Humans integrate multisensory information to reduce perceptual uncertainty when perceiving the world (1, 2) and self (3, 4) and it has been shown that two multisensory cues are combined and give rise to a single percept only if attributed to the same causal event (5,6,7). A growing body of literature studies the limits of such integration for bodily self-consciousness and the perception of self-location under normal and pathological conditions (8). We extend this research by investigating whether human subjects can learn to integrate two arbitrary visual and vestibular cues of self-motion due to their temporal co-occurrence.

We conducted two experiments (N=8 each) in which whole-body rotations were used as the vestibular stimulus and optic flow as the visual stimulus. The vestibular stimulus provided a yaw self-rotation cue, the visual – a roll (experiment 1) or pitch (experiment 2) rotation cue. Subjects made a relative size comparison between a standard rotation size and a variable test rotation size. Their discrimination performance was fit with a psychometric function and perceptual discrimination thresholds were extracted. We compared experimentally measured thresholds in the bimodal condition with theoretical predictions derived from the single cue thresholds.

Our results show that human subjects can learn to combine and optimally integrate vestibular and visual information, each signaling self-motion around a different rotation axis (yaw versus roll as well as pitch). This finding suggests that the experience of two temporally co-occurring but spatially unrelated self-motion cues leads to inferring a common cause to these two initially unrelated sources of information about self-motion.

multisensory integration; self-motion; visuo-vestibular