Interhemispherical comparisons in the processing of contour and random texture sinewave stereograms.

Abstract

Contour and random texture stereograms were developed, using continuously variable, vertically oriented, sinewave horizontal disparities and variable texture densities. The contours were both computer printed and generated, and presented on a dual-beam oscilloscope; the textures were generated on a UNIVAC-to-Calcomp plotter, photographed, and then presented as slides, via rear-view polarized screens, in both static and dynamic modes. By means of fixation control, in normal subjects, the images in the right and left visual fields (thus: left and right visual cortices) were studied either separately or together. Parameters such as apparent depth, rate of depth-phi-motion, target density (matching and mismatching), depth ripple rate, Panum's horizontal fusion-disparity limits, and imposed monocular vertical prismatic imbalance, were studied for the separate hemispheres. In all but a few instances, the results show comparable, thus symmetrical, performances for right and left visual cortices. In those few instances where we could say that clear inter-cortical differences were found, they were found with both contour and texture targets. Furthermore, the density range of the targets (from 0.5 down to 0.005) was chosen so as to cover the phenomenal and physical range from true textures, at the high density end, to single disparate dots, at the low disparity end. But no sharp flex points were found, for any of several parameters, when moving from textured to dot targets. Although generally, observer and hemispherical variance were greater with the higher densities, the curves (with density) were ogival or S-shaped in form and never discontinuous. These results are discussed in the context of two previous findings in the literature. We cannot support the claim that there is somehow a difference in the way in which the visual cortex processes localized dot or contour targets from the way in which it processes pattern or texture targets. Secondly, the literature tends increasingly to support the contention that right occipital injuries hinder the processing of texture stereograms but not that of dot or contour stereograms. Since we could find only a scattered enhancement of right hemispherical prowess in normal vision, with both sorts of stereograms, this suggests that, - should these effects be reliably found in such patients, - they would have a different and non-congenital basis.