PROJECT SUMMARY Stroke-causing illness, disability, and early death is set to double worldwide within the next 15 years. Despite physical therapy, about 50% of stroke survivors have impaired hand function, which strongly impacts activities of daily living and independence; novel treatment methods are urgently required. One of many predictors of chronically impaired hand function includes deficits in somatosensation. Here, we propose to use a systems neuroscience and `neural engineering' framework that captures dynamic interactions across somatosensory and distributed motor networks to develop novel neurophysiological based neuromodulation to enhance motor function. The recent study, Ramanathan et al., Nature Medicine 2018, in rats recovering from a stroke demonstrated that population dynamics linked to low-frequency oscillatory activity (0.5-4Hz “LFO”) tracked spontaneous recovery and cortical stimulation boosted LFO power and could also augment motor function. Essential translational steps involve testing whether this approach also works for gyrated brains during the performance of dexterous tasks. The main experimental approach of this proposal involves monitoring motor and somatosensory activity during dexterous control in unaffected and affected hemispheres of non-human primates (NHPs) following recovery from a unilateral cortical lesion. Linear state space models will be used to describe neural activity progression in perilesional motor cortices, and used to model the effects of somatosensory inputs on motor state. These models will be used to identify markers of when low-frequency network dynamics need strengthening in the recovery phase and chronic deficit phase. Informed by the models, both open-loop and closed-loop low- frequency cortical stimulation will be tested to determine if dexterity improves. Completion of these aims will provide critical models for designing therapeutic approaches that specifically target perilesional oscillatory activity with low frequency electrical stimulation, a new direction that could transform our ability to augment upper extremity function following stroke. In this fellowship period, additional training will be needed in primate anatomy and electrophysiology, creation of injury models in primates, and in material science and electrical engineering related to the design of a neuromodulation system. These areas of knowledge and skills are necessary for completion of the proposed aims, as well as development of an independent research program following the fellowship period. Finally, all work will take place at UCSF, renowned for its lab-to-clinic translational work and training of independent investigators.

PROJECT NARRATIVE Stroke is the leading cause of disability in the United States, leaving a substantial number of survivors with permanent deficits in dexterity. Targeted neuromodulation to restore neural patterns underlying somatosensory perception and motor responses is an innovative new direction to facilitate recovery of hand function. This proposal uses a multiscale model of electrophysiological activity underlying dexterous control to develop a closed-loop stimulation approach to strengthen network dynamics damaged by stroke.