Week-long visuomotor coordination and relaxation trainings do not increase sensorimotor rhythms (SMR) based brain-computer interface performance

L Botrel,A Kübler

DOI: https://doi.org/10.1016/j.bbr.2019.111993

2019-10-17

Abstract:Brain-computer interfaces (BCI) translate brain activity into control signals or commands for a device. Motor imagery of the limbs allows for modulating the sensorimotor rhythms (SMR), but there are up to 30% of the participants for whom electroencephalography (EEG) based SMR-BCI cannot detect any imagery-related changes. Individual variables, such as ability to concentrate on a task and error duration in a two-hand visuomotor coordination (VMC) task have been previously found to predict accuracy in an SMR-BCI. A first study attempted to substantiate those predictors by introducing a 30 min relaxation or VMC training period prior to an SMR-BCI session, but performance did not increase when compared to a control group. As the predictor training may have been too short, we applied 4 such training sessions on consecutive days in the current study. In a pre-post design, SMR-BCI accuracy of n = 39 participants increased from session 1 before to session 2 after the predictor training. While the manipulation of the predictor variables was successful, there was no effect on SMR-BCI performance. BCI accuracy correlated positively with the neurophysiological SMR predictor identified by Blankertz et al. [3], consolidating its predictive value, and with the state mindfulness scale. No other psychological predictor could be identified or replicated. Further studies should therefore focus more on delineating (partially) replicated or potential predictors such as VMC or mindfulness to help refining a sound model to predict SMR-BCI accuracy.