Error Augmentation Motor Learning Training Approach in Stroke Patients

Deficits in upper limb (UL) functional recovery persist in a large proportion of stroke survivors. Understanding how to obtain the best possible UL recovery is a major scientific, clinical and patient priority. We propose that UL motor recovery may be improved by training that focuses on remediating an individual's specific motor impairment. Our approach is based on evidence that deficits in the control of muscle activation thresholds (spatial thresholds) of the elbow in stroke underlie impairments such as disordered movement and spasticity. Our novel training program focuses on improving the individual's active elbow control range using error augmentation (EA) feedback. Since training intensity and lesion load are key factors in motor recovery that lack guidelines, we will also investigate effects of exercise dose and corticospinal tract (CST) injury on UL recovery. In this multicenter, double-blind, parallel-group, randomized controlled trial (RCT), patients with stroke will participate in an individualized intensive technology-assisted reaching training program, based on error augmentation (EA), in order to improve voluntary elbow function. They will practice robot-assisted reaching in a virtual reality (VR) game setting. We will identify if intensive training with feedback aimed at expanding the range of spatial threshold (ST) control at the elbow (experimental group) is better than intensive training with general feedback about task success (control group). We will also determine the patient-specific optimal therapy dose by comparing kinematic and clinical outcomes after 3, 6 and 9 weeks of intensive training, and again at 4 weeks after training to determine carry-over effects. We will quantify the severity of the participant's motor deficit, as the amount of cortico spinal tract (CST) injury due to the stroke (%CST injury) and relate training gains to their %CST injury. Results of this pragmatic trial will provide essential information for optimizing individu