Role of a Texture Gradient in the Perception of Relative Size

Abstract

Two theories regarding the role of a texture gradient in the perception of the relative size of objects are compared. Relational theory states that relative size is directly specified by the projective ratio of the numbers of texture elements spanned by objects. Distance calibration theory assumes that relative size is a product of visual angle and distance, once the distance is specified by the texture. Experiment 1 involved three variables: background (no texture, texture gradient patterns), the ratio of heights of the comparison stimulus to a standard (three levels), and angular vertical separation of the standard stimulus below the horizon (two levels). The effect of the retinal length of the comparison stimulus was examined in experiment 2. In both experiments, participants judged both the apparent size and distance of a comparison stimulus relative to a standard stimulus. Results suggest that the cues selected by observers to judge relative size were to some degree different from those used to judge relative distance. Relative size was strongly affected by a texture gradient and the retinal length of a comparison stimulus whereas relative distance perception was affected by relative height. When dominant cues that specify size are different from those which specify distance, relational theory might provide a better account of relative size perception than distance calibration theory.

Ebbinghaus Illusion

“…The other theoretical idea is that of relational theory (Tozawa, 2010). The relational theory specifies that the perceived size of objects is determined by how much of a texture is covered by an object.…”

The nature of perceptual constancies

“…However, at least with respect to some constancies exhibited by some perceptual systems, so‐called ‘direct’ theories are seen as viable alternatives to ‘indirect’ proposals, and may well turn out to be correct (cf. Epstein 1973; Shebilske & Peters 1995; Norman 2002; Tozawa 2010). Thus it may be argued that the information disentanglement account is too restrictive , which is why I will dub this problem ‘the restriction problem’.…”

“…What is more, it is an issue that must be decided separately for each constancy and each (type of) biological organism, and even a cursory glance at the empirical literature reveals that, in very many cases, the final verdict is not yet in (cf. Shebilske & Peters 1996; Norman 2002; Tozawa 2010). So if the information disentanglement account really entails what, prima facie, it seems to entail – namely, that an indirect theory will be true for every single constancy in nature –, then the account is in deep trouble.…”

Height Perception Influenced by Texture Gradient

“…This pattern was generated by the distance point construction method, a method of drawing horizontal lines and vertical lines based on two distance points (eg Sedgwick 1980;Gillam 1981). It is identical to the method employed by Tozawa (2010). To produce a textured background, first vertical lines that converge toward a vanishing point were constructed (these lines do not appear in figure 4).…”

“…This theory is referred to as the relational theory (see Hochberg 1971). Much research has shown that texture cues influence size perception (Gibson 1950;Blessing et al 1967;Vogel and Teghtsoonian 1972;Tozawa 2010). For example, Tozawa investigated the effectiveness of texture cues using horizontal lines with the same or different lengths displayed on a regularly textured background.…”