Bilingual children learn language rules more efficiently than monolinguals

Blogging on Peer-Reviewed ResearchAs Eddie Izzard notes in the video above, the English, within our cosy, post-imperialist, monolingual culture, often have trouble coping with the idea of two languages or more jostling about for space in the same head. "No one can live at that speed!" he suggests. And yet, bilingual children seem to cope just fine. In fact, they pick up their dual tongues at the same pace as monolingual children attain theirs, despite having to cope with two sets of grammar and vocabulary. At around 12 months, both groups produce their first words and after another six months, they know around 50.

Italian psychologists Agnes Melinda Kovacs and Jacques Mehler have found that part of their skill lies in being more flexible learners than their monolingual peers. Their exposure to two languages at an early point in their lives trains them to extract patterns from multiple sources of information.

Kovacs and Mehler demonstrated that by sitting a group of year-old infants in front of a computer screen and playing them a three-syllable word. The infants could use the word's structure to divine where a cuddly toy would appear on the screen - if the first and last syllables were the same ("lo-vu-lo"), it would show up on the right, but if the first and second syllables matched ("lo-lo-vu"), it appeared on the left. By watching where they were looking, the duo could tell if they were successfully predicting the toy's position.

Success depended on learning two separate linguistic structures over the course of the experiment. The infants had to discern the difference between 'AAB' words and 'ABA' words and linking them to one of the two possible toy locations. After 36 trials where they got to grips with the concept, Kovacs and Mehler tested the infants with eight different words.

The bilingual ones were much better at it. For both groups of words - AAB and ABA - they initially glanced over to the correct side more often than the wrong one, and they fixed their gaze on it for longer. Monolingual babies only accomplished this for the AAB words - unlike their bilingual peers, they couldn't learn both linguistic rules at the same time.


When Mehler and Kovacs paired the ABA and AAB words with distinctly pitched female and male voices, even the monolingual infants managed to learn the differences between them. So while both sets of children can happily associate two speakers with two different types of word (and two screen positions), only the bilingual infants could learn two linguistic structures at the same time.

This advantage could stem from a better ability at avoiding interference between the two three-syllable constructions.  Earlier this year, I wrote about other research from Agnes Melinda Kovacs and Jacques Mehler, who showed that infants raised in bilingual households have better executive functions - a catch-all term for advanced mental abilities that involve control, helping us to keep our goals and plans in mind while avoiding distraction from instinctive behaviours. These skills would certainly come into play when they tried to learn two separate languages at the same time. 

Reference: Kovacs, A., & Mehler, J. (2009). Flexible Learning of Multiple Speech Structures in Bilingual Infants Science DOI: 10.1126/science.1173947

More on bilinguals and child development:

i-77217d2c5311c2be408065c3c076b83e-Twitter.jpg i-3a7f588680ea1320f197adb2d285d99f-RSS.jpg

More like this

This is interesting, simply because it makes me think there must be a cost to being bilingual, simply because, if being bilingual gives these amazing abilities, why is it not the case that the natural state of human communities is to possess at least 2 languages? You could answer "Nationalism", but then why would a national group have a single national language, rather than 2 or more national languages (like, why do French people have French (or their local dialect) rather than French A, French B and French C)? If there is no monolingual advantage, then why is monolinguality the norm except where different ethnic groups who have different ethnic languages are in contact? Of course, maybe the "monolingual advantage" is just that where there is no communicative advantage, people cannot be bothered to make up multiple languages, as the cost outweighs the benefit.

Regarding executive function, has this been controlled against parental executive function or compared to non-bilingual sibs? It interests me if children who have parents with normal executive function have boosted executive function. If children become bilingual because they have bilingual parents, and their parents are bilingual because they have high executive function (if you follow the reasoning that being bilingual requires executive function), and executive function is heritable, this is still interesting, but not the same story. As I understand it, executive function is described as having high heritability, so this would make sense to check -….

very interesting i agree. as you probably know the US is rapidly becoming a dual english-spanish speaking country. despite my proficiency in spanish, i started learning too late to ever become fluent... now i have one more excuse for my stunted brain development :)

Most people learn the languages they need or want to learn, naturally. Our school based language learning has conditioned people to not like languages. This needs to change.

I must say this is a great article i enjoyed reading it keep the good work.Because languages is so important in this society.

"Our school based language learning has conditioned people to not like languages."

I went through the state school system here in the UK and came out the other side proficient in French and German, and I've gone on to study Chinese, Mongolian and Indonesian at university. I love learning languages, and I'm pretty good at it. I'd say that reducing the complex issue of why people in English-speaking countries don't learn other languages to the simple idea of language classes in schools being boring is misleading. And how do you propose to change it? Languages can't be learned without instruction, unless you live in an environment where the language is spoken all the time without alternatives, and even then, that leads to a skewed development. I know quite a few ex-pats in places like Taiwan who have learned to speak reasonably well and who may even have a convincing accent, but whose grammar is shoddy (in terms of being able to construct sentences themselves) and whose reading and writing, beyond basics (like "toilet"), is abysmal.
And having sat through languages classes in German schools, I can tell you that they're just as boring as anywhere, but their English and French are generally very good.

"why is it not the case that the natural state of human communities is to possess at least 2 languages?"

It depends on what you consider to be a language (as opposed to a dialect), but in some ways it is a natural state. Certainly, in a lot of developing countries it's the case. In the Philippines, most people speak at least rudimentary English, Filipino (Tagalog), and whatever the local language is. Children might have a Cebuano-speaking mother and a Waray-speaking mother, and at school they speak English and Tagalog. A lot of Africa is like that too. You could also consider the state of language in a lot of Indonesia - not only do people speak Indonesian reasonably well, but also the local language; and not only that, but with languages like Javanese and Balinese, there are three "levels of language", dependant on who you're speaking to, each with a different set of vocabulary (completely different, in most cases). So someone who speaks standard Indonesian and Javanese might speak four languages already.
It's interesting that you use French as an example, because until the late 19th century, most French citizens didn't speak standard French as a first language. There used to be a lot of variation, and no doubt people spoke both standard French and the local language to a high degree of proficiency. Italy was the same. Monolingalism is not the norm, in a sense. Or at least, it wasn't the norm until very recently, when media divided neatly along national lines.

@matt: "why is it not the case that the natural state of human communities is to possess at least 2 languages?"

How would these be defined as 'two languages'? Take for example the idea of having 'french A' and 'french B'. With everyone using them to communicate with each other, it would just become one language. French. The words would be used interchangably and grammer rules would merge.

Hightened learning ability is not necessary for continued survival and reproduction, therefore does not confir an evolutionary advantage. And as most human brains are, at base, a little lazy, learning only one set of language rules tends to be prefered, unless necessity, or personal preference, dictates the learning of two.

My understanding of worldwide bilingualism is that not so long ago the majority of people would have been bilingual due to trading. It's possible that the rise of the nation state has led to the reduction in languages worldwide and the reduction of bilingualism.

Bilingualism is also a tricky thing to define. I can speak a tiny bit of french but don't consider myself bilingual. At what point does one become bilingual? When you can arrange travel? (Common) When you can understand most of the utterances you hear? (Less common) When the native speakers of the 2nd language can't tell that you're not a native speaker?(rare. This of course does not include the complexities of when the 2nd language was learned.

By the way, this is an excellent blog.

By ex linguist (not verified) on 13 Jul 2009 #permalink

I just got back from Italy and it always amazes me that a lot of people there hardly speak any English. This makes it hard for me to talk to people, since I can't understand any Italian, even with the help of Google Translate (most sentences sound to me like one very long word).
I've run into the same problem in France.
I'm from The Netherlands, where we get taught English in school from an early age. I think other countries should follow this example, it would solve a lot of problems and apparently it's even good for the development of the brain.
It would also help a lot if these countries would stop dubbing TV shows and movies, Transformers: Vendetta del Caduto just doesn't sound right.

@matt: Where ever did you get the idea that monolingualism is the natural state of humans? Just because it happens to be that way in eg. the US doesn't mean it goes for the rest of the world. In many places, the "natural" state is to be bilingual (or tri-...); look at India, Indonesia, etc. etc. There are a lot of places where people grow up to be bilingual simply because the next village has a different language (I read somewhere about a place where it was only considered proper to marry someone who spoke a different language from yours, probably Papua New Guinea...).

Anyway, the point is that if you do grow up in a place where you hear two languages from people who you have to deal with (ie. within the group of parents, friends, educators there are at least two languages), any normal human will learn both. It just happens, without any more trouble than learning one language. The fact that many cultures now are monolingual is to be considered "accidental" â it has completely different causes.

And of course, enforcing one language is a great way of ensuring that minority groups stay subdued; this has happened in a lot of societies over the years (eg. in the past the Sámi people of northern Norway were not allowed to speak Sámi at school, nor keep any of their other symbols of their own culture â fortunately this is not the situation today).

Can we draw any correlations between countries who are generally "monolingual" vs. "multilingual" and their effect/impact/contribution/development in the world? If so, is this due at all or in part to those linguistic practices as a whole?

Does having multiple languages impede communication and thus impede development of the whole?

One more thing got me thinking in the article. What languages were represented in the "bilingual" children? Was there any correlation between the language's distance apart in language trees and their improved ability? Would there be furthur performance discrepancy between an English-Spanish environment and an English-Chinese environment?