Gaze Analysis in Early Warning Visual Feedback System for Hand Tracking Failures in Virtual Reality

Abstract

Hand-tracking failures significantly affect user performance and experience in VR, and thus, minimizing the negative effects of such failures remains a critical area of research. Although prior research introduced early-warning feedback systems to mitigate hand-tracking failures, it remains unclear why and how such feedback enhances user performance and reduces frustration. To address this gap, we investigated user gaze behaviors within a visual early-warning feedback system across three common hand-tracking failure scenarios: Low-Intensity Light Level, Out-of-Vision Hands, and Self-Occlusion. Our findings indicate that user attention toward feedback notifications varies depending on the type of simulated tracking failure. Moreover, providing early-warning feedback did not disrupt users’ attention to their primary tasks. Instead, users maintained their focus on object placement, resulting in fewer placement errors during pick-and-place tasks. These results contribute to a deeper understanding of early-warning visual feedback systems and provide valuable insights for designing feedback mechanisms that effectively improve performance without creating distractions in VR interactions.