A color code is a system for encoding and representing non-color information with colors to facilitate communication. This information tends to be categorical (representing unordered/qualitative categories) though may also be sequential (representing an ordered/quantitative variable).

25-pair color code chart used in certain kinds of wiring

History

edit

The earliest examples of color codes in use are for long-distance communication by use of flags, as in semaphore communication.[1] The United Kingdom adopted a color code scheme for such communication wherein red signified danger and white signified safety, with other colors having similar assignments of meaning.

As chemistry and other technologies advanced, it became expedient to use coloration as a signal for telling apart things that would otherwise be confusingly similar, such as wiring in electrical and electronic devices, and pharmaceutical pills.

Encoded Variable

edit

A color code encodes a variable, which may have different representations, where the color code type should match the variable type:

  • Categorical variable – the variable may represent discrete values of unordered qualitative data (e.g. blood type)
  • Quantitative variable – the variable represents ordered, quantitative data (e.g. age)
    • Discrete quantitative data (e.g. the 6 sides of a die: 1,2,3,4,5,6) are sometimes treated as a categorical variable, despite the ordered nature.

Types

edit

The types of color code are:

  • Categorical – the colors are unordered, but are chosen to maximize saliency of the colors, by maximizing color difference between all color pair permutations.
  • Continuous – the colors are ordered and form a smooth color gradient.
  • Discrete – only a subset of a continuous color code are used (still ordered), where each is distinguishable from the others.

Categorical

edit

When color is the only varied attribute, the color code is unidimensional. When other attributes are varied (e.g. shape, size), the code is multidimensional, where the dimensions can be independent (each encoding separate variables) or redundant (encoding the same variable). Partial redundancy sees one variable as a subset of another.[2] For example, playing card suits are multidimensional with color (black, red) and shape (club, diamond, heart, spade), which are partially redundant since clubs and spades are always black and diamonds and hearts are always red. Tasks using categorical color codes can be classified as identification tasks, where a single stimulus is shown and must be identified (connotatively or denotatively), versus search tasks, where a color stimulus must be found within a field of heterogenous stimuli.[3][2] Performance in these tasks is measured by speed and/or accuracy.[2]

The ideal color scheme for a categorical color code depends on whether speed or accuracy is more important.[3] Despite humans being able to distinguish 150 distinct colors along the hue dimension during comparative task, evidence supports that color schemes where colors differ only by hue (equal luminosity and colorfulness) should have a maximum of 8eightcategories with optimized stimulus spacing along the hue dimension,[3] though this would not be color blind accessible. The IALA recommends categorical color codes in seven colors: red, orange, yellow, green, blue, white and black.[4] Adding redundant coding of luminosity and colorfulness adds information and increases speed and accuracy of color decoding tasks.[3] Color codes are superior to others (encoding to letters, shape, size, etc.) in certain types of tasks. Adding color as a redundant attribute to a numeral or letter encoding in search tasks decreased time by 50–75%,[2]: Fig9  but in unidimensional identification tasks, using alphanumeric or line inclination codes caused less errors than color codes.[3][2]: 19 

Several studies demonstrate a subjective preference for color codes over achromatic codes (e.g. shapes), even in studies where color coding did not increase performance over achromatic coding.[2]: 18  Subjects reported the tasks as less monotonous and less inducing of eye strain and fatigue.[2]: 18 

The ability to discriminate color differences decreases rapidly as the visual angle subtends less than 12' (0.2° or ~2 mm at a viewing distance of 50 cm),[5] so color stimulus of at least 3 mm in diameter or thickness is recommended when the color is on paper or on a screen.[6] Under normal conditions, colored backgrounds do not affect the interpretation of color codes, but chromatic (and/or low) illumination of surface color code can degrade performance.[3]

Criticism

edit

Color codes present some potential problems. On forms and signage, the use of color can distract from black and white text.[7]

Color codes are often designed without consideration for accessibility to color blind and blind people, and may even be inaccessible for those with normal color vision, since use of many colors to code many variables can lead to use of confusingly similar colors.[7][8] Only 15–40% of the colorblind can correctly name surface color codes with 8–10 color categories, most of which test as mildly colorblind. This finding uses ideal illumination; when dimmer illumination is used, performance drops sharply.[8]

Examples

edit

Systems incorporating color-coding include:

See also

edit

References

edit
  1. ^ Transactions of the American Society of Civil Engineers: Volume 29 (1893), p. 507.
  2. ^ a b c d e f g Christ, Richard E. (December 1975). "Review and Analysis of Color Coding Research for Visual Displays". Human Factors: The Journal of the Human Factors and Ergonomics Society. 17 (6): 542–570. doi:10.1177/001872087501700602.
  3. ^ a b c d e f Jones, Mari Riess (December 1962). "Color Coding". Human Factors: The Journal of the Human Factors and Ergonomics Society. 4 (6): 355–365. doi:10.1177/001872086200400604.
  4. ^ IALA (December 2009). The Surface Colours used as Visual Signals on Aids to Navigation (2 ed.). International Association of Marine Aids to Navigation and Lighthouse Authorities. pp. 7–9.
  5. ^ Bedford, R. E.; Wyszecki, G. W. (1 February 1958). "Wavelength Discrimination for Point Sources". Journal of the Optical Society of America. 48 (2): 129. doi:10.1364/JOSA.48.000129.
  6. ^ Conover, Donald W.; Kraft, Conrad L. (1958). The Use of Color in Coding Displays. Wright Air Development Center, Air Research and Development Command, United States Air Force.
  7. ^ a b See, e.g., Michael Richard Cohen, Medication Errors (2007), p. 119.
  8. ^ a b Cole, Barry L (1 July 2004). "The handicap of abnormal colour vision". Clinical and Experimental Optometry. 87 (4–5): 258–275. doi:10.1111/j.1444-0938.2004.tb05056.x.
  9. ^ "Color-Coded Loot". Giant Bomb. Retrieved 13 June 2016.
edit