ИСТИНА

Brain-computer interface (BCI) enables an individual to send messages or control devices, directly using their EEG [1]. P300 BCI is based on the visual oddball paradigm, known to elicit the P300 component of the event-related potentials (ERP), when the user is consciously paying attention to the desired stimulus. However, the attention response to external stimulus can be obtained when the presented stimuli vary in their significance according to subject’s personal experience. To develop a P300 BCI-based system for detection of covert intentions or psycho-emotional states we used photos of neutral and emotional human faces. Fourteen healthy subjects took part in the experiment. The task in the ‘Emotional Unattended’ mode was just to view the center of the screen, where the photos were presented. The task in the ‘Emotional Attended’ and ‘Neutral Attended’ modes was to count the number of the target photo appearances among others (emotional among neutral ones or neutral among neutral ones). We discovered that using emotional faces as stimuli enhanced the amplitude of several ERP components. In the ‘Emotional Unattended’ condition the emotional faces were classified as target stimuli with the accuracy of 36.4%, which exceeds the random level more than twice. The mean classification accuracy in ‘Emotional Attended’ condition was 97.1%. Our results provide evidence that using emotional faces as stimuli increases the efficiency of P300 BCI operating and can be used for creating ERP-based systems for emotional focuses detection which can be applied in clinics as a part of the complex psychological conditions diagnostics and treatment assessment, as well as be used in the entertainment area. References 1. Kaplan, A., Shishkin, S., Ganin I., Basyu, I. and Zhigalov, A. (2013). Adapting the P300-based brain-computer interface for gaming: a review. IEEE Transactions on Computational Intelligence and AI in Games, 5(2): 141-149.