Изучены спектральные характеристики мю-ритма и корковые взаимодействия на частоте этого ритма у юношей принаблюдении, произнесении и мысленном воспроизведении неэмоционального слова. Обнаружены статистически значимые по сравнению с фоном изменения спектральной мощности ЭЭГ на частотах мю-ритма в центральных областях коры. Характер этих изменений зависит от частоты ритма: на одних частотах наблюдается рост спектральной мощности ЭЭГ, на других – снижение. Обнаружено также усиление уровней корковых связей на частоте мю-ритма между центральными и лобными, центральными и височными, центральными и затылочными зонами коры на этапах подготовки и выполнения коммуникативного действия. Полученные данные свидетельствуют о том, что мю-ритм включает ряд частот, которые имеют разное функциональное значение и отражают активность разных нейросетевых осцилляторов. Установлено, что корковые взаимодействия и спектральные характеристики мю-ритма отличаются при наблюдении, произнесении и мысленном воспроизведении неэмоционального слова.

The study of the role of mirror neurons in cognitive processes is an important problem of modern psychophysiology. Therefore, an urgent task is to search for and study EEG correlates of the activation of mirror neurons. The aim of the research was to study the cortical interactions at the frequency of the mu rhythm and the spectral characteristics of this rhythm in humans at different stages of activity related to observation, pronunciation and mental pronunciation of a non-emotional word. The studies involved volunteers, practically healthy young men (32 people) aged 18 to 23 years old (average age 21 ± 1.6 years), students of Tomsk universities, right-handers. All subjects gave informed consent to participate in this study, which was approved by the Commission on Bioethics of the Biological Institute of Tomsk State University (Tomsk, Russian Federation). Three series of experiments were carried out for the task. In the frst series (“Observation”), the subject observed the operator, who uttered the word “Raz” silently with his lips, when the stopwatch on the monitor crossed the divisions 0, 5, 10, 15, 20, 25, 30, 35, 40, 45, 50 and 55 s. Total stopwatch hand made 5 turns. During the experiment, the operator was located at the table, on which at a distance of 40-50 cm from him was a computer monitor. The subject, at the same time, was in front and to the right at a distance of 70-80 cm from him and watched only his lips. In the second series (“Speaking the word”), the subject himself performed the indicated activity, and in the third series (“The mental pronunciation of the word”), at specifed points in time he also mentally pronounced the word. During the pronunciation or mental reproduction of the word, the subject took the place of the operator and followed the movement of the stopwatch hand on the monitor screen. Before and during the execution of the activity, the monopolar EEG was recorded using the Encephalan-131-03 24-channel encephalograph-analyzer (Medicom company, Taganrog, Russia) in the frontal areas (F3, F4, Fz, F7, F8 ) central (C3, C4, Cz), temporal (T3, T4, T5, T6), parietal (P3, P4, Pz) and occipital (O1, O2) leads by the “10-20%” system. The leads A1 and A2 were used as referents. To exclude artifacts associated with eye movement and muscle activity, EOG and EMG of the neck and forehead muscles were recorded. The sampling rate of EEG recordings was 250 Hz. To study the cortical connections at the frequency of the mu rhythm, the EEG was pre-fltered: a 20-order Butterworth bandpass flter with a frequency suppression factor above 13 Hz at least 80 dB and frequencies below 8 Hz at least 40 dB). When processing the obtained data, we calculated the maximum values of crosscorrelation functions and spectral power estimates for short (1.5 s), devoid of artifacts, EEG recording periods before 3 s (background) and 1.5 s (preparation stage) with the stopwatch of the corresponding division and immediately after the specifed event (stage of execution). The obtained values of the correlation coeffcients and spectral power estimates were averaged separately for each stage of activity, for each series and over all subjects. When calculating the cross-correlation functions, we complied with the existing recommendations (Bendat SJ and Piersol AG.) that the maximum time shift should be no more than one-tenth of the length of the implementation, which was chosen to be greater than or equal to ten periods of mu rhythm (1.5 s). To describe the EEG power spectrum, a Fourier transform was used. The spectrum was calculated with an approximation for integer harmonics (8, 9, 10, 11, 12, 13 Hz), which made it possible to simplify the subsequent statistical processing and comparative analysis of the results. In the statistical processing of data, the MatLab v6.5 package, nonparametric analysis of variance, and the Wilcoxon criterion for related samples were used. The conducted research allowed to detect in the central areas of the cortex when observing, pronouncing and mentally pronouncing a non-emotional word, changes in the spectral power of the EEG, which are statistically signifcant compared to the background, at mu rhythm frequencies during the preparation and execution of the communicative action. We found that the nature of these changes depends on the frequency of the rhythm. In particular, in the “Observation” series, at the stage of performing the action in lead C3 compared to the background, a decrease in the spectral power of the EEG at a frequency of 8 Hz is statistically signifcant (p <0.05), and at a frequency of 10 Hz in the same lead there is a statistically signifcant (p <0.05) increase in the spectral power of the EEG (Fig. 1). In the series “Pronouncing the word” at the “Preparing” stage at a frequency of 9 Hz compared to the background, the decrease in the spectral power of the EEG in lead C4 is statistically signifcant (p <0.05), and at a frequency of 12 Hz in the same lead it was found signifcant compared to the background increase in the spectral power of the EEG (p <0.05) (Fig. 2). The obtained data indicate that the mu wave includes a series of frequencies which have different functional importance and reflect the activity of different neural oscillators. Apparently, the decrease in the spectral power of the mu rhythm observed at some frequencies reflects the activation of mirror neurons. Analysis of cortical interactions allowed to detect statistically signifcant (p <0.05), in a series of observations, pronouncing and mental pronouncing of a word compared to the background, enhancement of cortical connections at the mu rhythm frequency between central and frontal, central and temporal areas of the cortex at the stages of preparing and executing an action, as well as at the stage of executing an action in comparison with preparation. The nature of changes at the levels of cortical communication is different in the series of observation, pronunciation and mental pronunciation of the word. For example, in the Observation series, there was an increase in correlations between leads F3 and C3 at the preparation stage compared with the background (Fig. 3A), as well as increased correlations between leads F7 and C3 at the stage of performing the action compared with the background (Fig. 3B). An increase in the level of correlations at the frequency of the mu rhythm is also observed between leads C3 and C4 at the stage of executing a communicative action compared to its preparation (Fig. 3C). In the “pronunciation of the word” series, we observed statistically signifcant (p <0.05), compared with the background, increased correlations between leads F8 and C3 at the stage of preparation for action (Fig. 4A), as well as between leads F8 and C4, C3 and T5, C3 and O1 at the stage of performing the action (Fig. 4B). At the stage of performing the action, there is a statistically signifcant (p <0.05), compared with the preparation stage, increase in the correlation between the Cz and C4 leads (Fig. 4C). In the “Mental Pronunciation” series, there was a statistically signifcant (p <0.05), compared with the background, amplifcation of correlations between leads Fz and C3, F7 and C4, C3 and T6 at the preparation stage of the communicative action (Fig. 5A), as well as between leads F7 and C3 at the stage of performing the action (Fig. 5B). At the stage of performing the action compared to the stage of preparation, the increased correlations between the leads F7 and C4, Cz and C4, C4 and T5 (Fig. 5C) were statistically signifcant (p <0.05). Probably, the observed enhancements of cortical connections are due to the transmission of signals from the visual, auditory and somatosensory zones of the cortex to the ventral region of the premotor cortex and, close to it, the Broca zone, in which there are apparently communicative mirror neurons. Dispersion analysis revealed a statistically signifcant (p <0.05) influence of the “type of activity” factor on the spectral characteristics of the mu rhythm and cortical interactions at the frequency of this rhythm when observing, pronouncing and mentally pronouncing a non-emotional word. We established that the influence of this factor depends on the stage of the activity performed. So, if at the stage of “Preparation” in lead C3, the influence of the studied factor manifested itself only at frequencies of 8 and 13 Hz (p = 0.0004 ÷ 5.25 · 10-7), then at the stage of “Execution” in the same lead, EEG at all mu rhythm frequencies from 8 to 13 Hz (p = 0.012 ÷ 1.05 · 10-7). If at the “Preparation” stage, the factor under consideration has a statistically signifcant (p = 0.048 ÷ 6.29 · 10-7) effect on the majority of cortical connections (20) between the central and other areas of the cortex, then at the “Execution” stage, the number of such connections decreases twice. The results show that the spectral characteristics of the mu rhythm and cortical interactions at the frequency of this rhythm depend on the type and stage of the performed activity, and that the types of cognitive activity associated with observation, pronunciation and mental pronunciation of the word are provided by different functional systems that, as a mandatory component, include a subsystem of communicative mirror neurons. The paper contains 5 Figures and 20 References.