Abstract
We have previously shown that scene configuration can result in biases in relative depth perception. If a pair of points is separated both vertically, and by a depth difference some observers perceive greater depth between the points if the top is the farthest point, or vice-versa (Harris et al, VSS 2007). Here we explored whether individual differences in eye position, specifically in cyclovergence, could account for these biases.
In a single experiment, observers performed both a relative depth judgement and a psychophysical cyclovergence task. For the relative depth judgement, observers viewed two intervals. In each they saw a pair of dots, one above the other, with a depth separation between them. The task was to decide in which interval the depth separation was larger, or smaller. The dot pair could appear in two configurations: in one the upper dot was presented behind the screen plane, the lower dot presented in front (and vice-versa for the other). We compared conditions in which a test stimulus was presented in the same configuration as the fixed standard stimulus (e.g. test and standard: upper far, lower near), with conditions in which the configurations were opposite (e.g. test: upper far, lower near; standard: upper near, lower far). Immediately after each depth trial, cyclovergence was measured using a horizontal nonius line technique.
If an observer exhibited cyclovergence, one would expect a consistent bias to see one of the configurations as containing more depth than the other. Across a number of observers, we found only a very weak correlation between depth bias and cyclovergence. Further, the amount of cyclovergence measured was considerably smaller than that required to account for the biases. Our results therefore suggest that eye position differences due to cyclovergence cannot account for the large idiosyncratic biases found across observers.