Alexandr V. Vartanov

Four-dimensional spherical model of emotions was built with metric multidimensional scaling method. This model simultaneously describes emotional expressions in the oral human speech (based on the samples of the word “Yes”, spoken with different expressions) and vocalizations of rhesus macaque monkeys in the human listener’s perception. Based on statistical criteria, number of dimensions in the space of the perception of emotional human speech and monkey vocalizations samples was defined to be equal to 4. Stimuli points in the space are positioned in almost equal distance from the center of the hypersphere. Model space axes were interpreted in regard to the contents of human emotions. Orientation of perception of monkey vocalization samples space axes was conducted in a formal procedure based on direct experimental data — cross-assessment of the differences between some human speech samples and all monkey vocalizations samples. As a result, all studied monkey vocalizations samples were quantitatively presented in a unified system of the characteristics, defined for human emotions, and received an anthropomorphic interpretation. Obtained results are not controversial to the data from animal behavior observation and expert analysis of their vocalizations range. We revealed the similarity between emotional expressions in human speech and monkey vocalization range, which, in turn, demonstrates the relation of human and monkey emotions.

Relevance. The problem of identifying the cerebral mechanisms of internal speech (spoken words or even individual sounds — phonemes) based on the electrical activity of the brain is relevant and significant for a number of practical applications, in particular for improving the construction of brain-computer interfaces.

Target. Electroencephalographic research is aimed at identifying the cerebral mechanisms of internal speech, taking into account the modality of its initialization.

Methodology. In a study on 25 subjects, a 19-channel EEG was recorded in the process of performing three tasks on perception and internal pronunciation of seven phonemes of the Russian language: A, B, F, G, M, R, U, which were presented visually (in the form of letters) and audibly. The analysis of the obtained EEG data was carried out using a new author's method of localizing brain activity “virtually implanted electrode” (developed by A.V. Vartanov), which allows, according to scalp EEG data, to reconstruct electrical activity, the source of which is located in a certain place inside the head (activity was studied in 33 different points of the brain).

Results. The averaged evoked potentials were obtained for all the studied areas of the brain and the functional connections of these structures were revealed in the form of connectivity graphs for the corresponding conditions.

Conclusions. It was found that the processes of pronunciation, expressed in EP, differ depending on the modality of the initial form of the reproduced image — in visual perception, the supromarginal gyrus is more involved. The synchronized activity of a number of cerebral zones providing internal speech was revealed. A special role is played by the activity of the hippocampus and the pallidus (Gl. Palidus). The form of the EP when perceived is in many ways similar to its form when pronouncing. Activation at visual input is almost always significantly higher than at auditory input.

Brain mechanisms of subjective (internal) rhythmisation were examined in two experimental cases. One of the tasks was to combine acoustic rhythmic clicks as pairs; another task was to combine the same monotone clicks as triples. With the method of factorial analysis some types of specific variation of EVP form on the clicks which took different place in patterns were figured out. Analysis of dynamics of brain structures activity revealed some types of functional brain architectonics while subjects were performing a task of combining clicks. Specific cerebellum activation took place both in rhythm perception and in subject’s activity of integration of cognitive processes. This took place when subject used hierarchical strategy of rhythm perception.

We discussed the brain mechanisms of audiovisual integration and semantic comprehension as a direct process of comprehension (early integration) and process with second tuning (late integration) necessary in conflict situations. Existing of such audiovisual semantic conflict and four strategies to solve it were obtained in psychophysical experiment. Mathematic models were formulated according to that data. All models were described by linear equations, which were noted for the different weights of contributions of audio and visual modalities.

The article describes verification of a test diagnostics method that focuses on understanding of human biology, specifically the procedure and the results of verification. The study includes testing of 168 high school students and validation of the test by 55 experts. The results of the study indicate that the test can be recommended to diagnose how well high school students understand human biology. The difference in experts’ opinion regarding interpretation of the test questions has been analyzed.