Studies of neuroplasticity suggest that repetitious movements optimise changes at brain level, and that this response is optimised if the task is challenging and engaging. The evidence to date on physiotherapy interventions suggests that an increased intensity of therapy provides better outcomes and that exercise-based interventions demonstrate positive treatment effects. Robot-mediated therapy (RMT) is an innovative way of providing these components. This study investigated the effect of RMT on upper extremity function post-stroke.Single case study using an ABC design.Physiotherapy outpatient department.A 79-year-old female, 22 months following right cerebrovascular accident.Phase A consisted of a series of nine baseline measurements, phase B consisted of nine 30-minute sessions of RMT, and phase C consisted of nine 30-minute sessions of sling suspension.Fugl-Meyer Assessment, the Motor Assessment Scale and the Short-Form-36 (SF-36) questionnaire.The rate of recovery during the RMT phase B was greater than that with no treatment (A) and that with sling suspension (C) for the Fugl-Meyer Assessment and the Motor Assessment Scale. Improvement was seen only in those domains addressed by the RMT system. No change in quality of life as measured by the SF-36 was noted.Treatment delivered by this RMT system had a positive effect on the rate of recovery at the level of impairment of body function and at the level of activities. The superiority of RMT over sling suspension is consistent with the theories of neuroplasticity which suggest that repetitious movements must be challenging and meaningful. While these initial results concur with those of previous studies of RMT, further evidence is required before this form of intervention should be incorporated routinely into clinical practice.