Lightness, equivalent backgrounds, and anchoring

Abstract

Observers compared two center/surround configurations haploscopically. One configuration consisted of a standard surface surrounded by two, three, or four surfaces, each with a different luminance. The other configuration consisted of a comparison surface surrounded by a single annulus that varied in luminance. Center surfaces always had the same luminance but only appeared to have the same lightness with certain annuli (equivalent backgrounds). For most displays, the luminance needed to obtain an equivalent background was close to the highest luminance in the standard surround configuration. Models based on the space-average luminance or the space-average contrast of the standard surround configuration yielded poorer fits. Implications for computational models of lightness and for candidate solutions to the anchoring problem are discussed.

Measuring the meter: on the constancy of lightness scales seen against different backgrounds

“…The study by Jandó, Agostini, Galmonte, and Bruno (2003) dealt instead with the "unmatchable problem," i.e., the impossibility to match a test stimulus that is an increment to its background to a standard stimulus that is a decrement to its background, and vice versa. While the issue is problematic to both the matching paradigm and the adjustment matching methods (Bruno, 1992;Bruno, Bernardis, & Schirillo, 1997;Whittle & Challands, 1969), Jandó et al confronted the problem only with Munsell scales simulated on a CRT. In their study the simulated scales were seen on a black background, white background and a split black-white background.…”

Inhomogeneous surrounds, conflicting frameworks, and the double-anchoring theory of lightness

“…The data analyzed in the present article come from two works (Bruno et al, 1997;Schirillo & Shevell, 1996) in which complementary methods were used. In both, a patch centered on a uniform background was compared with a patch centered on a checkerboard.…”

“…Note that their subjects were asked to match the brightness of the comparison patch-that is, to adjust the patch on checkerboard until it appeared identical to the comparison. Bruno et al (1997) reported that (1) in the samedifferent task, any uniform decremental surround worked as a lightness-equivalent background and (2) in the lighter-darker task, the luminance of the lightnessequivalent background was approximately 76 cd/m 2 (i.e., it was essentially the same as the luminance of the patch). These data also fit the double-anchoring model.…”

“…In the first, I consider two important works on this topic (Bruno, Bernardis, & Schirillo, 1997;Schirillo & Shevell, 1996) and describe their findings, which, to date, have not received a comprehensive explanation. I also show that, under certain conditions (Melfi & Schirillo, 2000;Schirillo & Shevell, 1996), the differences between subjects are so remarkable as to demand interpretation.…”

Measuring surface achromatic color: Toward a common measure for increments and decrements

“…For instance, it is known that, in simple displays, increments cannot be matched to decrements (Whittle & Challands, 1969). Unmatchable surfaces have sometimes been reported when a simple disk-ring arrangement has been compared with a more complex stimulus (Bruno, 1992;Bruno, Bernardis, & Schirillo, 1997). Most important, a number of investigators have noted that within the same stimulus space, some surfaces can appear definitely too bright, almost self-emitting, for matching on a decremental scale to be possible (Bonato & Gilchrist, 1994;Heinemann, 1955;MacLeod, 1947).…”