Cognitive Daily

Can you tell if these two faces are the same or different?

How about these?

Or these?

If you’re like most adults, it will be easier for you to identify different adult faces compared to the infant faces in the second example, and even the small children’s faces in the third example.

This makes some sense, since most people don’t spend much time around a wide variety of infants. Even a new parent typically only sees one or two infants at a time. The infants you see out in public are usually wrapped up and often are difficult to see. Most of us see more children than babies, but unless you’re a preschool teacher or a parent, you still probably don’t spend as much time with kids as you do with adults.

It’s possible that this “other age effect” is similar to the “other race effect” we’ve discussed here before: If you live in a homogenous community, whether it’s all White people or all adults, it may be that you’re eventually going to become better at distinguishing between individual members of your community than you are to recognize others, whether they are Black people or babies.

Studies of the other race effect have found that it’s not automatic: if an Asian person lives for a long time among White people, she’ll be just as able to distinguish between White people as White people are. But studies about different-aged individuals have been inconclusive.

A team led by Dana Keufner showed 31 Italian university students a photo of an infant for 1 second at a time. The photo disappeared for a half-second, then was replaced by two photos, one of which was the baby they had previously seen. Their task was to identify the previously-seen face as quickly as possible. They repeated the task with upside-down babies, and finally with right-side-up and upside-down adult faces (the order was randomized). In all they saw 48 different faces in each category. Here are the results:

As you can see, they were significantly more accurate in identifying upright adult faces compared to infants. When the faces were upside-down, there was no difference in accuracy between adult and infant faces, or upright infant faces. There was a similar difference in reaction times (slower for upside-down and infant faces). This suggests that adults use a different mental process for identifying infant faces and adult faces. Inverting the faces equalizes the task. The researchers suggest that upright adult faces are being identified using a configural strategy, which has been well-documented in other studies. The inverted faces, as well as the infant faces, are recognized using a different, piecemeal strategy.

Next the researchers repeated the experiment with 3- to 4-year-old faces. Here are those results:

The results aren’t as dramatic, but they show a similar pattern. Adults are still significantly better at identifying adult faces compared to 3- and 4-year-olds.

So what about people who spend more time around children? Does this reduce the effect? Kuefner’s team repeated the experiment, using preschool teachers who spent at least 30 hours per week with small children. Here are the results:

This time there was no significant difference in accuracy between recognition of upright children’s faces and adult faces. So just spending more time around kids improves your ability to recognize kids’ faces. It’s not some ability that just fades away with age.

It would be interesting to see if children are similarly better-able to recognize kids’ faces — and perhaps worse at recognizing adults!

Dana Kuefner, Viola Macchi Cassia, Marta Picozzi, Emanuela Bricolo (2008). Do all kids look alike? Evidence for an other-age effect in adults. Journal of Experimental Psychology: Human Perception and Performance, 34 (4), 811-817 DOI: 10.1037/0096-1523.34.4.811

Comments

#1 gs
January 14, 2009

it actually looks to me like the preschool teachers aren’t too much better at the kids’ faces, but are a bit worse at the adult faces.
#2 Dave Munger
January 14, 2009

Look at the difference between upright and inverted faces — for preschool teachers, the difference is essentially the same either way, whereas for the university students, there is a much bigger difference for the adult faces — in this sense, their data is much closer to the infant data than the preschool teacher data.
#3 David
January 14, 2009

Bad charts! The effect is really exaggerated when the chart only goes from 80-100%. It’s only a 5% difference…
#4 Dave Munger
January 14, 2009

David,

You wouldn’t be able to see the effect if the scale wasn’t exaggerated. The effect is statistically significant, and the axes are clearly labeled. The difference between 90 and 95 percent accuracy is important: that’s twice as many recognition errors.
#5 kt
January 14, 2009

Why is white capitalized?
#6 Tony Jeremiah
January 15, 2009

This suggests that adults use a different mental process for identifying infant faces and adult faces. Inverting the faces equalizes the task. The researchers suggest that upright adult faces are being identified using a configural strategy, which has been well-documented in other studies. The inverted faces, as well as the infant faces, are recognized using a different, piecemeal strategy.

This seems to be supported by an earlier CogDaily article concerning different brain activation states that arise when adults look at infant vs. adult faces. So it would also be quite interesting to replicate this study with brain imaging data.
#7 Alexandra Cohen
January 15, 2009

From the 1st sight pics of babies and kids looked just the same to me while pics of adults seemed completelly different. Although from the second sight all of them differed from each other (Or is it the same person with different mimics?). But I think you used first opinion in your statisics right?
In fact I’ve never seen infants and rarely see kids in my life and belive that babies look all the same if you don’t examine their faces closely )
#8 Timothy (TRiG)
January 15, 2009

The faces all had different eyebrows and lips, so I’m fairly sure they’re all different. The lack of hair makes them slightly spooky!

I didn’t really notice any great difference in difficulty, and I don’t spend much time with kids these days.

TRiG.
#9 Samantha Vimes
January 15, 2009

There seem to be minor differences in eye shapes, too. And the ‘topography’ of the faces as suggested by shading. Look how much deeper one child’s eyes are set compared to the other.

But we aren’t looking at the same “test” as is described. We get to see the faces side by side, where we can compare, as opposed to one at a time and right side up and upside down.

I doubt I would get even the adults correctly when I can’t see them at the same time and have a limited period to view and decide.
#10 hillary james
January 15, 2009

To kt
The same reason Asian and Black are capitalized.
I notice that kt is not.
#11 bmn
January 15, 2009

Is there any evidence that babies and children may actually have more similar characteristics than adults which makes it harder to tell them apart? Don’t most babies have larger eyes and rounder faces to make them more attractive to their mothers so they do not get rejected?
#12 hillary james
January 15, 2009

Eyeballs appear large in babies because they are pretty much the size they are going be as an adult. They grow some, just much less proportionately than the rest of the body.