The ability to perceive color is a highly valued sensory capacity, and it has been a subject of experimental inquiry for over 200 years. Before the development of modern biological techniques, breakthroughs in color vision research stemmed from careful consideration of perceptual experiences. From color mixing and matching experiments, it was deduced that three different receptors in the eye, each maximally sensitive to a different region of the visible spectrum, were required to explain normal color vision. Thus, the three-component theory of Young and Helmholtz, as well as Hering’s conflicting hypothesis of three paired, opponent color processes, were developed in the 1800s, long before the three types of retinal cone photopigment were isolated and characterized. It is now understood that normal, or trichromatic, color vision is mediated by three types of cone photoreceptor—designated short- (S), middle- (M), and long- (L) wavelength sensitive—the activities of which are, indeed, combined in later opponent organization to provide color perception. The first direct measurements of the absorption properties of primate cone photopigments occurred in the 1960s using a technique called microspectrophotometry. Since then, additional techniques, including electrophysiological recordings from individual cones and molecular genetics, have allowed the refinement of the spectral sensitivity curves for the S, M, and L pigments, shown in Fig. 34.1 .
The fact that most humans have trichromatic color vision explains why computer monitors and televisions mix just three primary colors, red, green, and blue, to produce colors that match almost all those seen in the real world. Likewise, if humans were tetrachromatic, having four cone photopigments, televisions with four primaries would be required. The presence of three classes of cone photoreceptor also explains color blindness. That is, loss of function of each individual cone class is associated with an inherited form of color vision deficiency: protan defects involve loss of L-cone function, deutan defects with loss of M-cone function, and tritan defects with loss of S-cone function ( Box 34.1 ). Each type of color vision defect results from rearrangements, deletions, or mutations in the genes encoding the corresponding L, M, or S photopigment. Color vision tests have been developed to detect changes in the cone complement. Observations that cannot be fully explained just by the presence of three types of cone photoreceptor, however, involve the appearance of color. The neural circuitry for color vision extracts signals from only three cone types and yet gives rise to six color percepts arranged in opponent pairs: blue-yellow, red-green, and black-white. These topics, including the molecular genetics of color vision deficiencies, tests of color vision, and the neural circuitry of color perception, are discussed in detail in the following sections.
Protan defects are the least common form of red-green color vision defects affecting about 2% of males in the US. There are two categories: dichromatic (protanopia) and anomalous trichromatic (protanomaly), each affecting about 1% of males. The severity of protan defects varies over a relatively narrow range. Protan comes from the Greek root for “the first.” S-cones and M-cones mediate protan color vision.
Deutan defects are the most common form of red-green color vision defects, affecting about 6% of males in the United States. There are two categories: dichromatic (deuteranopia) and anomalous trichromatic (deuteranomaly). Deuteranomaly and deuteranopia affect about 5% and about 1% of males, respectively. The severity of deutan defects varies over a relatively broad range, from nearly normal to moderately red-green deficient. Deutan comes from the Greek root for “the second.” S-cones and L-cones mediate deutan color vision.
Tritan defects are relatively rare and are characterized by impaired blue-yellow color vision. Tritan defects are autosomal dominant, and they display “incomplete penetrance,” meaning there is variability in the degree to which color vision is impaired among individuals with the same underlying gene defect, even within a family. Tritan defects are also “acquired,” meaning they are not usually present from birth. Both of these observations are explained by the degenerative nature of tritan defects. Inheriting one mutant copy of the S-opsin gene causes the S-cones to degenerate. However, the cones function early in life because they express normal S-opsin from a normal copy of the gene.
Molecular genetics of color vision and color deficiencies
Although the tendency for color vision defects to run in families and to be more prevalent among men had long been recognized, a major advancement in understanding the biological basis of color vision and its defects came in 1986 when Jeremy Nathans and colleagues cloned and sequenced the genes encoding the S-, M-, and L-cone opsins and determined that the L- and M-opsin genes are in adjacent positions on the X-chromosome. Further breakthroughs in understanding differences in the quality of color vision among individuals with anomalous trichromacy followed with identification of the amino acids involved in spectral tuning, which are responsible for the approximately 30 nm difference in peak sensitivity of the L and M pigments underlying normal color vision.
Most inherited red-green color vision defects are caused by rearrangement and deletions of the L- and M-opsin genes on the X-chromosome that come about by meiotic recombination during oogenesis in females. A dichromatic phenotype results in males who inherit an X-chromosome in which all but one of the opsin genes have been deleted. A protanopic type defect is characterized by the absence of L-opsin gene expression or function. A deuteranopic type defect is characterized by the lack of M-opsin gene expression or function ( Box 34.2 ). Although humans often have more than just two opsin genes on the X-chromosome, only two are typically expressed, and these determine color vision phenotype. If meioic recombination creates an array in which the first two genes encode pigments with identical spectral properties, a male who inherits the array will be dichromatic, either protanopic or deuteranopic. However, a male with an array in which the first two genes encode opsins of the same class (M or L), but differ slightly in spectral sensitivities, will be an anomalous trichromat. He will be protanomalous if the genes encode two M-opsins or deuteranomalous if they encode two L-opsins. The extent of color vision loss in anomalous trichromats is determined by the degree of similarity in the spectral peaks of the two pigments.
Visual pigments, also known as photopigments, are composed of an apoprotein and an 11- cis retinal chromophore. The genes OPN1LW, OPN1MW, and OPN1SW each encode an apoprotein (termed opsin). The chromophore is a vitamin A derivative that absorbs ultraviolet light; when covalently bound to an opsin, the chromophore absorption spectrum is shifted to longer wavelengths. Amino acid differences among the opsins are responsible for the differences in the absorption spectra of the three cone classes ( Fig. 34.1 ). Within the L and M classes, variations or polymorphisms in the amino acid sequences produce relatively small spectral shifts, giving rise to L and M photopigment variants.
- •
OPN1MW: middle wavelength-sensitive cone opsin, expressed in M-cones, commonly referred to as “green” cones. OPN1MW is on the X-chromosome at location Xq28. Deutan defects are most commonly associated with the absence or lack of expression of the OPN1MW gene(s).
- •
OPN1LW: long wavelength-sensitive cone opsin, expressed in L-cones, commonly referred to as “red” cones. OPN1LW is on the X-chromosome at location Xq28 adjacent to OPN1MW. Protan defects are most commonly associated with the absence or lack of expression of the OPN1LW gene(s).
- •
OPN1SW: short wavelength-sensitive cone opsin, expressed in S-cones, commonly referred to as “blue” cones. OPN1SW is on chromosome 7 at 7q32.1. Tritan defects are caused by a missense mutation in one copy of the OPN1SW gene. A missense mutation is a change in a gene’s amino acid coding sequence that substitutes one amino acid for another in the encoded protein.
It has been assumed that the 30-nm spectral separation between the L and M pigments was optimized during evolution and that there is also an optimal ratio of L and M-cones. However, recent studies have shown that much smaller spectral separations between, and skewed proportions of, L and M-cones still provide robust red-green color vision. For example, observers do not show a reduction in color discrimination that can be reliably classified as just outside the normal range until the L/M separation is reduced to about 12 nm or less and anomalous trichromats with spectral separations between pigments on the order of only 3 nm still have 10 times better red-green color vision than the corresponding dichromat. Such anomalous trichromats can discern primary colors in their environment and perform basic color vision tasks. Another surprising finding occurred with the development of adaptive optics, a technique that allows visualization of the cone photoreceptor mosaic in living humans. Selective bleaching combined with adaptive optics imaging allowed identification of the different cone types and revealed that individuals with normal color vision have roughly the same number and arrangement of S-cones, but there is tremendous variation in the ratios of L- and M-cones. This was startling in light of the widely held belief that the quality of red-green color vision would be affected by extreme cone ratios. Instead, because normal color vision has low spatial resolution, it can be adequately served by a sparse mosaic of one of the cone types.
Inherited blue-yellow color vision, or tritan, defects are considerably more rare than red-green defects. They are caused by mutations in the S-opsin gene, and are inherited in an autosomal-dominant fashion, meaning that an individual who is heterozygous for an S-opsin gene mutation will often exhibit the phenotype. Evidence from adaptive optics imaging of tritan subjects indicates that the defect is an S-cone dystrophy, similar to the rod dystrophy caused by heterozygous mutations in the rod pigment rhodopsin in retinitis pigmentosa. Tritan defects are said to be incompletely penetrant, meaning that not everyone with an S-opsin gene defect exhibits the phenotype, but this is due to a progressive loss of S-cones, which must reach a critical threshold before the phenotype is manifested.
Tests of color vision
As described in the previous section, color vision defects result from genetic changes responsible for either the alteration or loss of cone photopigments. Because color vision is based on neural comparisons between different classes of cone photoreceptor, both the loss of a photopigment type and changes in the peak sensitivity of a photopigment such that there is a reduction in the differential responses of the two cone classes will result in a reduced ability to distinguish colors.
Pseudoisochromatic plate tests are perhaps the most familiar tool for diagnosing color vision deficiencies. Popular examples include Ishihara’s test of color vision and the Hardy, Rand, and Rittler (HRR) pseudoisochromatic plates. The plates consist of printed pages with dots of various colors and shades of gray, each containing a colored symbol(s) that is visible to individuals with normal color vision but invisible to those with certain color vision defects. Subjects are asked to identify the symbols. The specific pattern of errors in a test allows for diagnosis of the likely presence or absence of a color vision defect and an indication of the type and severity for some tests.
A fundamentally different testing strategy is used with arrangement tests, in which subjects are asked to arrange a series of colored discs “in order” so that each disc is placed adjacent to the discs that are most similar in color appearance. Misordering of the discs allows a diagnosis of the presence or absence of a color vision defect. Examples of arrangement tests are the Farnsworth-Munsell 100 Hue Test which involves arranging 85 discs (not 100 as the name suggests), and the much more practical Farnsworth-Munsell Dichotomous D-15 Test, which was designed to use the arrangement of 12 discs to separate strongly color deficient individuals from those with milder color vision deficiencies or normal color vision.
The Rayleigh color match is often referred to as the “gold standard” for color vision testing. It is performed on the anomaloscope, an instrument that contains an optical system that produces two side-by-side lighted fields. One field, the “test light,” is a monochromatic amber color, and the other is a mixture of red and green light. The person being tested adjusts the ratio of red-to-green light in the mixture until it exactly matches the amber test light. Setting a match with a higher or lower ratio of red-to-green light compared with the match made by a person with normal color vision is diagnostic of an inherited color vision abnormality caused by a shift in the spectral peak of either the L or M pigment. This test is extraordinarily sensitive and can detect genetically specified alterations in the spectral sensitivities of the photopigments, even in subjects in whom the alteration in the photopigment has little or no effect on the person’s ability to discriminate between different colors. Its extreme sensitivity in detecting the presence of anomalous photopigments is why the anomaloscope has been widely adopted as the “gold standard.” However, it must be emphasized that mild photopigment abnormalities may be associated with little loss in color discrimination ability. Even with shifted visual pigments, the quality of color vision may still be quite good.
Each of the widely used color vision tests has its distinct advantages and is considered to be best under different circumstances. Because of this, for general diagnosis of color vision defects, the results from a battery of tests can provide the most complete picture from which to make a differential diagnosis. Ideally, for diagnosis of color vision defects one should obtain information on the “type” of problem. Is it a red-green or blue-yellow deficiency or mixed? If it is red-green, is it a “protan” or “deutan” defect? Is it a genetic disorder, or is it caused by damage to the visual system by disease or injury? The test should also provide information about severity. Does the person have dichromatic color vision or a milder “anomalous” form of deficiency? If it is an anomalous form, is it very mild, mild, or moderate? Under some conditions, it is impractical to administer a battery of tests. Cole et al. recommended the fourth edition of the Richmond HRR Pseudoisochromatic test as the “one of choice for clinicians who wish to use a single test for color vision.” It has plates for detecting protan, deutan, and tritan defects, and its classification of mild, medium, and strong categories for deutan and protan defects is useful. Still, the HRR is not perfect at distinguishing dichromats from anomalous trichromats. The anomaloscope in the hands of a skilled practitioner is better at separating dichromats from anomalous trichromats, but it is not widely available. Administering the D-15 in conjunction with a pseudoisochromatic plate test can be helpful in discriminating dichromats from anomalous trichromats.
Genetic testing is the most reliable means of distinguishing acquired color vision defects from inherited color vision defects. Presently, genetic tests are not commercially available; however, the technology exists to perform such tests because the molecular genetics of inherited color vision defects is well understood, as described in the previous section.
Color appearance
The neural circuitry for color vision receives input from only three cone types ( Fig. 34.2 ) and yet gives rise to six hue percepts arranged in opponent pairs: blue-yellow, red-green, and black-white. The orthodox view has been that for blue-yellow opponency, signals from S-cones are combined antagonistically with an additive signal from M- and L-cones, abbreviated as S-(M+L); and for red-green opponency, signals from L-cones are combined antagonistically with those from M-cones (abbreviated as L-M). However, color vision testing in humans has established that the spectral signatures of hue perception actually involve neurons that receive contributions from all three cone types combined in the following manner: blue = (S+M)-L; yellow = L-(S+M); red = (S+L)-M; and green = M-(S+L). Therefore, although much progress has been made in identifying neurons in the visual system with color opponent responses, those matching the spectral signatures of human perception remain to be found. Results from molecular biology have shed new light on the evolution of primate color vision, imposing constraints on the possibilities for the underlying circuitry. Evolutionary constraints coupled with recent observations from human subjects with a retinal circuitry deficit are leading toward a new understanding of the color vision circuits.
