Super recognizers are good at remembering faces, but how do they do it?
A new study by Australian researchers has found that people who never forget faces “look smarter, but they’re not stubborn.” In other words, they naturally focus on a person’s most distinctive facial features.
Is this good news for the rest of us who want to know how to avoid future gaffes caused by misremembering an acquaintance? Unfortunately, not very often.
“Their skills are not something you can learn like a trick,” explains lead author James Dunn, a psychology researcher at the University of New South Wales (UNSW) in Sydney. “This is a way to automatically and dynamically detect what makes each face unique.”
RELATED: Having trouble recognizing faces? Facial blindness may be more common than scientists assume
To see what super-recognizers see, Dunn and his colleagues used eye-tracking technology to reconstruct how people explored new faces.
The researchers did this with 37 super-recognizers and 68 people with normal facial recognition skills, recording where and for how long the participants looked at pictures of faces on a computer screen.
The researchers then fed the data into a machine learning algorithm trained to recognize faces. The algorithm, a type known as a deep neural network, was responsible for determining whether two faces belonged to the same person.
“AI is very adept at facial recognition,” Dunn explains. “Our goal was to use this to understand which human eye patterns are most beneficial.”
Clearly, our brains play a major role in processing visual information. However, when considering eye-tracking data from the super-recognizers, the algorithm became more accurate at matching faces than when inputting data from people with typical facial recognition abilities.
“These findings suggest that the perceptual basis of individual differences in face recognition ability may originate at an early stage of visual processing, at the level of retinal encoding,” Dunn et al. write in their paper.
The study builds on previous work by the same team, which found that super-recognizers turn faces into something like a jigsaw puzzle. The super-recognizer splits a new face into parts, and the brain processes the parts into a composite image.
This “jigsaw” approach challenged the assumption that remembering faces well requires looking at the center of the face to see the entire face.
This new study extends these findings and suggests that superrecognizers do more than just recognize more Although we have more information about faces than others, we focus on features that provide more “clues.”
“It’s like a caricature; the idea is that exaggerating facial features can actually make you more recognizable,” Dunn explains. “Superrecognizers seem to do it visually. They focus on the most diagnostic features of a person’s face.”
Although this research could help improve facial recognition systems, the researchers say that humans still have an advantage over AI at the moment when it comes to facial recognition because humans use other cues in social situations.
But don’t be so bold as to think humans are special. Although evidence suggests that there is a strong genetic basis for remembering faces exceptionally well, Face recognition processing Also plays an important role in primate social behaviorso, The biological roots of this skill may not be unique human.
This study Proceedings of the Royal Society B: Biological Sciences.
