Feedback Modality Influences Students’ Surgical Skills Acquisition in Virtual Reality Simulation Training – a Randomized Clinical Trial

Introduction: Computer-assisted simulation tutoring systems can be implemented with a variety of feedback modalities to enhance trainee acquisition of surgical technical skills. However, the optimal method of feedback delivery to maximize skill acquisition needs to be explored. The goal of this study was to test the efficacy of four feedback protocols, including practice with no-feedback, to assess trainee response in learning bimanual skills in complex simulated brain tumor surgery.

Methods: In this 4-arm randomized controlled trial, 120 medical students performed five brain tumor resections. Participant groups are provided with (1) no-feedback, (2) numeric feedback, (3) visual feedback and (4) visuospatial feedback. Participants in Group-2 received printed numeric feedback with their scores compared to expert benchmarks in 14 performance metrics. Participants in Group-3 were provided with the same information in a colored format presented on the screen. In addition to this information, participants in Group-4 received 3D tumor and pia models to provide spatial information about their performance. The number of benchmarks achieved was calculated at each task and was compared from the first (baseline) to the last repetition of the task within groups and between groups to assess learning.

Results: All feedback groups significantly improved performance when compared to baseline performance. Visual and visuospatial protocols resulted in significantly better performance improvement in the second repetition. Visual feedback resulted in significantly better improvement in comparison to providing identical information in numerical format.

Conclusion : Autonomous visual assistance may provide efficient tools in teaching operative tumor resection skills while the skill acquisition can be enhanced by more engaging visual and visuospatial feedback delivery.