Abstract
Heading perception is inaccurate when the optical flow contains radial and lateral (rotational) motions as in the presence of eye-movement. Observers perceive a curved path if they are not moving their eyes and the optical flow is simulated by visual motion. However, they accurately perceive heading direction when the lateral motion is generated by an actual eye-movement. Thus, the extra-retinal information of eye-movement is utilized to perceive heading (Royden, 1994). We aimed to see whether the process for visual control of posture utilizes the extra-retinal information.
A linear viewpoint motion was simulated in a cloud of dots. Participants observed its visual image motion on a 120 inch screen at 130 cm distance while standing on a force plate to measure postural sway. For the real eye-movement condition, a red marker was set slightly left or right of the heading direction in the cloud of dots, and observers were asked to pursue the moving red marker with eye-movement. For the simulated eye-movement condition, a virtual camera pursued the red marker and its image motion was simulated. Thus, the red marker to be fixated was constant at the center of the screen. The image motions were identical on the retina for both conditions.
Observers inclined in the opposite direction to the visual simulation of linear viewpoint motion in sagittal direction for both conditions. Lateral sway was stronger in the simulated eye-movement condition than in the real eye-movement condition.
Though the retinal image motions were identical and contained both the lateral and radial motions, the postural sway to be induced by the lateral motion was inhibited when the observers actually moved their eyes. It is suggested that the process for the visual control of posture does not only due to the retinal image motion, but utilizes the extra-retinal eye-movement information.
This study was supported by Nissan Science Foundation and MEXT Japan.