Using quadratic Volterra model of the oculo-motor system in diagnostic research of psychophysiological state of human

Abstract

It contains the results of the study of the effectiveness of the application of approximation Volterra models of the first and second order of the oculo-motor system in diagnostic studies of the psychophysiological state of a person based on eye-tracking data. Visual stimuli are used as test signals, which are displayed on the monitor screen at different distances from the initial position, which formally corresponds to the action of step signals with different amplitudes at the OMS input. Experimental studies of the "input-output" of the OMS were carried out using the Tobii Pro TX300 eyetracker, and the first-order transient functions and the second-order diagonal intersections of the second-order transient functions were determined based on the eye-tracking data. The resulting transient functions are used to form the spaces of diagnostic features. The diagnostic value of all possible combinations of features in pairs according to the indicator of the probability of correct recognition (PCR) was studied. The research results were obtained using Bayesian classifier training in different spaces of the proposed features. A study of the robustness of features according to the PCR indicator was carried out, the combinations of features with the maximum and most stable PCR indicator were selected. Figs.: 13. Tabl.: 5. Refs.: 28 titles.