Когнитивная наука в Москве: новые исследования. Материалы конференции 15 июня 2017 г.
It was previously shown that the features of individual items retrieved from visual working memory (VWM) are systematically biased towards the mean feature of a sample set (Brady & Alvarez, 2011), suggesting hierarchical encoding in VWM. In our work, we investigated how hierarchical representations are stored over time. Observers were shown four differently oriented triangles for 200 ms and, after 1-, 4-, or 7-second delay, they had to report either one individual orientation, or the average orientation of all triangles, rotating a probe circle. Before set presentations, observers were informed that they had to remember one particular orientation, all four individual orientations, or the average orientation. Using the mixture model (Zhang & Luck, 2008), we estimated a probability of a tested representation being in VWM and its precision, as well as a systematic bias that would indicate hierarchical encoding. We found a strong bias towards the mean in the “remember four” condition, which provides evidence for hierarchical encoding in VWM. Our main result was the absence of significant changes in retaining the elements of a hierarchical representation (the mean and individual features). This supports an idea that hierarchical representations are related to encoding, rather than storing in VWM. Both fidelity and the probability of an item being in memory decrease over time. It supports "Sudden Death" and "Gradual Decay" accounts for storing hierarchical representations.
Currently there is a need for standardized language assessment test in Russian. Our group has developed Russian Aphasia Test (RAT) based on modern psycholinguistic models and psychometric principles, taking into account existing standardized tests in other languages. RAT allows to separately assess each level of linguistic processing: from phonemic perception to discourse. Here we present the design of the test and the first results of its approbation in groups of neurologically healthy participants and individuals with aphasia. Preliminary results demonstrated that the test is sensitive to language deficits and their severity. Thus, RAT is a practical instrument for language assessment in aphasia which can be used both in clinical practice and for research purposes. At present test standardization in a large group of participants with and without aphasia is on-going with the goal of developing appropriate clinical and age norms.
Alexander Luria is the founder of Russian neuropsychology, and his work is acclaimed worldwide. Since 2015, a research group from the Lomonosov Moscow State University Psychology Department and the Higher School of Economics' Neurolinguistics Laboratory has been working on a digital version of the Alexander Luria archive. This abstract contains descriptions of the key stages of the project: the digital system design development, systematization and annotation of the archive materials, document scanning, and the technical task and software development. We also present a showcase study of the unique archive materials that shed light on the history and methodology of neuropsychology, neurolinguistics, and cognitive science in general.
Numerous studies have found a reduced speed of linguistic processing in older adults, particularly in sentence processing. However, it has been suggested that the reduction of processing speed in older adults is a strategy that they adopt for reasons of caution in order to avoid potential error and/or to conserve "cognitive resources". This hypothesis has been tested in the linguistic domain only once. To test the hypothesis about the strategic nature of slower sentence processing speed in older adults, our experiment address whether they have a greater tendency than younger adults to use a processing speed that is slower than their performance limits. Participants were presented with syntactically complex sentences word-by-word, first in a self-paced mode and then in two externally-paced sessions: (1) at their individual median speed from the self-paced session and (2) twice as fast. After each sentence, participants answered a comprehension question with two response options. We assumed that an external speed increase will cause a greater decrease in response accuracy in younger than older adults, because older adults tend to choose a self-paced sentence processing speed that is further away from their performance limits. However, in the current sample, the hypothesis was not confirmed. Thus, it seems that the slowdown of language processing in older age is a compensatory mechanism, inevitable and necessary to provide the same level of language processing as at a younger age.
The mechanisms of lateralization of language processing are still not fully understood by neurolinguistics today. The current study aims to study the relation between language lateralization and such factors as individual handedness, familial sinistrality and tractography metrics of the corpus callosum (CC). We collected fMRI and DTI data, as well as information about individual handedness and familial sinistrality in 50 neurologically healthy Russian speakers. According to the results, language lateralization is related to the volume and fractional anisotropy of CC, as well as individual handedness. Specifically, people with greater right-hand preference and people with a larger volume and higher fractional anisotropy of CC have greater lateralization of language-related activation to the left hemisphere of a brain.
Age-related changes in language processing have not yet been as well-studied as those in perception, memory, attention or cognition. Specifically with regard to syntactic processing, it is still debatable whether only the processing speed or also accuracy decreases with age. The present study investigated the effect of age and individual differences on syntactic processing in healthy adults. Specifically, we tested the effect of age on the speed of reading syntactically complex sentences and the accuracy of their comprehension, and explored the neural correlates of individual differences in speed and accuracy when taking age into account. The analysis was limited to white matter and used diffusion tensor imaging and tract-based spatial statistics to analyze fractional anisotropy of white-matter tracts. The reading speed was found to become slower with age; however, sentence comprehension accuracy was unaffected by age. Thus, similar to the processing speed decrease in many other cognitive domains, a decrease in sentence processing speed seems to be a compensatory mechanisms that ensures that processing accuracy is maintained. The study did not find any significant correlates of individual differences in syntactic processing accuracy, which is likely due to small sample size.
Transcranial magnetic stimulation (TMS) is a modern non-invasive approach to study brain organization in humans. In TMS a time-varying magnetic fields generate induced electrical currents in the targeted brain regions with focal location of its maximum. Using of MRI navigation systems allows to fully realizing the advantages of TMS focality for brain mapping purposes. Due to this development, nowadays motor and speech nTMS mapping is becoming a routinely used procedure in neurosurgery. However, nTMS mapping for dynamic cortical assessment, for example, to study neuroplastic changes is still limited. An important reason for that is a lack of a standardized methodology for nTMS mapping results assessment. Here we propose TMSmap – a standalone graphical interface software for quantative analysis of the results of motor nTMS mapping (http://tmsmap.ru/), which allows considering both standard parameters like the size of the cortical muscle representation, the hotspot and the center of gravity location, as well as the additional ones such as the volume of the representation, the profile of the muscle cortical area and the overlap between the cortical representations and other user-defined parameters. The input data includes coordinates of the coil position and the response in each point of stimulation and individual structural MRI data.
The influence of a perceiver’s emotional state on the perception of emotion was studied. Experiments were focused on the emotion congruency effect. Emotional congruency is observed when an individual perceiving ambivalent facial expressions tends to see an emotion that corresponds to his/her momentary emotional state. In two experiments, emotional congruency was obtained for happy and sad emotional states. It was hypothesized that emotional congruency would be stronger in the earlier stages of information processing. To test this suggestion, four durations of stimulus presentation were used, namely 50, 100, 200, and 1000 ms. We found no dependence of emotion congruency from stimulus duration.
Previous studies showed that the brain response to a written word depends on whether the word is a target of a lexico-semantic task or is only read. Here we aimed to examine whether the task that uses the presented word not as the target but a cue to produce another word still modifies its recognition process. Using MEG and magnetic source imaging, we compared the spatio-temporal pattern of the brain responses elicited by a noun cue when it was read silently, either without an additional task (SR) or with a requirement to produce an associated verb (VG). We found that the task demands penetrated into early (200 - 300 ms) and late (500 - 800 ms) stages of written word processing by enhancing the brain response under the VG versus SR condition. The cortical sources of the early differential response were localized to the bilateral inferior occipito-temporal and anterior temporal cortices, suggesting elaborate orthographic and lexico-semantic analysis in the VG task. A late effect was observed in the middle and superior temporal gyri and the motor representation of articulators bilaterally and can be associated with enhanced sensorimotor transformations under the VG condition. Overall, our results suggest that written word processing depends on the task goal while intensified linguistic processing recruits bilaterally lateralized networks.
The theory of cognitive dissonance suggests that individuals prefer new incoming information to be consistent with already existing knowledge. Conflicting or inconsistent information results in an emotionally uncomfortable state called cognitive dissonance. Cognitive dissonance theory suggests that a choice between two similarly valued alternatives creates psychological tension (cognitive dissonance) that is reduced by a post-decision re-evaluation of the alternatives. According to the action-based model of cognitive dissonance, activity in the posterior medial prefrontal cortex (pMFC) underlies the detection of cognitive conflicts and the reduction of the dissonance. Nevertheless, the neurocomputational foundation of cognitive dissonance remains unclear. In this study, for the first time we show that cathodal transcranial direct current stimulation (tDCS) of the pMFC significantly reduced post-decision re-evaluation of the alternatives. An ongoing follow-up study that applied anodal tDCS to the pMFC preliminarily showed a tendency to increase choice-induced preference changes. Our results suggest that cognitive dissonance, underlined by the activity of the prefrontal cortex, is a part of the performance-monitoring circuitry
Activation of the limbic structures, and above all the amygdala, is believed to be evidence of a subjective emotional experience. However, a recent paper published by Boubela and colleagues (Boubela et al, 2015) discredits most of the fMRI research concerning BOLD-signal changes in the amygdala region. It was revealed that the BOLD signal increase during perception of affective stimuli might take its origin not from the grey matter itself, but from a large vessel situated near the amygdala: the basal vein of Rothental. In the present research, we attempted to estimate the contribution of veins and venules located near amygdala into BOLD signal changes from the region of interest. We found no systematic increase of BOLD signal in the group of veins within the region of interest in comparison with the BOLD signal in the actual gray matter of the amygdala. At the same time, correlations were found between the BOLD signal in the vessels and the gray matter. The obtained results allow us to interpret the BOLD signal from veins and venules in the region of interest as non systematic noise capable of masking or weakening the observed experimental effects. The method used in this work can be recommended for further fMRI studies of the amygdala.
Using resting state fMRI for individual brain mapping in preneurosurgical planning is discussed. The purpose of our study was to compare the localization of eloquent cortex (motor, speech and executive areas) obtained from resting state fMRI (rsfMRI) and from task based fMRI (tbfMRI). The average percent of overlap of motor areas obtained by the two fMRI methods was appreciable (median 41 - 70), but the overlap revealed for speech and executive areas was very small (median 0 - 23). A significant correlation was revealed between the laterality indexes obtained from tbfMRI and seed-based analysis of rsfMRI. The substantial inter-individual variability in overlap for areas of eloquent cortex mapped by the two different fMRI methods appeals to the necessity of verification of rsfMRI data by direct cortical electrostimulation before using this method in routine clinical applications.
Brain mapping is a complex scientific and practical challenge. One of the most important non-invasive methods of brain mapping is functional MRI (fMRI). FMRI has some limitations because of false positive BOLD-signal changes in region-of-interest draining veins. We can localize veins and venules after fMRI brain mapping, for example with MR-sequence Susceptibility Weighted Imaging (SWI; Haacke, Ye, 2004), in order to understand whether fMRI results are influenced by vein signals or not. We worked out a computer-aided protocol for individual veins and venules localization on the basis of SWI images for users of the SPM12 software package designed for the analysis of fMRI. We used the method described by Wilson (2014) with some modifications. The protocol was tested on data from 48 healthy volunteers.
This paper explores in more detail the phenomenon of deactivation of the auditory cortex evoked by the presentation of the affective pictures that we accidentally found in one of our recent studies (Litvinova et al., 2016). Data from two previous studies employing affective pictures (Litvinova et al., 2016 and Rozovskaya et al., 2014, 2016) were re-analyzed and demonstrated a similar pattern of results. Emotionally negative scenes elicited significant deactivation in the auditory cortex in passive viewing and during encoding of pictures into memory. Emotional valence (positive, negative or neutral) significantly affected BOLD signal change in the auditory cortex under conditions of passive viewing (with the most pronounced deactivation evoked by mutilation pictures), but not in the memory task condition. However, mere presentation of an affective picture is not enough to induce deactivation in the auditory cortex. Unlike encoding, retrieval of the affective pictures from working memory evokes a significant positive BOLD response. A pronounced leftward hemispheric asymmetry of this response suggests that it may reflect a role of verbal processes in the retrieval of information from the working memory. Overall, our results suggest that the observed phenomenon may reflect a neural signature of inattentional deafness (ID): a failure to perceive auditory stimuli which manifests under high visual perceptual load. Therefore, we predict that passive viewing and memorization of negative affective pictures, but not their retrieval from WM, would be accompanied by the ID effect at the behavioral level.
We analyzed brain functional connectivity using data from a task-based fMRI study of explicit categorization of neutral and emotional faces. During the perception of emotional faces, the correlation of BOLD signal in the right posterior superior temporal sulcus (STS) with other regions of the core face system increased, as compared to neutral faces. Conversely, during the perception of neutral faces, we found an increased correlation of STS and other core face system regions with brain areas outside the core system. The functional connectivity of STS shows lateralization due to the presence or absence of emotional expression. The obtained results support the key role of STS in facial expression recognition.
Despite being a routine technique for presurgical motor assessment, transcranial magnetic stimulation (TMS) mapping is underused for probing of neuroplastic brain changes. We investigated the test-retest reproducibility of the TMS cortical maps of several hand muscles using both standard and alternative parameters of the cortical representation.Pilot study results for four healthy right-handed male volunteers (19-33y.o.) are presented. Two TMS mapping sessions with the stimulation of the left motor cortex were performed within 5-10 days (Day1 and Day2). Day2 points repeated an exact order of the Day1. For quantative comparison of 3D profiles similarities earth mover's distance metrics was used. Analysis of nTMS maps was performed using custom-made software TMSmap (http://tmsmap.ru).The between-days difference in the area of cortical representation for four analyzed subjects was 14.5-30.4% for one and 3.9-11.2% for five repetitions of each cortical point. Considering 3D profiles of cortical representation, higher similarity was shown for the same muscles’ representations and their overlaps compared to the representations of the different muscles. The study is ongoing, further analyzed results will be present.