Neural adaptation of epidural electrocorticographic (EECoG) signals during closed-loop brain computer interface (BCI) tasks

Invasive BCI studies have classically relied on actual or imagined movements to train their neural decoding algorithms. In this study, non-human primates were required to perform a 2D BCI task using epidural microECoG recordings. The decoding weights and cortical locations of the electrodes used for control were randomly chosen and fixed for a series of daily recording sessions for five days. Over a period of one week, the subjects learned to accurately control a 2D computer cursor through neural adaptation of microECoG signals over ldquocortical control columnsrdquo having diameters on a the order of a few mm. These results suggest that the spatial resolution of microECoG recordings can be increased via neural plasticity.