'Substitute' knowledge of results provided through virtual reality enhances motor imagery-based learning

Improvements in motor performance achieved through motor imagery, the mental rehearsal of movement, are typically smaller than an equivalent dose of physical practice. One reason for this difference may be lack of feedback during motor imagery, integral to refining the motor plan, including knowledge of results. Here, we tested whether visual feedback provided via virtual reality impacted learning of a complex novel skill using motor imagery. Healthy novice participants (N = 48) were randomly assigned to one of three groups and engaged in four sessions of motor imagery practice of a golf putting task in a virtual environment. Following each motor imagery trial, participants observed one of three types of visual feedback: perfect feedback (100% success), erroneous feedback (30% errors, 70% success), or no feedback (control). Performance, measured by percentage of holed putts (%Success), mean radial error, and bivariate variable error, was assessed during three physical test blocks: before (pre), after (post), and minimum 24h after imagery practice (retention). Linear mixed effects models and accompanying Cohen's d effect sizes conducted to quantify changes in performance revealed improvements in %Success by all groups at both the post-test (particularly the erroneous feedback group) and retention test with no effects related to mean radial error or bivariate variable error. Findings suggest that supplementing motor imagery with visual feedback may lead to greater task success. Future research should explore impacts on movement quality (e.g., arm, trunk kinematics) and trajectory-based learning. Overall, this work has implications related to the use of virtual reality to enhance the effectiveness of motor imagery.