Experimental measures and theoretical account of hue scaling as a function of luminance

Functions relating the magnitude of chromatic response to stimulus luminance were determined at the unique green and yellow wavelengths. These measured functions were used to account for the shift from a preponderance of green at low stimulus luminances to a preponderance of yellow at higher luminances, for a stimulus wavelength of 550 nm. The two hue magnitude functions were found to have the same power-law exponents, and the hue shift was due to the difference in thresholds between the two opponent-color systems.     

Scaling of saturation and hue

The study concerns the relation of saturation to the purity and luminance of aperture colors viewed in a dark surround. For the primary hues, red, yellow, green, and blue, and the intermediate hues, orange and yellowish green, the saturations increased as power functions of colorimetric purity. An IS-dB increase in luminance caused a threefold increase in the exponent for yellow, but luminance had little effect on the exponents of the other colors. The direct heterochromatic matching of saturation to saturation confirmed the validity of the scales determined by magnitude estimation and led to the construction of families of saturation scales based on a common unit called a crome. Equisection and jnd scales were also determined. Their nonlinearity suggests that saturation is a prothetic continuum. It was found that mixing red or green with yellow behaves much the same as mixing red or green with achromatic light. The changes in hue behave as prothetic continua, for the equisection and jnd scales are nonlinearly related to the power-function scales obtained by magnitude estimation and matching.     

Are there nontrivial constraints on colour categorization?

In this target article the following hypotheses are discussed: (1) Colour is autonomous: a perceptuolinguistic and behavioural universal. (2) It is completely described by three independent attributes: hue, brightness, and saturation: (3) Phenomenologically and psychophysically there are four unique hues: red, green, blue, and yellow; (4) The unique hues are underpinned by two opponent psychophysical and/or neuronal channels: red/green, blue/yellow. The relevant literature is reviewed. We conclude: (i) Psychophysics and neurophysiology fail to set nontrivial constraints on colour categorization. (ii) Linguistic evidence provides no grounds for the universality of basic colour categories. (iii) Neither the opponent hues red/green, blue/yellow nor hue, brightness, and saturation are intrinsic to a universal concept of colour. (iv) Colour is not autonomous.     

Colour changes as a function of luminance contrast.

When spectral light increases in luminance, the hues change. Normally, long-wavelength light becomes increasingly yellow, and short-wavelength light turns blue or blue-green. This is known as the Bezold-Brücke hue shift. Less notice has been paid to the change in relative chromatic content (saturation or chromatic strength) that accompanies these shifts in hue. As luminance contrast increases from zero, chromatic strength increases to reach a maximum at a luminance that is wavelength dependent. Short-wavelength blueish light reaches this maximum at low relative luminances, whereas midspectral yellowish stimuli need several log units higher luminance. Red and green are somewhere in between. For luminances above this maximum, the chromatic content usually diminishes, and most light becomes more whitish in appearance. In this study it is demonstrated how the combined chromatic appearance of hue and chromatic strength change with intensity. Both phenomena find a common physiological interpretation in the nonlinear and nonmonotonic responses of colour-opponent P cells in the retina and lateral geniculate nucleus of the primate. A model that combines the outputs of six P-cell types accounts for observers' estimates of hue and chromatic strength.     

Identification and adaptation of hue: Parallels in the operation of mechanisms that underlie categorical perception in vision and in audition

Summary Four experiments were conducted to examine processes in identification and selective adaptation of hues in color perception that exactly parallel processes in identification and adaptation of auditory detectors that provide information for phonemic perception. The first experiment demonstrated an effect of adaptation on identification of blue and green when a hue category center was used as the adaptor; this experiment also assessed recovery from adaptation. Adaptation to one hue was found to shift identification to favor the alternative hue, implicating a single detector underlying hue categorization. The second experiment demonstrated similar effects of adaptation between green and yellow. The third experiment compared the magnitudes of shift following adaptation with a category center, a near-boundary hue, and variously graded adaptation series. Adaptation was found to be related to the category representativeness of the adaptor(s). Results of the third experiment also provided support for the view that adaptation, rather than response bias, is responsible for shifts in the position of identification functions following extended stimulus exposure. The fourth experiment explored the neural loci of adaptation by an interocular transfer test. Hue adaptation was found to occur at both central and peripheral loci. In the four main experiments, reaction times to identify hues in unadapted and adapted states were also analyzed and compared. Subsidiary experiments assessed the effects of stimulus luminance on the magnitude of adaptation. General principles of categorical perception and its underlying bases, including the sweep, magnitude, and symmetry of adaptation, are discussed. The principal findings of these studies provide new data on hue perception which strikingly parallel findings in speech perception.     

Color categories in macaques.

This experiment investigated whether macaque monkeys partition the photic spectrum into the same four basic hue categories that humans do, i.e., blue, green, yellow, and red. Monkeys were trained to respond in the presence of one chromatic stimulus and were tested, in extinction, for generalization to others. In extinction, the monkeys responded at similar and high levels to stimuli that fell in the same basic human hue category as the training stimulus and at similar and much lower levels to stimuli that fell in a different human hue category from the training stimulus. It was concluded that macaques and humans categorize the spectrum in a similar fashion.     

Luminance thresholds for the Bezold-Brücke hue shift

This study presents data on the luminance difference at which a hue difference is first perceived between two identical spectral lights. Thresholds were obtained for both luminance increments and decrements from a 2.00 log troland standard for wavelengths between 470 nm and 690 nm. A predicted luminance threshold for each wavelength was calculated based on wavelength discrimination data and the Purdy constant hue contours; i.e. that luminance at which the constant hue contour intersects the wavelength discrimination bound for a given wavelength. A generally good agreement was obtained between observed and predicted values.     

Hue discrimination as a function of stimulus luminance

The relation between hue discrimination and stimulus luminance was investigated. It was found that discrimination was best at the highest luminance and deteriorated at the lowest, except in the yellow region.     

Hue shift of induced colour as a function of luminance relations

The hue of induced colour was studied as a function of surround/test field luminance ratio using a chromatic surround and an achromatic central test field. The hue of the test field was determined by means of colour naming methods. Three inducing colours were used: blue (Wr No. 47), green (Wr No. 58), and red (Wr No. 25). The number of subjects was 9–11 in the two experiments. The luminance ratio (ranging from 0.07 to 17.1) was varied by varying the luminance of the test field (Experiment 1) or of the surround (Experiment 2). For the blue surround the results show a hue shift in accordance with the Bezold-Brücke phenomenon. For the inducing colours green and red the induced colours are weak, and the hue shifts are more or less unsystematic though there are individual subjects showing a trend in the Bezold-Brücke direction. It is concluded that the hue shifts depend on the luminance relations rather than on the test field luminance.     

Color measurement and color theory: II. Opponent-colors theory

Quantitative opponent-colors theory is based on cancellation of redness by admixture of a standard green, of greenness by admixture of a standard red, of yellowness by blue, and of blueness by yellow. The fundamental data are therefore the equilibrium colors: the set A₁ of lights that are in red/green equilibrium and the set A₂ of lights that are in yellow/blue equilibrium. The result that a cancellation function is linearly related to the color-matching functions can be proved from more basic axioms, particularly, the closure of the set of equilibrium colors under linear operations. Measurement analysis treats this as a representation theorem, in which the closure properties are axioms and in which the colorimetric homomorphism has the cancellation functions as two of its coordinates.Consideration of equivalence relations based on opponent cancellation leads to a further step: analysis of equivalence relations based on direct matching of hue attributes. For additive whiteness matching, this yields a simple extension of the representation theorem, in which the third coordinate is luminance. For other attributes, precise representation theorems must await a better qualitative characterization of various nonlinear phenomena, especially the veiling of one hue attribute by another and the various hue shifts.     

Color preferences in infants and adults are different

Adults commonly prefer blues most and greenish yellows least, but these hue preferences interact with lightness and saturation (e.g., dark yellow is particularly disliked: Palmer & Schloss (Proceedings of the National Academy of Sciences 107:8877–8882, 2010)). Here, we tested for a similar hue-by-lightness interaction in infant looking preferences, to determine whether adult preferences are evident early in life. We measured looking times for both infants and adults in the same paired-comparison task using all possible pairs of eight colors: four hues (red/yellow/green/blue) at two lightness levels (dark/light). The adult looking data were strikingly similar to other adults’ explicit preference responses, indicating for the first time that adults look longer at colors that they like. Infants showed a significant hue-by-lightness interaction, but it was quite different from the adult pattern. In particular, infants had a stronger looking preference for dark yellow and a weaker preference for light blue than did adults. The findings are discussed in relation to theories on the origins of color preference.     

Contingent capture in cueing: the role of color search templates and cue-target color relations

Visual search studies have shown that attention can be top-down biased to a specific target color, so that only items with this color or a similar color can capture attention. According to some theories of attention, colors from different categories (i.e., red, green, blue, yellow) are represented independently. However, other accounts have proposed that these are related—either because color is filtered through broad overlapping channels (4-channel view), or because colors are represented in one continuous feature space (e.g., CIE space) and search is governed by specific principles (e.g., linear separability between colors, or top-down tuning to relative colors). The present study tested these different views using a cueing experiment in which observers had to select one target color (e.g., red) and ignore two or four differently colored distractors that were presented prior to the target (cues). The results showed clear evidence for top-down contingent capture by colors, as a target-colored cue captured attention more strongly than differently colored cues. However, the results failed to support any of the proposed views that different color categories are related to one another by overlapping channels, linear separability, or relational guidance (N = 96).     

Episodic-like memory in pigeons

It has been proposed that memory for personal experiences (episodic memory, rather than semantic memory) relies on the conscious review of past experience and thus is unique to humans. In an attempt to demonstrate episodic-like memory in animals, we first trained pigeons to respond to the (nonverbal) question "Did you just peck or did you just refrain from pecking?" by training them on a symbolic matching task with differential responding required to the two line-orientation samples and reinforcing the choice of a red comparison if they had pecked and the choice of a green comparison if they had not pecked. Then, in Experiment 1, after providing the conditions for (but not requiring) the pigeons to peck at one new stimulus (a yellow hue) but not at another (a blue hue), we tested them with the new hue stimuli and the red and green comparisons. In Experiment 2, we tested the pigeons with novel stimuli (a circle, which they spontaneously pecked, and a dark response key, which they did not peck) and the red and green comparisons. In both experiments, pigeons chose the comparison appropriate to the response made to the test stimulus. Thus, the pigeons demonstrated that they could remember specific details about their past experiences, a result consistent with the notion that they have the capacity for forming episodic-like memories.     

Reaction times reveal the contribution of the different receptor components in luminance perception

Piéron (1914, 1920, 1952) demonstrated that simple reaction time decays as a hyperbolic function of luminance. Similarities between cell latencies and reaction time (RT) to luminance suggest that this relationship may be determined by retinal processes. If the exponent of the Piéron function is specific to a given sensory modality, as assumed by some authors (e.g., Bonnet, 1992a, 1992b; Norwich, 1987), it should reflect receptor activities. Consequently, functions with different exponents should fit data for different luminance ranges. In a contrast-discrimination experiment, we investigated this question with a large range of luminance levels in a two-alternative spatial forced-choice task. The results of the experiment show that three functions with different exponents fit RT to the three luminance ranges (scotopic, mesopic, and photopic). The exponent decreases with increasing luminance. The findings indicate that the exponent and the asymptotic latency of the RT function reflect receptor activities of the visual system.     

Color adaptation in a yellow-blue system: proof of a simple numerical relationship]

Three axes in color space are specified: a (unique) yellow - (unique) blue axis, a (unique) green - magenta axis and brightness. Based on the two chromatic axes two linear opponent colors systems are defined: a red/green-system and a yellowish/bluish-system. A numerical relation is presented to describe color adaptation for the yellowish/bluish-system under adaptation to (unique) yellow and (unique) blue: two pairs of color stimuli are equivalent with regard to the yellowish/bluish-system - consisting of a test stimulus and an adaptation stimulus, respectively - if the ratios from the yellowish/bluish-coordinates of test stimulus and adaptation stimulus are identical. A control of brightness and the red/green-system is presupposed. For several (unique) yellow and (unique) blue adaptation stimuli it is examined how a test stimulus that appears neither yellowish nor bluish changes its location on the (unique) yellow - (unique) blue axis within color space. Three observers take part in the experiment. For each observer a plane of constant brightness and the opponent colors axes are estimated experimentally. The data show that the ratios from the yellowish/bluish coordinates of test stimulus and adaptation stimulus are essentially constant. The results are compared with analogous data for the red/green-system. The findings provide evidence for the specification of the three phenomenal axes. The specification is discussed with regard to Hering's opponent colors theory and Krauskopf's three "cardinal" axes [1982, Vision Research, 22, 1123-1131].     

Colour matching of isoluminant samples and backgrounds: a model

A cone-opponent-based vector model is used to derive the activity in the red-green, yellow-blue, and achromatic channels during a sequential asymmetric colour-matching experiment. Forty Munsell samples, simulated under illuminant C, were matched with their appearance under eight test illuminants. The test samples and backgrounds were photometrically isoluminant with each other. According to the model, the orthogonality of the channels is revealed when test illuminants lie along either red-green or yellow blue cardinal axes. The red green and yellow-blue outputs of the channels are described in terms of the hue of the sample. The fact that the three-channel model explains the data in a colour-matching experiment indicates that an early form of colour processing is mediated at a site where the three channels converge, probably the input layer of V1.     

Age differences in what is viewed and remembered in complex conjunction search

Older and younger adults searched arrays of 12 unique real-world photographs for a specified object (e.g., a yellow drill) among distractors (e.g., yellow telephone, red drill, and green door). Eye-tracking data from 24 of 48 participants in each age group showed generally similar search patterns for the younger and older adults but there were some interesting differences. Older adults processed all the items in the arrays more slowly than the younger adults (e.g., they had longer fixation durations, gaze durations, and total times), but this difference was exaggerated for target items. We also found that older and younger adults differed in the sequence in which objects were searched, with younger adults fixating the target objects earlier in the trial than older adults. Despite the relatively longer fixation times on the targets (in comparison to the distractors) for older adults, a surprise visual recognition test revealed a sizeable age deficit for target memory but, importantly, no age differences for distractor memory.     

Age differences in what is viewed and remembered in complex conjunction search

Older and younger adults searched arrays of 12 unique real-world photographs for a specified object (e.g., a yellow drill) among distractors (e.g., yellow telephone, red drill, and green door). Eye-tracking data from 24 of 48 participants in each age group showed generally similar search patterns for the younger and older adults but there were some interesting differences. Older adults processed all the items in the arrays more slowly than the younger adults (e.g., they had longer fixation durations, gaze durations, and total times), but this difference was exaggerated for target items. We also found that older and younger adults differed in the sequence in which objects were searched, with younger adults fixating the target objects earlier in the trial than older adults. Despite the relatively longer fixation times on the targets (in comparison to the distractors) for older adults, a surprise visual recognition test revealed a sizeable age deficit for target memory but, importantly, no age differences for distractor memory.     

Newborn's discrimination among mid- and long-wavelength stimuli

Recent anatomical and behavioral evidence suggests that important visual functions like color vision may be severely limited at birth. Therefore, we examined human newborns' ability to discriminate among stimuli of different wavelength by habituating them to a 16 degrees chromatic square of widely varying luminance (as wide as 1.1 log cd/m2) and then testing for recovery of habituation to a different chromatic square of the same size. In the first experiment, newborns showed evidence of recovery to a 650-nm (peak wavelength) red square after habituating to 545-nm green squares, but they did not recover to the red or the green after habituating to 585-nm yellow squares. In a second experiment using the same procedure, newborns did not show evidence of discriminating the 650-nm red from the 545-nm green when the stimulus size was reduced to 8 degrees. These data suggest that human newborns are capable of making a chromatic discrimination within the spectral region above 540 nm (the Rayleigh region) but their ability is limited to chromatic stimuli of very wide spectral separation and of very large size. Possible neurological bases underlying this immaturity are discussed.     

Hue distinctiveness overrides category in determining performance in multiple object tracking

The visual distinctiveness between targets and distractors can significantly facilitate performance in multiple object tracking (MOT), in which color is a feature that has been commonly used. However, the processing of color can be more than “visual.” Color is continuous in chromaticity, while it is commonly grouped into discrete categories (e.g., red, green). Evidence from color perception suggested that color categories may have a unique role in visual tasks independent of its chromatic appearance. Previous MOT studies have not examined the effect of chromatic and categorical distinctiveness on tracking separately. The current study aimed to reveal how chromatic (hue) and categorical distinctiveness of color between the targets and distractors affects tracking performance. With four experiments, we showed that tracking performance was largely facilitated by the increasing hue distance between the target set and the distractor set, suggesting that perceptual grouping was formed based on hue distinctiveness to aid tracking. However, we found no color categorical effect, because tracking performance was not significantly different when the targets and distractors were from the same or different categories. It was concluded that the chromatic distinctiveness of color overrides category in determining tracking performance, suggesting a dominant role of perceptual feature in MOT.