Cortical control of many muscles and joints for voluntary movement may be simplified by the use of motor primitives, or muscle synergies. Muscle synergies have been found in a variety of voluntary movements, with only a few synergies (3-8) required to explain the activation patterns of many muscles. We used two types of intracortical microstimulation (ICMS) of single sites in the cortex to understand how synergies and individual joint degrees of freedom of the macaque forelimb may or may not be controlled from the primary motor cortex. High-frequency long-duration intracortical microstimulation (HFLD-ICMS) has been shown to hijack neural connections for muscle activity (Griffin et al., 2011) and move the limb through space such that the end-effector reaches a stable spatial endpoint (Van Acker et al., 2013c). Stimulus triggered averaging (StTA) slightly raises the associated motoneurons close to their firing threshold without causing overt discharge of the neurons.

The purpose of this study was to investigate the properties of muscle synergies and individual joint angle engineering degrees of freedom (eDOFs) represented in the output of primary motor cortex (M1), as they may be hardwired into cortical output modules. We applied single electrode HFLD-ICMS and StTA to M1 of two rhesus macaques (designated monkey A and monkey X) at 90-150 Hz and 90-150 μA for 1000 ms stimulus train length, which are parameters previously found to effectively produce stable spatial endpoints. We characterized synergies represented in M1 from HFLD-ICMS (for aims 1 and 3) and stimulus triggered averaging (StTA) (used in aim 2). In aim 3, we also characterized the five engineering degrees of freedom of the shoulder (flexion/extension, ad/abduction, internal/external rotation) and elbow (flexion/extension, supination/pronation) as they reached a steady state position during HFLD- ICMS to investigate the organization of steady state joint angles (eDOFs) in the forelimb region of primary motor cortex (M1).

To evoke a comprehensive representation of cortical output, stimulus was applied at 1 mm intervals across the precentral gyrus and 0.5 mm intervals in the depth of the central sulcus at 229 cortical locations in monkeys A and X. We subcutaneously recorded the EMG of 24 forelimb muscles and concurrently captured the kinematics of the forelimb using 20 motion capture markers on the forelimb. We also collected EMG associated with voluntary reaching tasks throughout the monkey's workspace. The mean EMG during the HFLD-ICMS and voluntary reaching movement were calculated (aim 1) and mean percent facilitation was calculated from the StTA data (aim 2). Non-negative matrix factorization was applied to the resulting values for each trial at all sites, and the weighting coefficients associated with each synergy were mapped to the cortical location of the stimulus. We found that 2-3 synergies accounted for 90% of the overall variation in the voluntary movement EMG, while 4-5 synergies were needed for HFLD- ICMS-evoked EMG, and 10-11 synergies were found in post-stimulus facilitation peaks of low intensity StTA EMG. When the calculated synergy weighting coefficients and steady state angles for the five engineering degrees of freedom (eDOFs) were mapped to each stimulus location in the cortex, we found patterns of synergy activation and steady state angle eDOF representations in the cortex were similar across monkeys and appeared to be somewhat regionally located in the cortex in manners consistent with previous mappings of stable spatial endpoints and stimulus triggered averaging of muscle representations (Van Acker et al., 2013a). The results discussed here suggest that HFLD-ICMS and StTA applied to the cortex activate focalized cortical output in a manner that may identify partial cortical organization of synergies and individual joint degrees of freedom, but how the CNS coordinates movements of the forelimb using these synergies in a functional manner requires further investigation.