Read our COVID-19 research and news.

Infants appear to break up the color spectrum into five categories—red, green, yellow, blue, and purple—suggesting these partitions are biologically hardwired.

Glasshouse Images/Alamy Stock Photo

Babies’ gazes suggest we are born understanding color

Unless you’re colorblind, you probably have a pretty good idea of what red, green, and blue are. Yet those labels are arbitrary divisions of the color spectrum; there’s no definitive point where the wavelengths of light we call orange turn into red. So cognitive scientists have long wondered whether we learn our labels from our culture or inherit them from our biology. Now, a study finds that infants see red, yellow, green, blue, and purple as different color categories, suggesting that at least some distinctions may be hardwired.

“I find it really compelling,” says Michael Webster, a psychologist who studies visual perception at the University of Nevada in Reno, who wasn’t involved in the study. “This isn’t going to immediately change anyone’s mind. But it’s another piece in the puzzle, and it’s a very nice piece.”

Scientists can’t just ask a newborn what it knows, so they use a trick known as “infant looking time” to figure out what’s in babies’ brains. The idea is that an infant’s gaze will linger on something unfamiliar for longer than something familiar, giving researchers a window into what babies expect—and what surprises them.

Applying this to color research, scientists led by Anna Franklin, a perception and cognition researcher at the University of Sussex in the United Kingdom, showed 179 infants aged 4-6 months 14 different swaths of color, each from a different part of the color wheel. Researchers showed one swath several times before displaying a hue from the next range over. If the infants looked at the new hue longer than the previous one, experimenters concluded that the babies considered it a different color.

The team found the infants reliably sorted the ranges of color into five categories: red, yellow, green, blue, and purple. Because these babies haven’t had much time to learn language—or even concepts—the researchers suspect these divisions of color are biologically innate, they report today in the Proceedings of the National Academy of Sciences.

A neural basis for the distinctions may lie in two neural pathways that carry light signals from the retina. One responds to wavelengths between cherry and teal and the other to the range between violet and chartreuse, roughly matching one of the color boundaries that the infants recognized, Franklin says. Why infants break up red and yellow into separate categories, whereas the brain's neural pathways do not, remains a mystery, however.

“Taken together, our findings suggest that how we divide up color isn't arbitrary, but that there are these universal, biologically derived fault lines," she says. Franklin says most cultures break up color in roughly the same way, according to data previously collected in the World Color Survey. "We're all working from the same template."

Webster says the study is “a really beautiful demonstration” of a broader idea in perception research: that the brain is especially sensitive to certain ranges along a continuum. Early studies showed that people, including infants, are more attuned to vertical and horizontal alignments of lines than to other angles. Being sensitive to categories of color and alignments may have helped our earliest ancestors survive, he says.

But cognitive scientists Rick Hanley and Debi Roberson at the University of Essex in the United Kingdom aren’t convinced that the findings demonstrate universal features of color perception. For one thing, some have criticized the World Color Survey for frequently misinterpreting communities’ responses, they wrote in an email to Science. What’s more, “The authors only tested the 6-month-old children of English-speaking parents and so the results cannot yet be generalized to infants in other cultures.”