Visual Motion And The Stabilization Of Gaze

National Eye Institute

Investigators

Linked publications & trials

Abstract

We sought to use the induced size effect of Ogle (1938) to dissociate the perceived slant of a surface from its horizontal disparity gradient. In this effect, a flat surface in the frontal plane appears slanted about a vertical axis when the image in one eye is vertically compressed/expanded relative to the image in the other eye even though this has no effect on the horizontal disparity gradient. Such changes in the vertical magnification of one image also reduce or enhance the perceived slants induced by horizontal compression/expansion of the other image (i.e., slants resulting from horizontal disparity gradients: the geometric effect). We previously reported that these changes in perceived slant caused by changes in vertical magnification influence the vergence eye movements associated with horizontal transfers of gaze between targets located on a surface plane. We now report on the horizontal eye movements of 4 human subjects who viewed a random dot pattern on a tangent screen in a dichoptic viewing arrangement, and shifted their gaze horizontally across the pattern, which was compressed horizontally at one eye by 3-6%. The patterns were extinguished as soon as the subjects initiated the gaze shift (open-loop responses). In half the trials, prior to the gaze shift, subjects were additionally required to vertically compress/expand the pattern at the other eye until the binocular image appeared to be fronto-parallel ("nulling"). We found that "nulling" reduced the vergence accompanying the horizontal transfers of gaze, on average, by 37% (SD+-8%, p<0.001). This reduced vergence was not simply due to degradation of the binocular depth signal(s) by the sometimes considerable vertical disparities: In a separate experiment, we showed that if these same vertical compressions/expansions were applied but with the opposite sign ("anti-nulling"), then gaze shifts were accompanied by an increase in vergence (21+-6%, p<0.001). We conclude that perceived slant can influence the planned changes in vergence that accompany gaze shifts across a surface.