Abstract
Perceived directions of self-motion can be multimodally integrated from orthogonally directed visual and vestibular stimulation (Sakurai et al., 2003, 2010). When observers passively experience real oscillatory forward/backward somatic motion while viewing leftward/rightward visual flow patterns consistent with rightward/leftward body motion, their perceived self-motion direction is intermediate to those specified by visual and vestibular information individually. To extend these studies, we introduced multiple levels of angular differences between body-motion direction and visually specified motion direction. We investigated the optimal range of the multimodal integration by measuring the perceived direction of self-motion and observers' confidence ratings. Participants were seated on a rotatable chair on a motor-driven swing and wore a head-mounted display. The visual stimuli consisted of translating vertical sine-wave gratings phase-locked to the swing motions. The vestibular stimulation was from somatic oscillatory motion with one of 13 orientations of the chair ( 0 to 180 degrees in 15-degree intervals) relative to the path of the swing. In the 0-degree condition the participants' leftward/rightward somatic motion and its phase were congruent with the visual stimuli, while they were incongruent in other conditions. Participants were sound-cued to indicate their perceived direction of self-motion via a rod-pointing task, and then indicated their confidence of judgments with a 5-point scale. The perceived directions of self-motion were intermediate to those specified by visual and vestibular information in 60-, 75- and 90-degree conditions in a weighted combination fashion, suggesting these are optimal conditions for integration. Average confidence ratings, however, were lower in these optimal conditions than those in other conditions. The perceived directions were larger than the real somatic motion directions in 15- and 30-degree conditions, while they were close to the real somatic motion directions in 120-, 135-, 150-, and 165-degree conditions, suggesting vestibular information was dominant in multimodal integration in these conditions.
Meeting abstract presented at VSS 2016