Global color impressions of multicolored textured patterns with equal unique hue elements

Abstract

A global color impression from a multicolored textured pattern can be identified. It is not clear, however, how such a single color impression can be determined from the elemental colors of the multicolored textured pattern. To investigate this question, two hypotheses were evaluated. The first hypothesis is that a single color impression is determined by the colorimetric average of the elemental colors in the textured pattern (colorimetric average hypothesis). The second hypothesis is that the impression is influenced by the color appearances of the elemental colors in the textured pattern (color appearance hypothesis). Using an asymmetrical color matching method, the authors obtained single color impressions for random-dot textured patterns consisting of two colors with the same unique hue and brightness but each with a different saturation. Our results showed that the matched colors were not located on the line connecting the two elemental colors of the pattern, but rather were on the curved unique hue loci line. Furthermore, the chromaticities of the matches shifted toward a higher saturation than the colorimetric averages. These results support the color appearance hypothesis and suggest that a single color impression from a multicolored textured pattern is determined by a mechanism that integrates the color appearances, i.e., hue, saturation, and brightness (or lightness), of the elemental colors in the pattern. In addition, it seems that the integration of the color appearances is not a simple process, because the apparent saturation of the color impression was higher than that of the colorimetric average and the average of the chromaticities of the colors in the pattern.

Accurate rapid averaging of multihue ensembles is due to a limited capacity subsampling mechanism

“…Kuriki [31] has shown that adjustments to mosaics with many tiny elements were not reflective of the colorimetric mean, being biased towards the most saturated element. However this was under continuous viewing conditions [see also 32], rather than the rapid-exposure of the ensemble perception paradigm.…”

Perceived Color Impression for Spatially Mixed Colors

“…Although rigorous studies have been carried out in the fields of detection and discrimination of spatial periodic patterns [3]- [6], there have not been many studies of perceived color impression from mixed patterns in the mid-range spatial frequency in which the discrimination of patterns is rather obscure [7]- [9]. Kuriki reported that the mosaic element with the highest saturation strongly influences the perceived average color to represent a mosaic pattern [8].…”

Effects of ensemble complexity and perceptual similarity on rapid averaging of hue

“…Previous research has also found that observers can approximate the colorimetric mean when adjusting a homogenous patch to represent a continuously presented multicolor mosaic but that these estimates are biased toward the salience or saturation of the mosaic elements (Kuriki, 2004) and the position of the unique hues (Sunaga & Yamashita, 2007)-i.e., the red, green, blue, or yellow that appears pure, unmixed with any other hue (see Kuehni, 2014). Such adjustments are also more variable when the perceptual distance between element colors is greater (Webster et al, 2014).…”

“…Studies that have probed directly for representations of the mean color (e.g., Kuriki, 2004;Sunaga & Yamashita, 2007;Webster et al, 2014) have allowed observers unlimited time to view the ensemble stimuli rather than using rapid presentation to encourage the deployment of distributed attention (e.g., Alvarez & Oliva, 2008;Baijal, Nakatani, van Leeuwen, & Srinivasan, 2013;Treisman, 2006). When rapid presentation has been used (e.g., de Gardelle & Summerfield, 2011;Maule et al, 2014;Michael et al, 2014), tasks have not probed directly for representations of the mean color.…”

“…For those interested in ensemble coding mechanisms and functions more generally, color is an ideal substrate for investigation. It is well described and characterized in terms of human perception, is continuous yet also subject to categorization (e.g., Bird, Berens, Horner, & Franklin, 2014;He, Witzel, Forder, Clifford, & Franklin, 2014;Roberson, Pak, & Hanley, 2008), and can help answer ecologically valid questions about the appearance of surface colors (Giesel & Gegenfurtner, 2010;Sunaga & Yamashita, 2007). Furthermore, hue (the subjective experience of which is qualitative), along with the other dimensions of color, saturation, and lightness (which are matters of magnitude), provide an opportunity to understand how ensemble coding deals with the integration of multiple perceptual dimensions.…”