While several brain regions involved in controlling movement have been identified, the neural signal processing that ultimately governs motor commands sent to muscles has remained stubbornly opaque. Synaptic interactions within and between neuronal populations are probable loci of this processing, but historically it has been extremely challenging in practice to observe these interactions at cellular and spike resolution during behavior. This has stymied the development of mechanistic models of motor system operation. Work in my lab aims to characterize interactions within and between motor system populations as a means to build and refine such models. During my talk, I will first discuss a line of research that addresses how motor cortical output engages downstream effector circuits. Analysis of motor cortical activity has consistently found, and imputed functional significance upon, signals that correlate with the totality of limb muscle activity or movement kinematics. Yet lesion studies are consistent with a selective role for motor cortex in driving certain aspects of muscle activity, a role which could rely on command signals related only to those aspects. We quantified the direct influence of forelimb motor cortex on muscle activity throughout a naturalistic climbing behavior, finding that this influence is selective for, and highly dependent upon, muscle activity states. We then used multielectrode array recording to identify components of motor cortical activity that align with its influence on muscles. Our results here reveal a direct motor cortical influence that is selective within a motor behavior and relies on a previously undescribed neural activity subspace. I will also discuss another line of research addressing the common view of a functional hierarchy in motor cortex where a premotor cortex primarily plans movements, while a downstream primary motor cortex is involved chiefly with movement execution. We used multielectrode array recording and optogenetics together with methods for quantifying neuronal interactions to assess functional hierarchy among motor cortical areas. Our results confirm the existence of hierarchy on fast timescales but indicate that it does not manifest as expected in neural activity. Here neural network modeling has provided new insight into the expression of functional hierarchy at the level of single unit activity.

   ZOOM LINK