Motor imagery during action observation in virtual reality: the impact of watching myself performing at a level I have not yet achieved

Feedforward modeling, the creation of one's own behaviour that is potentially achievable in the future, can support motor performance and learning. While this has been shown for sequences of motor actions, it remains to be tested whether feedforward modelling is beneficial for single complex motor actions. Using an immersive, state-of-the-art, low-latency Cave Automatic Virtual Environment (CAVE), we compared motor imagery during action observation (AOMI) of oneself performing at one's current skill level against AOMI of oneself performing at an achievable future skill level. We performed 3D scans and created a ready-to-animate virtual human of each participant. During acquisition, participants observed an avatar of themselves performing either one of their own previously executed squats (Me-Novice) or observed an avatar of themselves performing a skilled squat (Me-Skilled), whilst simultaneously imagining the feelings and sensations associated with movement execution. Findings revealed an advantage for the Me-Skilled group as compared to the Me-Novice group in motor performance and cognitive representation structure, while self-efficacy improved in both groups. In comparison to watching and imagining oneself performing at the current novice skill level, watching and imagining oneself performing at a more advanced skill level prevented from making errors in motor performance and led to perceptual-cognitive scaffolding as shown by functional changes in underlying representations. Simultaneous imagery whilst observing future states of action may therefore help to establish cognitive prerequisites that enable better motor performance. To this end, virtual reality is a promising tool to create learning environments that exceed an individual's current performance level.