Guiding locomotion with a vibrotactile belt: Sensory substitution for visual-spatial information
Poster Presentation 23.463: Saturday, May 16, 2026, 8:30 am – 12:30 pm, Pavilion
Session: Action: Navigation, locomotion
Schedule of Events | Search Abstracts | Symposia | Talk Sessions | Poster Sessions
There is a Poster PDF for this presentation, but you must be a current member or registered to attend VSS 2026 to view it.
Please go to your Account Home page to register.
William Warren1 (bill_warren@brown.edu), Sofia Ruiz1, Michael Vail1, Benjamin Kimia1, Sina Feldmann1; 1Brown University
Can sensory substitution be used to enhance the guidance of locomotion for people with impaired vision? In previous research with a vibrotactile belt (VSS 2025), we demonstrated high accuracy and precision in judging the egocentric direction of single-tactor vibrations. Here we test whether such vibrations can guide active walking to targets in virtual reality (VR). Participants (N=16) wore a Quest 3 VR headset that tracked head position and orientation, and a vibrotactile belt that contained 16 pager motors spaced 22.5˚ apart. On each trial, the participant walked to a sequence of two targets using either visual or tactile information. Visual targets appeared as green poles, while tactile targets were specified by single-tactor vibrations in 0.5s bursts at 225 Hz. The burst rate was either (a) constant, with a 0.5s pause between bursts, or (b) increased in 3 steps (0.5s, 0.1s, 0.01s pause) as the target was approached (3m and 1.5m). Target position was manipulated, with eight egocentric directions (45˚ intervals) and two distances (4m, 8m). Walking performance was only moderately worse with tactile compared to visual information. Mean stopping distance from the final target was slightly larger for tactile (0.34m) than visual (0.15m) targets, p<0.001, but there was no effect of tactile burst rate on stopping distance (constant 0.35m, increasing 0.33m). Mean tortuosity (normalized path length) was slightly greater with tactile (1.14) than visual (1.03) targets, p<0.001. However, mean walking time to 2 targets was much longer with tactile (31.8s) than visual (19.3s) targets, p<0.001. We are currently analyzing walking speed, path curvature, maximum path deviation, and energetic cost. The results indicate that tactile guidance increases path length by only 10% but walking time by 65% compared to sighted guidance; the latter may improve with training. These findings demonstrate the potential of the vibrotactile belt for guiding locomotion.
Acknowledgements: Supported by NIH R01EY029745