The place of white in a world of grays: A double-anchoring theory of lightness perception.

Abstract

1The specific grey shades in a visual scene can be derived from relative luminance values only when an anchoring rule is followed. The double-anchoring theory I propose in this paper, as a development of the anchoring theory of , assumes that any given region (a) belongs to one or more frameworks, created by Gestalt grouping principles, and (b) is independently anchored, within each framework, to both the highest luminance and the surround luminance. The region's final lightness is a weighted average of the values computed, relative to both anchors, in all frameworks. The new model not only accounts for all lightness illusions that are qualitatively explained by the anchoring theory but also for a number of additional effects, and it does so quantitatively, with the support of mathematical simulations.Keywords: lightness illusions, simultaneous lightness contrast, highest luminance, anchoring theories, illumination discounting All at once his white shirt blazed out, and I came out after him from shadow into full sunlight... -Ursula LeGuin, The Left Hand of DarknessOur visual world can be treated as a collection of groups of surfaces that belong together, either by design, like the black and white stripes on the zebra's back or the four regions in Figure 1a, or by accident, like the combination of sky fragments and tree branches through my window every morning at eight o'clock. Groups such as these have been called frameworks . Depending on the structural complexity of its context, a region can simultaneously belong to two or more nested frameworks, as is indeed the case with most objects in our visual experience. In the example of Figure 1a, what we may call the global 1 This article is a result of the cognitive dissonance stemming from my enchantment with the anchoring theory and my subsequent, and unplanned, experience of double-increment illusions. I am indebted to Peter Kramer, Dejan Todorović, Dave Rose, Ennio Mingolla, Luigi Burigana, Joe Cataliotti, Fred Bonato, Nicola Bruno, Tom Troscianko, Elias Economou, Suncica Zdravkovic, Piers Howe, Daniele Zavagno, and Fred Kingdom for their critical reading of one or the other of the several drafts of this manuscript. Special thanks to Alan Gilchrist for innumerable exchanges-an honor and a pleasure.Correspondence concerning this article should be addressed to Paola Bressan, Dipartimento di Psicologia Generale, Università di Padova, Via Venezia 8, 35131 Padova, Italy. E-mail: paola.bressan@unipd.it DOUBLE-ANCHORING THEORY OF LIGHTNESS framework (by temporarily ignoring the page and the visible rest of the room) is composed of two side-by-side local frameworks, each containing a grey patch on a uniform background. The interesting aspect of this familiar display is that the two patches are the same physical shade of grey, but the one in the black field appears lighter than the other. In spite of its apparent irrelevance to the issue of lightness, the multiple-framework idea, developed by Gilchrist and his collaborators, explains precisely why this should be so. The ...

Von Bezold Assimilation Effect Reverses in Stereoscopic Conditions

“…Kanizsa did not expand this intuition any further; but it seems clear from his words that perceptual belongingness may generate stronger contrast effects when it favours the figure/ground stratification process. With regards to this point, Bressan (2006) proposed a lightness perception model that, although it does not include a luminance decomposition process, assumes that the figure/ground segregation process affects the contrast phenomenon.…”

The subjective duration of ramped and damped sounds

“…It is wellknown that human memory for magnitude of stimulations within an intensive continuum (i.e., those continua in which the signal involves a change in stimulus energy-e.g., loudness, lightness) is rather poor (see Garner, 1953, for an example involving lightness). On the contrary, maximal or minimal (e.g., silence) stimulations can validly serve as anchors for estimations (see, e.g., Bressan, 2006, andGilchrist et al, 1999).…”

A Novel Tone Mapping Based on Double-Anchoring Theory for Displaying HDR Images

“…In this paper, we propose a novel compression operator for tone mapping of HDR images, which is based on two anchoring values obtained by the Double-anchoring theory [3]. We adopt an accelerated K-means algorithm based on triangle inequality to decompose HDR images into frameworks, which can reduce the amount of time needed.…”

“…Since we employ the double anchoring theory, we need to find the highest luminance value and surrounding luminance that would be perceived as white. According to the surrounding luminance represented in paper [3], there was no specific definition for complex image. In DABTM, we propose a new method to define the surrounding luminance.…”

“…This effect is known as simultaneous lightness contrast. In [3], in order to relate the luminance to the lightness, a criterion between the luminance value and the perceived gray shades is defined. The criterion is called anchor.…”