Off the shelf: Investigating transfer of learning using commercially available virtual reality equipment

Abstract

The evolution of virtual reality (VR) has created the opportunity for a relatively low-cost and accessible method to practice motor skills. Previous studies have demonstrated how motor skill practice in non-immersive virtual environments transfers to physical environments. Though minimal research has investigated motor learning and transfer within immersive VR, multiple experiments provide empirical evidence of positive transfer effects. To enhance the similarities between virtual and physical environments, most studies have used software engines and modified hardware. However, many learners and practitioners are currently using commercially available VR with the goal of enhancing real-world performance, though there is very little evidence to support the notion of positive transfer for these systems. Therefore, the purpose of this experiment was to investigate how motor skill practice using a commercially available VR system improved real-world performance and how that compared to physical practice. Participants (n = 68) were randomly selected into one of two groups: virtual reality (VR) practice (n = 33) or real-world (RW) practice (n = 35). The experiment took place over two consecutive days with a pretest, posttest, and practice phase occurring on both days. The pre- and post-testing phases were identical for both groups and consisted of putting a golf ball 10 times on a carpeted surface towards the center of a target. The practice phases consisted of 60 total putts per day in the respective environment (VR or RW). Participants continuously alternated golf putting holes from three different distances until they accrued 60 total putts. Participants in the RW group performed golf putts to three targets. Participants in the VR group also performed golf putts on three different miniature golf putting holes, using the commercially available Oculus Rift and the Cloudlands VR Minigolf game. The VR putting targets were designed to replicate the putting holes in the physical environment. Separate 2 (condition) x 4 (test phase) repeated measures ANOVAs were used to assess accuracy and club head kinematics. The results revealed a significant main effect for test phase, but not for condition. Post hoc analyses revealed both groups significantly improved their putting accuracy and club head kinematics at similar rates. The results from this study indicate that the transfer of learning that occurred from the commercially available VR practice was equally effective when compared to RW practice.