One of the most fascinating windows scientists have into the human mind comes from watching babies learn to interact with the world around them.
LiveScience spoke to Werker at the recent meeting of the American Association for the Advancement of Science in Washington, D.C., about what bilingual children can teach us about how the mind works.
LiveScience: So where did this idea that being bilingual might confuse the brain come from?
Janet Werker: I'm not quite sure where the idea came from, but it's something that North Americans in particular seem to worry about. Because elsewhere, bilingualism — even in Canada — is just considered natural, because in most places in the world people speak more than one language.
Sometimes immigrant families are told that if they speak their heritage language that their baby may not learn the dominant language as well, and if they speak two languages to their children that they're going to be confused.
There just isn’t any really good evidence of language confusion. Of course there's language mixing. It's called code-switching and it's rule-governed. So any community of two languages will have rules about how much and when they can switch. Babies who grow up in this kind of environment, by the time they're 2 1/2 or 3, do the code-switching [going back and forth between the languages] themselves. If one parent is bilingual, they'll code-switch with them, and if another parent is monolingual they won't code-switch with them. So these are kinds of rules that they figure out.
So you have found a perceptual sensitivity in bilingual babies, where they can not only distinguish between two languages when hearing them, but also when watching muted videos of the same person speaking two different languages? What's going on here, and how long does this sensitivity last?
That's a great question. We've only tested babies on the visual-language discrimination at 4, 6 and 8 months of age. And 4- and 6-month-olds can discriminate two languages visually whether they're familiar with only one of them, or both of them.
But by 8 months, our previous work has suggested that they have to be familiar with both languages to keep them apart. Whereas our more recent work is showing that if you're bilingual you can discriminate even two unfamiliar languages [at 8 months old].
We haven’t tested babies after 8 months, so I'm not sure how long it's maintained. We have tested adults, and what we find is that as an adult, you do better again if you're bilingual. You can still perform better than chance but only slightly, so if you're familiar with one of the languages. But, in work we haven’t published yet, we've shown that you have to have had exposure to one of those languages by about age 4 or 5 to continue to show even this slightly-better-than-chance discrimination as an adult.
So we think there's a lasting influence from this early exposure.
OK. But the peak of this visual sensitivity to languages occurs when they're doing the bulk of this language learning?
We think the peak of this sensitivity occurs very early.
It's not necessarily the case that an 8-month-old couldn’t learn this. The fact that adults, when they're trying hard, can do a little bit better than chance, suggests there's some latent sensitivity there. We think what's going on is the perceptual system is getting tuned. It becomes more proficient at using that kind of information that it might require. And if you're growing up in a monolingual environment, and you haven't been experiencing the variability in input, well then treating all visual languages the same is probably not unadaptive, [in other words, wouldn't cause any disadvantage].
So this is the first I'd heard that bilingual children have other skills besides extra sensitivity to sound – that they, in fact, have heightened sensitivity to visual cues as well. So what other differences between bilingual and monolingual babies are known?
There's some great work by Aggie Kovács and Jacques Mehler that shows that at 7 and 12 months of age, babies growing up bilingual are better able to switch rules. So if a baby is taught to turn their head in one direction in order to hear or see something interesting, they'll do that well. But a bilingual baby at 7 months can then reverse the rule and learn to turn their head in another direction better than a monolingual baby can. And similarly, at 12 months they're better [able] to learn two sets of rules.
So it seems that babies who are growing up bilingual are learning the perceptual properties of each of their languages. They're learning to pay attention to perceptual cues that might be important for distinguishing things in the world, beyond distinguishing two languages and that they're able to switch between paying attention to one kind of property and paying attention to another.
Is there any reason to think these differences could affect other types of learning, beyond language?
Well, yeah. I think Aggie Kovács and Jacques Mehler's work suggests it can lead to more flexibility in learning more generally. In learning one rule, and then learning a second rule. So that's really interesting.
So I think there is evidence that growing up with two languages confers certain cognitive advantages. But I wouldn't go so far as to say you have to grow up bilingual to have those cognitive advantages. I think this is one natural route. And I think more generally what the work shows is that babies are just as prepared to learn two languages from birth as they are one, and that if parents speak two languages in the home they should be comfortable continuing to do so.
Based on what you know, would you tell your friends and family to try to raise their own children to be bilingual if at all possible?
I would tell my friends and family that if they have a baby and if they speak two languages in the home, to feel comfortable speaking both of those languages. I wouldn't say that they should now start introducing some other language that they don't really know yet.
What about sending babies to language schools, or hiring foreign au pairs?
You know, nobody's done any work on au pairs or even grandparents in the home, and I think that's a really interesting question. Babies learn the languages they want to learn. So even if bilingual families maintain the two languages that they have in the home, and even if babies are learning those two languages, once they start going even to preschool, if only one language is being used, they'll often stop using the nondominant language, and they'll even stop using it at home. And so that's a frustration, I think, for a lot of families.
So the attempt to introduce a second language in an unnatural fashion, or in a natural fashion but with somebody who isn't Mom or Dad – more work is needed to see what kind of impact this has. Does it confer the perceptual advantages and the cognitive advantages, even though the child might refuse to maintain that language? That we don't yet know.
What are some of your biggest unanswered questions about how humans learn language?
There are so many questions about language learning. I think what really drives me is I'm really interested in the preparation we have at birth for language learning. How the perceptual system – cause that's really all that we have at birth – we don't know any words yet, we don't know any concepts yet, we don't know sentence structure yet of our native language and so we have to get it all through listening and watching – and just how we do that is what really fascinates me.
Why can babies learn second languages without "foreign" accents, but adults rarely can?
I think what the research thinking on accents is, is that we've already established one representational system, the sound properties in both of the individual consonant and vowel sounds as well as the rhythmical properties of a first language. And then when we start putting a second language on top of that, if it's past this kind of sensitive period that people have talked about, it's more difficult.
I think the debate in the literature now is, is this sensitive period one that is in the brain – so are there structures or connections that are just difficult or impossible to change after a certain time – or is it continued interference with the first language? Because usually when somebody learns a second language, they're still speaking their first language. And so the properties of the first language will be influencing and maybe getting in the way and interfering with the second.
There's some work by Christophe Pallier and others that suggests that if you remove the first language entirely – so work, for example, with people who were adopted from Korea at 8 years of age, into different villages in France where they had no more contact with Korean speakers – suggests that without any interference from the first language, then more acquisition in an accent-free fashion might be possible.
It's still up in the air, because there are hardware changes in the brain as well. It's still an ongoing question.
Related on LiveScience: