EEG-BASED BRAIN COMPUTER-INTERFACE (BCI) - SEARCH FOR OPTIMAL ELECTRODE POSITIONS AND FREQUENCY COMPONENTS

Several laboratories around the world have recently started to investi gate EEG-based brain computer interface (BCI) systems in order to crea te a new communication channel for subjects with severe motor impairme nts. The present paper describes an initial evaluation of 64-channel E EG data recorded while subjects used one EEG channel over the left sen sorimotor area to control on-line vertical cursor movement. Targets we re given at the top or bottom of a computer screen. Data from 3 subjec ts in the early stages of training were analyzed by calculating band p ower time courses and maps for top and bottom targets separately. In a ddition, the Distinction Sensitive Learning Vector Quantizer (DSLVQ) w as applied to single-trial EEG data. It was found that for each subjec t there exist optimal electrode positions and frequency components for on-line EEG-based cursor control.