“Did she have a parrot? Cases are much easier to solve when there’s a parrot.”Cartoon by Sophie Lucido Johnson and Sammi SkolmoskiCopy link to cartoonCopy link to cartoonLink copiedShopShopRoorda was the first guinea pig in the Oz research project. In a dark lab, he sat in a chair, eyes dilated, with his head position stabilized by a bite bar. His wife and his daughter had sometimes teased him, saying that, though he had spent years studying color vision, he still didn’t notice much of a difference between purple and magenta. “I think they just grew up paying a little more attention than I did,” he said. So perhaps he would not be a great test subject. Or maybe there would be a technical failure. Or the experiment would succeed in exclusively stimulating M cones, but the brain would not know what to do with input that nothing in evolution had primed it to interpret.Hannah Doyle, a graduate student who had been key in developing the experimental protocol and the software, was operating what they called the Oz Vision system, monitoring the optical correction and making sure Roorda maintained alignment. “We were using 543-nanometre light, green light,” Roorda explained—a wavelength that both L and M cone cells respond to strongly. “While we were getting set up, everything just kind of looked green.” Technical challenges dictated that only a thousand M cone cells would be stimulated, enough to see a patch of color about the size of a thumbnail viewed at arm’s length. As soon as the experiment began, the regular green that Roorda was seeing “immediately changed to this deep, more saturated color. Like a deep teal.” The experiment had worked. When Roorda pitted that green “against the most saturated natural color you could ever see, which would be a monochromatic wavelength”—in this case, one of around 510 nm., the visible color that best matched the experiment’s “greenest green”—“the monochromatic wavelength just looked really pale by comparison,” he said. Roorda and Ng’s team had their novel color. But it lacked a name. James Fong, another graduate student deeply involved in the project, came up with “olo,” after the 0-1-0 coding shorthand for the cone types that the laser targeted: L cones (off), M cones (on), S cones (off).No printer can print olo, and no camera can photograph it, and there can never be an olo-colored pigment or dye. And yet the study participants—there were four additional researchers, including Ng—did experience a color that you and I have not. This seeming contradiction emerges from the haziness of what we mean by “color.” As Isaac Newton noted more than three hundred years ago in “Opticks: or, a Treatise of the Reflections, Refractions, Inflections and Colours of Light,” rays of light “to speak properly are not coloured. In them there is nothing else than a certain power and disposition to stir up a sensation of this or that Colour.” Kreimer showed me an image of a simple red bird with a pinpoint black eye. After I stared into the eye for a minute, I looked away, and a turquoise bird came into view. “So that also tells us about how our vision works,” Kreimer said. The retina’s L and M cone cells—the more red-tuned ones—are thought to tire out, in a sense, while staring at the red bird. When one finally looks away, some of these cones remain inhibited, tilting the color balance toward the S cone cells. The turquoise bird is your mind trying to interpret what its cone cells are telling it. Olo is something like that turquoise bird.In a 1794 paper, “Extraordinary Facts relating to the Vision of Colours,” the English schoolteacher (and meteorologist and chemist and founder of atomic theory) John Dalton wrote that he had long felt that “several colours were injudiciously named.” Pink, for example, was sometimes called red. “I thought it highly improper; it should have been blue,” he noted. Pink and blue looked “very nearly allied” to him, and pink and red appeared to have “scarcely any relation.” The deep green of a laurel leaf, on the other hand, appeared to Dalton to be a close match to a stick of red sealing wax, and the paler side of the leaf resembled “the lighter red of wafers.”Dalton undertook an investigation into this conundrum. He compared how various items looked to the people around him: grass, woollen cloth, paper “heated almost to ignition.”His brother saw the world much as he did, and so did a small number of his students, all of whom were male. Dalton speculated that a blue tint to the vitreous humor of his eye might explain the peculiarities of how he visualized things. When he died, he wanted his doctor to examine his eyes carefully. The doctor dutifully did so. The vitreous humor of Dalton’s eyeballs was clear. The Manchester Literary and Philosophical Society then preserved Dalton’s eyes and, in 1995, scientists genetically analyzed a sample to confirm what they already suspected from Dalton’s written accounts: he lacked the gene for M cone cells. Color blindness for a time was termed Daltonism.Some seventy years after Dalton’s death, Shinobu Ishihara, a Japanese ophthalmologist, was asked to design a test to screen military recruits for color blindness—soldiers must be able to quickly and easily distinguish the contour lines of topographical maps, the reds and greens of signal lights, the colors of flags. Working with a color-blind assistant, he created what are known as Ishihara plates: circles containing variably sized dots in different hues, the contrast of which makes a number or a letter visible to some and invisible to others. Ishihara plates remain a common test for color blindness and have also become a standard fixture of an art-school education.Steve Antony, a color-blind artist and children’s-book author, recalled that in elementary school he’d had trouble with anything color coded: “Pie charts, flags, geography maps, sports-team colors—I knew I didn’t see colors in the same way that most others do.” Now, as an adult, he finds the red/green band on the locks of public bathrooms unhelpful, and he hates having to turn the handle to check whether anyone is inside. “But color blindness never really hindered my creativity,” he said. In fact, it has given him distinct perceptual powers. He’s especially sensitive to value, to contrast, and to patterning, maybe because he has had to follow such cues more closely than a person with typical color vision would. “Pop art was a real discovery for me as a kid,” he said. “Then, later, the geometric shapes and bright contrasts of suprematism.” Kazimir Malevich’s “Eight Red Rectangles” was the inspiration for Antony’s children’s book “Green Lizards vs. Red Rectangles”—itself an echo of his red-green color blindness. “I always talk about color blindness when I visit classrooms, because people don’t know much about it, even though, in most classrooms, at least one of the boys has it,” he said. “There’s this misperception that what a color-blind person sees is gray—and that’s not it.”More than ninety per cent of people with color blindness are male. But not all of them have only two cone types, like John Dalton; the majority have three, but one is atypical in what wavelength of light it responds to best. An atypical M cone cell might be most sensitive to a wavelength of 555 nm., for example, instead of the more standard 530 nm. Color blindness is primarily an X-linked trait—a trait for which women have two copies of a gene and men one—so a mother who has one copy of a gene for a typical M cone cell and another copy for an atypical M cone cell might pass on the atypical gene to her son, who would then see colors differently than most of his peers. But what kind of vision would the mother have? In 1948, the Dutch scientist Hessel de Vries published a study of color-blind men in which he included a brief speculation: there were likely women directly related to them who had all three normal cone-cell types plus the atypical one that ran in their family. Would such tetrachromats, as de Vries called them, have an augmented color palette? It wasn’t clear. De Vries was in love with Anneke Hogeveen, his lab assistant, and, in 1959, before he could settle the question, he murdered her and killed himself. Inquiry into tetrachromacy remained dormant for many years.In the nineteen-eighties, Gabriele Jordan, a Ph.D. student in experimental psychology at Cambridge University, was looking into natural variation in normal vision. The majority of people have L cone cells most sensitive to wavelengths of light of about 560 nm., M cone cells sensitive to those of about 530 nm., and S cells sensitive to those of about 430 nm. The “about” was what Jordan was studying. Using an instrument called an anomaloscope—with which a user mixes red and green light to match a standard yellow—Jordan discovered that, even among people with typical trichromatic vision, one person’s red is not necessarily the same as another person’s. Your S cone cells might be most sensitive to 429-nm. light and mine to 431-nm. Our reds would be siblings but not twins.During her research, Jordan worked under John Mollon, who had studied the color vision of squirrel monkeys. Like most New World monkeys, squirrel monkeys typically have two cone-cell types, unlike Old World monkeys, who have three. But Mollon and his colleagues demonstrated, through color-matching experiments and then through genetics, that some female squirrel monkeys were trichromatic. He hypothesized that, in the wild, trichromats could lead the foraging efforts for food—perhaps because they would have a more nuanced read on the ripeness of fruit. These studies recalled de Vries’s speculation that some human females might be tetrachromats. “It was literally, Well, if it works for monkeys, it might work for humans,” Jordan told me.Tetrachromatic women would be difficult to identify, though. A tetrachromatic person would be even more unlikely than a color-blind one to know that her visual world was anything other than ordinary—it would be different in a way that she had never heard about. But Jordan began testing boys for color blindness at a local school, using both Ishihara plates and the anomaloscope, which detects more subtle color-vision deficiencies. When Jordan found a color-blind student, she would reach out to the boy’s mother to ask if she was open to having her color perception tested. One mother, subject cDa29, when asked to use the anomaloscope, was never satisfied that she had made a perfect match to a standard yellow. But when she was asked to discriminate two identical colors from a very similar color—imagine two cobalts and a lapis—she could do so effortlessly, with exceptional accuracy and speed. Subsequent genetic testing showed that cDa29 had genes for four cone-cell types.“It was a once-in-a-lifetime feeling of true discovery,” Jordan said. Ultimately, she found more functional tetrachromats among the mothers of the schoolboys. “I wanted to know, of course, what they had to say about their perceptual worlds,” she told me. None of the women had a sense of themselves as having supernormal vision, although cDa29 was a prominent hematologist, working in a profession that requires detecting subtle anomalies in slides of blood smears. At least two others worked in medicine. One was a midwife, who recalled an experience from her training. Meconium, an infant’s first feces, usually passes after birth, but sometimes it passes while a baby is still in utero, and that can cause respiratory distress. The tetrachromatic midwife detected meconium in amniotic fluid, “whereas the qualified midwives training her couldn’t see anything at all,” Jordan told me.Several of the women said that finding a tiny object on a colorful background—like an earring on a patterned carpet—was a particularly easy task for them, because the object would stick out the way a red apple on a tree might to most people. “Quite a few of them talked about color memory,” Jordan said. They didn’t need to bring a sweater from home when going to buy yarn to match it, or a paint swatch to match a wall.Roorda and Ng are following up their olo work by using the Oz platform to simulate trichromacy in a dichromatic color-blind subject and tetrachromacy in a trichromatic subject. (Roorda, who is now a faculty member at the University of Waterloo, in Canada, retains an affiliation with Berkeley.) The success of olo suggests a certain amount of neuroplasticity in color vision, in that the mind made sense of unprecedented visual information. Roorda believes, however, that further experiments would be necessary to know that with confidence. “This is sort of why we think this is interesting, scientifically, and why we continue to pursue these experiments,” he said. “How do percepts of color develop?” He talked about how newborns cannot perceive shapes or colors well, and how they have trouble recognizing faces. “The optics are fine when they’re born, but the brain is not ready to make sense of the input,” he said. “My belief is that, although in the early olo experiments I was comfortable describing olo as a teal-like color, if I was exposed to it enough I would probably start saying that the word ‘teal’ is not doing it for me anymore.”In February, 2015, an amateur photograph of a dress went viral on the internet because people disagreed, strongly, about its color. Some said that it was white with gold, and others that it was blue with black. Like millions of people, I looked at the photo. The dress was white and gold. This was so obvious that I assumed people were pretending it looked blue and black as a psychological prank. I asked my partner and our daughter what color the dress was. They agreed with me that the controversy was nonsense. Without a doubt, the dress was blue and black.“Please take some leftovers so we can pawn off our worst tote bags on you.”Cartoon by Jeremy NguyenCopy link to cartoonCopy link to cartoonLink copiedShopShopBevil Conway, a neuroscientist who studies color vision, was grading student lab reports when he received an e-mail from Adam Rogers, a science reporter at Wired at the time, asking for his thoughts about the dress controversy. Conway gave the photograph a look. “To me, it appeared periwinkle blue and bronze,” he said.For a few years, Conway had been playing with the hypothesis that, he said, “you have in your brain a ‘prior’ about lighting conditions.” In other words, your brain makes assumptions about whether what you are looking at is lit by warm or cool light. This makes some intuitive sense: your mind wants to understand a red wheelbarrow as red regardless of whether it is viewed in dim light or bright light, at sunset or at noon. So your brain must have some way of “solving” for color constancy. “A whole bunch of pennies dropped,” Conway said. He thought that the photo of the dress “might be the little piece of proof I had been waiting for.”He suggested that Rogers ask Wired’s photo team to run the dress image through its color corrector, first adjusting for warm lighting, then for cool. In effect, this meant telling the photo software whether it should assume there was indoor lighting (warm) or natural sunlight (cool). The dress in the photograph is closely cropped and a bit blurred. It is not clear whether it is indoors or outdoors. “Your brain has to make a choice,” Conway said. “If it thinks it’s warm lighting, it subtracts orange, which makes the dress look blue and black. If it thinks it’s cool lighting, it subtracts blue, which makes the dress look white and gold.” The dress was, in fact, blue and black. It was photographed indoors.I asked Conway, who now heads the Section on Perception, Cognition, and Action, at the National Eye Institute, for his thoughts about olo. He was impressed by the Roorda lab’s ability to stimulate select cones, but he hesitated to call olo a new color; it is a supernaturally saturated version of an extant hue. “I would say you might get olo by staring at a saturated red spot and then looking at a green screen of a very saturated bluish-green,” he told me. This would place the afterimage of the red spot over an already maximally green screen, potentially leading to an experience of a supersaturated green. “Color isn’t simply what the eye tells the brain,” Conway said. “You can’t ask if piano music is a key, a hammer, a wire, or the vibrating air. Because it’s all those things.”A painter can distinguish cadmium red and cobalt violet solely by the weight of the paint tube, the artist Amy Sillman observed in the opening of her talk “Color as Material,” given at the Whitney Museum of American Art, in 2014. There is a Greek word, pharmakon, which can mean color, but which also historically meant poison, medicine, charm, intoxicant, and paint. (“Every color class across the country probably shows that lecture—it’s great,” Kreimer told me, of “Color as Material.”) I visited Sillman recently in the Bushwick neighborhood of Brooklyn at her studio, the walls of which were covered in paintings about as tall as a person. Sillman typically applies, scrapes, obscures, re-scrapes, and reapplies paint on her canvases so many times that often the early form of the painting is to the finished work as an ancestor is to a younger incarnation. She is a celebrated colorist, one who came up through the art world at a time when painting had fallen out of favor and color was dismissed, by some, as bourgeois and conventional.Sillman explained that her Whitney talk was informed by a reading group on color that had met for years. The group read Aristotle (who wrote that color is an intrinsic property of objects), Isaac Newton (that white light comprises all the colors of the rainbow), Johann Wolfgang von Goethe (that all the colors combined make black, and that colors have emotional qualities), and Albert Munsell (who, finding the names of colors chaotic and silly, ordered them with numbers and letters corresponding to hue, value, and saturation). They read Frank O’Hara (“Why I Am Not a Painter”), Robert Hass (“The Problem of Describing Color”), Esther Leslie (“Synthetic Worlds: Nature, Art and the Chemical Industry”), and Herman Melville (the “Whiteness of the Whale” chapter from “Moby-Dick”). They read art historians and anthropologists and artists. But the group never agreed on how best to describe what color is. “I realized eventually that there are earth people and there are sky people and that never the twain shall meet,” Sillman said. Earth people thought of color as minerals, mud, pigments—as touchable things. Sky people were focussed on light.Sillman said that she was more earth than sky, but her main conclusion was that “only Josef Albers was right,” referring to the Bauhaus artist and pedagogue, who said that understanding color should follow from working with it, not theorizing about it. “The person you should really talk to is my former student Roger Carmona, who runs New York Pigment Co.,” she said. “He knows more about color than anyone.”I visited Carmona’s shop, inconspicuously situated on a scrubby commercial block in Chelsea; it looked like an enchanted apothecary. Wooden shelves held bags and bottles of colored powder. Samples of minerals (azurite, malachite, selenite) rested near the cash register. Painted rectangles of color hung on the walls, displaying the characteristics of pigments in watercolor paints versus oil versus acrylic. For nineteen years, Carmona had worked for the New York branch of Kremer Pigments, which was founded in the seventies by a chemist from Germany named Georg Kremer. Carmona started out fulfilling orders, then eventually taught make-your-own-paint workshops—it had been an incidental apprenticeship. When Kremer closed up the New York store last year, Carmona, in close consultation with him, opened his own shop, which carries Kremer’s pigments. “Georg used to literally knock on people’s doors and ask them if he could dig in their back yards” for new pigments, Carmona said. More recently, Georg’s son David, who shares his father’s dedication, spent two weeks exploring the distinctive rocks of Iceland, where he unearthed three natural-earth pigments, one of which—Snæfellsjökull red, found near the Snæfellsjökull volcano—Carmona now held in his hand.Carmona began mixing his own paints in college. “Every pigment has a personality,” he explained, placing a small vial of dark powder in my palm. “For example, cobalts tend to be bullies”—they obscure what they are painted over, whereas other pigments have more transparency, which he described as shyness. “Cote d’Azur violet has a silky or velvety personality,” Carmona said. Studio Pigment violet, a lab-produced color made with dioxazine violet and calcium carbonate, “looks synthetic,” he added, indicating a bag that appeared as if it had been filled with crushed candy. “Synthetic pigments refract light differently.” What a pigment is suspended in alters the color further. Egg-tempera paint uses egg yolk, casein uses milk proteins, and oil paint uses linseed oil. “The way the particles are suspended in the oil allows more light to pass through the oil film,” Carmona said, explaining the distinctive shine of oil paints. “That gives depth, which is what the Renaissance painters realized.” Before the Renaissance, egg tempera was dominant. “It has a mother-of-pearl quality.”Carmona seemed only mildly interested in the olo experiment when we spoke, though he later read up on colors invisible to humans, and a recent exhibition at the shop included work proposing a chart of “unseen colors.” The day I visited, however, he introduced me to a new color of a different kind—a new pigment—handing me a black-capped clear bottle that held a blue powder. “YInMn blue was discovered at Oregon State University in 2009,” he said. It is the first new inorganic blue pigment in two centuries, made of the elements yttrium, indium, manganese, and oxygen. Andrew Smith, a graduate student working in the lab of the materials-science chemist Mas Subramanian, was heating manganese oxide to two thousand degrees, as research into the compound’s electrical properties, when an unexpected, luminous, almost phosphorescent blue came out of the furnace. Crayola has since replaced one of its crayons with a YInMn hue, named Bluetiful. “You can see how intensely saturated it is,” Carmona observed. It looked to me like an olo of blue. “The refractive index is similar to that of lapis lazuli,” he said, referring to the storied pigment that was once found almost exclusively in the high mountains of northeast Afghanistan. YInMn has “the personality of lapis,” he went on. “But not the history.” ?