The Visual and Vestibular Perception of Passive Self-Rotation

This study investigated human sensitivity to detect conflicts between visual and vestibular information about angular displacement during passive rotations around the earth-vertical (yaw) axis. During a passive whole-body rotation primarily the optic flow and displacement information from the visual system and the signals from the semicircular canals in the vestibular system play a role [1]. In our experiments, physical yaw rotations were presented using a Stewart Motion Platform and visual stimuli were shown on a projection screen. The gain (vis./vest.) – ratio between the visual and physical rotation angle—was varied so that the visual stimuli were either faster or slower than the physical rotation. The task was a psychophysical conflict detection task (2AFC) where different gain factors (0.25–2.75) of the visual and vestibular rotation speed were presented in a pseudo-random order and participants had to judge whether the visual motion was faster or slower than the real physical motion. The rotation angle of the platform was always 30 and a raised cosine velocity profile was used. Two different kinds of visual stimuli were used to compare the impact of visually rich and immersive visual stimuli (“Market place of Virtual Tuebingen (VT)”) to that of a simple, abstract visual stimuli (“Limited life-time Random Dots (RD)”). The results show following effects. First, the psychometric function shows that the PSE (Point of Subjective Equality) is around the gain factor 1.6 (Vis./Vest.) across all subjects. This implies an underestimation of the visual speed: The visuals had to be 1.6 times faster than the physical motion to be perceived as equally fast. This is consistent with other studies which also show that there is an underestimation of the visual rotation speed [2, 3] in similar experimental conflict situations. Secondly, there is no significant difference of the PSE and JND (Just Noticeable Difference) between the VT- and RD-condition which could imply that the visual richness is not a significant factor to detect a conflict between the visual and vestibular inputs. Further control studies will be conducted to investigate whether different angular displacements and higher physical rotation speed will influence the PSE and the thresholds