Abstract
How does the visual system determine whether a surface is in plain view or is partially obscured by a transparent surface (or media)? In contoured displays, it has been proposed that spatial variations in luminance contrast are used to infer the presence of transparent surfaces. Specifically, it has been argued that the highest contrast segment of a continuous contour (or texture) is seen as a surface in plain view (i.e., the highest contrast contour segment is “anchored” to appear as a surface unobscured by a transparent filter). Lower contrast regions are seen to contain transparent surfaces, and the opacity of these surfaces is determined by the magnitude of the contrast change. Here, we tested the generality of this anchoring principle by modulating contrast in both time and space. Our stimuli consisted of random dot textures against a uniform gray background. The contrast of the textures was modulated spatially, temporally, or both. We found that purely temporal contrast modulations invoked only a weak sense of transparency, whereas spatial contrast modulations induced a strong sense of transparency. However, when temporal modulations were combined with spatial modulations, a compelling sense of transparency was experienced in regions that were previously seen as surfaces in plain view. In particular, a surface region that appeared in plain view would appear to be seen behind some transparent media when it was temporally preceded by a texture that was higher in contrast. These results suggest that the anchoring of perceived transmittance has both a spatial and a temporal component.