Effects of a Mirror-Induced Visual Illusion on a Reaching Task in Stroke Patients

Background. Although most mirror therapy studies have shown improved motor performance in stroke patients, the optimal mirror training protocol still remains unclear. Objective. To study the relative contribution of a mirror in training a reaching task and of unilateral and bimanual training with a mirror. Methods. A total of 93 stroke patients at least 6 months poststroke were instructed to perform a reaching task as fast and as fluently as possible. They performed 70 practice trials after being randomly allocated to 1 of 5 experimental groups: training with (1) the paretic arm with direct view (Paretic-No Mirror), (2) the nonparetic arm with direct view (Nonparetic-No Mirror), (3) the nonparetic arm with mirror reflection (Nonparetic Mirror), (4) both sides and with a nontransparent screen preventing visual control of paretic side (Bilateral-Screen), and (5) both sides with mirror reflection of the nonparetic arm (Bilateral-Mirror). As baseline and follow-up, patients performed 6 trials using only their paretic side. Primary outcome measure was the movement time. Results. We found the largest intervention effect in the Paretic-No Mirror condition. However, the Nonparetic-Mirror condition was not significantly different from the Paretic-No Mirror condition, while the Unaffected-No Mirror condition had significantly less improvement than the Paretic-No Mirror condition. In addition, movement time improved significantly less in the bimanual conditions and there was no difference between both bimanual conditions or between both mirror conditions. Conclusion. The present study confirms that using a mirror reflection can facilitate motor learning. In this task, bimanual movement using mirror training was less effective than unilateral training.