Abstract
Spurious motions are commonly produced at points of occlusion by moving objects. The visual system must discount these motions to successfully interpret image motion; this necessitates form information. To help characterize the form analysis affecting the discounting process, we used stimuli based on a cross moving within an occluding aperture. The two bars of the cross cohere or move separately depending on the context; in accord with prior literature, coherence depends in part on whether the bar endpoints appear to be occluded. We explored the dependence of the discounting process on the junctions generated at potential points of occlusion. In some cases, junction categories seemed to have a large effect on whether motion was discounted; in others they made little difference. For instance, coherence depended strongly on whether the junctions formed at the cross intersection were Ls rather than Ts, but was little affected by the category of the junction at the cross endpoints. Further experiments suggested that what matters is not junctions per se, but the relative genericity of a display's various interpretations, which in our stimuli is determined by illusory edges. Changing the junction category seems to matter only when it affects the relative genericity of the interpretations of a stimulus. Parsimony thus favors an explanation of these phenomena in terms of the genericity of layered surface interpretations, with no explicit reference to junctions.