Poor BCI performers still could benefit from motor imagery training

Abstract

Nowadays, there is a growing number of studies suggesting that coupled with the brain-computer interface (BCI) the motor imagery practice could be a helpful tool in neurorehabilitation therapy, but the actual neurophysiological correlates of such exercise are poorly understood. In this study we examined two of the most notable neurophysiological effects of motor imagery – the EEG murhythm desynchronization and the increase in cortical excitability assessed with transcranial magnetic stimulation (TMS). We have found that subjects’ BCI performance was highly correlated with mu-rhythm features and was not associated with the cortical excitability increase. Subjects with the lowest accuracy in BCI all had a statically significant excitability raise during motor imagery and did not differ from better performers. Our results suggest that poor BCI performers with weak EEG response still could benefit from the motor imagery training, and in that case cortex excitability level had to be considered for the control measurement.