Representation of Limb Movement in Somatosensory Cortex and its Application to a Bidirectional Brain-Machine Interface

Control of reaching movements requires an accurate estimate of the state of the limb. This need is present for natural as well as prosthetic limbs. I investigate the ways in which the brain represents the sensation of limb state in healthy macaques. Further, I enquire how such knowledge might be used to supply feedback from prostheses to the brain. I expand upon previous results by quantifying the response properties of individual neurons to limb movements that are either self-generated (active) or imposed by the environment (passive). I recorded several tens of neurons simultaneously and was able to decode continuous kinematics from the firing rates of these neural populations. Additionally, I particularly focused on the representation of force in S1, discovering that many neurons in somatosensory cortex respond to non-linear combinations of force and velocity. These complex interactions explain one method by which the brain can distinguish passive movements, which may need to be corrected, and passive movements, which do not. Finally, I electrically stimulate proprioceptive regions of somatosensory cortex and determine that a monkey can perceive these stimulations.