Temporal properties of stereopsis
ABSTRACT The goal of the research presented in this thesis was to investigate temporal properties of disparity processing and depth perception in human subjects, in response to dynamic stimuli. The results presented in various chapters, reporting findings about different temporal aspects of disparity processing, are based on psychophysical experiments and computational model analysis. In chapter 1 we investigated which processes of binocular depth perception in dynamic random-dot stereograms (DRS), i.e., tolerance for interocular delays and temporal integration of correlation, are responsible for the temporal flexibility of the stereoscopic system. Our results demonstrate that (i) disparities from simultaneous monocular inputs dominate those from interocular delayed inputs; (ii) stereopsis is limited by temporal properties of monocular luminance mechanisms; (iii) depth perception in DRS results from cross-correlation-like operation on two simultaneous monocular inputs that represent the retinal images after having been subjected to a process of monocular temporal integration of luminance. In chapter 2 we examined what temporal information is exploited by the mechanisms underlying stereoscopic motion in depth. We investigated systematically the influence of temporal frequency on binocular depth perception in temporally correlated and temporally uncorrelated DRS. Our results show that disparity-defined depth is judged differently in temporally correlated and uncorrelated DRS above a temporal frequency of about 3 Hz. The results and simulations indicate that: (i) above about 20 Hz, the complete absence of stereomotion is caused by temporal integration of luminance; (ii) the difference in perceived depth in temporally correlated and temporally uncorrelated DRS for temporal frequencies between 20 and 3 Hz, is caused by temporal integration of disparity. In chapter 3 we investigated temporal properties of stereopsis at different spatial scales in response to sustained and transient presentation of DRS. For both sustained and transient presentations of the stimuli, the results show that: (i) stereopsis has similar temporal properties at coarse and fine spatial scales; (ii) interaction between spatial scales depends on their relative sizes. The results indicate a strong inhibitory influence of rivalry at a coarse scale on stereopsis at a fine scale, and just a weak inhibitory influence of rivalry at a fine scale on stereopsis at a coarse scale. This study provides experimental evidence for a hierarchical organisation of spatial scales in stereoscopic vision based on neural interaction instead of vergence eye movements. In chapter 4 we examined how binocular visual system interprets the depth of monocular random-dots superimposed on stereoscopic surfaces, when disparity and monocular depth result from elements of different size. We also examine the perceptual effects of dot density. We found that depth of monocular surfaces was affected by the disparity-defined surfaces and it changed gradually with dot density. Strength of the effect depended on dot density and relative angular size of the dots. In chapter 5 we examined how visual perception changes over time in the presence or absence of perceived stereoscopic depth in rivalrous images. We found that the presence of disparity-defined depth did not influence significantly the perceptual dominance durations of binocular rivalry. This indicates that stereopsis and binocular rivalry at the level of textured surfaces follow from separate processes.