Phase-amplitude coupling between mu- and gamma-waves to carry motor commands

The development of brain computer interfaces, especially the ones related to controlling exoskeletons and neurorehabilitation of stroke patients, strongly relies on our understanding of motor system and its neuronal mechanism. Motor imagery (MI) has turned out to be one of the most popular experimental regimes to study this system. Kinesthetic imagery (KI) is a kind of MI which shares a large portion of its neuronal pathway with real movements, except for having an additional inhibitory mechanism to prevent movement execution. Our magnetoencephalographic (MEG) experiments with ten untrained subjects revealed that this inhibitory control implied local neuronal desynchronization. We found that the motor-related communication between the inferior parietal cortex and the prefrontal cortex was carried out using the mu-frequency range. Additionally, three gamma frequencies were also pinpointed that encode the motor command specifics. Using artificial neural networks (ANNs) we classified left- and right-hand MI which reached maximal accuracy when we included these three gamma frequencies in the input signal for ANN. We suggest that mu-activity acts as a carrier of gamma-activity between inferior and parietal areas utilising phase-amplitude coupling.