The goal of the work on neuro-computational models is to understand the neural bases of motor learning. We are notably investigating motor plasticity in the cerebellum, cortical map plasticity and reorganization in the sensorimotor cortex, multiple task learning, and adaptive decision-making during motor learning in healthy and lesioned brains. When appropriate, we test our predictions by conducting behavioral and/or brain imaging (fMRI and TMS) experiments either at USC or with our collaborators at ATR in Japan or at INSERM in France.

The goal of the work on learning optimization is to enhance-learning of motor skills in patients with stroke. Despite great progress in psychology and neuroscience, physical therapists treating patients with stroke rely on non-specific guidelines to determine task practice schedules for functional motor skill re-acquisition. Using algorithms that combine neuroscience-based models and artificial intelligence, we aim at defining and testing adaptive practice schedules, with particular emphasis on the micro-schedules of the practice.
The Computational Neuro-Rehabilitation Lab is closely associated with the Motor Behavior and Neurorehabilitation Laboratory and is part of the USC Department of Biokinesiology and Physical Therapy (ranked #1 by U.S. News and World Report).