Global Motion Distorts Perceived Shape: An Investigation of the Relationship Between Motion and Form

This dissertation, "Global Motion Distorts Perceived Shape: an Investigation of the Relationship Between Motion and Form" by Wang-on, Li, 李允安, was obtained from The University of Hong Kong (Pokfulam, Hong Kong) and is being sold pursuant to Creative Commons: Attribution 3.0 Hong Kong License. The content of this dissertation has not been altered in any way. We have altered the formatting in order to facilitate the ease of printing and reading of the dissertation. All rights not granted by the above license are retained by the author. Abstract: Abstract of thesis entitled Global Motion Distorts Perceived Shape: An Investigation of the Relationship Between Motion and Form Submitted by Li Wang On for the degree of Doctor of Philosophy at the University of Hong Kong in February 2007 Coding of spatial information by the visual system was once considered to be an automatic consequence of where light falls on retina. This topographic map was thought of as constituting the necessary spatial information used for further processing to, for example, identify an object. Recent psychophysical and physiological studies have shown that retinal projection coupled with fidelity in topographic transmission through the visual pathway may not be the sole source for spatial perception. Indeed, a topographic map constitutes an implicit visual code, whereas it is increasingly clear that, as David Marr eloquently argued in his 1982 monograph, it is explicit coding that is important for vision. The apparent position of an object possessing motion information is often shifted in the direction of motion; subsequent spatial processing operates as if the object is located at its apparent, and not retinal, position. The major research question of the present dissertation is whether a motion-induced position shift leads to form distortion. If so, is the consequence a challenge to traditional views on modular processing of motion and form? The effect of global motion on the perception of three types of form, including forms generated by motion trajectories, illusory forms, and motion-contrast defined forms, was investigated. The results show that motion distorts the perception of form as if its position information within the motion field had been induced by the motion signal to a position in ithe direction of motion. An apparent elliptical motion trajectory appeared to be circular when its major axis, containing its apses, is located within a contracting motion field and its minor axis is located within an expanding motion field. A Kanizsa figure appears to be "fatter" when it is located in an expanding motion field. The boundaries of a motion- defined form also appear to be shifted in the direction of motion, and lead to a distortion in the perceived shape. However, only motion owned by the target of interest (or when the motion ownership cannot be clearly distinguished) substantially affects its form perception. The distortion effect vanishes when the target can be segregated from its motion background; segregation cues include polarity and disparity, and speed and direction of motion. Motion and form processing are not as independent as once thought; motion distorts perceived shape by inducing a position shift. Figure-ground segmentation seems to distinguish between which motion is relevant to the spatial perception of the figure; motion in the background has little or no effect on the spatial perception of the figure. The implication of this result is that shape distortion is a not a global effect that arises from a motion-induced distortion of visual space. One conclusion drawn in that, the visual system should be considered as a unified computational system. Spatial perception does not result from topographical mapping; it is the computational result of information of different types and at different levels. A primary concern may not be what types of information exist in an image, but whether the information is owned by the target