Perceiving Safety: User Preferences and Perception of Multimodal Virtual Boundary Cues in VR

In immersive virtual reality (VR) environments, users often face safety risks due to limited spatial awareness, such as unintentionally approaching or crossing physical boundaries. To address this challenge, we propose a novel boundary warning risk assessment system to enhance spatial perception and operational safety. Our system continuously computes a risk score based on four key kinematic features: the user’s distance to the boundary, velocity, acceleration, and the angle between movement direction and boundary orientation. Based on this score, we designed and compared four feedback strategies: a progress bar indicating risk level, a color-coded circular indicator with gradient transitions, auditory cues with varying frequencies, and a combined multimodal approach integrating visual and auditory feedback. A user study (n = 24) was conducted where participants navigated a 4 \(\times \) 4 m virtual space with the objective of collecting as many targets as possible. We measured both objective safety metrics and subjective user experience ratings to assess the user perception of our proposed methods. Results showed that different feedback strategies led to noticeable differences in safety performance. Of the four methods tested, the Multimedia Cue was the most effective, significantly reducing boundary intrusions and receiving the highest user preference ratings.