ionpsych.com

It’s a good thing that the movie Inglourious Basterds has subtitles, because everyone except the Americans are always switching languages. Now imagine that you see a silent clip from the movie — do you think you’d be able to tell when someone starts speaking a different language? If you’ve ever tried figuring out what someone is saying just from looking at their mouth, you’ll know it’s not easy. What if you showed Inglourious Bastards to a baby — would it be able to detect a language switch if the sound were turned off? Since babies often seem more like blobs than humans, you might think not. Amazingly, we’ve recently discovered that 8 month-old infants CAN tell the difference — but only if they’re bilingual.

How can babies, who can’t even understand what their parents are saying, tell when a silent face switches languages? And how can you test a baby to figure this out? Whitney Weikum and her colleagues had babies sit on a parent’s lap and watch silent videos clips of 3 different women reading children’s stories. Babies saw lots of similar video clips all in the same language (English) until they stopped looking at them as much. Just like when you look away from something when you’re bored with it, so will babies.

Suddenly, the language in the videos switched: although the same 3 women were reading stories in silent videos, now they were reading them in a different language. Now French-English bilingual babies returned to looking at the videos. They noticed that the language changed, and became interested in the videos again. Since the videos were silent, the only clues they could have used were the way the women’s faces and mouths moved. So babies who can’t even talk were using these very subtle differences to tell languages apart.

Not all babies were equally good at this test, though: monolingual English-speaking 8 month-olds couldn’t tell the difference when the languages switched. But when monolingual babies were younger (4-6 months old), they could. Why is this? It seems like all babies start out with the ability to tell languages apart just from looking at people’s faces when they’re speaking. But since this skill is only really helpful for babies exposed to more than one language on a regular basis, bilingual babies use it and monolinguals loose it.

Rather than being confused by constantly switching languages at home, bilingual babies are actually quite good at telling them apart — even just from tiny differences in people’s facial expressions. When it comes to language learning, babies are very intelligent blobs.

Weikum, W., Vouloumanos, A., Navarra, J., Soto-Faraco, S., Sebastian-Galles, N., & Werker, J. (2007). Visual Language Discrimination in Infancy Science, 316 (5828), 1159-1159 DOI: 10.1126/science.1137686

What do you see in this picture?

Now ask yourself: What do you know about this creature?

You probably came up with at least a couple facts: it’s a kind of reptile, it’s huge, it went extinct millions of years ago. And you’re right, because this is the newly discovered dinosaur “Brontomerus” (apparently, it had frighteningly powerful thighs). Unless you follow this kind of archaeology news, you probably hadn’t ever seen this creature before — and yet you were able to figure out some of its properties. Because you recognized the picture as belonging to the group or “category” of dinosaurs, you could use your knowledge of dinosaurs in general to predict things about the one in the picture. Using categories like this, to make intelligent guesses about something new, is crucial to human survival. (If you see a huge cat that kind of looks like a lion, you know not to get too close to it). How do we form these kinds of categories in the first place?

We seem to have this evolutionarily adaptive grouping skill very early in life. If you show 4 month-old babies lots of cat pictures, and then show them a picture of a dog, they will look at the dog for a long time — longer than they would look at a new cat picture (Quinn, Eimas & Rosenkrantz, 1993). Babies pick up on the similarity between all the cats they’ve seen, so when they suddenly see a dog, it’s is more different and interesting to look at than another cat.

A fascinating twist to this story is that babies are much better at noticing these kinds of categories when they hear language. A recent experiment by Ferry, Hespos and Waxman showed 4 month-olds pictures of dinosaurs. Half of the babies heard someone call the dinosaur by a made-up name (“Look at the toma. Do you see the toma?”). After seeing just 8 dinosaurs, these babies looked longer at a picture of a fish than at a picture of a new dinosaur. But if babies heard beeping instead of human language while seeing the dinosaur pictures, they didn’t look longer at the fish, as if they didn’t notice the similarity between all the dinosaurs they had just seen. Only the babies that heard language were able to pick up on the dinosaur category.

Why does this happen? When a person uses a new word like “toma” while showing a bunch of pictures, this suggests that they’re referring to a new category in those pictures. But if you’ve spent any time around a 4 month-old, you should be scratching your head right now. Babies this young don’t seem to understand language at all. They don’t even realize that “Look” and “toma” are separate words, much less realize that “toma” is a new word and could be referring to the category “dinosaur”. In fact, it seems like hearing anything that sounds like language will get very small babies to notice categories better — it doesn’t even have to be human language. Recent experiments similar to the one with dinosaurs find that babies who hear monkey calls are also better at picking up on categories. It’s as though any human-like noise sends a signal to very little babies to “pay attention!” — and specifically, to pay attention in a way that makes them notice how things can be grouped together in meaningful ways.

Why and how this happens in such young children is still a big mystery in psychology. But at least now we know that it doesn’t really matter what we say to our babies. If you feel silly enough to hoot like a monkey, no worries — your baby will still be learning something.

Ferry, A., Hespos, S., & Waxman, S. (2010, November). Tuning the Link Between Words and Categories: Primate Vocalizations Facilitate Object Categorization in 3-Month-Old, But Not 12-Month-Old Infants. Poster presented at Boston University Child Language Development Conference, Boston, MA.

Ferry, A., Hespos, S., & Waxman, S. (2010). Categorization in 3- and 4-Month-Old Infants: An Advantage of Words Over Tones Child Development, 81 (2), 472-479 DOI: 10.1111/j.1467-8624.2009.01408.x

Quinn, P., Eimas, P., & Rosenkrantz, S. (1993). Evidence for representations of perceptually similar natural categories by 3-month-old and 4-month-old infants Perception, 22 (4), 463-475 DOI: 10.1068/p220463