Babies Recognize Faces Better Than Adults, Study Says

Hillary Mayell
for National Geographic News
May 22, 2005
Human babies start out with the ability to recognize a wide range of faces, even among races or species different from their own, according to a new study.

The researchers focused on face processing—the ability to recognize and categorize faces, determine identity and gender, and read emotions. Their findings suggest that, in humans, this skill is a case of "use it or lose it."

In the study six-month-old infants were able to recognize the faces of individuals of a different species—in this case, monkeys. Babies who received visual training retained the ability. But those with no training lost the skill by the time they were nine months old.

Led by Olivier Pascalis, a psychologist at England's University of Sheffield, the team reported their findings this week in the journal Proceedings of the National Academy of Sciences.

Training Babies for Face Recognition

To study early infant abilities, Pascalis and his team first tested a group of six-month old infants by showing them a series of pictures of the faces of Barbary macaques. Parents of some of the infants then regularly showed their children photographs of six Barbary macaque faces over a period of three months. The other babies were the control group and were not shown the monkey photos again.

The infants were retested at nine months of age. When babies in the control group were shown paired photos of mankey faces, the babies looked at each picture for the same amount of time. Babies who had been regularly shown monkey pictures by their parents spent more time looking at the pictures of new, unknown monkeys.

"A standard behavioral measure of familiarity with infants in visual processing is a measure of looking time," said Michael J. Wenger, a neuroscientist at Pennsylvania State University, who was not involved in teh study.

"One of the very regular things associated with visual recognition of any kind of object with infants is that they tend to prefer novelty. If you give them a choice between a novel and a familiar stimulus, they will typically always look more at the novel stimulus."

By looking at the novel monkeys longer, the visually trained babies demonstrated the ability to discriminate between the pictures of the monkeys they knew and pictures of monkeys they'd never seen before.

The study suggests that babies are born with a broad idea of what a face is. By the time they're nine months old, though, face processing is based on a much narrower model, one that is based on the faces they see most often.

This more specialized view in turn diminishes our early ability to make distinctions among other species, and possibly other races. For instance, if an infant is exposed to mainly Asian faces, he or she will grow to become less skilled at discerning among different, say, Caucasian faces.

The authors suggest that broad exposure to other races and species in infancy may prevent that loss of ability

Taking it one step further, Pascalis, teh lead study author, noted that even adults who work with primates have trouble recognizing individual animals by their faces.

"Primatologists do recognize monkeys [based] on body posture or spots on fur but can hardly recognize them [based] on faces," he said.

Nature or Nurture?

The question remains: Are our face-processing abilities inborn, learned, or some combination of the two?

"Basically, we have very little knowledge on how an adult brain is made," Pascalis said." We know that part of the [brain's]development is genetically determined and that the environment is going to influence part of its development too. However, we don't know for which cognitive function environmental inputs are crucial nor when they are important."

Previous studies on the role of experience in brain specialization have shown that early visual experience is important to development of face-processing skills.

Wenger, the Penn State neuroscientist, said that there is a hypothesis that, among infants, "the human visual system is specialized in a physical way for the processing of faces," Wenger said.

The alternative to the human brain being hardwired for processing faces "is that we use generalized mechanisms that apply to any visual object, and that over time our face-processing skills become very good, because we get a lot of practice," he continued. "It's an incredibly messy field [of study]. You can probably find almost exactly the same number of studies that argue both sides of the question."

Changing Views of the Brain

"All of these questions intersect with what we've learned in the last 10 to 15 years. Our view of what is modifiable in terms of brain structure has changed dramatically," Wenger said. "Ten years ago the accepted wisdom was that brain maturation was done by adolescence, and that is simply not the view any longer."

Scientists now know that adults can still modify synapses and change the functional characteristics of the cortex by learning a language, taking up a musical instrument, or undertaking other activities that are repeatedly performed.

Wenger's own research, which has been conducted in adults, leans toward a nonhardwired view. His findings suggest that we use generalized vision processes that become highly specialized with particular classes of stimuli.

According to Wenger, we see lots of faces, so we get lots of practice. Therefore, he said, we learn to discern among the types of faces we see the most—whether they be African, Caucasian, Asian, or even macaque.

Don't Miss a Discovery
Sign up our free newsletter. Every two weeks we'll send you our top news by e-mail (see sample).

© 1996-2008 National Geographic Society. All rights reserved.