Date of Award

Spring 2011

Project Type


Program or Major


Degree Name

Doctor of Philosophy

First Advisor

William Wren Stine


The visual system can use small disparities between the stimuli seen by each eye as a binocular cue for depth. Geometric disparities are the most commonly used cue for stereopsis, but disparities in contrast and luminance can also lead to the perception of depth and both can cancel perceived depth from a geometric disparity. Humans are able to perceive changes in depth from contrast disparities at frequencies up to approximately 1.4 Hz, whereas changes in depth from geometric disparities are visible at frequencies above 5 Hz (Dobias, 2008). This dissertation is divided into two parts. The first part (experiments 1 & 2) explores the dynamics of canceling the perceived depth from a geometric disparity using a contrast or a luminance disparity. The second part (experiments 3-5) explores the role of individual differences in the dynamics of perceived depth from contrast, luminance, and geometric disparities. Dobias (2008) and results in experiments 3 & 4 show that the probability of reporting depth from contrast and luminance disparities diminish at frequencies much lower than from a geometric disparity (critical frequency). A generalized difference model (simple low-pass filter), does not account for the data from experiment 1. However, a gated generalized difference model, which applies low-pass filtering to the amplitude of the modulation but not its rate of change, does account for the data. Results from part two show that the frequency at which depth changes from contrast and luminance disparities (experiment 3) differs from the frequency at which depth is no longer visible for a geometric disparity (experiment 4), that the frequency can differ depending on which eye views the changing stimulus (experiments 3 & 4), and that either visual masking or eye movements are likely not causing the decrease at lower frequencies for contrast and luminance disparities (experiment 5). The finding that results from part one follow the gated model, and the differences depending on which eye views the dynamic stimulus found in part two, suggest that the visual system processes depth from contrast and luminance disparities much differently from what would be expected based on our typical understanding of depth perception.