Some Women Have a Fourth Color Receptor. Almost None of Them See More Color.

Concetta Antico looks at a shadow on the ground and sees a mosaic. "Around the edge I'll see orange or red or purple in the shadow," she has said. "You might see dark green but I'll see violet, turquoise, blue." Antico is a painter, and in 2012 researchers confirmed what her canvases had been arguing for years: she has four types of color receptors in her eyes instead of the usual three. Depending on who you ask, that means she can see somewhere around 100 million colors. Most people top out around 1 million.

That number gets repeated a lot, and it's the least interesting thing about tetrachromacy. The stranger fact is this: a real, working fourth color channel is vanishingly rare, even though the genetic mutation that could produce one shows up in a meaningful slice of the population. Millions of women are walking around with four kinds of cone cells in their retinas. Almost none of them see any more color than you do. The gap between having the hardware and using it is where the actual story lives, and it says something useful about how color vision, and color design, actually work.

The one-million-color baseline

Standard human color vision runs on three types of cone cells, each tuned to a different band of the visible spectrum: roughly short (blue), medium (green), and long (red) wavelengths. Your brain doesn't see wavelengths directly. It compares the output of those three receptor types against each other, and that comparison is what you experience as hue. Researchers estimate that this three-way comparison lets a typical person distinguish somewhere around a million distinct colors, though the exact figure depends heavily on how you test it.

Three cones, three inputs, one shared coordinate system. That system is why an RGB screen can fool your eyes into seeing millions of colors while only ever lighting up three types of pixels. It's also why the whole apparatus of modern color work, hex codes, Pantone chips, sRGB gamuts, is built around three channels. Nobody designed it that way on purpose. It's just what a trichromatic eye needs to be convinced.

Why a fourth cone shows up at all

The genes for the red and green cone pigments, OPN1LW and OPN1MW, sit on the X chromosome. Men have one X chromosome, so they get whatever opsin variant that chromosome carries. Women have two, and thanks to a process called X-inactivation, each individual cone cell in a woman's retina randomly switches off one of her two X chromosomes and uses only the opsin gene on the other.

If a woman's two X chromosomes carry different variants of the red or green opsin gene, the math changes. Instead of three cone types, her retina ends up as a mosaic of four: the usual blue cone, plus two now-distinct versions of red or green, plus whichever standard cone she already had. This is exactly the situation facing an obligate carrier of red-green color blindness: a woman known through her family tree, most reliably by having a colorblind son, to carry the gene without showing the condition herself. That family-tree logic is why studies of human tetrachromacy have concentrated on mothers of colorblind sons. Maternal carrier status is a straightforward way to find these women in the first place, since the same X-linked inheritance that gives a son color blindness can give his mother a fourth cone type.

Estimates of how common this genetic setup is vary by study and by how strictly "four photopigments" is defined, but the carrier population isn't small. It's a meaningful fraction of women. The interesting question was never whether the extra cone exists. It's whether having it changes what you see.

Having four cones isn't the same as seeing in four dimensions

For a long time, the working assumption was that it didn't matter. Standard color vision tests kept finding that women with four cone types matched colors the same way trichromats did, which suggested the fourth signal wasn't doing anything the brain could use.

That assumption started to crack in 2001, when researchers Kimberly Jameson, Susan Highnote, and Linda Wasserman published a study in Psychonomic Bulletin & Review comparing color descriptions given by women with four-photopigment genotypes against trichromat controls. The four-photopigment group reported noticeably richer, more varied color experiences. The study's other conclusion was just as important: the standard tests used to screen for color vision were built around trichromatic assumptions, so they weren't equipped to detect a fourth dimension even if one was there.

That left an open question. Genetically, four-cone women are not rare. Functionally, was anyone actually using the extra channel?

The test that found subject cDa29

In 2010, Newcastle University neuroscientist Gabriele Jordan, working with Samir Deeb, Jenny Bosten, and John Mollon, published the answer in the Journal of Vision. Their study, "The dimensionality of color vision in carriers of anomalous trichromacy," took 24 obligate carriers of deuteranomaly, meaning women confirmed to carry a red-green color vision variant, and ran them through tests designed specifically to catch a working fourth channel rather than just infer one from genetics.

The tests combined classical color-matching (a Rayleigh match, the same basic technique used to diagnose color blindness), a discrimination task using precisely controlled wavelengths, a multidimensional scaling test meant to reveal an extra dimension in color space if one existed, and molecular genetic analysis to pin down the exact peak sensitivity of each woman's cone pigments.

Twenty-three of the 24 carriers performed like trichromats. One did not. Participant cDa29 had three distinct, well-separated cone pigments in the long-wavelength region of the spectrum, on top of her ordinary short-wavelength blue cone, four functioning types in total. And she could do something the other 23 women couldn't: reliably tell apart an orange light around 590 nanometers from specific mixtures of red light around 670 nanometers and green light around 546 nanometers, combinations that look identical to a trichromat no matter how you tune the ratio. After two decades of research into the possibility, cDa29 was the first person confirmed to pass every test for functional tetrachromacy.

What it's like on the inside

Which brings the story back to Concetta Antico, a working artist rather than a lab subject with a code name. Cognitive scientist Kimberly Jameson, the same researcher behind the 2001 study, spent roughly a year studying Antico alongside Alissa Winkler of the University of Nevada, Reno, after Antico was identified as a tetrachromat in 2012. Jameson determined that Antico's fourth cone peaks in what she describes as a reddish-orangey-yellow range of wavelengths, sitting between the standard red and green cones most people have.

Antico paints what she sees rather than what other people expect to see, and her canvases are full of the mosaic she describes in shadows and along edges. "It's shocking to me how little color people are seeing," she has said. Her case isn't purely a story about genetics, either. She has painted since childhood, starting around age seven, and researchers who have studied her think that years of deliberately looking closely at color likely helped her visual system learn to exploit a channel that, in another four-cone woman who never trained her attention on hue, might have stayed functionally silent. The hardware may be necessary. It doesn't appear to be sufficient on its own.

More receptors doesn't automatically mean more color

It's tempting to assume that more color receptors always means richer color vision, but the animal kingdom argues otherwise. Mantis shrimp are the famous extreme case. Depending on the species, they carry as many as 16 types of photoreceptors, several times what even a functional human tetrachromat has. When Hanne Thoen and colleagues at the University of Queensland tested mantis shrimp color discrimination in a 2014 Science paper, the animals performed worse than ordinary trichromatic humans at telling closely related hues apart. The leading explanation is that mantis shrimp aren't comparing wavelengths the way our brains do. They may be scanning an object across their many receptor types to identify its color almost instantly, more like a satellite sensor sorting light into preset bands than a human eye running a fine-grained comparison. More channels, but a completely different strategy for using them.

Birds offer a cleaner contrast. Many species, including the plum-headed finch, are genuine functional tetrachromats, with four cone types tuned across the visible spectrum and into the ultraviolet. Unlike the accidental, still-being-mapped case of human tetrachromacy, bird color vision had tens of millions of years to co-evolve hardware and wiring together, which is presumably why it works so reliably. Human tetrachromacy, by contrast, looks like a genetic possibility that mostly arrived faster than the neural infrastructure needed to use it.

Your screen was never built for a fourth channel

Even someone with a fully functional fourth cone doesn't get to use it when she's looking at a screen. Every monitor, phone, and printer is calibrated to fool three cone types with three primaries. A tetrachromat viewing an sRGB image is still only receiving red, green, and blue light mixed to trick a trichromatic eye, because that's the only kind of eye the format was ever designed to satisfy. Whatever her fourth cone is doing when she looks at a real rose or a real shadow, a photo of that rose on a phone screen doesn't give it anything to respond to.

That's a useful thing to sit with even if you'll never meet a tetrachromat. Every color system you work in encodes an assumption about the eye looking at it. Usually that assumption holds well enough not to notice. But it is an assumption, tuned to a specific, common, and still not universal version of human color vision. The eye you're designing for is doing a lot more interpretive work than the hex code suggests, and for a small number of people, the code is leaving real, functioning color receptors with nothing to do.