Predicting intended movement direction using EEG from human posterior parietal cortex

Abstract

The posterior parietal cortex (PPC) plays an important role in motor planning and execution. Here, we investigated whether noninvasive electroencephalographic (EEG) signals recorded from the human PPC can be used to decode intended movement direction. To this end, we recorded whole-head EEG with a delayed saccade-or-reach task and found direction-related modulation of event-related potentials (ERPs) in the PPC. Using parietal EEG components extracted by independent component analysis (ICA), we obtained an average accuracy of 80.25% on four subjects in binary single-trial EEG classification (left versusright). These results show that in the PPC, neuronal activity associated with different movement directions can be distinguished using EEG recording and might, thus, be used to drive a noninvasive brain-machine interface (BMI). © 2009 Springer.