A growing number of infant studies show that well before very young babies begin forming words, simply hearing language boosts their thinking. We’ve demonstrated at our own research center that speech helps infants as young as 3 months old learn about the world: babies who are shown a group of objects while listening to speech successfully categorize those objects, recognizing what makes them form a cohesive group. Infants who hear other interesting sounds while looking at the exact same set of objects, however, fail to learn which ones go together. These findings constitute powerful evidence that language can support infants’ thinking from the first months of life. They also raise a pressing question: What’s so special about language?
As you might imagine, this is a difficult question to test! We can compare how infants perform at learning tasks when hearing language versus other interesting sounds, but we don’t know why the language improves their success; could it be something as simple as greater familiarity or heightened interest? One of our researchers, Brock Ferguson, speculated that in fact it’s the communicativeness of language that facilitates learning, and we came up with an interesting way to test this: instead of pitting speech against non-speech sounds, we could pit identical non-speech sounds against each other!
The key, we decided, was that in one scenario, we would convince infants that this non-speech sound was communicative; in the other scenario, we would leave its function ambiguous. Next, after this pre-exposure to the sound as either communicative or non-communicative, we could compare these two groups’ learning while listening to the sound. If communication is what matters, we should see that infants who were led to believe the sound was communicative learned more successfully than those infants who weren’t clued in to its function.
For a non-speech sound, we used sine-wave tones because we know from previous research that these sounds never provide a boost to babies’ learning. The tones sound a little musical and a little mechanical; imagine R2D2 singing a lullaby and you’ll be pretty close. But how could we make some of the babies think that these sounds were communicative and some think that they weren’t?
To accomplish this, we divided 6-month-old babies into two groups. In one, they watched a video of two women having a “conversation”: one woman was shown speaking English and the other woman’s responses were dubbed over with the sine-wave tones, as though this were her own language. In this scene, the tones were communicative. A second group of babies watched a video of the same two women cooperating in an activity that looked somewhat like cooking—stirring, passing utensils, et cetera—while the tone conversation from the first video played in the background like a radio. The babies heard the exact same audio input in group two as they did in group one, but this time the English-tones “conversation” was uncoupled from the actors onscreen; in this scene, the tones were non-communicative.
Afterward, both groups of babies listened to the tones while performing the same categorization task, in which they saw a series of images that all belonged to one category (e.g., dinosaurs). Then, we showed them two images side by side in silence: one was a member of the now-familiar category (e.g., another dinosaur) and the other was a member of an entirely new category (e.g., a fish). In studies like this, babies who form a category – who recognize that these objects all have some important relationship to one another –prefer to look at the new object over the familiar one. Just like adults, when babies see a lot of the same kind of thing, they prefer to look at something new when given the chance. Critically, in this task, recognizing the object at test as “the same kind of thing” means they’ve formed the category! By contrast, babies who don’t recognize the category – who are just enjoying the pictures – look equally to both test objects. They don’t recognize one as another member of the category because they didn’t form the category in the first place! Most importantly, if communicative status is what’s special about language – and why it promotes learning at this age – then infants who were exposed to the tones as a communicative signal should show evidence of categorization, while those who were exposed to the tones without this communicative boost should fail.
Our results suggested exactly this: the infants who saw the tones in the communicative scene successfully performed the categorization task, while those who saw the non-communicative scene did not! This dramatic conclusion offers two new important insights into infants’ cognitive development: First, even very young infants can flexibly identify which signals in their environment are communicative simply by watching the ways by which people communicate around them. Second, once a sound is identified as communicative, it can promote infants’ learning in fundamental ways. These insights in hand, we finally have an answer to the question we posed at the outset: What’s so special about language? It’s communicative!
Ferguson, B. &Waxman, S.R. (in press). What the [beep]? Six-month-olds link novel communicative signals to meaning. Cognition.