differences were found between stimulation conditions. These results powerfully indicate that cognitive control (with a key player within the rDLPFC) is a central requirement for overcoming unwanted emotional responses and regulating one’s actions when displaying pro-social behavior.

Poster Session II

Wednesday, 27 June, 16:15-17:30

Poster Number


Title with Abstract

Neuroimaging Physics and Signal Processing


Movahedian Attar, F.

Towards optimisation of a dedicated preprocessing pipeline for intra-cortical fibre and sub-cortical U-fibre mapping using diffusion MRI


Rose, D.

Informing delay in neural mass models with estimates of myelination and axonal diameters based on quantitative structural and diffusion-weighted MRI


Vaculčiaková L.

Towards high resolution cortical myelination mapping


Waschke, J.

Detection of unlabeled cells in 3D bright-field images

Language and Communication


Maran, M.

The neural basis of phrasal building


Musiolek, L.

Modeling the N400 as semantic Bayesian surprise


Numssen, O.

Localization and uncertainty of TMS effects during motor cortex stimulation


Qi, T.

Gray matter structural covariance changes during language comprehension development


Rysop, A.U.

Modulating neural network dynamics of speech comprehension – the role of the angular gyrus

Social Cognition and Emotion


Linz, R.

Changes in serum BDNF levels during acute psycho-social stress negatively relate to salivary cortisol


Ma, R.

Real-life social contact increases emotional well-being, relates to amygdala volume and interacts with early adversity


Nguyen, T.

Brain-to-brain synchrony during mother-child interactions: The role of maternal caregiving and attachment


Overbye, K.E.O.

Development of ERN and Pe from childhood to adulthood: A multimodal EEG and MRI study

Clinical Neuroscience


Gong, R.

Movement control task for detecting voluntary movement impairment of Parkinson disease


Herzog, N.

Shared and differential neurocognitive mechanisms in obesity and binge eating disorder


Eken, A.

Task- based functional connectivity based classification of fibromyalgia using fNIRS


Hidalgo-Lopez, E.

Differential effects of menstrual cycle and individual inhibitory control on activation and connectivity of the basal ganglia during a Stop Signal Task.


Maier, C.A.

The impact of sex hormones on reward sensitivity in women

Perception, Action, and Movement


Lubinus, C.

The role of action-based and temporal predictability in sensory attenuation on a behavioural and neural level


MacIntyre, A.D.

P-centres, beats, and the rhythms of movement


Murali, S.

The reciprocal relationship between eye blinks, perceptual changes and motor responses in the visual and auditory modality


Wen, S.

Preferences for motor acts are matched for execution and observation in a subpopulation of F5


Tostova E.A.

The Influence of physical activity on psychophysical characteristics of the subject within virtual environment


Wheatley, S. H.

Measuring oscillatory behaviour of the primary visual cortex following saccadic eye movements

memory and Attention


Meyer, A.-K.

Tracking the impact of retrieval suppression on individual memory representations


Paulus, P.C.

Episodic simulations reveal the structure of affective representations in ventromedial prefrontal cortex


Szabó, D.

Human-analogue resting-state networks in the dog brain

Poster Session III

Wednesday, 27 June, 17:30-18:45

Neuroimaging Physics and Signal Processing

Poster Number


Title with Abstract


Zarubin, G.

Development of a tACS-EEG closed loop system in order to understand and utilize the neuromodulatory role of tACS


Zoraghi, M.

An integrated approach to layer-specific analysis of multi-resolution MRI data


Jamshidi Idaji, M.

A novel framework for the detection of non-linear interactions in the human brain


Kalloch, B.

Semi-automated generation of individual surface-based computational models of the human head and torso from MR images

Language and Communication


Tippmann, J.

Developmental changes in neuronal processing of irregular morphosyntactic rules during childhood


Villar González, P.

Brain lateralization of a whistled language


Sandoval, M.

To dub or not to dub - that is the question!


Tromp, J.

Combining Virtual Reality and EEG to study pragmatic language processing in a naturalistic environment


Shcherbakova, O.

Unconscious detection of verbal and non-verbal ambiguous stimuli

Social Cognition and Emotion


Schleihauf, H.

Dual-mode model for over-imitation


Schliephake, L.

The role of the lateral geniculate nucleus in autism spectrum disorder


Verim, B.

The relationship between Theory of Mind (ToM) deficits and neurocognitive functions in euthymic patients with bipolar disorder


Swidrak, J.

The virtual midas touch in the ultimatum VR game.


Tetereva, A.O.

Resting state functional connectivity of amygdala followed the extinction training after fear learning


Liu, T.

The role of arousal in pro-social behavior after exposure to relaxing music: Evidence with self-report and physiological measures


Langeloh, M.

Is learning all about Theta? Infants encode unexpected events at the 4 Hz theta rhythm


Puhlmann, L.

Can a Nine-Month Contemplative Mental Training Intervention increase your lifespan?

Clinical Neuroscience


Molloy, E.

PRESS & MEGA-PRESS: Proof of Concept for Imaging GABAergic & Glutamatergic Metabolism at 3T

Ripp, I.

Integrity of neurocognitive networks in dementing disorders as measured with functional and metabolic neuroimaging


Schulz, C.C.

Effects of child maltreatment on adolescent brain structure and function: Roles of threat and deprivation


Steinhardt, J.

Mechanisms of body weight gain in patients with Parkinson´s disease after deep brain stimulation: A study design presentation


Zsido, R.

Hormone-mediated network changes driving depression susceptibility across the female lifespan


Kroczek, A.M.

Increasing ecological validity in neuroscientific measurements

Memory and Attention


Emch, M.F.

Neural correlates of verbal working memory: An ES-SDM meta-analysis


Baczkowski, B.M.

Do conditioned threat responses generalize according to pre-existing relational organization of episodic memories?


Avendaño Diaz, J.C

Attention performance in dyads: Insights from dual-EEG


Stephani, T.

Probing cortical excitability with somatosensory evoked potentials
Go to Editor View
loading content