How Does Motion-to-Photon Latency Influence the Perception-Action Loop

Abstract

The perception-action loop plays an essential role in sports performance. Virtual reality is increasingly used to train this loop because it provides a controlled setting. However, motion-to-photon latency, the delay between the user’s real and virtual motion, may disrupt user experience by impairing the perception of timing. This effect may be further amplified by the limited field of view, which also constrains spatial and body awareness. Previous studies have shown that motion-to-photon latency impacts user performance and motion in various tasks. Nevertheless, its combined effect with viewpoint on the perception-action loop remains unexplored, particularly in tasks requiring precise timing. To investigate this, we conducted an experiment in which participants intercepted virtual bouncing balls under varying motion-to-photon latencies and two viewpoints: first-person and third-person viewpoints. Participants wore Xsens inertial sensors to control an avatar in VR. In first-person, the viewpoint matched the avatar’s head position; in third-person, the viewpoint was positioned three meters behind the avatar. We evaluated perceptual sensitivity to motion-to-photon latency, performance (success rate, racket trajectory and speed), and the sense of agency. Results showed that motion-to-photon latency degraded both performance and sense of agency. In first-person viewpoint, participants achieved higher success rates but were more perceptually sensitive to motion-to-photon latency. Notably, performance declined even before participants could consistently detect motion-to-photon latency. Overall, these findings highlight the interactions among the perception-action loop, motion-to-photon latency and viewpoint, with important implications for designing VR training systems.