Answers are grounded in this paper and cite their sources.
Gibson (1950a) has suggested that a gradient of retinal stimulation from an irregularly textured surface is used in the estimation of slant. There are several measures that could form such a gradient. Eight pictures of slanting surfaces were constructed with the aid of a computer. These consisted only of texture elements: in this case, outline circles and ellipses. Subjects studied pairs of pictures and indicated which had the greater tilt. By varying separately the gradients of element size and element density, it was possible to show that only the former contributed appreciably to the impression of slant. Only some of the measures considered are consistent with this result.
No takes yet. Share an insight, caveat, or question.
R. J. Phillips (1970) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: