Abstract
When a brief offset vernier target is followed by a bar grating of non-offset vernier elements, the target is sometimes not masked but instead is perceived as being in a foreground plane separate from the mask elements in the background plane (Herzog & Koch, 2001). We analyzed the dynamics of the 3D LAMINART (Cao & Grossberg, 2005) cortical model of depth perception and found that it explains the appearance of shine-through for these stimuli. The model explanation proposes that shine-through is due to a combination of false binocular disparity matches between the target and the central element of the mask, and a weakening of between-disparity-plane competition that occurs from spatial competition and boundary grouping. Simulations of the model demonstrate that it closely matches empirical data on the properties of shine-through, including effects of the number of elements in the mask, the influence of a gap in the mask grating, and long-range effects from elements around the mask grating. The model is contrasted with an alternative explanation of shine-through, and novel mask conditions are identified that allow for empirical tests of the model hypotheses.