The Nature of Colour As Light

Light, the Eye, and Colour
For most of us, the perception of colour occurs so easily and naturally that it seems to suggest that colour resides fully and directly in the objects that we see. In fact, colour perception is the creative act of our uniquely configured visual system.

We will begin our exploration of colour perception with a discussion of colour's underlying physical basis - the electromagnetic energy. This will be followed by a discussion of spectral sensitivity, and then an explanation of colour mixing.

Light originates from a very small portion of the electromagnetic (EM) spectrum, a spectrum that also includes x-rays, gamma-rays, radio waves. Our visual system is able to convert electromagnetic wavelengths between 380 nm and 700 nm into light (1.0 nm = one-billionth of a metre); we are blind to all the rest of the EM spectrum. In short, without an eye, there is no light, only EM energy of different wavelengths.

As shown in the accompanying figure, wavelength refers to the distance between corresponding points of the wave cycle. Within the visible portion of the electromagnetic spectrum, wavelength is measured in billionths of a metre (nm). Alternatively, you can refer to a light's frequency (measured in Hz), which is simply the reciprocal of its wavelength.

The primate eye's overall daylight (i.e., photopic) spectral sensitivity function is based on the more limited sensitivities of short (S), medium (M), and long (L) wavelength cones (see figure). Although, as shown in the figure, each cone type is sensitive over a range of wavelengths if provided with sufficient electromagnetic energy, each has a peak sensitivity (i.e., requires the least amount of energy) at a specific wavelength. S cones are maximally sensitive to wavelengths of 419 nm, M cones to 531 nm, and L cones to 558 nm. Because there are relatively few S cones, however, overall photopic sensitivity is due largely to the summed responses of the M and L cones, accounting for the overall sensitivity peak at 555 nm.

Because photopigment molecules do not respond to these wavelengths, humans are relatively insensitive to ultraviolet and infrared light.

The Experience of Colour
Colour perception can be described in terms of three dimensions: hue, brightness, and saturation. These three dimensions can be represented in simple terms as a spindle-shaped "colour space."

The hue of a colour refers the colour name (e.g., red or blue), and is a function of the wavelength(s) reaching the eye.

Brightness refers to the intensity of a colour, and is roughly proportional to the amplitude of the incident wavelength(s).

Saturation refers to the "depth" or "purity" of a colour, and is related inversely to the number of different incident wavelengths. At the brightness extremes (i.e., black and white), the spindle comes to an infinitely small point, indicating a lack of apparent hue and thus complete desaturation.




Mixing Colours: Lights vs. Filters
The colour that we see is not the "colour" of a surface per se, but rather the mixture of wavelengths that are reflected back to the eye. For example, an orange appears orange because its skin absorbs very short and medium wavelengths, while reflects medium (yellow) and long (red) ones.

In short, colour perception is a function of both lights that are added up (i.e., additive mixing) and those that are absorbed (i.e., subtractive mixing).

Additive Colour Mixing
If two or more light sources of different hues are added by being directed to the same white surface, the colour experienced would be different from that elicited by either hue alone. The hue that results from the mixture of red, green, and blue additive primaries is shown in the accompanying diagram. For example, if red and green spotlights were shone on a white dress, it would look yellow. If red, green, and blue lights of equal intensity illuminated the same dress, however, it would still look white. In other words, its brightness would be greater but its "colour" would be unchanged.


Additive mixing of this type is the basis for colour on t.v. and computer screens. You can test this yourself by examining a yellow patch on your computer monitor with a powerful magnifying glass; you'll see that it's composed of yellow and green dots. Additive mixing by the placement of small colour patches next to one another is referred to as partitive mixing.

In addition to t.v. and monitors, partitive mixing can occur with fabrics as well as mosaics. Certain fabrics (e.g., tweeds) are woven with threads of different colour, but from a distance, the eye perceives a global additive colour. For example, some wedding dresses are woven with very desaturated primary colours (red, green, blue usually) to give a shimmering white appearance (much like a pearl). In clothing design this fabric is called "incandescent." 

Depicted in the image at left is another type of fabric, called "shot", which is composed of interwoven threads of two (or more) different colours. In the figure, the purple and yellow fibres do not actually appear as you see them in the image, but rather the eye combines these to produce a brown hue (a camera cannot accurately capture the appearance of shot). 

Partitive mixing can also be used as a technique in art. In this technique called pointillism, small dabs of paint are placed adjacent to one another to produce a brighter, more vibrant work. This form of art and was influential in France, Holland, and Italy in the 1890s and 1900s. The image below is a pointillist rendering of a summer scene at a beach. When viewed from a distance so that the dabs of paint cannot be discriminated, a somewhat different global perception of colour results. Even when the colour dabs can be seen, pointillist paintings tend to be more lively and luminous than those based completely on subtractive mixture.

Subtractive Colour Mixing
Subtractive colour mixing is based on the selective removal of wavelengths from light to produce a different colour. Common examples of this include paint, dyes, inks, and colour filters. For example, red paint is a paint that reflects red while absorbing all other wavelengths. The accompanying figure illustrates subtractive mixing.

As you can see, a green "filter" is one that transmits wavelengths in and around green; likewise, a blue "filter" transmits the shorter wavelengths associated with blue. The only wavelengths that would be transmitted by the two filters placed in sequence are those not absorbed by either filter. Thus, when green and blue filters are used together a dimmer, a dimmer colour intermediate between them results, in this case teal.

Paints work similarly. Colour particles are suspended in a clear medium (e.g. linseed oil). As light passes through that clear medium, it is transmitted selectively by the colour particles. The wavelengths that are reflected back to the eye (i.e., not absorbed) determine the colour seen.

Thus, mixing paints of different colour is completely analogous to placing several filters of different colour in a sequence.


Bases of Colour Vision created by 
Brian Thomas Wagner 
and Donald Kline
University of Calgary