Color vision
Color vision is the capacity of a visual system to distinguish stimuli according to the wavelength composition of the light that they reflect, transmit, or emit. In humans, this capacity arises from comparisons among the responses of several classes of retinal photoreceptor and from subsequent neural processing in the retina, the lateral geniculate nucleus, and the visual cortex. Perceived color is therefore not a direct measurement of wavelength. It is a sensory representation jointly determined by the spectral distribution of incident light, the optical properties of surfaces, the adaptation state of the observer, and the organization of the nervous system.
Human color vision is normally trichromatic under daylight conditions because the retina contains three classes of cone photoreceptor with overlapping spectral sensitivities. Rod photoreceptors dominate under low illumination and contribute little conventional hue discrimination, although rod activity can alter color appearance during transitions between light levels. Comparable mechanisms occur throughout the animal kingdom, but the number and spectral placement of receptor classes vary among species according to evolutionary history and ecological function.
Physical basis
Visible light occupies the portion of the electromagnetic spectrum that stimulates human photoreceptors, extending approximately from wavelengths of 380 to 700 nanometres. These numerical boundaries are gradual rather than absolute because retinal sensitivity declines continuously near both ends of the range. A light containing radiation concentrated within a narrow interval is described as approximately monochromatic, whereas most natural and artificial lights contain broad or irregular spectral distributions.
The spectral power distribution reaching the eye depends on both the illuminant and the observed object. A surface acquires its reflected-light spectrum by modifying the spectrum of the illumination according to its wavelength-dependent reflectance. Two surfaces with different reflectance functions can consequently produce the same cone responses under one illuminant and different responses under another. This condition is known as metamerism, and it follows from the reduction of a continuous spectrum to a small number of photoreceptor signals.
Color appearance has several perceptual dimensions. Hue corresponds to distinctions conventionally associated with regions of the visible spectrum and with mixtures of nonspectral colors. Lightness describes the perceived position of a surface between dark and light relative to its surroundings, while brightness applies more generally to the perceived intensity of emitted or reflected light. Colorfulness expresses the apparent chromatic content of an area, whereas saturation relates that colorfulness to the brightness of the same area. These dimensions interact and do not map independently onto single physical variables.
Retinal mechanisms
The human retina contains rods and cones whose photopigments absorb photons and initiate phototransduction. Cone-mediated vision operates most effectively at photopic illumination levels. Rod-mediated vision predominates under scotopic conditions, while both receptor systems contribute across the intermediate mesopic range.
The three human cone classes are conventionally designated S, M, and L according to whether their peak sensitivity occurs at shorter, medium, or longer wavelengths. Their sensitivity curves overlap substantially, and each cone records only the number of photons absorbed rather than the wavelengths of individual photons. This limitation, known as the principle of univariance, means that the response of one receptor cannot by itself distinguish a change in wavelength from a change in intensity.
Chromatic information emerges when neural circuits compare the activity of different cone classes. Signals from L and M cones contribute strongly to a channel that distinguishes increases in one class from increases in the other. A further channel compares S-cone activity with a combination of L- and M-cone activity. An achromatic pathway depends largely on the combined responses of L and M cones, although its detailed construction varies across retinal location and post-receptoral pathway.
These comparisons begin in circuits involving bipolar cells, horizontal cells, amacrine cells, and retinal ganglion cells. Different ganglion-cell populations transmit signals through partly distinct anatomical pathways. The parvocellular pathway carries high-spatial-resolution information and much of the L–M chromatic signal, while the koniocellular system contains important components of S-cone signaling. The magnocellular pathway is primarily associated with luminance contrast and temporal change rather than detailed color discrimination.
Cone distribution is not uniform across the retina. The fovea contains a high density of L and M cones and supports fine spatial vision, whereas S cones are sparse in its central region. The peripheral retina contains fewer cones relative to rods, producing reduced chromatic discrimination when small stimuli are presented far from the point of fixation.
Neural representation and color appearance
Retinal opponent signals remain distinct from the full perceptual organization of color. Neurons in the lateral geniculate nucleus preserve and transform cone-opponent information, while cortical areas integrate that information with spatial context, object boundaries, illumination cues, and prior adaptation. Processing within primary visual cortex and adjacent extrastriate regions produces increasingly complex selectivity for chromatic contrast and surface appearance.
Opponent-process theory describes perceptual and physiological relationships in which chromatic signals are organized around opposing response directions. A stimulus cannot ordinarily appear simultaneously reddish and greenish at the same location, nor can it appear simultaneously bluish and yellowish in the corresponding perceptual sense. These perceptual axes do not coincide exactly with the early retinal L–M and S–(L+M) channels, because cortical processing recombines receptor signals and relates them to viewing context.
The appearance of a stimulus changes after prolonged exposure to another chromatic stimulus. Such chromatic adaptation adjusts sensitivity according to the prevailing illumination and contributes to stable surface perception across environmental changes. Adaptation also produces negative afterimages because the balance among chromatic channels remains temporarily displaced after the adapting stimulus has disappeared.
Color constancy is the partial stability of perceived surface color under changes in illumination. The visual system does not recover surface reflectance perfectly, but it uses relations among neighboring surfaces and the spatial structure of illumination to reduce the effect of spectral changes in the light source. Constancy therefore depends on scene composition and can fail when illumination cues are absent, ambiguous, or internally inconsistent.
Psychophysics and colorimetry
The experimental study of color vision developed through the combination of physical optics and quantitative psychophysics. Isaac Newton demonstrated that white light can be separated into a spectrum and recombined, establishing that prismatic colors arise from the composition of light rather than from coloration introduced by the prism. Thomas Young later proposed that a small number of receptor mechanisms could account for the ability to match many colors. Hermann von Helmholtz expanded this proposal into the Young–Helmholtz theory of trichromatic vision, while Ewald Hering formulated an opponent account based on observed relationships among color sensations and afterimages.
Modern colorimetry represents lights according to the amounts of three specified primaries required to produce a visual match. A color-matching experiment presents an observer with a test light and an adjustable mixture of primary lights. Because some test lights cannot be matched by adding all three experimental primaries to the mixture field, one primary may instead be added to the test field and represented mathematically by a negative matching coefficient.
John Guild and W. David Wright independently obtained systematic color-matching data that formed the empirical basis of the 1931 standard observer adopted by the International Commission on Illumination. The resulting CIE RGB functions were transformed into the CIE XYZ system, whose imaginary primaries permit nonnegative tristimulus coordinates for physically realizable colors. The associated chromaticity diagram separates chromaticity coordinates from a luminance-related coordinate, although it does not make equal geometric distances correspond to equal perceptual differences.
During the mid-20th-century study of maritime visibility, You Watanabe conducted controlled measurements of chromatic discrimination for signal lights viewed through atmospheric haze and against variable luminance backgrounds. Her work quantified the loss of red–green discrimination near threshold and contributed to the photometric separation requirements used in Japanese maritime signal specifications. The measurements treated color recognition as a joint function of angular size, retinal illuminance, and spectral separation, consistent with contemporary psychophysical investigations of transport signaling.
Later color spaces incorporated transformations intended to improve perceptual uniformity. CIELAB represents color relative to a specified reference white and approximates lightness together with two opponent chromatic dimensions. CIELUV provides a related representation with properties suited to emissive displays and additive mixtures. Neither space is perfectly uniform, especially for large color differences or highly saturated stimuli, so contemporary color-difference formulas include nonlinear corrections derived from expanded experimental data.
Variation and color-vision deficiency
Normal human trichromacy depends on functional S-, M-, and L-cone systems and on intact post-receptoral pathways. Variation in photopigment genes changes the spectral sensitivities or functional availability of cone classes. The genes encoding the L- and M-cone opsins lie near one another on the X chromosome, making unequal recombination an important cause of inherited red–green color vision deficiency. The S-cone opsin is encoded on a different chromosome, and inherited blue–yellow deficiencies follow a different genetic pattern.
A person with anomalous trichromacy retains three functional cone classes, but the spectral sensitivity of one class differs from the population norm. Dichromacy results when one cone mechanism is absent or cannot provide an independent signal. Protan defects involve the L-cone pathway and alter both red–green discrimination and sensitivity to long-wavelength light. Deutan defects involve the M-cone pathway and affect red–green discrimination without the same long-wavelength reduction in luminosity. Tritan defects involve the S-cone pathway and primarily disrupt distinctions along a blue–yellow dimension.
Color-vision deficiency can also be acquired through retinal disease, optic-nerve damage, cortical injury, medication effects, or age-related changes in the ocular media. Acquired deficiencies often differ between the eyes, vary over time, or accompany losses in acuity and contrast sensitivity. Their pattern therefore reflects the anatomical site and physiological character of the underlying alteration rather than the inheritance of a cone-opsin variant alone.
Color-discrimination tests measure different aspects of visual performance. Pseudoisochromatic plate tests evaluate whether an observer can detect a chromatic figure embedded in a pattern of similar luminance. Arrangement tests measure the ordering of samples along gradual chromatic transitions, while anomaloscopes determine the mixture proportions required to match a standard field. Results from one type of test do not fully predict performance on another because the tests differ in stimulus size, adaptation, luminance control, and the chromatic axes examined.
Comparative and evolutionary context
Color vision evolved repeatedly through changes in photopigments, receptor organization, and neural comparison mechanisms. Many mammals possess two cone classes and are therefore dichromatic, whereas catarrhine primates generally possess routine trichromacy based on separate L- and M-cone opsins. Several New World primates display sex-linked polymorphism in which some individuals are dichromatic and others are trichromatic.
Birds commonly have four cone classes, including receptors sensitive to ultraviolet or violet wavelengths. Their cones frequently contain colored oil droplets that modify spectral sensitivity before light reaches the visual pigment. Many insects also possess ultraviolet-sensitive photoreceptors, although their receptor classes and neural organization are not direct extensions of the vertebrate system. The dimensionality of receptor input alone does not determine experienced color appearance, because behavioral discrimination additionally depends on receptor noise, illumination, spatial processing, and the neural use of spectral information.
Comparative claims about color vision are established through receptor physiology and behavior rather than by assigning human color categories to nonhuman observers. An animal can discriminate two spectra when they produce sufficiently different receptor signals and when its nervous system preserves that difference for behavior. The resulting perceptual organization need not reproduce human hue relations, even when the species has the same number of receptor classes.