Motor learning: its relevance to stroke recovery and neurorehabilitation

John W Krakauer

DOI: https://doi.org/10.1097/01.wco.0000200544.29915.cc

Abstract:Purpose of review: Much of neurorehabilitation rests on the assumption that patients can improve with practice. This review will focus on arm movements and address the following questions: (i) What is motor learning? (ii) Do patients with hemiparesis have a learning deficit? (iii) Is recovery after injury a form of motor learning? (iv) Are approaches based on motor learning principles useful for rehabilitation? Recent findings: Motor learning can be broken into kinematic and dynamic components. Studies in healthy subjects suggest that retention of motor learning is best accomplished with variable training schedules. Animal models and functional imaging in humans show that the mature brain can undergo plastic changes during both learning and recovery. Quantitative motor control approaches allow differentiation between compensation and true recovery, although both improve with practice. Several promising new rehabilitation approaches are based on theories of motor learning. These include impairment oriented-training (IOT), constraint-induced movement therapy (CIMT), electromyogram (EMG)-triggered neuromuscular stimulation, robotic interactive therapy and virtual reality (VR). Summary: Motor learning mechanisms are operative during spontaneous stroke recovery and interact with rehabilitative training. For optimal results, rehabilitation techniques should be geared towards patients' specific motor deficits and possibly combined, for example, CIMT with VR. Two critical questions that should always be asked of a rehabilitation technique are whether gains persist for a significant period after training and whether they generalize to untrained tasks.