Color Discrimination Indices

One way to characterize human color vision is to ask how good we are at making color discriminations. Figure 3 summarizes results from four different tests of human color vision. Just as it does for all other animals, the sensitivity of humans to light varies as a function of the wavelength of the light. The results shown in Fig. 3A illustrate this point. The continuous line is called a spectral sensitivity function and it plots

Color Spectral Measurements

Figure 3 Four measurements of human color discrimination abilities. (A) Spectral sensitivity function. The continuous line plots the reciprocal of the intensity of spectral lights required for detection under daylight test conditions. (B) Wavelength discrimination function. Plotted is a measure of how much wavelength has to be changed in nanometers at each spectral location in order to yield a perceptible color difference. Higher values indicate poorer discrimination. (C) Spectral saturation. The continuous line shows the variation in the saturation of spectral lights as a function of the wavelength of the light. High values indicate greater degrees of saturation. (D) Spectral hues. Plotted as a function of spectral wavelength are the percentages of cases in which each of four separate hue categories (blue, green, yellow, and red) was employed to describe the colors of various lights.

Figure 3 Four measurements of human color discrimination abilities. (A) Spectral sensitivity function. The continuous line plots the reciprocal of the intensity of spectral lights required for detection under daylight test conditions. (B) Wavelength discrimination function. Plotted is a measure of how much wavelength has to be changed in nanometers at each spectral location in order to yield a perceptible color difference. Higher values indicate poorer discrimination. (C) Spectral saturation. The continuous line shows the variation in the saturation of spectral lights as a function of the wavelength of the light. High values indicate greater degrees of saturation. (D) Spectral hues. Plotted as a function of spectral wavelength are the percentages of cases in which each of four separate hue categories (blue, green, yellow, and red) was employed to describe the colors of various lights.

the inverse of the intensity of light required for detection under daylight test conditions. This is a measure of the degree of achromatic variation across the spectrum. The sensitivity variations shown in Fig. 3A predict that colors produced by lights of different wavelengths but equal intensities vary greatly in brightness, and so they do. This characteristic feature of the visual system is so important that a standard spectral sensitivity curve representing the averaged values for many subjects has been derived. This curve has come to be used to provide another specification of the brightness dimension—a metric referred to as luminance. Because luminance has formal properties that brightness does not possess (e.g., luminance units can be additively combined, whereas units of brightness cannot), vision scientists typically prefer luminance as a means of specifying the visual effectiveness of lights.

As noted previously, color depends centrally on the wavelength of light. Figure 3B illustrates how sensitive humans are to changes in wavelength. Plotted is a measure of the size of the wavelength difference (Al) required to yield a discriminable change in the appearance of a light for all locations across the spectrum. This result is sometimes called a hue discrimination function, although in fact wavelength change may induce both hue and saturation differences. Note that people are exquisitely sensitive to wavelength changes in two parts of the spectrum. Indeed, under stringent test conditions humans can reliably discriminate some wavelength differences that amount to no more than fractions of a single nanometer. Saturation can be measured in several ways, but they all indicate that spectral lights differ greatly in the degree to which they yield a saturated color. As illustrated in Fig. 3C, lights of both long and short wavelength (usually having an appearance of red and blue, respectively) are seen as highly saturated; lights of about 570-580 nm (seen as yellow) are very unsaturated. The results shown in Figs. 3A-3C imply that people are acutely attuned to detecting differences in several color dimensions when these dimensions are probed separately. However, how good are we at detecting differences when all three perceptual dimensions can vary as they do in normal viewing? One way to appreciate human color vision is to ask how many different colors people can see. A recent estimate suggests that those of us with normal color vision can discern in excess of 2 million different surface colors. Clearly, the human color palette is very extensive.

The results of Figs. 3A-3C are discrimination measures in which people were asked to operate as instruments solely designed to detect differences, entirely ignoring how these lights actually appear. In our ordinary experience with color the rich medium of language is most often employed to give objects color names. It turns out that people can apply color names to lights varying in wavelength and intensity in a sufficiently systematic fashion to generate very reliable indications of color appearance. Figure 3D shows the results from a so-called hue-scaling experiment in which people were asked to name the colors of lights varying in wavelength. They were allowed to use only four different color names (red, yellow, green, and blue), either singly or in pairs; the latter being the case in which one of the four names could be used to modify another (e.g., yellowish red). Not only can people do this reliably but also, perhaps surprising, there is very high consistency of the use of hue names among individuals. There are two important aspects of these results. First, note that in some parts of the spectrum hue changes rapidly with changes in wavelength (i.e., there are well-defined transitions between the color names used by the subjects). Not surprisingly, these regions are those where wavelength discrimination is most acute. Therefore, for example, at approximately 580 nm there are abrupt changes in the hue names given to lights of different wavelengths and this coincides with one of the locations where wavelength discrimination is most acute (compare Fig. 3D with Fig. 3B). Second, there are apparently mutually exclusive categories of hue appearance: A light may be seen to yield both red and yellow components (i.e., reddish yellow or yellowish red) or green and yellow components, but people almost never describe lights as containing both red and green components. The same conclusion holds for yellow and blue. This mutually exclusive pairing of color sensations has long been known and traditionally interpreted as reflecting mutually antagonistic interactions between color processing mechanisms somewhere in the visual system. The antagonistic hue pairs (red/green and yellow/blue) are usually called opponent colors.

The utilization of names to group colors into coherent categories seems universal across all human populations. Results from a survey conducted by Berlin and Kay in 1969, involving an analysis of more than 100 languages, indicated that there are only 11 basic color terms (the English equivalents are white, black, red, yellow, green, brown, purple, pink, orange, and gray). The survey further suggested that these basic color terms have "evolved" among human populations in a reasonably predictable sequence in the sense that rules can be established to specify which names are present in languages that do not contain the full set of color terms. For example, languages that contain only two basic color terms have black and white, and those with three color terms have black, white, and red. These observations have been taken to imply that basic color terms could reflect universal properties of the organization of the human nervous system. This conclusion has not escaped criticism, particularly from professional linguists, and although attempts have been made to connect these universal color categories to features of visual system physiology, no completely compelling linkages have been established.

Was this article helpful?

0 0
Do Not Panic

Do Not Panic

This guide Don't Panic has tips and additional information on what you should do when you are experiencing an anxiety or panic attack. With so much going on in the world today with taking care of your family, working full time, dealing with office politics and other things, you could experience a serious meltdown. All of these things could at one point cause you to stress out and snap.

Get My Free Ebook


Post a comment