Emotionally Dependent Oculographic and Brain-Computer Interfaces Based on Additional Channel of Human-Computer Communication

The paper presents a model for evaluating the emotional state based on the involuntarily controlled signals of the user of oculographic and neurocomputer interfaces. For such signals were selected the parameters of heart rate variability, whose recording is well-proven both at the firmware level and from the point of view of processing and interpretation of the results. As markers for evaluating the user's emotional state as a compromise between the interface rate, the accuracy of the obtained results and the resource intensity, were chosen the values of the mean-square deviation of the pulse indicators (RR intervals, peak-to-peak segments of the pulse waves) that can be associated with the spectral estimate of the total power of the heart rate variability spectrum. Based on the game theory, a model was proposed for selecting by the firmware part of the interface of one of the several competing algorithms for processing the signals coming from the user. The possible parameters for evaluating the "gain" of the firmware suite were considered. It was demonstrated that the option containing the indicator that reflects the value of the difference between the dispersion of the heart rate variability at rest and when working with interfaces, in the process of execution of the strategy, weighted for the time of realization of this strategy, is more preferable for formation of the channel controlling the selection of the algorithms of processing the patient signals by a firmware suite.