Testing models of cerebellar ataxia via dynamic simulation

Abstract

Patients with damage to the cerebellum make reaching movements that are uncoordinated or "ataxic." One prevailing hypothesis is that the cerebellum functions as an internal model for planning movements, and that damage to the cerebellum results in movements that do not properly account for arm dynamics. An exoskeleton robot was used to record multi-joint reaching movements. Subsequently, joint-torque trajectories were calculated and a gradient descent algorithm found optimal, patient-specific perturbations to actual limb dynamics predicted to reduce directional reaching errors by an average of 41%, elucidating a promising form of robotic intervention and adding support to the internal model hypothesis.