Relevance. P300-based brain-computer interface (P300 BCI) enables a user to choose commands by consciously attending to certain visual stimulus. However, P300 BCI-based systems for detection of covert focuses of human interest, including emotional stimuli, can be built and applied in systems for monitoring human’s normal and pathological state.

Objective. The objective was to investigate the properties of perception of stimuli with different subjective emotional significance for the person and to evaluate the possibility of P300 BCI-based detection of covert emotional focuses of attention.

Methods. 14 healthy subjects participated in the EEG-study with consequent presenting of visual stimuli with emotional or neutral human faces photos. In the first block the participants’ task was just to look at the screen with appearing stimuli (passive attention) while one of six stimuli in each trial was emotional. In the other two blocks the task was to silently count the number of the target stimulus presentations among non-target stimuli (active attention) with emotional or neutral target stimulus depending on the block. We analyzed the event-related potentials (ERP) for stimuli of different classes and the classifica-tion accuracy for target (or emotional) stimulus in P300 BCI-similar fashion.

Results. In this study we showed that the accuracy of presented in passive viewing paradigm emotional stimuli recognition exceeded the random level more than twice. Also we found characteristic features of the potentials in response to stimuli with various significance.

Conclusions. The obtained results create opportunities for developing the clinically applicable P300 BCI-based systems for detection of covert attentional focuses.