Contours information and the perception of various visual illusions

The simplicity principle states that the human visual system prefers the simplest interpretation. However, the conventional coding models could not resolve a contradiction between the global minimum principle and the local minimum principle. By quantitatively evaluating information content along the contours under our framework, it is shown that the most plausible interpretation of a stimulus (the one correctly predicted by local minimum principle) possesses the lowest information globally, hence resolves the contradiction. This model could also explain the perception of various visual illusions including Kanizsa illusions, Ehrenstein illusions and Rubin’s vase via estimating the lower bound of the spread parameter of the von Mises distribution governing human visual expectation. This provides new insight into the celebrated simplicity principle and could serve as a fundamental explanation of the perception of illusory boundaries and the bi-stability of perceptual grouping.