Transparent objects are those that allow light to flow through them, and researchers normally anticipate the light to match the color of the item. When the object is put on a colored surface, the light that returns to the eyes is a combination of surface, illumination, and transparency qualities. Despite this, human eyes can frequently distinguish separate layered and overlying layers with different hues.
Although it was uncertain how neurons divide information to extract the transparent layer, the physical properties of transparent filters create geometrical and color features in retinal pictures, which might give indications for layer separation. In forced-preference psychophysics studies, investigators assessed the relative value of such cues in a perceptual scale for transparency, utilizing stimuli with X- or T-junctions, various relative movements, and consistent or inconsistent colors cooperating or competing. Maximum-likelihood Thurstone scaling indicated that motion enhanced transparency for X-junctions while decreasing transparency for T-junctions by producing the illusion of an opaque patch.
However, if the motion of a filter revealed a dynamically changing but stationary pattern, having the same destiny as the surrounding but creating T-junctions, the likelihood of detecting transparency was nearly as great as for moving X-junctions, despite the stimulus being physically implausible. Furthermore, geometric signals significantly outweighed color inconsistencies.
Finally, a linear model of transparency perception as a result of relative movements between the filter, overlay, and surrounding layers, as well as contour continuity and color consistency, quantified a hierarchy of latent impacts on whether the filter is perceived as a distinct transparent layer.
Reference:jov.arvojournals.org/article.aspx?articleid=2778812
