MOTOR IMAGERY BASED BRAIN COMPUTER INTERFACE WITH VIBROTACTILE INTERACTION

Motor imagery based brain computer interface (MI BCI) is aimed to allow a person to control external devices by consciously engaging with his or her body movements' mental representations. BCI systems are currently being developed in many variations but all of them require visual input channel to operate in the control loop. In this paper we propose the new paradigm of MI BCI based on vibrotactile stimulation that eliminates the need in the visual channel. Eleven healthy subjects participated in the study that included 7 sessions, each held on a different day. Their task was to learn how to perfoim motor imagery of fingers movements and to use it to control the BCI system in response to the stimuli of vibrotactile stimulators installed on the subjects' bodies. Such an approach allowed us to examine the effectiveness of motor imagery training with eyes closed. Over the course of the study all subjects were successful in mastering imagery of the proposed movements. Thus the possibility of user-training for BCI system with eyes closed and vibrotactile stimuli as signals to command switching was demonstrated. It was found that the EEG patterns associated with motor imagery, the features of corticospinal excitability changes during MI, the temporal (session-by-session) dynamics and the accuracy level in the BCI obtained while using the proposed system at least do not differ from those in the traditional approach, in which visual stimulus are used. Moreover, for some users the new paradigm grants substantial advantages over the standard approach in terms of BCI-accuracy and corticospinal excitability increase. The proposed BCI system provides new practical applications and may also be used by people with impaired vision.