Huff, Samuel (2002) An empirical and theoretical study of stereoscopic illusory contours and surfaces. PhD thesis, James Cook University.

Preview

PDF (Thesis) - Submitted Version Download (14MB)

Abstract

Part 1: The first part of this thesis outlines major concepts important for later analyses. Chapter 1 introduces concepts of stereoscopic vision. Topics include retinal coordinate geometry, retinal disparity, its neutral substrates, disparity gradients, the correspondence problem and a contemporary information-processing model of binocular vision. Chapter 2 presents some of the diverse literature that has addressed illusory contours, discusses the notions of modal and amodal completion and the experiments that have revealed the psychophysical character of illusory contours. Suggested neutral substrates and Grossberg's explanation of 2-D illusory contours are also presented. In chapter 3 recent research in stereoscopic ilusory contours and surfaces is examined. Stimuli including the stereoscopic Kaniza square, Gulick and Lawson's sparse texture matrices, stereoscopic contrast spreading and stereo capture effects are discussed. The chapter develops theoretical distinction between a Surface Heuristic view and a Form Computation view of the mechanisms that underpin stereoscopic illusory contours and surfaces.

Part 2: The second part of the thesis presents studies designed to explore the binocular vision processes that extract vision from the 2-D layout of half-images in a way that yields 3-D illusory contours and surface. The aim of the series of experiments was to reveal binocular image processing mechanisms able to recover stereoscopic depth in the absence of texture. Chapter 4 describes and inital experiment designed to establish the relationship between the phenominal characteristics of a 3-D illusory figure and stereopsis. Experiment 1 employed three metrics: seen depth, seen slant and lightness judgement. These three metrics suggest that the appearance of illusory 3-D percept might be a product of binocular image processing mechanisms. Chapter 5 proposes mechanisms that could account for the stereoscopic Kaniza percepts. These are tested in Chapter 6 and Chapter 7. Finally the generality of the mechanisms is tested in several experiments concerning stereoscopic Ehrenstein figures in Chapter 8.

Part 3: Part 3 of the thesis reviews the possible generality of processes identified in the BIPASS model for perception of a range of stimuli generating 3-D illusory percepts. It then summarises experimentation and makes concluding remarks. In Chapter 9 the BIPASS model is reviewed and used to offer an alternative functional account of percepts for several key stimuli. These stimuli have been principally proposed by the Surface Heuristic approach. The chapter attempts to describe how the 2-D structure of the half-images is related to the 3-D illusory percepts. Grossberg's approach is also reviewed. It is suggested that a gap in his theory appears to be the apparently mechanistic spearation of surface layers in stereopsis. Finally, Chapter 10 summarises the project and proposes some possible implications for natural binocular vision. Several stereo photographs are used to demonstrate the possible relevance of the BIPASS model for natural vision. It is suggested the outcome of binocular vision may be a complex product of 2-D relationships between point disparities, disparate subtense and the topographic architecture of the binocular sensory array.

Item ID:	27720
Item Type:	Thesis (PhD)
Keywords:	stereoscopic vision; binocular vison processes; illusory contours; illusory surfaces; binocular sensory array
Date Deposited:	26 Jun 2013 03:39
FoR Codes:	17 PSYCHOLOGY AND COGNITIVE SCIENCES > 1701 Psychology > 170112 Sensory Processes, Perception and Performance @ 90% 17 PSYCHOLOGY AND COGNITIVE SCIENCES > 1702 Cognitive Science > 170299 Cognitive Science not elsewhere classified @ 10%
SEO Codes:	97 EXPANDING KNOWLEDGE > 970117 Expanding Knowledge in Psychology and Cognitive Sciences @ 100%
Downloads:	Total: 254 Last 12 Months: 2
	More Statistics