Abstract
Introduction. Psychophysical and physiological studies of stereopsis have demonstrated the existence of luminance based (1st-order) and contrast-based (2nd-order) processing. The 2nd-order mechanism is thought to provide depth information when the luminance information is unreliable or unavailable. A number of investigators have suggested that 2nd-order stereopsis provides only qualitative depth information, but this proposal has not been tested empirically. The aim of this set of experiments is to evaluate this claim and in doing so clarify the contribution of 2nd-order processing to human depth perception.
Methodology. We have used a novel two-temporal alternative forced-choice procedure and a method of constant stimuli. This task was designed to avoid the separation confound inherent in discrimination tasks using diplopic stimuli. Observers were required to judge which of two intervals contained the largest difference in depth between a reference + disparity pedestal and a target + disparity pedestal. We assessed performance across a large range of fused and diplopic disparities, and measured diplopia for each observer. In addition we used stimuli designed to favour 1st-order, or isolate 2nd-order processing, to permit comparison of the relative contributions of these two mechanisms.
Results. All observers were able to perform the 2IFC depth interval judgment using both 1st and 2nd-order stimuli, though there were large and consistent differences between these conditions. The most notable result is that observers perceived quantitative depth from targets that isolated 2nd-order processing. Results from the 1st-order condition show a clear transition from high-resolution performance in the fused range to coarser low-resolution depth perception in the diplopic range, which is likely mediated by 2nd-order processing. This study provides the first definitive evidence that quantitative depth can be provided by both 1st- and 2nd-order mechanisms in the fused range, but only the 2nd-order signal is used when stimuli are diplopic.