Depth magnitude and binocular disparity: a closer look at patent vs. qualitative stereopsis

Abstract

Ogle (1952; 1953) used measurements of perceived depth as a function of disparity to divide human stereopsis into patent (quantitative) and qualitative categories. Patent depth percepts result from a range of disparities within and outside Panum's fusional zone, while qualitative percepts result only from very large disparities well beyond the fusional limit. While this dichotomy is widely recognized, it is not clear if it is merely descriptive, or if it reflects an underlying neural dichotomy. If the latter is true, then patent and qualitative depth percepts should be associated with other distinguishing properties. In this series of experiments we evaluate the possibility that the 1^st /2^nd -order dichotomy proposed by Hess & Wilcox (1994) maps onto Ogle's patent/qualitative distinction. We used a magnitude estimation technique to evaluate the amount of depth perceived from test disparities within and beyond the fusable range. In separate blocks of trials we used stimuli designed to activate either the luminance-based 1^st-order or the contrast-based 2^nd-order system. The stimuli were windowed, 1D luminance noise patches that were presented either as correlated or uncorrelated stereopairs which activated 1^st and 2^nd-order stereopsis respectively. As anticipated, we find that at small disparities our 1^st-order stimuli provide patent depth percepts that follow geometric predictions. However, our data also reveal that quantitative depth percepts are provided by 2^nd-order stereopsis at small disparities, but the amount of depth is less than predicted by viewing geometry. Further, depth percepts become qualitative as the stimuli become diplopic and are mediated by solely 2^nd-order mechanisms. Our results show that Ogle's qualitative stereopsis reflects the operation of a distinct neural mechanism designed to provide crude depth estimates for diplopic stimuli. The situation for stimuli within Panum's area is not as straightforward, as both 1^st and 2^nd-order mechanisms provide quantitative depth information in this range.

Footnotes

 NSERC to LMW.