Абстракты
INTRODUCTION
From computational models to neurotechnology
Vasily Klyucharev
National Research University Higher School of Economics, Moscow, Russia
HSE aims to create centers of excellence and to disseminate the experiences of these centers throughout the university. We have created a new strategic academic unit focusing on cognitive sciences. We build a research and educational center of excellence that integrates cognitive neuroscience and information science in order to construct interdisciplinary models of human behavior and to implement them in the cutting-edge neurotechnologies.
KEYNOTE PRESENTATIONS
Who is in control of our attention?
Artem V. Belopolsky
Vrije Universiteit Amsterdam, The Netherlands
The question of how visual attention is controlled has occupied the minds of researchers for several decades. Are attention and the eyes controlled by events in the physical world (salience) or by our own free-will? The typical answer is that our attention can be controlled by both types of control. Recently, it has been proposed that such top-down vs. bottom-up dichotomy cannot account for a large range of effects in which our attention seems to be drawn by our previous attentional experience. In my talk I will present some recent evidence that provides further insight into how our previous experience with reward and fear-conditioned stimuli shape attentional selection. Finally, I will present recent evidence that clarifies the relationship of selection history to the classical notions of top-down and bottom-up control.
The contribution of neurodegenerative disorders to cognitive neuroscience
Stefano F. Cappa
IUSS Pavia and IRCCS S. Giovanni di Dio, Brescia, Italy
The study of the consequences of brain disease remains one of the cornerstones of cognitive neuroscience. The classical approach of neuropsychology, based on focal lesions of the brain, usually stroke, is now augmented by the investigation of cognitive changes in neurodegenerative diseases. The recognition of selective, early changes of episodic memory, language and social cognition, combined to quantitative structural and functional imaging methods has provided novel insights into the neural basis of cognition and behavior, which complements the information derived from functional imaging in normal subjects and from computational neuroscience. Besides the theoretical interest, these advances have important translational implications for early diagnosis, prevention and rehabilitation of common conditions, such as Alzheimer’s disease and movement disorders.
Monetary validity and the visual brain
Sacha Bourgeois-Gironde
Université Panthéon Assas, Paris, France
Money, when used as an incentive, activates the same neural circuits as rewards associated with physiological needs. However, unlike physiological rewards, monetary stimuli are cultural artifacts: how are monetary stimuli identified in the first place? How and when does the brain identify a valid coin, i.e. a disc of metal that is, by social agreement, endowed with monetary properties? We took advantage of the changes in the Euro area in 2002 to compare neural responses to valid coins (Euros, Australian Dollars) with neural responses to invalid coins that have lost all monetary properties (French Francs, Finnish Marks). We show in magneto-encephalographic recordings, that the ventral visual pathway automatically distinguishes between valid and invalid coins, within only 150 ms. This automatic categorization operates equally on coins subjects were familiar with as on unfamiliar coins. These results could suggest the existence of a generic, all-purpose neural representation of money that is independent of experience. The ventral visual pathway, a system previously thought to analyze visual features such as shape or color and to be influenced by daily experience, could also able to use conceptual attributes such as monetary validity to categorize familiar as well as unfamiliar visual objects. I will conclude by suggesting some possible operationalization of this neural marker of monetary validity in view of testing economic models of money-emergence such as Kiyotaki and Wright's (1989) or Banerjee and Maskin's (1996).
The theory of critical brain dynamics
Klaus Linkenkaer-Hansen
Dept. Integrative Neuroscience, Center for Neurogenomics and Cognitive Research, Neuroscience Campus Amsterdam, VU University Amsterdam
Several lines of evidence have pointed to the presence of critical dynamics in neural networks at different levels of organization, where activity tends towards the balance between order and disorder. Different computational models have been created that each exhibit some aspect of this critical behavior. These were used to hypothesize that the critical state is optimal for information processing such as memory storage or response to stimuli.
To investigate the relationship between critical-state dynamics and oscillations in the processing of stimuli, we used a previously developed neuronal network model that exhibits neuronal avalanches on short timescales and long-range temporal correlations of oscillations on long timescales (Poil et al., 2012). By altering the excitatory/inhibitory connectivity balance, networks with sub-, critical or super-critical dynamics were created. These networks were probed with varying intensities of stimulation, and their responses analyzed in terms of the dynamic range of post-stimulus phase locking and how post-stimulus effects were modulated by the amplitude of pre-stimulus oscillations.
Our results indicate that networks with critical oscillations show the largest dynamic range of post-stimulus phase locking and, interestingly, that critical networks exhibit the most reliable pre-stimulus amplitude dependence of post-stimulus response. These results show that critical networks go through periods of being more or less sensitive to stimuli based on their ongoing activity, which could be important for attentional mechanisms.
To test the predictions derived from the model, we studied the relationship between criticality of neuronal oscillations and post-stimulus phase locking in source modeled M/EEG data of healthy participants performing a threshold-stimulus detection task. We observed that individual differences in phase locking can be understood in terms of individual differences in critical brain dynamics. The promise of “critical brain dynamics” for explaining individual variation in brain functioning is discussed.
References
Poil S-S, Hardstone R, Mansvelder HD, Linkenkaer-Hansen K (2012) Critical-State Dynamics of Avalanches and Oscillations Jointly Emerge from Balanced Excitation/Inhibition in Neuronal Networks. J Neurosci., 32: 9817–9823.
Practice makes perfect in free memory recall
Misha Tsodyks
Weizmann Institute of Science, Rehovot, Israel
A large variability in performance is observed when participants recall briefly presented lists of words. The sources of such variability are not known. I will show that most of variability can be accounted for by a simple recall model with random representations of words in long term memory. However, our analysis of a large data set of free recall revealed a small fraction of participants that reached an extremely high performance, including many trials with the recall of complete lists. Moreover, some of them developed a number of consistent input-position-dependent recall strategies, in particular recalling words consecutively (“chaining”) or in groups of consecutively presented words (“chunking”). The time course of acquisition and particular choice of positional grouping were variable among participants. Our results show that acquiring positional strategies plays a crucial role in improvement of recall performance.
Mechanisms and dynamical structure of brain rhythms: from rest to perception
Sylvain Baillet
McGill University, Montreal, Canada
One broad objective in neuroscience is to comprehend the mechanisms of large-scale, oscillatory neural dynamics: how they enable functions by shaping communication in brain networks, and how the earliest detection of their alterations in disease can contribute to improved healthcare prevention and interventions. To contribute toward this goal, our approach is to combine imaging methods and experimental neuroscience with computational and disease models, neuromodulation techniques, and translational arms to the clinic and industry.
Our rationale is that the ubiquitous polyrhythmic activity of the brain has been approached empirically, with underlying mechanisms that remain not understood. This hinders our comprehension of how 1) perception and behaviour emerge from brain network activity, and 2) the pathophysiological developments of brain and mental-health disorders increasingly studied as network diseases, affect large-scale neural communication.
Our vision is that these difficult questions require a bottom-up approach: We aim to understand how basic physiological factors of neural integrity and function shape the dynamical structure of oscillatory brain rhythms, such as their interdependence across multiple frequencies through cross-frequency coupling. These phenomena represent a deep source of uncharted markers of neural excitability, activity and connectivity. I will illustrate these principles with our latest results concerning the resting brain, multimodal perception and pathophysiological markers of epilepsy and neurodegenerative syndromes.
CCDM PRESENTATIONS
Role of cortical theta and gamma endogenous oscillations in speech encoding
Boris Gutkin
Ecole Superior Normal, Paris, France
National Research University Higher School of Economics, Moscow, Russia
Many environmental stimuli present a quasi-rhythmic structure at different timescales that the brain needs to decompose and integrate. Cortical oscillations have been proposed as instruments of sensory de-multiplexing, i.e., the parallel processing of different frequency streams in sensory signals. Yet their causal role in such a process has never been demonstrated. Here, we used a neural microcircuit model to address whether coupled theta–gamma oscillations, as observed in human auditory cortex, could underpin the multiscale sensory analysis of speech. We show that, in continuous speech, theta oscillations can flexibly track the syllabic rhythm and temporally organize the phoneme-level response of gamma neurons into a code that enables syllable identification. The tracking of slow speech fluctuations by theta oscillations, and its coupling to gamma-spiking activity both appeared as critical features for accurate speech encoding. These results demonstrate that cortical oscillations can be a key instrument of speech de-multiplexing, parsing, and encoding.
Interfaces between brain, cognition and communication
Yury Shtyrov & Andriy Myachykov
Aarhus University, Denmark
Northumbria University Newcastle, UK
National Research University Higher School of Economics, Moscow, Russia
The aim of our research is to investigate the brain basis of communication, its control and its interactions with other neurocognitive systems. We will present a snapshot of recent research carried out by the group members in collaboration with other local, national and international collaborators. This includes studies of language, attention and executive control, multilingualism, spatial and motor cognition, pragmatics, communicative disorders, memory and metacognition. Our projects employ a wide range of methods, including behavioural, eye-tracking, electrophysiological (EEG, MEG) and neuro-stimulation (TMS, tDCS) techniques. The results help us build a comprehensive picture of dynamic interactions between neurocognitive systems and functions in carrying out communicative and other types of human behavior.
Effects of background neuronal states on cortical reactivity and task performance
Vadim Nikulin
Charite – Medical University Berlin, Germany
National Research University Higher School of Economics, Moscow, Russia
Neuronal activity is remarkably flexible and demonstrates rich repertoire of distinct patterns relating to different behavioral elements. Interestingly, this flexibility is juxtaposed with the strong influence of the background neuronal states on the brain responses in different experimental conditions. This influence can be expressed in the effects of instantaneous pre-stimulus neuronal state on the stimulus or task-related activity. Moreover, even neuronal activity recorded at rest can predict cortical responsiveness and behavior. In my talk I will present our recent neuroimaging studies demonstrating the influence of background neuronal states on cortical excitability and task performance.
The BDNF val66met polymorphism affects the Level of Processing effect of memory: A deep and shallow rTMS study
Matteo Feurra
National Research University Higher School of Economics, Moscow, Russia
A polymorphism in the BDNF (BDNF val66met; rs6265) gene causing a valine (val)-to-methionine (met) substitution at codon 66 results in altered intracellular trafficking and packaging of BDNF, and in a reduction of its regulated secretion. BDNF-val/val carriers usually showed a better cognitive performance in terms of motor, attention and memory task with respect to those who are met-carriers. So far it is still unclear if Val/Val subjects are susceptible to better plasticity and if their cognitive performance is a matter of better connectivity. Here we investigated how the Val66Met polymorphism of the BDNF gene, affects plasticity and long-term memory performance. By using perturbation repetitive Transcranial Magnetic Stimulation (rTMS) we aimed to investigate the causal role of dorsolateral prefrontal cortices (DLPFCs) underlying retrieval processes accordingly to the Val66Met polymorphism. Thus, we adopted a classical deep and shallow memory task; episodic encoding was TMS-free while 10Hz neuronavigated rTMS was delivered during retrieval over the left and right DLPFC, Vertex (control site) and a no-TMS condition. Our data show that, left DLPFC-rTMS at retrieval abolished the beneficial effect of deep encoding. This effect is specific only for ValVal carriers group.
MEG and EEG based imaging of true cortical zero-lag coupling
A. Ossadtchi1 & D. Altukhov1,2
1National Research University Higher School of Economics
2MEG Centre, Moscow State University of Psychology and Education, Moscow, Russian Federation
Cortical networks is the dominating concept in the modern functional neuroscience. Currently, the major concerns in the network centered studies are the poor reproducibility of the results across subjects and inability to image networks with zero-phase lag. To address these issues we have developed a novel method for non-invasive imaging of the dynamic networks from MEG and EEG data. The developed method allows to perform a global fit to the generative model of the sensor-space cross-spectrum and assess the amount of variance explained by the solution found. Our approach is insensitive to the volume conduction (VC) and in contrast to the other existing techniques allows to explore the dynamics of zero-phase coupled networks in both sensor and source space. When applied to the real MEG data collected from 10 subjects involved into a language task the use of the real-part of the cross-spectrum projected away from the VC yielded significantly higher reproducibility of the observed networks. Additionally, we show that the use of the real part of the VC projected cross-spectrum allows to improve the accuracy of motor-imagery states classification.
Reinforcement learning mechanisms of Cognitive Dissonance
Marco Colosio, Anna Shestakova, Vadim Nikulin, Anna Shpektor, & Vasily Klucharev
National Research University Higher School of Economics, Moscow, Russia
Cognitive dissonance theory suggests that our preferences are modulated by the mere act of choosing. A choice between two similarly valued alternatives creates a psychological tension (cognitive dissonance) that is reduced by a post-decisional re-evaluation of the alternatives. Based on our EEG observation that the associated with stronger cognitive dissonance choices triggered a larger negative fronto-central evoked response similar to the previously associated with incorrect responses the error-related negativity (ERN) component of ERP, we proposed that the general performance-monitoring circuitry is implicated in the CD behavior. More specifically, our EEG results showed the amplitude of the evoked response correlated with the re-evaluation of the alternatives. Pursuing another goal of finding neuropredictors of individual CD performance, we suggested that the resting state neuronal dynamics as measured by the long-range temporal correlations- LRTC (LRTC) of neuronal oscillations can predict the CD magnitude. Recent neuroimaging studies have shown that the performance in the motor (Smit et al., 2013) and perceptual tasks (Palva et al., 2013) can relate to the long-range temporal correlations (LRTC) of neuronal oscillations recorded at rest conditions. Importantly, LRTC indicate a presence of a scale-free structure of neuronal activation on multiple time scales that is vital for optimal neuronal processing in the human brain (Linkenkaer-Hansen et al., 2001; Hardstone et al., 2012; Palva et al., 2013). Indeed, for the first time in a decision making experiment we found that individuals with stronger resting-state LRTC demonstrated a larger post-decisional re-evaluation of the alternatives. By and large, our results suggest that cognitive dissonance can be reflected in both resting-state and choice-related activity of the prefrontal cortex as a part of the general performance-monitoring circuitry.
CCDM PARTNER PRESENTATIONS
Training an efficient visual search: Can we create new feature detectors?
Igor S. Utochkin1 and Jeremy M. Wolfe2
1National Research University Higher School of Economics, Russia
2Harvard Medical School and Brigham & Women’s Hospital, USA
It was shown in classical studies (e.g. Shiffrin & Schneider, 1974) that inefficient visual searches (when every additional item in a display yields cost in the reaction time (RT) associated with the serial deployment of attention to that item) can become highly efficient (no RT cost with additional items) after extended practice. In our study we tested a hypothesis suggested by Treisman (2006) that this high post-practice efficiency is due to the formation of new preattentive feature detectors that selectively respond to trained complex targets in the same way that primary sensory detectors respond to basic visual features (color, orientation, motion, etc.). We trained our participants for 16 days (25,600 trials per participant) to search for an artificially created complex feature (“openness to the right”). Different groups searched for the feature in objects with different shapes (curved “amoebas” or straight line “trees”). Control participants were trained to search for color-color conjunctions.
To estimate the effects of training, we tested the participants on all three tasks (“amoebas”, “trees”, and color-color conjunctions) before training (pre-test) and after (post-test). We estimated traditional criteria of feature search such as overall search efficiency and search asymmetry. We also assessed transfer of training from “amoebas” to “trees” and vice versa to estimate the generality or specificity of the “feature”. Critically, we tested whether the trained feature is susceptible to selective adaptation – this is a criterion that can provide evidence for a feature detector independently from visual search (Treisman, 2006). We found strong effects of training on search efficiency and asymmetries, as well as cross-transfers supporting the hypothesis of new features. However, the independent selective adaptation test failed to find any effects. We hypothesize that training substantially refined participants’ use of existing features, but did not create a new one.
Maximum likelihood estimation of discriminability d’ when hit or correct-rejection rates are 0% or 100%
Tadamasa Sawada
National Research University Higher School of Economics, Moscow, Russia
Signal detection theory is used to estimate discriminability d’ separately from a response criterion in a psychophysical experiment. Subjects in the experiment are often instructed to maintain neutral criteria because the estimated d’ is inflated to infinity whenever hit or correct-rejection rates are 0% or 100%. I will propose a maximum likelihood method that estimates d’ by combining: (i) a 2D likelihood distribution of d’ and the criterion and (ii) a prior distribution of the criterion. The distributions (i) and (ii) are combined using Bayes’ theorem. The prior distribution is derived by assuming that hit and false-alarm rates are never 0% or 100% and is based on the instruction to the subjects for the neutral criterion. The proposed method will be compared with existing methods that compute d’ by adjusting the measured hit or correct-rejection rates.
How automatic is exogenous spatial attention?
Joseph W. MacInnes
National Research University Higher School of Economics, Moscow, Russia
Spatial attention is often manipulated in one of two ways: With an (exogenous) external event like a sudden visual onset, or through internal (endogenous) expectations such as manipulating the likelihood of a target at a location. Exogenous orienting has long been believed to be fast, reflexive and automatic even when the viewer is aware that the causal cue does not produce reliable information about the location of potential targets. Recent studies however, have shown limitations on the automaticity of endogenous cuing effects and this has led to revised theories on the underlying mechanism of spatial attention. I will present a series of studies using a continuous time-course paradigm to show problems with existing theories and revise an older explanation based on attentional control settings.
Atypical orientation sensitivity points to deficiency of early experience-dependent brain plasticity in children with ASD
Stroganova T.A., Sysoeva O.V., Galuta I.A., Davletshina M.A., Orekhova E.V.
Autism Research Laboratory, Center for Neurocognitive Research (MEG Center)
Moscow State University of Psychology and Education, Moscow, Russia
Influence of past experience on neural activity of primary visual cortex is most spectacularly reflected in the phenomenon of the oblique effect, i.e., much better orientation discrimination along cardinal (vertical and horizontal) than along oblique axes, as reported in a neurotypical population (Appelle, 1972). The development of orientation selectivity depends on an interaction between innate neural mechanisms and visual experience (Tanaka et al., 2009). Within this framework the anisotropy of orientation discriminability (i.e., the “oblique effect”), which is suggested to be a consequence of the prevalence of vertical and horizontal orientations in the environment (Girshick et al., 2011), may reflect developmental plasticity of the visual cortex.
According to the predictions of the deficient divisive normalization account, ASD individuals should have a reduced oblique effect due to atypically increased line orientation discrimination threshold along cardinal axes (Rosenberg et al., 2015). Given that the oblique effect is a basic feature of visual processing reflecting E/I balance, this effect’s investigation in ASD subjects may shed light on the nature of visual sensory abnormalities frequently reported in this population. A predicted reduction of oblique effect in children with ASD, if found, could illuminate the abnormal experience-dependent plasticity as underlying causes of atypical developmental trajectories of perceptual skills observed in infants later diagnosed with ASD (Sacrey et al., 2014).
The current investigation’s objective was to address this gap in the knowledgebase concerning ASD. The approach was to examine vertical and oblique line orientation discrimination sensitivity in children with ASD and in typically-developing (TD) control participants aged 7–15 years. Children were asked to detect the direction of tilt of a high-contrast black-and-white grating relative to vertical (90◦ ) or oblique (45◦ ) templates. The oblique effect was reduced in children with ASD as compared to TD participants, irrespective of their IQ. This reduction was due to poor orientation sensitivity along the vertical axis in ASD children, while their ability to discriminate line orientation along the oblique axis was unaffected. We speculate that this deficit in sensitivity to vertical orientation may reflect disrupted mechanisms of early experience-dependent learning that takes place during the critical period for orientation selectivity.
Crime without punishment: an experimental study of nature and causes of punishment in public goods games
Alexey Belyanin
National Research University Higher School of Economics, Moscow, Russia
The public goods (PG) game is one of the main workhorses of modern experimental research, and one of the major tools for investigation of social cooperation. It has been extensively studied in a number of contexts and under a variety of institutions, including the possibility of punishment of some participants by the other participants. Such punishments are known to be efficient as a tool of sustaining cooperation in the game; yet the exact nature and reasons why people punish each other to a large extent remains unexplored. In particular, in a series of recent experiments Simon Gaechter and Benedikt Herrmann observe a puzzling phenomenon of spitefulpunishment: a nonnegligible fraction of players tend to punish those who have contributed more than they did. This observation is remarkably robust in some countries, and casts doubts as to whether punishments in the PG context really serve the goal to express one's emotional dissatisfaction with low level of cooperation. In this work we study the punishment strategies in a systematic way, disentangling several possible explanations for this behaviour: 1) competitive, that is trying to gain more than other players in the experiment; 2) retaliation, or dissatisfaction with others' behaviour vs own contribution; 3) jealousy, or disapproval of deviation of one's contribution from the social norm; 4) preemptive, trying to retaliate for the possible punishment on the part of the other player; 5) availability, punishment because the experimental design suggests this option; 6) tolerance towards punishment; finally, 7) availability of punishment option. We set and ran a series of experiments in the various cities of Russia, which establishes that most important determinant of antisocial punishment is fear of being punished (preemption), followed by willingness to outperform one's peers (competitiveness); for prosocial punishment major motives include deviation from own contribution and normative standard. To qualify these conclusions, we use multilevel structural models (gllamm) which confirm the prevalence of these phenomena over the classical explanations of emotional dissatisfaction with others' behaviour, and in particular, their role in determining spiteful punishments.
The role of prefrontal cortex in mathematical cognition
Marie Arsalidou
National Research University Higher School of Economics, Moscow, Russia
Mathematical problem solving relies on core cognitive abilities such as working memory, mental attention, inhibition and updating. These processes often elicit activity in the prefrontal cortex. Although the prefrontal cortex consistently activates during math problem solving, it has received little attention in neuroimaging studies of mathematical cognition as compared to parietal regions. Data from a total of 101 eligible fMRI studies that examined mathematical cognition with adult participants were compiled. Using meta-analytical methods, data were analyzed based on several mathematical problem-solving categories that included tasks with and without formal mathematical operations (i.e., addition, subtraction and multiplication). Concordant activity across studies shows that prefrontal regions are differentially implicated in mathematical cognition as a function of process-based task complexity. Findings will be discussed in terms of a new model of the hierarchical organization of the prefrontal cortex in mathematical cognition.
Building a Human-like Robot: Automatic Text Processing and Gesture Synthesis
Artemy Kotov
National Research Center “Kurchatov Institute”, Moscow, Russia
We design a set of linguistic technologies, which should allow construction of an “understanding” human-like robot for natural speech and non-verbal communication with humans in different environments. In particular, the robot should react on incoming text with gestures and speech, simulating rich emotional dynamics. We solve this problem in three steps. On the first step a syntactic parser makes tokenization and constructs dependency syntactic trees with semantic representation. On the second step – each semantic representation is processed by a number of scripts (similar to scripts described in classical frame semantics). In particular, a subset of scripts is responsible for the emotional processing – reactions like: Nobody loves me! You care only about yourself! etc. Balancing between several competing scripts simulates emotional dynamics and enriches robot behavior. On the third step – gestures and output phrases, invoked by the activation of scripts, are transferred for execution to a simple robot, while output speech is simultaneously processed by a third-party speech synthesizer. In this attempt we try to verify theoretical cognitive architectures by the designed computational model for a human-like robot. As we expect this model can also be implemented in the field of applied robotics.
Categorical and non-categorical referential choice
Andrej A. Kibrik 1,2, Mariya V. Khudyakova3, Grigory B. Dobrov4, Anastasia Linnik5, Dmitrij A. Zalmanov2
1Institute of Linguistics, Russian Academy of Sciences, Moscow, Russia
2Lomonosov Moscow State University, Russia, Moscow, Russia
3National Research University Higher School of Economics, Moscow, Russia
4Consultant Plus, Moscow, Russia
5University of Potsdam, Potsdam, Germany
This is a study of referential choice in discourse production, understood as the choice between various types of referential devices, such as pronouns and full noun phrases. Our goal is to predict referential choice, and to explore to what extent such prediction is possible. Our approach to referential choice includes a cognitively informed theoretical component, corpus analysis, machine learning methods and experimentation with human participants. Machine learning algorithms make use of 25 factors, including referent’s properties (such as animacy and protagonism), the distance between a referential expression and its antecedent, the antecedent’s syntactic role, and so on. Having found the predictions of our algorithm to coincide with the original almost 90% of the time, we hypothesize that fully accurate prediction is not possible because, in many situations, more than one referential option is available. This hypothesis was confirmed by an experimental study, in which participants answered questions about either the original text in the corpus, or about a text modified in accordance with the algorithm’s prediction. Proportions of correct answers to these questions, as well as participants’ rating of the questions’ difficulty, suggested that divergences between the algorithm’s prediction and the original referential device in the corpus occur overwhelmingly in situations where the referential choice is not categorical.
Modulating lexical retrieval with tDCS
Svetlana Malyutina
National Research University Higher School of Economics, Moscow, Russia
Transcranial direct current stimulation (tDCS) is a promising tool for many clinical applications, including language therapy in patients with aphasia. But any applications of tDCS to language therapy need to be informed by studies on its effects on language processing in healthy speakers. In this study, healthy participants received high-definition tDCS over either Broca’s area or the left angular gyrus, followed by naming and lexical decision tasks with single-word verb and noun stimuli. We found that cathodal stimulation over both Broca’s area and the left angular gyrus increased naming speed for both verbs and nouns. This finding challenges the traditional view of cathodal stimulation as suppressive or leading to decreased performance, and suggests that cathodal stimulation may also enhance language performance. Additionally, effects of specific stimulation types depended on the order of their administration. This may be due to possible physiological carry-over and/or task novelty effects and suggests that within-subject designs may not always be ideal in brain stimulation research.
POSTERS
Fast and slow responses: two scenarios of performance monitoring
N.A. Novikov1,2, A.A. Lapina1, J.M. Nurislamova1, & B.V. Chernyshev1
1Laboratory of Cognitive Psychophysiology,
2Centre for Cognition & Decision Making,
National Research University Higher School of Economics, Moscow, Russian Federation
(bchernyshev@hse.ru)
Cognitive control includes two basic aspects: maintenance of task-specific processes (including attention), and non-specific control of the motor threshold. In line with this distinction, performance errors may be explained by two alternative mechanisms: fast errors are believed to result from inappropriate action impulses that were allowed due to a failure in keeping the motor threshold sufficiently high, while slow errors may be caused by failures in specific task-relevant information processing, leading to uncertainty. Correct responses are not uniform as well – they may be preceded by subthreshold attempts to execute an erroneous response, thus evidencing uncertainty and resulting in increased reaction time (RT).
We expected that RT might be a valid proxy that will allow distinguishing trials with high and low level of uncertainty. We hypothesized that on fast-RT trials an internal outcome detection would occur following response commission, while on slow-RT trials the low level of certainty would hinder the internal detection, and only an external feedback signal would lead to processing the trial outcome.
EEG data were collected from 50 healthy right-handed adults. We used an auditory condensation tasks involving a two-alternative choice. For each participant, responses were classified into fast and slow relative to individual median RTs. Visual feedback was given 500 ms after the response.
We found a much greater error-related theta power increase at fronto-central sites on fast-RT trials compared with slow-RT trials. Only on slow-RT correct trials, positive feedback induced a feedback-related increase in prefrontal beta power. Late parietal suppression of alpha oscillations was also prominent on slow-RT erroneous trials.
Our findings are compatible with the view that fast responses involve little internal uncertainty, thus allowing internal error detection. Slow responses occur in conditions of higher internal uncertainty, making the external feedback unpredictable and thus informative in terms of reinforcement learning.
The study was implemented in the framework of The Basic Research Program at the National Research University Higher School of Economics in 2015-2016.
Two-levels of feature binding in the auditory modality: reconciliation of opposing views
B.V. Chernyshev1, D.V. Bryzgalov1,2, & I.E. Lazarev1
1Laboratory of Cognitive Psychophysiology,
2Centre for Cognition & Decision Making,
National Research University Higher School of Economics, Moscow, Russian Federation
Feature binding is believed to be critical for perception, while mechanisms of this process remain under debate. Studies measuring mismatch negativity (MMN) demonstrate that binding occurs at a low level of the cortical hierarchy, while behavioral experiments suggest a much higher integrative level. The current study aimed at testing the hypothesis that processing of feature conjunctions may involve both levels.
EEG was recorded while participants were engaged in an auditory two-alternative choice task, which was a combination of the oddball and condensation tasks. Two types of deviant target stimuli were used – complex stimuli, which required feature conjunctions to be identified, and simple stimuli, which could be identified by a single feature.
Responses to complex stimuli were slower and less accurate than responses to simple stimuli. For simple stimuli, errors were associated with increased response time, while there was no such effect for complex stimuli. MMN was prominent and its amplitude did not differ between simple and complex stimuli – although the stimuli deviated from standards either in a single feature or in two features correspondingly. Errors in response only to complex stimuli were associated with decreased MMN amplitude. P300 amplitude was greater for complex stimuli than for simple stimuli. For simple stimuli, P300 amplitude was reduced on erroneous trials.
Thus, we replicated within a single experiment the major effects reported in two opposing lines of binding research. Our MMN data hint that the neuronal population encoding feature conjunction is closely associated with (or coincides with) the neuronal population that generates MMN. Our P300 data are compatible with the explanation that higher processing levels receive sensory representations of conjoined features as well as of separate features. Thus, the increased informational load created by complex stimuli prolongs processing time – leading to increased response time for the stimuli that require feature conjunction to be identified. Our findings provide resolution to conflicting views concerning the nature of feature binding and support the notion that feature binding is a distributed multi-level process.
This study was supported by the Russian Foundation for Humanities, research project No 15-06-10742.
Crosslinguistic interaction between phonology and semantics in a masked priming task: an ERP study
Novitskiy, N1, Myachykov, A.1,2 & Shtyrov, Y.1,3
1Center for Cognition and Decision Making, Higher School of Economics, Russia
2Department of Psychology, Northumbria University, Newcastle upon Tyne, UK
3Center of Functionally Integrative Neuroscience, Institute for Clinical Medicine, Aarhus University, Denmark
We investigated crosslinguistic phonological and semantic similarity effects on the activation of bilingual lexicon in the human brain. Russian-English late unbalanced bilinguals were tested in a masked priming paradigm with L1 (Russian) words as masked primes and L2 (English) words as targets. The experimental task was to maintain the targets in memory until the next trial and match them against occasionally presented catch stimuli. In different randomly presented pairs, the primes and the targets either overlapped (1) phonetically, (2) semantically, (3) both phonetically and semantically, or (4) did not overlap. In parallel, we continuously recorded 128-channel electroencephalogram (EEG) and measured the event-related potentials amplitude at N170 and N400 latencies. Phonological similarity between primes and targets reduced the task performance for both semantically related and unrelated pairs and reduced the amplitude of N400 in the semantically unrelated condition. This interaction between semantics and phonology was underpinned by activity in the left inferior temporal gyrus in the EEG source reconstruction. In conclusion, phonological decoding of prime-target pairs interacted with their semantic decoding. Phonetically similar words in the two languages shared short-term memory representations that caused ambiguity during the recall of the written form as well as reduced performance that accompanied this ambiguity.
ERP research of the incongruent price
Kislov, A.A. & Levchenko, E.S.
Center for Cognition and Decision Making, Higher School of Economics, Russia
Nowadays, there are a lot of fundamental works on the valuation process, which was provided using fMRI, However, findings of these works cannot be always used on practice. On the other hand, there are a lot of EEG-based works which showed detection of ERP N400 in incongruent condition. The aim of this work was to show N400 response to incongruent price and to understand whether we can estimate subjective price using EEG data or not. We demonstrated images of beverages of different prices, namely, common price, overcharges, and very overcharges, to subjects of the study. As we supposed, N400 was detected in incongruent conditions. The next step is to develop algorithm which will allow to calculate subjective price based on ERP data automatically.
Continuous theta-burst stimulation of ventromedial prefrontal cortex enhances schema-linked encoding
Alicia Vorobiova, Anna Shpektor, & Matteo Feurra
Center for Cognition and Decision Making, Higher School of Economics, Russia
It has long been established that the existence of prior knowledge (or schema representation) to which new information can be related, facilitates memory encoding, consolidation and retrieval. Nowadays schema is interpreted as a network of strongly interrelated neocortical representations, which can influence information processing. Recent studies show the important role of the medial prefrontal cortex (vmPFC) for establishing interrelationships between representations of objects based on their congruency to either prior knowledge or to each other by facilitating memory functions. Here, we used navigated continuous theta-burst stimulation (cTBS) on vmPFC to modulate encoding of object-scene paired associates of different congruency (i.e. differently fitting to subject’s knowledge about real world). According to our preliminary results, cTBS applied on vmPFC before encoding selectively increased retrieval memory performance (item recognition and associative memory) only for congruent items (compared to incongruent and intermediate) with respect to sham (placebo) stimulation.
More ambiguity of a response is produced by a “simpler” stimulus
Vasiliy Minkov1, Sergey A. Kutylev2, Tadamasa Sawada1
1School of Psychology, Higher School of Economics
2Big Data and Information Retrieval School, Higher School of Economics
Theoretical understanding of a stimulus and of a subject's task in a cognitive psychological experiment is critical for controlling the experiment and for interpreting its results. Such understanding often encourages cognitive psychologists to use "simple" stimuli because simple stimuli are expected to make understanding the results easier. But, if the stimuli are too simple and if they are impoverished, it has been pointed out that the cognitive system may be forced to use a less effective, unnatural, mechanism to perform a given task. For example, a triangle (or its representation by 3 points) has been used in many studies of 3D perception based on a single image or from a stereo-pair of images because the 3 points of a triangle are the minimum number needed to define a shape in a 3D scene. But, computational theories suggest that the 3D interpretation of an image becomes more ambiguous when the information in the image is too limited. We will analyze visual stimuli used in existing studies and discuss how difference of the visual stimuli affects interpretations of their results.
Memorability in vertical space
Beatriz Martín-Luengo1, Karlos Luna2, Andriy Myachykov1,3, & Yury Shtyrov1,4
1National Research University Higher School of Economics, Moscow, Russia
2School of Psychology, University of Minho, Portugal
3Northumbria University Newcastle, UK
4Aarhus University, Denmark
Vertical orientation was shown to affect different numerical judgments. In addition, physical dimensions, such as weight or font size, affect judgments of learning (JOLs). In two experiments we tested the hypothesis that vertical orientation may affect perceived words’ memorability, i.e., JOLs. Participants provided JOLs for the words presented centrally or in the upper or lower parts of a computer screen. In Experiment 1, JOLs were collected after the presentation of each word and in the centre of the screen. In Experiment, 2 JOLs were collected along the perceived word with either lower or upper presentation. In general, Bayesian analyses showed evidence in support for the lack of an effect of vertical orientation in JOLs. We interpret our results as indicating that the effects of physical dimensions on JOLs are mediated by subjective importance– the information vertical orientation fails to convey.
Different answers to different audiences: Effects of social context on the accuracy-informativeness trade-off
Beatriz Martín-Luengo1, Yury Shtyrov1,2, Karlos Luna3, & Andriy Myachykov1,4
1National Research University Higher School of Economics, Moscow, Russia
2Aarhus University, Denmark
3School of Psychology, University of Minho, Portugal
4Northumbria University Newcastle, UK
Research about conversational exchanges shows that people aim to optimize response relevance when they know the answer for certain (e.g., “What time is it?”). However, such certainty is often unavailable while speakers may still be under social pressure to provide an answer. We investigated how social context influences the informativeness level when answering questions under uncertainty. In two experiments participants answered difficult general-knowledge questions placed in different social contexts (formal vs. informal). Additionally, in Experiment 2 an informativeness payoff was introduced. Participants generated and decided on the number of alternatives in an answer (plurality option) and whether it should be reported or withheld (report option). Participants reported more answers independently of the informativeness level in the informal context. In the formal context, single answers were preferred to multiple ones and they were reported more often. We conclude that social context influences the level of informativeness in a conversation affecting achievable accuracy. Our results show that our conversational exchanges are sometimes driven by metacognitive components, in particular in informal contexts, but that sometimes they are driven by social norms and situational demands.
A latent variable model for state prediction in the asynchronous BCI paradigm
Nikolay Dagaev, Ksenia Volkova and Alex Ossadtchi
National Research University Higher School of Economics, Moscow, Russia
Many BCI algorithms based on machine leaning explicitly or implicitly employ a probabilistic model of data distribution. Most often, for simplicity and computational convenience, states' distribution is assumed to be directly conditioned on EEG patterns, without any intermediate or additional variables. However, given the complexity of relations between the EEG patterns and subject's states, a more flexible model with additional variables may provide for a more accurate classification.
We propose a probabilistic latent variable approach which implies that there is a discrete latent variable associated with each data point. That is, in addition to predictor variables (EEG pattern), the distribution of a target variable (state of a subject) is conditioned on an unobserved value of the corresponding latent variable. Importantly, the latter itself is conditioned on predictors. When a latent variable is marginalized out, the distribution of a target still remains conditioned on the predictors.
Practically, the optimal number of mixture components can be chosen by a cross-validation. A maximum likelihood solution for the model's free parameters can be obtained iteratively by a simple EM-algorithm. In the classification mode (targets are unknown), the most probable state distribution for each data point is chosen from the finite set of components on the basis of posterior probability value. Then, using the corresponding conditional distribution, we predict the target solely based on the predictors' values. Using real data from the asynchronous motor-imagery BCI paradigm, we demonstrate that the use of our probabilistic model yields increased classification accuracy as compared to the standard single component model.
Cross-frequency synchronization of brain rhythms
Denis Volk, Boris Gutkin, & Vadim Nikulin
National Research University Higher School of Economics, Moscow, Russia
Rhythmic activity in the central nervous system is important for various processes, such as: perception, movement organization, cognitive processes. It is known that when two distinct brain regions participate in the same cognitive process, their rhythms tend to synchronize. Therefore it is important to have reliable tools to detect the presence of such a synchronization, as well as the location and neuronal composure of the synchronized regions.
There are many methods to detect synchronization of brain regions in the same frequency band. However, cross-frequency interactions are more difficult to tackle. We have developed a new robust method of detecting cross-frequency synchronized components of brain activity. Our approach is an extension of the Cross-Frequency Decomposition technique [Nikulin et al, 2012]."
"The idea of CFD is to solve a substitute problem: fix a reference signal, frequency warp it to match the others’ frequency, and find a spatial component with the best linear least squares fit to the reference. Often, the solution turns out to be in a very good synchrony with the reference. In a general case of frequency ratio p:q, the linear fit is not enough. To overcome this difficulty, we augment the signals to create their complex-valued versions while preserving all the information relevant to synchronization. In the complex domain the problem reduces to a polynomial optimization problem, which could be solved either by standard methods or a custom solver.
Erasing of working memory trace by periodic input to a bi-stable neural network
Nikita Novikov
National Research University Higher School of Economics, Moscow, Russia
Working memory (WM) is the ability of the brain to retain information not currently presented in the environment in such form that this information is available for ongoing cognitive processes. A neuronal correlate of WM retention is self-sustained spiking activity in cortical networks. Many studies suggest that, in addition to the increased firing rate, oscillatory neural activity is modulated during WM retention, with theta, beta and gamma activity usually elevated in task-related brain areas, and alpha activity elevated in task-unrelated areas. We suggest that the role of alpha oscillations in disengagement of irrelevant brain areas during working memory tasks is related to its specific capability to terminate self-sustained spiking activity. This hypothesis is based on the experimental evidence that (1) self-sustained activity can be terminated by a strong synchronous excitatory pulse, and (2) cortical excitatory neurons have resonance in the alpha band (so the excitatory subnetwork can be potentially synchronized by the alpha activity). Here we present a model of spiking working memory network with periodic external input. Without input, the network is bistable and can be switched by excitatory pulses between the background state (with low spiking activity) and the memory state (with high spiking activity). We demonstrate that in the appropriate range of parameters, the network can be switched from the memory state to the background state by a periodic input in the alpha band, but not by an input in the theta or gamma bands of the same amplitude. We also show that this switching can be prevented by increasing NMDA to AMPA ratio, which occurs in the natural conditions as a result of specific neuromodulation in the neocortex.
Changes in Resting State Network Functional in Post-Stroke Aphasia
Vladislav Balaev1, Alexey Petruchevsky2, & Olga Martynova1,3
1Institute of Higher Nervous Activity and Neurophysiology Russian Academy of Sciences, Moscow, Russian Federation
2Center for Speech Pathology and Neurorehabilitation, Moscow, Russian Federation
3Centre for Cognition and Decision Making, National Research University Higher School of Economics, Russian Federation
To evaluate the influence of post-stroke aphasia on the functional association of widespread large-scale neuronal networks, we analyzed functional connectivity between resting state brain networks (RSNs) in aphasic patients (N = 15) and in healthy volunteers (N = 17) of the same age using resting state functional MRI. As a result, six RSNs were isolated and cross-correlation matrices were computed for their time courses. Aphasic patients showed decreased correlations between posterior part of the default mode (pDMN) and both auditory (AUD) and right frontoparietal (RFP) networks. Additionally, we calculated regions of interest based functional connectivity (ROI-FC), gray and white matter volumes in the ROIs overlapping with pDMN, AUD and RFP. ROI-FC analysis showed decreased FC between the right pars triangularis and both right middle frontal gyrus and right superior frontal gyrus. The decreased pDMN-RFP connectivity in patients is likely to reflect the changes in FC of these nodes. The lesion in the regions overlapping with pDMN and AUD networks lead to the significantly decreased pDMN-AUD connectivity. Our results suggest that abnormal functional connectivity in stroke patients may reflect the impairment of activity not only in the regions directly affected by stroke lesion in the left hemisphere, but also in the homotopic regions of the intact right hemisphere. The increase of gray and white matter volume in the right supramarginal gyrus, the functional hub of pDMN, AUD and RFP networks, correlated with less speech impairment. This increase might reflect a right hemisphere neuroplasticity process to compensate the impaired function of the homotopic region of LFP, pDMN and AUD in the left hemisphere. The presented results contribute to the hypothesized compensative role of the transfer of attention and executive functions from the damaged areas in the left hemisphere to the right homotopic areas, accompanied by more preserved language skills at the chronic stroke stage.
DTI-TMS assessment of corticospinal tract and corpus callosum integrity in ischemic stroke patients with relation to their motor outcome at the chronic stage
S. Kulikova1, M. Piradov2, R. Konovalov2, A. Limonova2, V. Nikulin3,4, M. Nazarova2,3
1National Research University Higher School of Economics, Laboratory of Interdisciplinary Empirical Studies, Perm, Russian Federation
2Research Center of Neurology, Moscow, Russian Federation
3Centre for Cognition and Decision Making, National Research University Higher School of Economics, Moscow, Russian Federation
4Charité - University Medicine Berlin, Berlin, Germany
Introduction
Adequate assessment of individual functional motor potentials is important for developing appropriate rehabilitation strategies in ischemic stroke [1]. Microstructural changes in corticospinal tract (CST) and corpus callosum (CC) were repeatedly correlated to post-stroke outcome [2,3]. However, relationship between them and functional recovery remains unclear. Here we investigated relationship between integrity of CST and CC assessed with diffusion tensor imaging (DTI) and brain functional state assessed with navigated transcranial magnetic stimulation (nTMS) in chronic ischemic supratentorial stroke.
Subjects and Methods
35 patients (45±9y) from different motor outcome groups: I – favorable (N=8); II – moderate (N=6); III - unfavorable (N=21) were compared to 30 age-matched healthy volunteers. CST and CC integrity was characterized using fractional anisotropy (FA) from DTI (1.5T Siemens, 20 directions, b-value=1000). Voxels with reduced FA were correlated with motor outcome and nTMS measures (Nexstim): resting motor thresholds (RMT) for Abductor Pollicis Brevis(APB) and Extensor Digitorum Communis(EDC) and intracortical inhibition (SICI). FA asymmetry was calculated in the internal capsule (FA-IC) and between corresponding points along ipsilesional and contralesional CST. Ipsilesional and contralesional (CORRp) or control (CORRc) FA profiles for CST were also correlated. Data was processed in BrainVISA and SPM8 software.
Results
In patients compared to volunteers FA was decreased over the whole CC and bilaterally in CST, while differences between groups I and III were localized in callosal motor fibers and ipsilesional CST (Fig.1). FA within IC negatively correlated with RMTs in both hemispheres. The best predictors of unfavourable outcome were FA-IC asymmetry, average asymmetry for 3 specific points on FA profiles, CORRp, CORRc, FA in selected voxels on CC and CST, and absence of motor evoked potentials (MEP) (Fig.2). Neither DTI nor TMS parameters could discriminate patients between the 3 groups. Surprisingly, this could be done with a 86% success rate using a combination FA-IC+CORRp+CORRc.
Discussion
Although low CST and CC integrity and MEPs absence in the affected hemisphere predicted unfavorable outcome, only combination of several parameters could differentiate between the 3 groups suggesting that evaluation of recovery potentials may benefit from multiparametric strategies. Interestingly, unfavorable outcome group demonstrated also lower integrity of the unaffected CST than healthy individuals. Correlation between FA in IC and RMTs was stronger in the ipsilesional side, while in the contralesional side it was significant but weak [4]. Changes in unaffected CST and CC highlight importance of interhemispheric interactions during stroke recovery, however no correlation between CC integrity and TMS parameters was found.
References
[1] Auriat et al. (Frontiers in Neurology, 2015)
[2] Borich et al. (BMC Neuroscience, 2012)
[3] Lindenberg et al. (HBM, 2012)
[4] Klöppel et al. (Neuroimage, 2008)
Pre–stimulus Alpha Oscillations and Inter-Subject Variability of Motor Evoked Potentials in Single- and Paired-pulse TMS Paradigms
Zafer Iscan1*, Maria Nazarova1,2*, Tommaso Fedele1,3, Evgeny Blagovechtchenski1,4, & Vadim V. Nikulin1,5
*Authors contributed equally to the study.
1Centre for Cognition and Decision Making, National Research University Higher School of Economics, Moscow, Russian Federation
2Research Center of Neurology, Moscow, Russian Federation
3Department of Neurosurgery, Unispital Zurich, University of Zurich, Switzerland
4Laboratory of Neuroscience and Molecular Pharmacology, Institute of Translational Biomedicine, Saint Petersburg State University, Saint Petersburg, Russian Federation
5Neurophysics Group, Department of Neurology, Charité - University Medicine Berlin, Berlin, Germany
Inter- and intra-subject variability of the motor evoked potentials (MEPs) to TMS is a well-known phenomenon. Although a possible link between this variability and ongoing brain oscillations was demonstrated, the results of the studies are not consistent with each other. Given that alpha oscillations were shown to reflect cortical excitability, we hypothesized that their power and variability might explain subject-specific motor evoked responses to single- and paired-pulse TMS (spTMS, ppTMS, respectively). Neuronal activity was recorded with multichannel EEG. We used spTMS and two ppTMS conditions: intracortical facilitation (ICF) and short-interval intracortical inhibition (SICI). Spearman correlations were calculated within and across subjects between MEPs and the pre-stimulus power of alpha oscillations in low (8–10 Hz) and high (10–12 Hz) frequency bands. Coefficient of quartile variation was used to measure variability. Across-subject analysis revealed no difference in the alpha power among the TMS conditions. However, the variability of high-alpha power in spTMS condition was larger than in the SICI condition. In ICF condition high-alpha power variability correlated positively with MEP amplitude variability. No correlation has been observed between the pre-stimulus alpha power and MEP responses in any of the conditions. Our results for the first time show that the variability of the alpha oscillations can be more predictive of TMS effects than the commonly used power of oscillations and we provide further support for the dissociation of high and low-alpha bands in predicting responses produced by the stimulation of the motor cortex.
Steady state visual evoked potentials based BCI performance under different perturbations
Zafer Iscan1 & Vadim V. Nikulin1,2
1Centre for Cognition and Decision Making, National Research University Higher School of Economics, Moscow, Russian Federation
2Neurophysics Group, Department of Neurology, Charité - University Medicine Berlin, Berlin, Germany
Electroencephalogram (EEG) based BCI designs have become popular due to their safety and high time resolution. Among them, steady state visual evoked potentials (SSVEPs) are particularly attractive since they provide high signal to noise ratio (SNR). In this study, we proposed a four-class BCI design based on SSVEPs to study the BCI performance under different perturbations. Four circles with individual flickering frequencies were presented to healthy participants on an LCD monitor while multi-channel EEG was recorded. In the offline task, subjects focused on the randomly presented circle indicated by a red oval frame for three seconds (100 trials). In the online tasks, subjects focused on one of the four circles for three seconds (100 trials). In the end of three seconds, the classification result was presented and subjects gave feedback with the keyboard. The online task was performed under four perturbation conditions randomized across subjects:
1) No count:
2) Speaking: Subjects counted from 1 to 10 (repeated) verbally.
3) Thinking: Subjects counted from 1 to 10 (repeated) mentally.
4) Listening: Subjects listened to their pre-recorded voice when they were counting from 1 to 10.
Decision tree, Naïve Bayes and K-Nearest Neighbor classifiers were used to evaluate the classification performance using features generated by canonical correlation analysis. In the preliminary analysis we have seen that perturbations might even improve or degrade classification accuracies across subjects. However, more subjects are needed to have statistical significance for the differences among conditions. Moreover, a combination with the eye-tracker is needed to obtain reliable results.
Neural mechanisms underpinning decision-making of bargaining under social competition
M. Martinez-Saito1*, A. Shestakova1, B. Gutkin1,2, & V. Klucharev1
1Centre for Cognition and Decision Making, National Research University Higher School of Economics, Moscow, Russian Federation
2Group for Neural Theory, LNC INSERM U960, PSL* Research University Ecole Normale Superieure, Paris, France
Behavioral economics has extensively studied how people make economic decisions and how are they biased by social preferences. However, the neural mechanisms underpinning such decisions remain unclear. Here we study the neural mechanisms underlying decisions in different conditions of economic competition. Additionally, we aim to investigate the learning processes that lead to adaptive bargaining strategies and how these are modulated by the degree of economic competition.
Twenty-nine subjects played the role of buyers in simultaneous games against different numbers of prerecorded buyers and sellers. Subjects played repeatedly a Double Auction paradigm in 50-minute 3T functional magnetic resononace imaging (fMRI) scanning sessions. Overall, the game allowed us to identify the effects of competition (number of sellers and buyers) on subjects’ willingness to pay (the size of bids).
Behavioral results demonstrated that subjects adjusted their trading price during the game based on the perceived competitiveness of the environment. We observed a progressive, yet incomplete convergence towards the optimal strategy predicted by a game-theoretic analysis, and separable learning processes involved: subjects’ overall scales of bid values are mainly influenced by the market environment, whereas subjects trial-by-trial adjustment of bid values instead display a skewed distribution modulated by the outcome of the previous trial. fMRI data analysis showed significant differential activations and dynamics in the anterior and posterior striatum, corresponding to different learning signals: reward prediction error and binary outcome (acceptance or rejection) respectively.
The results indicate that people learn to alter trading price based on the perceived competitiveness of the environment and suggest that the brain uses a simplified heuristic adapted specifically to bargaining.
Long-Range Temporal Correlations in the amplitude of alpha oscillations predict and reflect strength of intracortical facilitation: combined TMS and EEG study
Tommaso Fedele1,2*, Evgeny Blagovechtchenski1,3*, Maria Nazarova1,4, Zafer Iscan1, Vadim V. Nikulin1,5
* These authors contributed equally to the paper
1Centre for Cognition and Decision Making, National Research University Higher School of Economics, Russian Federation
2Department of Neurosurgery, Unispital Zurich, University of Zurich, Switzerland
3Laboratory of Neuroscience and Molecular Pharmacology, Institute of Translational Biomedicine, Saint Petersburg State University, Russian Federation
4Research Center of Neurology, Moscow, Russian Federation
5Neurophysics Group, Department of Neurology, Charité - University Medicine Berlin, Germany
(vadim.nikulin@charite.de)
Variability of the motor responses to TMS is widely acknowledged, however, little is known about its central origin. It is possible that such variability may relate to different neuronal states defining the reactivity of the cortex to TMS. In this study intrinsic spatio-temporal neuronal dynamics were estimated with Long-Range Temporal Correlations (LRTC) in order to predict the inter-individual differences in the strength of intra-cortical facilitation (ICF) and short-interval intracortical inhibition (SICI) produced by paired pulse TMS (ppTMS) of the left primary motor cortex. LRTC in the alpha frequency range were assessed from multichannel EEG obtained at rest before and after the application of ppTMS protocols. For the EEG session, preceding TMS application, we showed a positive correlation across subjects between the strength of ICF and LRTC in the fronto-central and parietal areas. This finding indicates the existence of subject-specific neuronal phenotypes defining the reactivity of the brain to ppTMS. Moreover, we also showed that ICF was associated with the changes in neuronal dynamics after the application of the stimulation. This result provides additional evidence for the recent observations that the cortical stimulation with sparse non-regular stimuli might have considerable long-lasting effects on the cortical activity.
The role of targets similarity in subsequent search misses
Alyona A. Lanina & Elena S. Gorbunova
Centre for Cognition and Decision Making, National Research University Higher School of Economics, Russian Federation
(lanina-al@yandex.ru)
Subsequent search misses effect refers to decrease in accuracy at detecting a second target after a first target has been found in visual search. In this study we investigated the role of targets similarity in this effect. The subject’s task was to find all the even or odd digits, among respectively, odd or even distracters. On each trial, it could be one high-salient target, one low-salient or two targets (one
high salient and one low salient), or no targets at all (catch-trials). In dual-target condition, targets could be identical (e.g. 2 and 2) or different (e.g. 2 and 4). The results of our study revealed significant differences between conditions with one target and two identical targets (t = -7.11; p < .000), with one target and two different targets (t = -7.51; p < .000). Differences between two identical and two different targets are not significant (t = 1.26; p < .223).These results suggest that the targets similarity is not the main factor in subsequent search misses.
Transcranial alternating current stimulation (tACS) of the primary motor cortex revealed no after effects.
Ivan Pozdniakov, Alicia Vorobiova, Matteo Feurra
Centre for Cognition and Decision Making, National Research University Higher School of Economics, Russian Federation
(bucherr@yandex.ru)
Transcranial alternating current stimulation (tACS) is a non-invasive brain stimulation method that allows to interact with brain oscillatory rhythms. tACS represents a potential substitute of neurorehabilitation treatment with drugs or invasive brain stimulation, as a tool for cognitive enhancement and for basic research in humans.
However, physiological mechanisms of tACS effects are still not understood. Online tACS revealed its efficiency in the primary motor cortex modulation during stimulation in frequency-dependent and state-dependent way. Whereas studies on tACS aftereffects (offline protocol) demonstrated mixed results.
Here, we conducted a sham-controlled tACS study with a neuronavigated transcranial magnetic stimulation (TMS) to test tACS aftereffects of the primary motor cortex (M1) on the TMS-induced motor evoked potentials (MEPs). tACS was delivered at 10 Hz, 20 Hz as well as in sham modality by 1 mA intensity (electrodes size: 35 cm2) during 15 min of MEPs recording. The target tACS electrode was located over the M1 hotspot corresponding to the first dorsal interosseous (FDI) muscle, the reference electrode (reference) was placed over the ipsilateral shoulder.
Our results showed no aftereffects of tACS on MEPs amplitude for both frequencies (10 Hz and 20 Hz) as well as for sham. The present findings suggest that tACS entrainment of the primary motor cortex activity is vanished after the end of tACS session, thereby confirming tACS as an online technique.
Software for quantative assessment of the results of navigated TMS mapping
Pavel Novikov1*, Maria Nazarova2,3*, Vadim Nikulin2,4
* These authors contributed equally to the paper
1Bauman Moscow State Technical University, Moscow, Russian Federation
2Centre for Cognition and Decision Making, National Research University Higher School of Economics, Moscow, Russian Federation
3Research Center of Neurology, Moscow, Russian Federation
4Charité - University Medicine Berlin, Berlin, Germany
(novikov-p-a@yandex.ru)
Transcranial magnetic resonance (TMS) is a modern non-invasive approach to study brain organization in humans. Recently, navigation based on individual magnetic resonance images (MRI) became widely available for TMS use making real the technique called navigated TMS (nTMS). However, wider use of nTMS mapping for research purposes, for example, to study motor cortex organization, neuroplastic changes of the areas after rehabilitation or special training or other dynamic assessment is still limited. There are several reasons for that. First, even for motor mapping there is no general agreement which parameters of nTMS maps should be assessed. Second, in nTMS motor mapping studies some well-known principles of the motor cortex organization such as divergence and convergence are generally ignored. We developed a software for quantative analysis of the results of motor nTMS mapping (http://tmsmap.ru/), which allows considering not only standard parameters like the size of the cortical muscle representation, the hotspot and the center of gravity location, but as well the volume of the representation considering the amplitude of MEPs in each stimulation spot, the shape of the area, the profile/landscape of the muscle cortical area and the overlap between the representations. The software was written in C# on the Windows platform, it is totally independent and has graphical interface. The input data includes coordinates of the EF maximum, coil position information and the response (for instance, MEPs peak-to-peak amplitude) in each point of stimulation and individual structural MRI data. The results of area calculation were compared with the results received using the method of spline interpolation, which was shown to have a good repeatability in the previous studies (Julkunen, 2014). The software was used for the analysis of the nTMS mapping data for several upper extremity muscles in healthy volunteers and stroke patient with hand motor deficit. Now the software is validated in test-retest nTMS motor mapping study. Nearest steps of the development will include adaptation of the software for speech nTMS mapping and elaboration of the sample brain (MNI version) making possible comparison of nTMS maps of different subjects.
Reference list:
Julkunen, Petro. 2014. “Methods for Estimating Cortical Motor Representation Size and Location in Navigated Transcranial Magnetic Stimulation.” Journal of Neuroscience Methods 232.Elsevier B.V.: 125–33. doi: 10.1016/j.jneumeth.2014.05.020.
Test-retest reliability of nTMS multi-muscle motor mapping
Maria Nazarova1,3, Pavel Novikov2, Ivan Gusarovas2, Zafer Iscan1, Evgeny Blagoveshchensky1, & Vadim Nikulin1,4
* These authors contributed equally to the paper
1Centre for Cognition and Decision Making, National Research University Higher School of Economics, Moscow, Russian Federation
2Bauman Moscow State Technical University, Moscow, Russian Federation
3Research Center of Neurology, Moscow, Russian Federation
4Charité - University Medicine Berlin, Berlin, Germany
(mnazarova@hse.ru)
Rigid somatotopical organization of M1 which has been a dominant concept for several decades after the iconic works of H. Jackson and Penfield & Boldrey has been proven to be an oversimplified view. Using micro-scale approaches like microstimulation it was shown that functional aspects of movement such as direction, velocity or joint angles could be also represented in the M1. At the macro-scale level one of the closest approaches to microstimulation is an approach of focal non-invasive brain stimulation such as transcranial magnetic stimulation (TMS), especially MRI navigated TMS (nTMS). One of the main issues is a nTMS motor maps reproducibility, without understanding of which the study of the dynamical changes is hard. Another question is whether additional parameters of nTMS motor cortical representations and/or their overlaps may have any additional informational value. We hypothesized that the overlap of cortical muscle representations may be linked to the performance of fine motor movements and to the excitation/inhibition balance in the brain, which is based on our previous results from the unaffected hemispheres of stroke patients (Nazarova M., PhD thesis). The aim of the current study is to assess the test-retest reproducibility of the nTMS motor maps of several hand muscles' cortical representations and their overlaps during rest and isometric contraction of one of the investigated muscles. First part of the project includes evaluation of test-retest reliability of multi-muscle nTMS motor maps during rest. Healthy right-handed male subjects are enrolled. Sessions consists of multi-muscle nTMS mapping of the cortical representation of the three upper limb muscles: abductor pollicis brevis (APB), abductor digiti mininmi (ADM) and extensor digitorum communis (EDC). Two mapping sessions are performed within 5-8 days of each other. Paired pulse TMS phenomena – short interval cortical inhibition (SICI) and intracortical facilitation (ICF) are assessed in all volunteers. In addition, all subjects undergo EEG investigation during rest and self-paced movements performance. The reproducibility of the profiles of the muscle cortical representation and their overlaps are analyzed using custom-made software for TMS mapping analysis (http://tmsmap.ru/) with the metrics used for the comparison of two probability distributions – earth mover's distance (EMD. The results including the repeatability of standard parameters like the size of the cortical muscle representation, hotspot and center of gravity location and additional parameters such as the volume of the representation, the shape of the area, and the profile of the muscle cortical representation and their overlap would be presented.
Predicting the unpredictable: learning morphological alternations in Russian-speaking
children
Ekaterina Tomas1, Ruben van de Vijver2, Katherine Demuth3, & Peter Petocz3
1Center for Language and Brain, National Research University Higher School of Economics,
2Heinrich Heine University Düsseldorf,
3Macquarie University
Morphophonological alternations interfere with young children’s production of grammatical morphemes. In addition, the ability to generalise even systematic patterns (e.g., final devoicing) is challenging for children under 5 years (Kerkhoff, 2007; Zamuner et al., 2011; Tomas, Demuth & Petocz, in submission) and it is unclear whether less predictable/regular patterns remain lexically restricted. This paper explored these effects in Russian-speaking children (aged 4;0–7;11), using a ‘wug’ test with real and nonce words. Depending on phonological structure, the participants were expected to delete target vowels (e.g., ko'mokNom,sg – kom'kaGen,sg) or preserve them (e.g., pji'lotNom,sg – pji'lotaGen,sg). The results showed that children’s sensitivity to morphophonological patterns increases with age, with 4- year-olds tending to preserve vowels across in both real and nonce words, and older children demonstrating highly variable behaviour when producing the latter. Children’s growing accuracy with real words indicated high lexicalisation of the vowel deletion pattern. Interestingly, the data also suggested that morphophonological patterns interact, with stressed vowels being more frequently correctly alternated or preserved.
References:
Kerkhoff, A. (2007). Acquisition of Morpho-Phonology: The Dutch voicing alternation. PhD Dissertation. Utrecht: LOT.
Tomas, E., Demuth, K. & Petocz, P. (in submission). Perception and production of English present and past tense allomorphs in children with and without SLI. Journal of Speech, Language & Hearing Research.
Zamuner, T. S., Kerkhoff, A., & Fikkert, P. (2011). Phonotactics and morphophonology in early child language: Evidence from Dutch. Applied Psycholinguistics, 33(03), 481–499.http://doi.org/10.1017/S0142716411000440
Object and action naming in Russian individuals with epilepsy
Anna Yurchenko1, Alexander Golovteev2,3, Dmitry Kopachev2, & Olga Dragoy1
1National Research University Higher School of Economics, Moscow, Russia
2Burdenko Neurosurgical Institute, Moscow, Russia
3Epilepsy Center, Moscow, Russia
Language assessment at the single word level in individuals with epilepsy is usually limited to comprehension and production of nouns. The goal of the present study was to investigate production as well as comprehension of both nouns and verbs in Russian individuals with temporal lobe epilepsy (TLE) as compared to healthy controls. The results showed that difficulties in single word processing are more prominent in individuals with the left TLE than with the right TLE. Moreover, comprehension and production of verbs in individuals with TLE is impaired to the same or even greater extent as compared to comprehension and production of nouns. The observed single word processing impairments in individuals with TLE may be related to structural changes and brain network reorganization, which involve both hemispheres – ipsi- and contralateral to the epileptogenic focus.
Нашли опечатку?
Выделите её, нажмите Ctrl+Enter и отправьте нам уведомление. Спасибо за участие!
Сервис предназначен только для отправки сообщений об орфографических и пунктуационных ошибках.