Ivan A. Basyul
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Ivan A. Basyul has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Alexander V. Zhegallo
Institute of Psychology of the Russian Academy of Sciences
9 shared publications - Kristina I. Ananyeva
Institute of Psychology of the Russian Academy of Sciences
5 shared publications - Alexander A. Demidov
Moscow Institute of Psychoanalysis
4 shared publications - Alexander N. Kharitonov
Institute of Psychology, Russian Academy of Sciences, 129366 Moscow, Russia
3 shared publications - Elena A. Lupenko
Moscow State University of Psychology and Education
3 shared publications - Evgeny G. Khoze
Moscow State University of Psychology and Education
3 shared publications
Latest Publications
- Open Access
- 2 October 2026
- Published by MDPI AG in Journal of Eye Movement Research
- Vol. 19 (5) , 112
Through the Looking Glass: Eye Tracking as a Cognitive Benchmark for Language-Model Semantics and Observer-Specific Semantic Alignment
Assessing semantic representations learned by language models remains challenging because conventional intrinsic benchmarks capture limited aspects of meaning and often correlate weakly with downstream performance. Eye tracking offers complementary, cognitively grounded evidence: gaze measures recorded during reading and scene viewing index online processing and attentional selection, although they should not be treated as readouts of semantic representations. This narrative review and methodological framework synthesizes research linking distributional and contextual language representations to eye-movement behavior, focusing on semantic relatedness, predictability, surprisal, model scale, layer-dependent alignment, and model–human correspondence. Evidence from naturalistic reading corpora shows that embedding-derived relatedness and language-model surprisal explain variance in fixation measures, while recent work demonstrates that cognitive alignment depends on model size, layer, tokenization, task, and measurement regime. We propose extending this approach from population-level model evaluation to observer-specific gaze-derived relational structure. For visual scenes, gaze transitions can be represented as candidate choices and relational graphs and compared with multiple semantic-relation matrices derived from language or vision–language models. Because gaze transitions are influenced by spatial distance, object size, salience, task demands, context, and preprocessing decisions, these factors must be controlled and, where causal isolation is intended, manipulated independently of semantic relations. We propose combining candidate-normalized hierarchical choice models, residualized gaze relations, representational similarity analysis with reliability and uncertainty assessment, and required independent behavioral validation to characterize convergence and divergence between model semantic geometry and observer-specific relational organization.
- 2 October 2026
- Published by MDPI AG in Journal of Eye Movement Research
- Vol. 19 (5) , 112
Open AccessThrough the Looking Glass: Eye Tracking as a Cognitive Benchmark for Language-Model Semantics and Observer-Specific Semantic Alignment
Assessing semantic representations learned by language models remains challenging because conventional intrinsic benchmarks capture limited aspects of meaning and often correlate weakly with downstream performance. Eye tracking offers complementary, cognitively grounded evidence: gaze measures recorded during reading and scene viewing index online processing and attentional selection, although they should not be treated as readouts of semantic representations. This narrative review and methodological framework synthesizes research linking distributional and contextual language representations to eye-movement behavior, focusing on semantic relatedness, predictability, surprisal, model scale, layer-dependent alignment, and model–human correspondence. Evidence from naturalistic reading corpora shows that embedding-derived relatedness and language-model surprisal explain variance in fixation measures, while recent work demonstrates that cognitive alignment depends on model size, layer, tokenization, task, and measurement regime. We propose extending this approach from population-level model evaluation to observer-specific gaze-derived relational structure. For visual scenes, gaze transitions can be represented as candidate choices and relational graphs and compared with multiple semantic-relation matrices derived from language or vision–language models. Because gaze transitions are influenced by spatial distance, object size, salience, task demands, context, and preprocessing decisions, these factors must be controlled and, where causal isolation is intended, manipulated independently of semantic relations. We propose combining candidate-normalized hierarchical choice models, residualized gaze relations, representational similarity analysis with reliability and uncertainty assessment, and required independent behavioral validation to characterize convergence and divergence between model semantic geometry and observer-specific relational organization.
- Open Access
- 30 June 2026
- Published by Moscow State University of Psychology and Education
- Vol. 19 (2) , 67-90
Typology of facial expressions in study participants during joint solution of the discriminatory AB-X task mediated by video communication
In a paired experiment, subjects solved a discriminatory AB-X task. One subject was briefly shown two similar images of transient emotional expressions, A and B, while the other was shown a target image X, matching either A or B. The participants' facial expressions were videotaped throughout the study. For further analysis of facial expressions, we used facial landmarks marking, individual principal component analysis, and a further search for facial expression patterns common to the study participants using k-means cluster analysis. The initial data for the cluster analysis were the loading vectors of individual principal components (up to the 16th component), describing individual complex facial expression patterns. Based on the results of the cluster analysis, 10 clusters were described, fully including the first 5 principal components and partially including components 6 through 12. The identified clusters correspond to the following changes: high-intensity horizontal, vertical, and diagonal shifts of facial position; medium-intensity smile with squinting of the eyes and lowering of the eyebrows; medium-intensity head tilts to the left and right; medium-intensity mouth opening/closing; low-intensity compression of the mouth with simultaneous squinting of the eyes; low-intensity changes in the eyebrow and mouth area; low-intensity left-right rotations around the vertical axis. Further study of changes in the intensity of the generalized principal components during the solution of the target task is planned. It is assumed that this will make it possible to assess the functional role of facial expressions in the conducted model experiment.
- 30 June 2026
- Published by Moscow State University of Psychology and Education
- Vol. 19 (2) , 67-90
Open AccessTypology of facial expressions in study participants during joint solution of the discriminatory AB-X task mediated by video communication
In a paired experiment, subjects solved a discriminatory AB-X task. One subject was briefly shown two similar images of transient emotional expressions, A and B, while the other was shown a target image X, matching either A or B. The participants' facial expressions were videotaped throughout the study. For further analysis of facial expressions, we used facial landmarks marking, individual principal component analysis, and a further search for facial expression patterns common to the study participants using k-means cluster analysis. The initial data for the cluster analysis were the loading vectors of individual principal components (up to the 16th component), describing individual complex facial expression patterns. Based on the results of the cluster analysis, 10 clusters were described, fully including the first 5 principal components and partially including components 6 through 12. The identified clusters correspond to the following changes: high-intensity horizontal, vertical, and diagonal shifts of facial position; medium-intensity smile with squinting of the eyes and lowering of the eyebrows; medium-intensity head tilts to the left and right; medium-intensity mouth opening/closing; low-intensity compression of the mouth with simultaneous squinting of the eyes; low-intensity changes in the eyebrow and mouth area; low-intensity left-right rotations around the vertical axis. Further study of changes in the intensity of the generalized principal components during the solution of the target task is planned. It is assumed that this will make it possible to assess the functional role of facial expressions in the conducted model experiment.
- Open Access
- 22 December 2025
- Published by Moscow State University of Psychology and Education
- Vol. 18 (4) , 30-46
Organization of oculomotor activity of a person in the joint solution of a discriminatory AB-X task through video communication
In a paired experiment, the subjects solved a discriminatory AB-X task. One of the subjects was briefly shown two similar images of transitional emotional expressions A and B, and the other was shown a target image of X that matches either A or B. The study participant who was shown the target image of X should give an answer to which of the images of A or B corresponds to X. An asymmetry of erroneous responses was found for the extreme pairs of the transition series; the images at the extreme of the transition series are preferably selected as the answers. Since the task is not purely perceptual, the detected effect should be considered as related to the interpretation of verbal descriptions of images given by the participants. The analysis of eye movements showed a high proportion of the time during which the communicant's face is addressed during the discussion: for the role of X, Me = 0.74; for the role of AB, Me = 0.97. The proportion of time during which mutual gaze is simultaneously addressed to each other's faces is also quite high, Me = 0.55. When addressing the communicant's face, the lower part of the face dominates in terms of viewing time and duration of fixations. There are fundamental differences here from the task of identifying emotional expressions, in which the upper part of the face is dominant in terms of viewing time. The obtained results raise the question of the functional role of expressions of the observer's face and gaze in communication.
- 22 December 2025
- Published by Moscow State University of Psychology and Education
- Vol. 18 (4) , 30-46
Open AccessOrganization of oculomotor activity of a person in the joint solution of a discriminatory AB-X task through video communication
In a paired experiment, the subjects solved a discriminatory AB-X task. One of the subjects was briefly shown two similar images of transitional emotional expressions A and B, and the other was shown a target image of X that matches either A or B. The study participant who was shown the target image of X should give an answer to which of the images of A or B corresponds to X. An asymmetry of erroneous responses was found for the extreme pairs of the transition series; the images at the extreme of the transition series are preferably selected as the answers. Since the task is not purely perceptual, the detected effect should be considered as related to the interpretation of verbal descriptions of images given by the participants. The analysis of eye movements showed a high proportion of the time during which the communicant's face is addressed during the discussion: for the role of X, Me = 0.74; for the role of AB, Me = 0.97. The proportion of time during which mutual gaze is simultaneously addressed to each other's faces is also quite high, Me = 0.55. When addressing the communicant's face, the lower part of the face dominates in terms of viewing time and duration of fixations. There are fundamental differences here from the task of identifying emotional expressions, in which the upper part of the face is dominant in terms of viewing time. The obtained results raise the question of the functional role of expressions of the observer's face and gaze in communication.
- Open Access
- 24 October 2024
- Published by Moscow State University of Psychology and Education in Experimental Psychology (Russia)
- Vol. 17 (3) , 16-29
Verbal Portraits of Strong and Weak Emotional Facial Expressions
The article presents the results of an analysis of verbal data obtained when people compared images of emotional expressions of different intensities. The assumption is confirmed that the perceived quality of the same expression (in accordance with the categorization given in Ekman’s papers), but with different degrees of its severity, can reflect the perception of different objects. This assumption turned out to be true for both facial and affective verbal portraits - empirical referents of the perceived quality of expressions. The results obtained confirm the conclusion made in previous works that the comparison of data on the perceived quality of expressions and their list presented in Ekman’s papers is not completely legitimate, since we are not talking about the perception of the corresponding image, but about performing the task of identifying a specific expression. Accordingly, in the first case, the result is characteristics of perceived quality, in the aggregate of which the sign of a given expression will not necessarily come to the fore. In the second case, the image is assessed by the researcher himself based on data on the degree to which the subject recognizes the expression, the content of which is specified by the instructions.
- 24 October 2024
- Published by Moscow State University of Psychology and Education in Experimental Psychology (Russia)
- Vol. 17 (3) , 16-29
Open AccessVerbal Portraits of Strong and Weak Emotional Facial Expressions
The article presents the results of an analysis of verbal data obtained when people compared images of emotional expressions of different intensities. The assumption is confirmed that the perceived quality of the same expression (in accordance with the categorization given in Ekman’s papers), but with different degrees of its severity, can reflect the perception of different objects. This assumption turned out to be true for both facial and affective verbal portraits - empirical referents of the perceived quality of expressions. The results obtained confirm the conclusion made in previous works that the comparison of data on the perceived quality of expressions and their list presented in Ekman’s papers is not completely legitimate, since we are not talking about the perception of the corresponding image, but about performing the task of identifying a specific expression. Accordingly, in the first case, the result is characteristics of perceived quality, in the aggregate of which the sign of a given expression will not necessarily come to the fore. In the second case, the image is assessed by the researcher himself based on data on the degree to which the subject recognizes the expression, the content of which is specified by the instructions.
- Open Access
- 16 July 2024
- Published by Moscow State University of Psychology and Education in Experimental Psychology (Russia)
- Vol. 17 (2) , 29-51
Relationship of Mimic and Affective Characteristics in Verbal Comparison of Faces
The results of the analysis of verbalizations obtained in the situation of verbal comparison of images of emotional expressions from the VEPEL database are presented. It has been shown that specific information about expression is quantitatively presented in the verbal comparison data. The analysis of verbalizations, performed using the tools of the perceived quality paradigm, made it possible to differentiate the features of expressions, presenting them in two types of verbal portraits: affective and mimic. Comparison of these two types of verbal portraits allows to interpret affective judgments in mimic terms, and vice versa. The verbal features included in mimic verbal portraits were compared with descriptions of mimic characteristics of basic expressions according to P. Ekman. It is shown that many of the features identified in the situation of verbal comparison coincide with the descriptions proposed by P. Ekman, who, however, enumerated them without indicating their relative value.
- 16 July 2024
- Published by Moscow State University of Psychology and Education in Experimental Psychology (Russia)
- Vol. 17 (2) , 29-51
Open AccessRelationship of Mimic and Affective Characteristics in Verbal Comparison of Faces
The results of the analysis of verbalizations obtained in the situation of verbal comparison of images of emotional expressions from the VEPEL database are presented. It has been shown that specific information about expression is quantitatively presented in the verbal comparison data. The analysis of verbalizations, performed using the tools of the perceived quality paradigm, made it possible to differentiate the features of expressions, presenting them in two types of verbal portraits: affective and mimic. Comparison of these two types of verbal portraits allows to interpret affective judgments in mimic terms, and vice versa. The verbal features included in mimic verbal portraits were compared with descriptions of mimic characteristics of basic expressions according to P. Ekman. It is shown that many of the features identified in the situation of verbal comparison coincide with the descriptions proposed by P. Ekman, who, however, enumerated them without indicating their relative value.
- Open Access
- 27 October 2023
- Published by Moscow State University of Psychology and Education in Experimental Psychology (Russia)
- Vol. 16 (3) , 53-68
Predictors of Nonverbal Recognition of Emotional Facial Expressions
The work was carried out in the paradigm of analyzing the structure of erroneous responses when performing the identification of emotional facial expressions. The subjects performed the task of non-verbal recognition of facial expressions at tachistoscopic exposure time (16 ms, 8 ms, 4 ms). The search for predictors that determine the partial proportions of responses for different combinations (response, expression shown) was performed among the subscales of the Boyko empathy questionnaire, the EmIn questionnaire, the analytical-holistic scale, and the Alekseev thinking styles questionnaire. The study involved students from Moscow universities. Sample size: 49 people, 9 men, 40 women, average age m=22.2 years, sd=4.45.The identified predictors are specific (not universal) in nature, different for different combinations of the presented image and the assessment performed. We believe that this result should be interpreted as indicating the presence of competing parallel processes associated with testing for the presence of different emotional expressions. The selected system of predictors is not complete. A promising direction for further research is expanding the scope of the search for predictors through methods that measure individual personality characteristics and the level of anxiety.
- 27 October 2023
- Published by Moscow State University of Psychology and Education in Experimental Psychology (Russia)
- Vol. 16 (3) , 53-68
Open AccessPredictors of Nonverbal Recognition of Emotional Facial Expressions
The work was carried out in the paradigm of analyzing the structure of erroneous responses when performing the identification of emotional facial expressions. The subjects performed the task of non-verbal recognition of facial expressions at tachistoscopic exposure time (16 ms, 8 ms, 4 ms). The search for predictors that determine the partial proportions of responses for different combinations (response, expression shown) was performed among the subscales of the Boyko empathy questionnaire, the EmIn questionnaire, the analytical-holistic scale, and the Alekseev thinking styles questionnaire. The study involved students from Moscow universities. Sample size: 49 people, 9 men, 40 women, average age m=22.2 years, sd=4.45.The identified predictors are specific (not universal) in nature, different for different combinations of the presented image and the assessment performed. We believe that this result should be interpreted as indicating the presence of competing parallel processes associated with testing for the presence of different emotional expressions. The selected system of predictors is not complete. A promising direction for further research is expanding the scope of the search for predictors through methods that measure individual personality characteristics and the level of anxiety.
- Open Access
- 15 August 2023
- Published by Moscow State University of Psychology and Education in Experimental Psychology (Russia)
- Vol. 16 (2) , 203-217
The Relations of Constructs Measured by the Boyko Empathy Questionnaire and the EmIn Questionnaire
A comparison was made between each other at the level of subscales of the Boyko empathy questionnaire and the EmIn questionnaire. Additionally, the subscales of the questionnaires were compared with the results of the genetic analysis of COMT, DRD4, 5HTR2A. It is shown that the subscales of the questionnaires are mutually complementary. The revealed relationships of gene polymorphism with a number of subscales of both the Boyko empathy questionnaire and the EmIn questionnaire as a whole do not contradict existing data on the mechanisms in the regulation of which the studied genes are involved. At the same time, empathy, as measured by Boyko's questionnaire, is obviously not a monolithic construct. We believe that the final indicator of EEI, measured by the EmIn questionnaire, should also not be considered as a monolithic construct.
- 15 August 2023
- Published by Moscow State University of Psychology and Education in Experimental Psychology (Russia)
- Vol. 16 (2) , 203-217
Open AccessThe Relations of Constructs Measured by the Boyko Empathy Questionnaire and the EmIn Questionnaire
A comparison was made between each other at the level of subscales of the Boyko empathy questionnaire and the EmIn questionnaire. Additionally, the subscales of the questionnaires were compared with the results of the genetic analysis of COMT, DRD4, 5HTR2A. It is shown that the subscales of the questionnaires are mutually complementary. The revealed relationships of gene polymorphism with a number of subscales of both the Boyko empathy questionnaire and the EmIn questionnaire as a whole do not contradict existing data on the mechanisms in the regulation of which the studied genes are involved. At the same time, empathy, as measured by Boyko's questionnaire, is obviously not a monolithic construct. We believe that the final indicator of EEI, measured by the EmIn questionnaire, should also not be considered as a monolithic construct.
- 1 January 2022
- Published by Moscow Institute of Psychoanalysis in Cognition and Experience
- Vol. 3 (3) , 64-86
Natural dynamic expressions of children’s basic emotions: induction and self-assessment
В работе представлены результаты выполнения исследовательского проекта, направленного на проработку проблем диагностики и коррекции эмоциональной сферы детей. С учетом возрастных особенностей разработаны процедуры индукции и самооценки эмоциональных состояний у детей. Подготовлен и апробирован с участием детей и подростков визуальный контент для индукции и самооценки эмоциональных состояний детей. В результате получены динамические и статические естественные экспрессии индуцированных базовых эмоций. Подготовлены уникальные динамические и статические слабо- и ярко выраженные экспрессии естественных индуцированных эмоций детей, предназначенные для использования в качестве стимульного материала в экспериментальных исследованиях, изучающих особенности восприятия и выражения базовых...
- 1 January 2022
- Published by Moscow Institute of Psychoanalysis in Cognition and Experience
- Vol. 3 (3) , 64-86
Natural dynamic expressions of children’s basic emotions: induction and self-assessment
В работе представлены результаты выполнения исследовательского проекта, направленного на проработку проблем диагностики и коррекции эмоциональной сферы детей. С учетом возрастных особенностей разработаны процедуры индукции и самооценки эмоциональных состояний у детей. Подготовлен и апробирован с участием детей и подростков визуальный контент для индукции и самооценки эмоциональных состояний детей. В результате получены динамические и статические естественные экспрессии индуцированных базовых эмоций. Подготовлены уникальные динамические и статические слабо- и ярко выраженные экспрессии естественных индуцированных эмоций детей, предназначенные для использования в качестве стимульного материала в экспериментальных исследованиях, изучающих особенности восприятия и выражения базовых...
- Open Access
- 1 January 2022
- Published by Moscow State University of Psychology and Education in Experimental Psychology (Russia)
- Vol. 15 (3) , 79-87
Paired Model Experiment “Believe/Do Not Believe” — The Ability to Highlight Non-Verbal Signs Of Reliability/Unreliability of the Information Reported
As markers of reliability/inaccuracy of the reported information, there have been repeated attempts to use the partial facial features identified by experts. Our research proposes a transition to analysis at the level of automatically identified integral mimic patterns. To solve this problem, a model paired experiment was conducted in which participants had to repeatedly make a decision to tell the truth/lie. The analysis of facial expressions was performed by automatically calculating the coordinates of key points and further extracting integral facial expression patterns using the method of principal components. At the level of loads of the main components, signs of making a decision “to lie/tell the truth” for the sample as a whole and for individual subjects are highlighted. It was found that holistic patterns of facial expressions associated with the reliability/unreliability of the reported information at “low rates” are characterized by high individual variability. As a direction for further research on facial expressions, a transition to the study of the individual repertoire of integral patterns of facial expressions is proposed.
- 1 January 2022
- Published by Moscow State University of Psychology and Education in Experimental Psychology (Russia)
- Vol. 15 (3) , 79-87
Open AccessPaired Model Experiment “Believe/Do Not Believe” — The Ability to Highlight Non-Verbal Signs Of Reliability/Unreliability of the Information Reported
As markers of reliability/inaccuracy of the reported information, there have been repeated attempts to use the partial facial features identified by experts. Our research proposes a transition to analysis at the level of automatically identified integral mimic patterns. To solve this problem, a model paired experiment was conducted in which participants had to repeatedly make a decision to tell the truth/lie. The analysis of facial expressions was performed by automatically calculating the coordinates of key points and further extracting integral facial expression patterns using the method of principal components. At the level of loads of the main components, signs of making a decision “to lie/tell the truth” for the sample as a whole and for individual subjects are highlighted. It was found that holistic patterns of facial expressions associated with the reliability/unreliability of the reported information at “low rates” are characterized by high individual variability. As a direction for further research on facial expressions, a transition to the study of the individual repertoire of integral patterns of facial expressions is proposed.
- Open Access
- 1 January 2022
- Published by Moscow State University of Psychology and Education in Experimental Psychology (Russia)
- Vol. 15 (3) , 100-120
Specifics of Perception of Verbal Indications and Graphic Expressions of Basic Emotions by Children of Elementary and Middle School Age
The work studied the components of emotional competence of children of primary and secondary school age using the method of semantic differential. An analysis was made of differences in the distribution of verbal designation assessments and a comparison of the assessment profiles of verbal designations and graphic expressions (of different levels of severity) of 9 basic emotions. It is shown that the lowest level of differences in the assessments of verbal designations of children of primary and secondary school age is observed for the emotions of “joy”, “anger”, “fear” and “disgust”; the most different assessments are “interest”, “contempt” and “shame”; “guilt”, “surprise” and “sadness” occupy an intermediate position. The affective vocabulary of children of primary school age is represented by affective concepts understood at the level of children of secondary school age: “joy”, “anger” and “fear”; as well as characteristics that are extremely pronounced in terms of valence — “friendly / hostile”, and associated with communication needs — “social / unsociable”. The results obtained can be useful in the further use of graphic expressions of basic emotions as a stimulus material for studying the characteristics of perception and in the procedures for self-assessment of emotions when working with children of primary and secondary school age.
- 1 January 2022
- Published by Moscow State University of Psychology and Education in Experimental Psychology (Russia)
- Vol. 15 (3) , 100-120
Open AccessSpecifics of Perception of Verbal Indications and Graphic Expressions of Basic Emotions by Children of Elementary and Middle School Age
The work studied the components of emotional competence of children of primary and secondary school age using the method of semantic differential. An analysis was made of differences in the distribution of verbal designation assessments and a comparison of the assessment profiles of verbal designations and graphic expressions (of different levels of severity) of 9 basic emotions. It is shown that the lowest level of differences in the assessments of verbal designations of children of primary and secondary school age is observed for the emotions of “joy”, “anger”, “fear” and “disgust”; the most different assessments are “interest”, “contempt” and “shame”; “guilt”, “surprise” and “sadness” occupy an intermediate position. The affective vocabulary of children of primary school age is represented by affective concepts understood at the level of children of secondary school age: “joy”, “anger” and “fear”; as well as characteristics that are extremely pronounced in terms of valence — “friendly / hostile”, and associated with communication needs — “social / unsociable”. The results obtained can be useful in the further use of graphic expressions of basic emotions as a stimulus material for studying the characteristics of perception and in the procedures for self-assessment of emotions when working with children of primary and secondary school age.