Bilingual Infants Learning New Words
How do bilingual infants manage to learn similar sounding words?
Posted August 2, 2017
Interview conducted by François Grosjean
One of the most amazing phenomena in young children learning a language is how well and how fast they learn new words. There comes a time in their development when we, adults, can hardly keep up with their increasing vocabulary. And yet, this is far from evident, especially when new words only differ from other words by one sound, either a consonant or a vowel. Our amazement increases several folds when we observe young bilingual children doing the same thing, but in several languages, and doing it as well. In this post, Dr. Krista Byers-Heinlein, of Concordia University in Montreal, a leading expert in language development in bilingual infants, has very kindly accepted to answer a few of our questions. We thank her wholeheartedly.
Can you first tell us what is involved in the learning of new words in infants and notably how they acquire words that sound similar?
Children often seem to learn words effortlessly, but there are actually a lot of different aspects to successfully learning a new word. At a minimum, infants have to identify and remember the sounds of the word, figure out what the word refers to, and associate the word and its meaning. For similar-sounding words, infants have to pay special attention to very subtle differences in sound. For example, “cat” and “cut” differ only in their vowel. If infants only remember the general outline of the word, that it starts with “c” and is one syllable long, for example, they won’t be able to tell these two words apart.
In what way is this challenge even greater in bilingual infants?
Every language has different sound distinctions that are meaningful. For example, English has a “th” sound, so “three” has a different meaning from “tree.” French does not have a “th” sound, and adult French-speakers often have difficulty with this sound when they try to learn English. Whereas monolingual infants have to figure out what sounds to pay attention to in their single language, bilingual infants have to figure all of this out for two different languages (if not more). This is a lot to keep track of especially while trying to learn a new word.
One approach that is used to study new word acquisition is called the "switch approach." Can you tell us how it works?
When researchers want to study what babies know and how they learn, it can be challenging because babies can’t really follow instructions, respond verbally, or press a button. Researchers have to devise clever ways to indirectly figure out what they’re thinking.
In the “switch approach,” we first teach infants two new words – usually ones we have invented so that we know they haven’t heard them before. Sitting on their parent’s lap, infants see two different objects presented on a computer screen and hear their labels. Thus, for example, they see a crown moving back and forth and hear "kem," or they see a molecule made of building blocks and hear "gem."
This teaching phase continues for several minutes until infants get bored, and start to look away from the screen. Then we test their learning with a “switch.” We show them the first object but accompany it with the second label, i.e. the crown and the word "gem,” and vice-versa. If infants have learned the correct names, they should be surprised by the switch and stare longer at the screen than when we provide a correct label.
With your colleague, Christopher Fennell of the University of Ottawa, you studied whether monolingual and bilingual infants are equally sensitive to new words that differ on the very first consonant. How did you proceed?
In this study, we tested two groups of 1.5-year-old infants — monolingual English-learners, and bilingual French-English learners. We used the switch approach to teach them two similar-sounding words like the ones above. Half of the infants heard the words spoken by a speaker who was herself bilingual, and the other half heard them spoken by a speaker who was monolingual.
What was the result you obtained and did it surprise you?
We found that both monolingual and bilingual infants were sensitive to a change in sound and could use this difference to successfully learn the two new words. However, they succeeded only when the words were articulated by the speaker of their own language background. That is, bilingual infants could learn from the bilingual speaker, and monolingual infants could learn from the monolingual speaker, but not the other way around. This was surprising because the difference between the two speakers was very subtle – adult listeners could not tell the speaker’s background.
Basically, infants are very well adapted to their own language-learning environments. Many bilingual infants have bilingual parents, and many monolingual infants have monolingual parents. However, this doesn’t mean that none of the babies could learn from the other speaker. We also found that the bilingual babies who had a monolingual parent did better learning from the monolingual speaker than the ones who didn’t.
Keep in mind that our participants were only 1.5 years old, an age when they are just beginning to build their vocabularies. We think that older infants would have done fine learning from either type of speaker.
A more recent study by Leher Singh and her colleagues, at the National University of Singapore, seems to show that bilingual infants might even have an advantage over monolinguals. Can you first tell us how different their study was to yours?
Their study used a very similar procedure to ours, with three main differences. First, whereas the words in our study differed only in their initial consonant (“kem” and “gem"), theirs differed only in the middle vowel (“min” and “mun”). Second, there were differences in the groups of infants tested. Whereas we only tested English monolinguals, along with the bilinguals, they tested both English and Mandarin monolinguals. Third, our bilingual groups were also learning different language pairs: ours French-English, and theirs Mandarin-English.
What results did they obtain?
Once again, bilingual infants learned the words just fine from the bilingual speaker. However, surprisingly, monolingual infants were not able to learn the words, whether they were pronounced by a monolingual or a bilingual speaker. The researchers concluded that bilingual infants might have an overall advantage in learning similar-sounding words that differ only in their vowel.
Where do you see the word learning research in bilingual infants going in the future?
The studies we have just discussed are inspired by the work of Dr. Janet Werker and her colleagues (see here), who first observed how challenging learning similar-sounding words can be for monolingual infants. Going forward, word-learning research with bilingual infants is beginning to ask questions that are unique to this population. For example, how do bilingual infants manage to learn words in each of their languages? In challenging word-learning situations, do bilingual infants prioritize one language over the other? And how do they learn words when a speaker switches back and forth between their languages?
For a full list of "Life as a bilingual" blog posts by content area, see here.
Photo of boy with abacus from Shutterstock.
Fennell, Christopher and Byers-Heinlein, Krista (2014). You sound like Mommy: Bilingual and monolingual infants learn words best from speakers typical of their language environments. International Journal of Behavioral Development, 38(4), 309-316.
Singh, Leher, Fu, Charlene S. L., Tay, Zhi Wen and Golinkof, Roberta Michnick (2017). Novel word learning in bilingual and monolingual infants: Evidence for a bilingual advantage. Child Development, doi:10.1111/cdev.12747.
François Grosjean's website.