ABSTRACT
Occluded contours are very common in the visual environment. They occur each time an object is partially occluded by other objects in the foreground. The author test the hypothesis that, in vision, the first stage of computing the occluded edge from the visible part-edges yields a contour that is separated from the visible contours, even if occluded contours are necessarily the by-product of the visible contours. Occluded contours are bound together with visible parts by means of a secondary process. As counterintuitive as this hypothesis appears at first sight, it may represent a necessary step for the visual system to be able to provide alternating interpretations for the occluded contour. The results are used to reconcile conflicting evidence concerning the level of processing at which contour completion occurs. In addition to patient HJA, young and elderly control participants, placebo and lorazepam-treated participants were also evaluated.
