Science

Tetrachromats and the colours most people cannot see

Graph of the three human cone types and how strongly each responds across the visible spectrum

Colouriest Potion Pack - 4 x 100ml - World's colouriest acrylic paints by Stuart Semple Stuart Semple

PHAZE - colour changing paint - Purple Haze to Pinkest Pink Stuart Semple

YOLO - newly discovered colouriest colour Stuart Semple

A small number of women carry genes for a fourth kind of colour receptor, and at least one has been shown to use it. What the research found, where the famous "100 million colours" figure comes from, and why it is shakier than the headlines suggest.

A fourth cone

Most people see colour with three kinds of cone cell in the retina. Each is tuned to a different band of wavelengths, short, medium and long, loosely called blue, green and red. The brain compares the signals from the three and builds every colour we experience out of that comparison. The genes for the red and green pigments sit on the X chromosome, and about 6% of men carry an altered version of one of them, which shifts the sensitivity of a cone and gives a mild colour anomaly.¹

Women have two X chromosomes, so a woman can carry a normal gene on one and an altered gene on the other. Around 12% of women do, and in principle that lets them make four kinds of cone instead of three. The Newcastle University researchers who study this point out that a son or father with a mild colour anomaly is a good clue that a woman may be a carrier.¹ Having the genes for a fourth cone is a long way from seeing with it, though. For a woman to be a functional tetrachromat, her brain has to treat the extra signal as a separate channel of colour information, and that turns out to be rare.²

Subject cDa29

Gabriele Jordan at Newcastle University spent years looking for a woman who could actually use a fourth cone. In a paper published in the Journal of Vision in 2010, she and her colleagues Samir Deeb, Jenny Bosten and John Mollon tested 24 women who were known carriers of deuteranomaly, a mild red-green anomaly. They used a Rayleigh test, in which a person judges mixtures of red and green light against a yellow, and a second task that measured how many dimensions each woman's colour space had. Most of the carriers showed no sign of four-dimensional colour vision. One woman, recorded as cDa29, behaved as a tetrachromat on every test, and her genetics suggested three distinct long-wave cone pigments alongside her short-wave cone.² Press coverage has described her as a doctor living in northern England.³

The Rayleigh test gets at what makes her unusual. Two lights or two paints can reflect quite different mixtures of wavelengths and still look identical to someone with three cones. A tetrachromat can sometimes tell such pairs apart.²

How many colours?

Articles about cDa29 usually say she can see around 100 million colours, against roughly 1 million for everyone else.³ The arithmetic behind those numbers is simple. If each type of cone lets a person tell apart about 100 steps, three cones give 100 times 100 times 100, which is 1 million. A fourth cone gives 100 million.⁴ Researchers who work on tetrachromacy are wary of this. The Human Tetrachromacy Research Collaborative, hosted at the University of California, Irvine, says there is "no scientific basis" for the 100 million figure, because it comes from a theory that assumes perfect neural processing.⁵ Jordan's study showed that cDa29 could separate colours that ordinary trichromats cannot.² How much larger her range is, and what those extra colours are like for her, nobody has measured.

An artist with the genes

The case artists are more likely to have heard about is Concetta Antico, a painter who grew up in Australia and runs a painting studio in San Diego. Genetic testing showed she has a normal gene on one X chromosome and a variant on the other, and she made no errors on the Farnsworth-Munsell 100 hue test, which asks people to arrange coloured caps in order of hue.⁶ Kimberly Jameson of UC Irvine and Alissa Winkler of the University of Nevada, Reno have studied her colour vision.⁷ Antico says she sees purples, blues and reflected colours in dark shadows. Her paintings put purple into Balboa Park and orange glimmers into the night sky.⁶

Jameson has been careful about what this means. She has said the difference is "not as dramatic" as the gap between colour-blind and normal vision, and that existing tests were not designed for people with more than three pigments.⁷ Scientists have not confirmed that Antico's fourth receptor works as a separate channel, and it is unclear whether her eye for colour comes from her genes, from a lifetime of painting, or from both.⁶

Seeing the same painting

Tetrachromacy makes an old idea about colour hard to ignore. A surface reflects some wavelengths and absorbs others, and colour is what a particular eye and brain make of the light that comes back. A functional tetrachromat standing in front of a canvas may split apart colours that the person beside her sees as one. She also has no easy way to show anyone else what she sees. Screens build their colours from three primaries, matched to three-cone vision.⁶ Antico's purple shadows are her attempt to put on canvas, for viewers with three cones, something only she can check. Every painter mixing a colour to match what they saw is doing a version of the same thing, with less certainty than they might like about whether anyone else sees it the same way.