Cognitively driven brain machine control using neural signals in the parietal reach region

Abstract

This study demonstrates that the spiking and local field potential (LFP) activity in the parietal reach region (PRR) of the macaque monkey can be jointly used to control the location of the computer cursor when the correct target location must be inferred symbolically, e.g., leftward arrow for the leftward target, etc. The average correct target acquisition rate during this brain machine control task without actual movements was 86% for the six discrete target locations when using spikes and LFPs from 16 electrodes. This performance was significantly better than using spikes or LFPs alone. These results, together with our previous findings, suggest that a single decoder based on both spikes and LFPs in PRR can robustly provide the subjects’ motor intent under varying contexts for neural prosthetic applications.