Adaptive virtual reality-based training: a systematic literature review and framework

Abstract

Virtual reality (VR) provides the capability to train individuals to deal with complex situations by immersing them in a virtual environment. VR-based training has been used in many domains; however, in order to be effective, the training should be adapted based on user’s capabilities, performance, and needs. This study provided a framework for adaptive VR-based training including performance measures, adaptive logic, and adaptive variables. A systematic review of literature was conducted using Compendex, Web of Science, and Google Scholar databases to identify the adaptive VR-based training approaches used in different domains. Results revealed that adaptive VR-based training can be improved by using real-time kinematic/kinetic data and physiological measures from the user, incorporating offline measures such as trainee’s profile information, providing adaptations on controlled elements in the simulation, adjusting feedback content, type, and timing, and using reinforcement learning algorithms. The recommendations provided in this study need to be further validated using longitudinal studies comparing adaptive and non-adaptive training approaches.