RESUMEN
OBJECTIVE: Stroke is one of the most debilitating neurological disorders that commonly results in both cognitive and motor dysfunctions. Although the recovery of gait is one of the main goals of patients with stroke, only 50-60% of the patients commonly reach this target. This study aimed to evaluate the effects of action observation training, based on mirror neurons, as an add-on therapy to the conventional physical rehabilitation on the gait performance of patients with stroke. METHODS: Fourteen patients with chronic stroke were randomly assigned to the sham or the experimental group. Both groups received a 40-min conventional physical training following a 12-min observation training depicting exercises for the experimental group but nature pictures for the sham group each session. The patients' walking was recorded using a motion analysis system at baseline and after the 12-session intervention. Spatiotemporal parameters of gait and ground reaction forces were measured. RESULTS: Significant improvements were found in most measured spatiotemporal parameters of gait on the unaffected side of the patients in the experimental group, while in the sham group, the recovery was observed only in the percentage of the stance phase. Regarding the affected side, the stride and step length of the patients in the experimental group were parameters with a significant amelioration. CONCLUSION: The results of this study showed that the action observation training had the potential to improve the walking quality of the patients with hemiplegia in the chronic phase of stroke.