Abstract
Klymenko and Weisstein (1980) were the first to report the percept of an illusory contour that occurs when individuals observe an animation of a wire-cube—with one of its vertical edges removed—rotating in depth back-and-forth along its vertical axis. As the cube rotates it creates the impression that the missing vertex is visible: the Motion-Induced Contour. In a series of experiments, Klymenko and Weisstein examined the boundary conditions for perceiving the contour. Critically, even in the absence of the entire cube, a Motion-Induced Contour can be observed by aligning two vertices like a pair of chevrons and rotating them in depth such that they resemble the open spine of a book. Here, we revisit the original Motion-Induced Contour illusion using insights gained from research on illusory contours and structure from motion since the initial discovery. In a manner analogous to the definition of the 'support ratio' for stationary Kanisza figures, we sought to determine the relationships between the lengths of the explicit contours in the MIC stimuli, their distance to each other, and the subjective strength of the Motion Induced Contours. Participants viewed a series of displays containing two inducing vertices that rotated in depth along their vertical axis. Vertex-length, the distance between vertices, and the orientation of the inducers were manipulated as participants judged the strength of the illusory contour perceived. Object orientation had no impact on the strength of the illusion, but both vertex-length and the distance between vertices influenced the illusory percept. Specifically, the strength of the illusion increases as a function of the length of the vertices and decreases as a function of the distance between the vertices. These results additional insight into the processes influencing illusory contours in addition to clarifying the relationship between motion-induced contours and related illusions.
Meeting abstract presented at VSS 2018