Purchase this article with an account.
Kai M. Schreiber, Heather Rose, James M. Hillis, Clifton M. Schor, Marty S. Banks; Eye position and the 2D pattern of retinal correspondence. Journal of Vision 2004;4(8):595. doi: https://doi.org/10.1167/4.8.595.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
When we view a distant scene, all objects project onto geometrically corresponding retinal points. When the eyes are in a converged position, geometrically corresponding points can in general not be stimulated from a single location in space. However, for each retinal position in one eye, there is a spatial location that projects closest to the geometrically corresponding point in the other eye. These spatial locations define the surface that comes closest to projecting to geometrically corresponding points for this converged eye position. It is well known that empirically corresponding points do not coincide with geometrically corresponding points along the eyes′ horizontal and vertical meridians. It is possible that pattern of empirical corresponding points across the retinas is optimized for fusion of a particular surface shape or eye position. To test this hypothesis, we measured the 2D distribution of empirically corresponding points over the central 8 deg of the visual field. Observers fixated a large pattern of radial lines with a central Nonius stimulus in order to maintain accurate binocular fixation. Observers adjusted the horizontal and vertical positions of flashed dichoptic targets until perceived target movement was nulled. Settings were performed along 16 radii (0–337.5°) and at 3 eccentricities (1–8°). Corresponding retinal points for all observers are offset along the horizontal meridian in a pattern consistent with the Hering-Hillebrandt deviation, and along the vertical meridian in the previously referenced horizontal shear pattern. The overall distribution does not appear to be optimal for any particular surface or eye position. Intuitively we would expect the pattern of retinal correspondence to be shaped to the demands of near vision. The absence of such an adaptation might be attributed to having a fusional range large enough to compensate for correspondence offsets due to eye-position changes.
NIH EY 0-8882, NIH EY 0-6266, Emmy-Noether-Programm of the Deutsche Forschungsgemeinschaft, AFOSR F49620-98
This PDF is available to Subscribers Only