Induction in variants of White's effect: common or separate mechanisms?

Empirical evaluation of computational models of lightness perception

Lightness of a surface depends not only on its physical characteristics, but also on the properties of the surrounding context. As a result, varying the context can significantly alter surface lightness, an effect exploited in many lightness illusions. Computational models can produce outcomes similar to human illusory percepts, allowing for demonstrable assessment of the applied mechanisms and principles. We tested 8 computational models on 13 typical displays used in lightness research (11 Illusions and 2 Mondrians), and compared them with results from human participants (N = 85). Results show that HighPass and MIR models predict empirical results for simultaneous lightness contrast (SLC) and its close variations. ODOG and its newer variants (ODOG-2 and L-ODOG) in addition to SLC displays were able to predict effect of White's illusion. RETINEX was able to predict effects of both SLC displays and Dungeon illusion. Dynamic decorrelation model was able to predict obtained effects for all tested stimuli except two SLC variations. Finally, FL-ODOG model was best at simulating human data, as it was able to predict empirical results for all displays, bar the Reversed contrast illusion. Finally, most models underperform on the Mondrian displays that represent most natural stimuli for the human visual system.

Contrasting a Misinterpretation of the Reverse Contrast

The reverse contrast is a perceptual phenomenon in which the effect of the classical simultaneous lightness contrast is reversed. In classic simultaneous lightness contrast configurations, a gray surrounded by black is perceived lighter than an identical gray surrounded by white, but in the reverse contrast configurations, the perceptual outcome is the opposite: a gray surrounded by black appears darker than the same gray surrounded by white. The explanation provided for the reverse contrast (by different authors) is the belongingness of the gray targets to a more complex configuration. Different configurations show the occurrence of these phenomena; however, the factors determining this effect are not always the same. In particular, some configurations are based on both belongingness and assimilation, while one configuration is based only on belongingness. The evidence that different factors determine the reverse contrast is crucial for future research dealing with achromatic color perception and, in particular, with lightness induction phenomena.

Layer and framework theories of lightness

Lightness (the perceived dimension running from black to white) represents a problem for vision science because the light coming to the eye from an object totally fails to specify the shade of gray of the object, due to the confounding of surface gray and illumination intensity. The two leading approaches, decomposition theories and anchoring theories, split the retinal image into overlapping layers and adjacent frameworks, respectively. Because each approach has important strengths and some weaknesses, an integration of them would mark an important step forward for the lightness theory. But the problem remains how this integration can actually be realized.

Dissecting the influence of the collinear and flanking bars in White's effect

In White's effect equiluminant test patches placed on the black and white bars of a square-wave grating appear different in brightness. The illusion has generated intense interest because the direction of the brightness effect does not correlate with the amount of black or white border in contact with the test patch, or in its general vicinity. Therefore, unlike brightness induction effects such as simultaneous contrast, White's effect is not consistent with explanations based on contrast or assimilation that depend solely on the relative amounts of black and white surrounding the test patches. We independently manipulated the luminance of the collinear and flanking bars to investigate their influence on test patch matching luminance (brightness). The inducing grating was a 0.5c/d square-wave and test patches measured 1.0° in width and either 0.5° or 3.0° in height. Test patches measuring 0.5° in height had more extensive contact with the collinear bars and test patches measuring 3.0° in height had more extensive contact with the flanking bars. The luminance of the collinear (or flanking) bars assumed twenty values from 3.2 to 124.8cd/m(2), while the luminance of the flanking (or collinear) bars remained white (124.8cd/m(2)) or black (3.2cd/m(2)). Under these conditions the influence of the collinear and flanking bars was found to be purely in the direction of contrast. The effect was dominated by contrast from the collinear bars (which results in White's effect), however, the influence of the flanking bars was also in the contrast direction. The data elucidate the luminance relationships between the collinear and flanking bars which produce the behavior associated with White's effect as well as that associated with "the inverted White effect" which is akin to simultaneous contrast.

An exponential filter model predicts lightness illusions

Lightness, or perceived reflectance of a surface, is influenced by surrounding context. This is demonstrated by the Simultaneous Contrast Illusion (SCI), where a gray patch is perceived lighter against a black background and vice versa. Conversely, assimilation is where the lightness of the target patch moves toward that of the bounding areas and can be demonstrated in White's effect. Blakeslee and McCourt (1999) introduced an oriented difference-of-Gaussian (ODOG) model that is able to account for both contrast and assimilation in a number of lightness illusions and that has been subsequently improved using localized normalization techniques. We introduce a model inspired by image statistics that is based on a family of exponential filters, with kernels spanning across multiple sizes and shapes. We include an optional second stage of normalization based on contrast gain control. Our model was tested on a well-known set of lightness illusions that have previously been used to evaluate ODOG and its variants, and model lightness values were compared with typical human data. We investigate whether predictive success depends on filters of a particular size or shape and whether pooling information across filters can improve performance. The best single filter correctly predicted the direction of lightness effects for 21 out of 27 illusions. Combining two filters together increased the best performance to 23, with asymptotic performance at 24 for an arbitrarily large combination of filter outputs. While normalization improved prediction magnitudes, it only slightly improved overall scores in direction predictions. The prediction performance of 24 out of 27 illusions equals that of the best performing ODOG variant, with greater parsimony. Our model shows that V1-style orientation-selectivity is not necessary to account for lightness illusions and that a low-level model based on image statistics is able to account for a wide range of both contrast and assimilation effects.

Local computation of lightness on articulated surrounds

Lightness of a grey target on a uniform light (or dark) surround changes by articulating the surround (articulation effect). To elucidate the processing of lightness underlying the articulation effect, the present study introduced transparency over a dark surround and investigated its effects on lightness of the target. The transparency was produced by adding a contiguous external field to the dark surround while keeping local stimulus configuration constant. Results showed that the target lightness did not change on the articulated surround when a dark transparent filter was perceived over the target, although it did on the uniform surround. These results suggest that image decomposition into a transparent filter and an underlying surface does not necessarily change lightness of the surface if the surface is articulated. Moreover, the present study revealed that articulating the surround does not always enhance lightness contrast; it can reduce the contrast effect when the target luminance is not the highest within the surround. These findings are consistent with the theoretical view that lightness perception on articulated surfaces is determined locally within a spatially limited region, and they also place a constraint on how the luminance distribution within the limited region is scaled.

The Wertheimer–Benary Effect Does Not Invert, and a Nulled Wertheimer–Benary Effect

More than three-quarters of a century ago Wertheimer and Benary demonstrated an ingenious and clear, though, interestingly, small effect: a grey triangle just inside an arm of a black cross on a white background appears slightly lighter than an identical triangle immediately adjacent to the cross, despite both triangles having the same perimeter exposure to black and white. Over a generation ago White discovered an apparently related, but far stronger effect: when short grey (test) bars are placed onto either black or white alternating long bars, the short test bars placed on the long black bars appear much lighter than those placed on the long white bars. A decade ago Spehar, Gilchrist, and Arend found that, enigmatically, if the short test bars in White's effect are the lightest stimulus in a figure, then the relative lightness of the test bars inverts compared with the standard version of White's effect. Here we show that the Wertheimer–Benary effect does not invert, but instead produces a very weak version of the standard effect. We also demonstrate a novel, nulled Wertheimer-Benary effect.

Standard definitions of chromatic induction fail to describe induction with S-cone patterned backgrounds

Inducing patterns that selectively stimulate the S cones can induce large shifts in color appearance. For example, a "peach" test-ring presented within contiguous purple and non-contiguous lime inducing rings appears pink while the physically identical peach test-ring appears orange when presented within contiguous lime and non-contiguous purple inducing rings (Fig. 1c). These shifts have been accounted for by a neural substrate which predicts that chromatic assimilation and simultaneous contrast can operate synergistically to produce large shifts with these patterns [Monnier, P., & Shevell, S. K. (2004). Chromatic induction from S-cone patterns. Vision Research, 44, 849-856]. Here, induction was measured for test-rings that stimulated the S cones either more or less than did the inducing rings. According to standard definitions of induction, color shifts for test s-chromaticities either lower or higher than both inducing chromaticities should be attenuated compared to test-rings of intermediate S-cone stimulation. On the other hand, a previously proposed model of induction predicted independence of the color shifts with test-ring s-chromaticity. Consistent with standard definitions of induction, a reduction in the magnitude of the color shifts for test-ring chromaticities either lower or higher in S-cone excitation than the inducing chromaticities was observed. Additional measurements with patterns that have been shown to isolate assimilation and simultaneous contrast were conducted. For these patterns, expectations based on standard definitions of induction suggested that the magnitude of the color shifts should be monotonic with the S-cone stimulation of the test-ring, and the direction of the color shift should reverse for test-ring chromaticities either lower or higher than both inducing chromaticities compared to test-rings of intermediate chromaticity. In contrast, the previously proposed model of induction based on a receptive-field with S-cone spatial antagonism predicted the color shifts should be independent of the test-ring chromaticity (Monnier & Shevell, 2004). Color shifts were generally independent of the level of the test-ring chromaticity, supporting the S-cone antagonistic model of induction.

[The effects of the shape of stripes and the length of target area on White's effect and inverted White's effect]

Two experiments were designed to differentiate White's effect and an inverted White's effect. In White's effect, a gray target bordered by black stripes longer than white stripes appears darker than a target of the same physical luminance bordered by white stripes longer than black stripes when the luminance of the target area lies between the minimum and maximum luminance values of the inducing stripes. In an inverted White's effect, the direction of the effect is reversed when the target is either lighter or darker than both inducing stripes (Spehar, Clifford, & Agostini, 2002). Twenty undergraduates were asked to rate the brightness of a target area using a Munsell scale on the display. The results showed that both White's effect and an inverted White's effect were reproduced in any condition of Experiment 1 and 2. Experiment 1 showed that an inverted White's effect was affected by the shape of stripes, while White's effect was not Experiment 2 showed that an inverted White's effect was affected by the length of target area, while White's effect was not. The results support Spehar et al. (2002) that White's effect and an inverted White's effect are not mediated by the same mechanism.

Influence of target size and luminance on the White-Todorovic effect

Variants of a lightness effect described by [Todorovic's, D. (1997). Lightness and junctions. Perception, 26, 379] were studied to quantify the failure of lightness constancy as a function of target luminance and target size. Todorovic's effect is similar to White's effect. Simultaneous lightness contrast appears to operate selectively between stimuli belonging to the same perceptual group, and not between stimuli of equal proximity belonging to different perceptual groups. We found that mid-gray targets grouped with a white contextual stimulus were matched on average to a darker-than-veridical gray. Those grouped with a black contextual stimulus were matched on average veridically. This is consistent with 'anchoring' effects observed in simple two-stimulus displays. However, target luminance had an effect that was not captured by mid-level target luminance data or data averaged across target luminances. For both white and black contextual stimuli, light-gray targets were matched to a darker-than-veridical gray and the direction of this error shifted toward the lighter-than-veridical direction as the luminance of the target was lowered. The result was a constant difference between the perceived lightnesses of targets presented with white and black contextual stimuli. Target size had no effect on perceived lightness. These data imply that the Todorovic-White effect can be characterized as lightness assimilation rather than as lightness contrast. By accounting for compression as well as the Todorovic-White effect, assimilation is the more general explanation.

The statistical structure of natural light patterns determines perceived light intensity

The same target luminance in different contexts can elicit markedly different perceptions of brightness, a fact that has long puzzled vision scientists. Here we test the proposal that the visual system encodes not luminance as such but rather the statistical relationship of a particular luminance to all possible luminance values experienced in natural contexts during evolution. This statistical conception of vision was validated by using a database of natural scenes in which we could determine the probability distribution functions of co-occurring target and contextual luminance values. The distribution functions obtained in this way predict target brightness in response to a variety of challenging stimuli, thus explaining these otherwise puzzling percepts. That brightness is determined by the statistics of natural light patterns implies that the relevant neural circuitry is specifically organized to generate these probabilistic responses.