Primate Motor Cortex: Individual and Ensemble Neuron-Muscle Output Relationships

Abstract

The specific aims of this study were to: 1) investigate the encoding of forelimb muscle activity timing and magnitude by corticomotoneuronal (CM) cells, 2) test the stability of primary motor cortex (M1) output to forelimb muscles under different task conditions, and 3) characterize input/output relationships associated with different intracortical microstimulation (ICMS) methods. Neuronal recording and stimulating methods were used in combination with electromyographic (EMG) recording of 24 forelimb muscles to investigate questions related to M1 control of forelimb muscles. Target muscles of CM neurons were identified by the presence of post-spike facilitation (PSpF) in spike-triggered averages (SpTA) of EMG activity. Post-stimulus output effects were obtained with three different ICMS methods; stimulus-triggered averaging (StTA) of EMG activity, repetitive short duration ICMS (RS-ICMS) and repetitive long duration ICMS (RL-ICMS). Our results demonstrate that CM cells exhibit strong and consistent coactivation with their target muscles. Further, the summed activity of populations of identified CM cells was a better predictor of the common muscle's EMG activity than individual neurons. Our data support the view that M1 output encodes muscle activation related parameters. Regarding stability, we found that output effects in StTAs of EMG activity are remarkably stable and largely independent of changes in joint angle, or limb posture. This further validates the use of StTA for mapping and other studies of cortical motor output. RL-ICMS evoked EMG activity was also stable in sign, strength and distribution independent of starting position of the hand. Our data support a model in which RL-ICMS produces sustained co-activation of multiple agonist and antagonist muscles which then generates joint movements according to the length-tension properties of the muscles until an equilibrium position is achieved. Further, RL-ICMS evoked EMG activity did not sum with the existing level of activity; rather the stimulus forced a new EMG level that was independent of existing voluntary background. Our results further show that post-stimulus output effects on muscle activity obtained with StTA and RS-ICMS closely resemble one another. However, RL-ICMS produces effects that can deviate substantially from those observed with StTA.