# Non-immersive virtual reality telerehabilitation for motor accuracy and precision in individuals with Cerebral Palsy: A non-randomized clinical trial

## TL;DR

A home-based VR telerehab program improved motor accuracy and precision in people with Cerebral Palsy over 10 days, with lasting effects.

## Contribution

Demonstrates effectiveness of non-immersive VR telerehabilitation for motor learning in CP individuals.

## Key findings

- Participants with CP showed significant improvements in motor accuracy and precision over 10 days of VR training.
- Improvements were maintained at follow-up, indicating retained gains.
- VR performance in CP participants approached control levels in some target positions.

## Abstract

This study aimed to describe motor performance, accuracy and precision, in individuals with Cerebral Palsy (CP) compared to typically developing individuals (controls), and to examine longitudinal changes during a non-immersive virtual reality (VR) telerehabilitation protocol. Methods: The final sample included, 38 with CP and 21 controls. Controls completed a single practice session, while 29 participants with CP completed the 10-day home-based protocol, and 20 returned for a follow-up session to assess retention. Analyses followed the intention-to-treat principle, including all 38 CP participants. The intervention used the MoveHero game, performed two to three times per week, with sessions lasting at least 9 minutes under remote therapist supervision. Motor performance was assessed in a coincident timing task using absolute error (AE, accuracy), variable error (VE, precision), and percentage of hits. The study was registered in the Brazilian Clinical Trials Registry (ReBec: RBR-4kf52bv). Results: The CP group had higher mean AE and VE values than controls but showed progressive improvements across practice days. Mean accuracy improved across most target positions (p < 0.05), with mean AE decreasing from 1350 ms on Day 1–895 ms on the best practice day. Controls had a mean AE of 584 ms, and participants with CP achieved similar performance in two of the four target positions. Mean VE improved notably for Target 3 (1124 ms on Day 1–811 ms on Day 9, p = 0.007), and the mean percentage of correct responses increased from 45.8% to 58.5% (p = 0.019), while controls reached 69.1%. No significant differences were found among GMFCS subgroups. Gains were maintained at follow-up (p > 0.05). Conclusion: Home-based non-immersive VR telerehabilitation using MoveHero effectively enhanced upper-limb motor performance in individuals with CP, regardless of functional level. These findings support VR as an accessible, motivating, and adaptable tool for promoting motor learning and continuity of care in home-based rehabilitation.