Modulatory effects of the supplementary motor area on primary motor cortex outputs.

Premotor areas of primates are specialized cortical regions that can contribute to hand movements by modulating the outputs of the primary motor cortex (M1). The goal of the present work was to study how the supplementary motor area (SMA) located within the same hemisphere [i.e., ipsilateral SMA (iSMA)] or the opposite hemisphere [i.e., contralateral (cSMA)] modulate the outputs of M1. We used paired-pulse protocols with intracortical stimulations in sedated capuchin monkeys. A conditioning stimulus in iSMA or cSMA was delivered simultaneously or before a test stimulus in M1 with different interstimulus intervals (ISIs) while electromyographic activity was recorded in hand and forearm muscles. The pattern of modulation from iSMA and cSMA shared some clear similarities. In particular, both areas predominantly induced facilitatory effects on M1 outputs with shorter ISIs and inhibitory effects with longer ISIs. However, the incidence and strength of facilitatory effects were greater for iSMA than cSMA. We then compared the pattern of modulatory effects from SMA to the ones from the dorsal and ventral premotor cortexes (PMd and PMv) collected in the same series of experiments. Among premotor areas, the impact of SMA on M1 outputs was always weaker than the one of either PMd or PMv, and this was regardless of the hemisphere, or the ISI, tested. These results show that SMA exerts a unique set of modulations on M1 outputs, which could support its specific function for the production of hand movements.NEW & NOTEWORTHY We unequivocally isolated stimulation to either the ipsilateral or contralateral supplementary motor area (SMA) using invasive techniques and compared their modulatory effects on the outputs of primary motor cortex (M1). Modulations from both SMAs shared many similarities. However, facilitatory effects evoked from ipsilateral SMA were more common and more powerful. This pattern differs from the ones of other premotor areas, which suggests that each premotor area makes unique contributions to the production of motor outputs.

Contrasting Modulatory Effects from the Dorsal and Ventral Premotor Cortex on Primary Motor Cortex Outputs.

The dorsal and ventral premotor cortices (PMd and PMv, respectively) each take part in unique aspects for the planning and execution of hand movements. These premotor areas are components of complex anatomical networks that include the primary motor cortex (M1) of both hemispheres. One way that PMd and PMv could play distinct roles in hand movements is by modulating the outputs of M1 differently. However, patterns of effects from PMd and PMv on the outputs of M1 have not been compared systematically. Our goals were to study how PMd within the same (i.e., ipsilateral or iPMd) and in the opposite hemisphere (i.e., contralateral or cPMd) can shape M1 outputs and then compare these effects with those induced by PMv. We used paired-pulse protocols with intracortical microstimulation techniques in sedated female cebus monkeys while recording EMG signals from intrinsic hand and forearm muscles. A conditioning stimulus was delivered in iPMd or cPMd concurrently or before a test stimulus in M1. The patterns of modulatory effects from PMd were compared with those from PMv collected in the same animals. Striking differences were revealed. Conditioning stimulation in iPMd induced more frequent and powerful inhibitory effects on M1 outputs compared with iPMv. In the opposite hemisphere, cPMd conditioning induced more frequent and powerful facilitatory effects than cPMv. These contrasting patterns of modulatory effects could allow PMd and PMv to play distinct functions for the control of hand movements and predispose them to undertake different, perhaps somewhat opposite, roles in motor recovery after brain injury.SIGNIFICANCE STATEMENT The dorsal and ventral premotor cortices (PMd and PMv, respectively) are two specialized areas involved in the control of hand movements in primates. One way that PMd and PMv could participate in hand movements is by modulating or shaping the primary motor cortex (M1) outputs to hand muscles. Here, we studied the patterns of modulation from PMd within the same and in the opposite hemisphere on the outputs of M1 and compared them with those from PMv. We found that PMd and PMv have strikingly different effects on M1 outputs. These contrasting patterns of modulation provide a substrate that may allow PMd and PMv to carry distinct functions for the preparation and execution of hand movements and for recovery after brain injury.

Modulatory Effects of the Ipsi and Contralateral Ventral Premotor Cortex (PMv) on the Primary Motor Cortex (M1) Outputs to Intrinsic Hand and Forearm Muscles in Cebus apella.

The ventral premotor cortex (PMv) is a key node in the neural network involved in grasping. One way PMv can carry out this function is by modulating the outputs of the primary motor cortex (M1) to intrinsic hand and forearm muscles. As many PMv neurons discharge when grasping with either arm, both PMv within the same hemisphere (ipsilateral; iPMv) and in the opposite hemisphere (contralateral; cPMv) could modulate M1 outputs. Our objective was to compare modulatory effects of iPMv and cPMv on M1 outputs to intrinsic hand and forearm muscles. We used paired-pulse protocols with intracortical microstimulations in capuchin monkeys. A conditioning stimulus was applied in either iPMv or cPMv simultaneously or prior to a test stimulus in M1 and the effects quantified in electromyographic signals. Modulatory effects from iPMv were predominantly facilitatory, and facilitation was much more common and powerful on intrinsic hand than forearm muscles. In contrast, while the conditioning of cPMv could elicit facilitatory effects, in particular to intrinsic hand muscles, it was much more likely to inhibit M1 outputs. These data show that iPMv and cPMv have very different modulatory effects on the outputs of M1 to intrinsic hand and forearm muscles.

Whole-brain mapping of long-range inputs to the VIP-expressing inhibitory neurons in the primary motor cortex.

The primary motor cortex (MOp) is an important site for motor skill learning. Interestingly, neurons in MOp possess reward-related activity, presumably to facilitate reward-based motor learning. While pyramidal neurons (PNs) and different subtypes of GABAergic inhibitory interneurons (INs) in MOp all undergo cell-type specific plastic changes during motor learning, the vasoactive intestinal peptide-expressing inhibitory interneurons (VIP-INs) in MOp have been shown to preferentially respond to reward and play a critical role in the early phases of motor learning by triggering local circuit plasticity. To understand how VIP-INs might integrate various streams of information, such as sensory, pre-motor, and reward-related inputs, to regulate local plasticity in MOp, we performed monosynaptic rabies tracing experiments and employed an automated cell counting pipeline to generate a comprehensive map of brain-wide inputs to VIP-INs in MOp. We then compared this input profile to the brain-wide inputs to somatostatin-expressing inhibitory interneurons (SST-INs) and parvalbumin-expressing inhibitory interneurons (PV-INs) in MOp. We found that while all cell types received major inputs from sensory, motor, and prefrontal cortical regions, as well as from various thalamic nuclei, VIP-INs received more inputs from the orbital frontal cortex (ORB) - a region associated with reinforcement learning and value predictions. Our findings provide insight on how the brain leverages microcircuit motifs by both integrating and partitioning different streams of long-range input to modulate local circuit activity and plasticity.

Autonomous optimization of neuroprosthetic stimulation parameters that drive the motor cortex and spinal cord outputs in rats and monkeys.

Neural stimulation can alleviate paralysis and sensory deficits. Novel high-density neural interfaces can enable refined and multipronged neurostimulation interventions. To achieve this, it is essential to develop algorithmic frameworks capable of handling optimization in large parameter spaces. Here, we leveraged an algorithmic class, Gaussian-process (GP)-based Bayesian optimization (BO), to solve this problem. We show that GP-BO efficiently explores the neurostimulation space, outperforming other search strategies after testing only a fraction of the possible combinations. Through a series of real-time multi-dimensional neurostimulation experiments, we demonstrate optimization across diverse biological targets (brain, spinal cord), animal models (rats, non-human primates), in healthy subjects, and in neuroprosthetic intervention after injury, for both immediate and continual learning over multiple sessions. GP-BO can embed and improve "prior" expert/clinical knowledge to dramatically enhance its performance. These results advocate for broader establishment of learning agents as structural elements of neuroprosthetic design, enabling personalization and maximization of therapeutic effectiveness.

Hierarchical Bayesian Optimization of Spatiotemporal Neurostimulations for Targeted Motor Outputs.

The development of neurostimulation techniques to evoke motor patterns is an active area of research. It serves as a crucial experimental tool to probe computation in neural circuits, and has applications in neuroprostheses used to aid recovery of motor function after stroke or injury to the nervous system. There are two important challenges when designing algorithms to unveil and control neurostimulation-to-motor correspondences, thereby linking spatiotemporal patterns of neural stimulation to muscle activation: (1) the exploration of motor maps needs to be fast and efficient (exhaustive search is to be avoided for clinical and experimental reasons) (2) online learning needs to be flexible enough to deal with noise and occasional spurious responses. We propose a stimulation search algorithm to address these issues, and demonstrate its efficacy with experiments in the motor cortex (M1) of a non-human primate model. Our solution is a novel iterative process using Bayesian Optimization via Gaussian Processes on a hierarchy of increasingly complex signal spaces. We show that our algorithm can successfully and rapidly learn correspondences between complex stimulation patterns and evoked muscle activation patterns, where standard approaches fail. Importantly, we uncover nonlinear circuit-level computations in M1 that would have been difficult to identify using conventional mapping techniques.