Haptic Adaptive Feedback to Promote Motor Learning With a Robotic Ankle Exoskeleton Integrated With a Video Game

<b>Background:</b> Robotic devices have been used to rehabilitate walking function after stroke. Although results suggest that post-stroke patients benefit from this non-conventional therapy, there is no agreement on the optimal robot-assisted approaches to promote neurorecovery. Here we present a new robotic therapy protocol using a grounded exoskeleton perturbing the ankle joint based on tacit learning control. <b>Method:</b> Ten healthy individuals and a post-stroke patient participated in the study and were enrolled in a pilot intervention protocol that involved performance of ankle movements following different trajectories via video game visual feedback. The system autonomously modulated task difficulty according to the performance to increase the challenge. We hypothesized that motor learning throughout training sessions would lead to increased corticospinal excitability of dorsi-plantarflexor muscles. Transcranial Magnetic Stimulation was used to assess the effects on corticospinal excitability. <b>Results:</b> Improvements have been observed on task performance and motor outcomes in both healthy individuals and post-stroke patient case study. Tibialis Anterior corticospinal excitability increased significantly after the training; however no significant changes were observed on Soleus corticospinal excitability. Clinical scales showed functional improvements in the stroke patient. <b>Discussion and Significance:</b> Our findings both in neurophysiological and performance assessment suggest improved motor learning. Some limitations of the study include treatment duration and intensity, as well as the non-significant changes in corticospinal excitability obtained for Soleus. Nonetheless, results suggest that this robotic training framework is a potentially interesting approach that can be explored for gait rehabilitation in post-stroke patients.