Colour Theory

I'd like to have a colour theory that someone designing, say, a form for payroll input, can make use of.

Modern formal scientific colour theory has its roots in the experiments of Isaac Newton (1642-1727) who found that by shining sunlight through a glass prism he would get different colours projected onto his laboratory wall. The beam of sunlight gets split into red, green and blue coloured beams leaving the prism.

In the early 1800s Thomas Young discovered that he could combine the coloured lights emanating from the initial prism and so get an additive colour mix. Putting the red, green and blue beams back together gives a white light, and combining the red and green gives yellow – a most suprising result when you first come across it .

Later on that century, Hermann von Helmholtz began spinning coloured discs and looking at the outcome, from which he surmised that the human eye must contain three different receptors, one detecting red, one green, and one for blue. Ever since, this has been assumed to be so, yet that no-one has identified these discrete receptors in the eye and numerous experiments undertaken to seal the proof once and for all have done little more (to me at least) than make the theory seem ever more shaky.

In the 1950s, Edwin Land published the results of an experiment that was so simple, so much like a Victorian party trick, it could only be really important. He photographed a scene onto black and white film, through a red filter. Then the same scene onto black and white film through a green filter. Then he projected the two images superimposed onto a screen, the red filtered image through a green filter and the green filtered image through a red filter and lo! Full colour. Remove one of the filters (either) from the projector and – still full colour (though slightly washed out). No magic, you can do it yourself at home.

I have long thought, that whatever the truth or not of the red, green, blue receptor theory, I don't know what to do with this information. How do you apply it when something needs designing?

I find it much more useful to think of the eye as really have just one sort of colour receptor, which sees things most brightly at around the mid-range of the visible spectrum, in the green, and tails off towards the longer wavelength end, the red, and the shorter wavelength end, the blue.

With this as a basic principle, though it may not be scientifically proven, or at least not yet scientifically accepted, you can start to do some useful things, particularly around the area of legibility.

The practicalities of working seem to fit much better with Edwin Land’s experiments than with what was – still is by many – the accepted view of how we perceive colour, of which there are numerous descriptions available on the web, for example, Why are Things Coloured?, which is very thoroughly presented, but which, yet again, I have to wonder what practical use I’m to make of.

And what’s more, it seems from recent analysis of the morphology of the retina that there may be good scientific reason for looking more to Land than to Helmholz. See the work of Gerald Huth

Until I get round to writing more on this, see the References on Colour in Screen Design page and please have some fun with my Text colour over background using sliders page.

^top