Abstract
When a target is observed binocularly, the viewing direction (e.g., 30 deg rightward against straight ahead) causes both horizontal and vertical disparities. Some studies showed the effect of retinal cue in perceiving direction of viewing, however, the gain was low (Banks, et al., Berends, et al.). The oculomotor cue probably suppresses the retinal cue. This research attempts to reduce the influence of the oculomotor cue and examine the retinal cue as a cue to perceive direction of viewing. An excessive usage of extraocular muscle probably deteriorate accuracy of signaling eye position (i.e., proprioception); for example, one holds his eyes onto a limit position of the range of eye movement for a while, then moves his eyes toward straight ahead. In our experiment, a subject was asked to keep his eyes onto a limit position, upper right or upper left, for 15 seconds before each presentation of a test stimulus. The stimulus was random-dot stereogram displayed on a rear-projected screen in a dark room. The subject observed the stimulus through stereo shutter glasses. The center of the stimulus was always on the median plane of the subject, however, a disparity pattern consisted of horizontal and vertical disparities was given to the stimulus to simulate direction of viewing. The subject was asked to adjust the direction of an unseen pointing stick by hand to indicate perceived direction of the center of the stimulus. The results were as follows: the retinal cue is effective in perceiving direction, the effect is weak without pre-observing eye movement, and pre-observing eye movement increases the effect of retinal cue on direction perception.