Development of brain-computer interface: preliminary results

Reports on the development of an experimental setup and initial results for the evaluation of electronencephalogram (EEG) signals for control and communication. In our experiments, we recorded and analyzed surface EEG signals above sensory-motor areas, while the subjects were attempting to use only mental activities to modulate their EEG signals, resulting in desired movements of an animated object on the feedback computer screen. To discover possible communication channels based on EEG signals, we asked our subjects to determine which mental activity produced reproducible control actions. We calculated the power spectral density (PSD) of the recorded EEG signals and used sensory-motor rhythm (SMR) as the feedback-generating variable, i.e. the object´s movement direction and speed depended on the integrated PSD in the SMR range. In our initial experiments with three subjects, during the first session all three demonstrated the ability to determine which mental activity resulted in the desired movements. Off-line analysis showed that 60% classification accuracy can be achieved using a linear classifier on a subject with only two previous training sessions. These results are very encouraging and provide a good basis for the development of a direct brain-computer interface using cognitively modulated EEG signals