RESUMO
PURPOSE: To evaluate the performance improvement of individuals with hemiparesis cerebral palsy (CP) using a virtual task. PARTICIPANTS AND METHODS: Twenty individuals were selected and distributed into two groups. The experimental group (CP group) comprised ten individuals with a medical diagnosis of CP, and ten individuals with typical development (sex- and age-matched) composed the control group (TD group). Both groups followed the same intervention protocol, which included a virtual coincident timing task: the participants performed upper limb movements in front of a computer's webcam and interacted with the task with the aim of virtually intercepting spheres that fell in four rows following the rhythm of a pre-selected song during an 8-minute period. To verify the influence on a real task, pre- and posttests were performed in a similar task, but with physical contact (using the spacebar on the keyboard of a computer). To analyze the data, we evaluated the variable, constant, and absolute errors during the task and in the pre- and posttests. RESULTS: The results showed that there was an improvement in performance between the pre- and posttests; that is, after practicing the task in an environment without physical contact, there was a performance improvement in posttests in the real task, but only for the CP group. Moreover, there were significant differences in precision and accuracy between the two groups, with worse performance in the CP group. CONCLUSION: Individuals with CP presented better performance in the real task after practice in a virtual reality task, albeit with worse performance compared with individuals with TD. This is an interesting result that supports the possible use of virtual tasks for the rehabilitation of individuals with CP.