‘Baby Talk’ Can Help Kids Learn Language (Oh *Yes* It *Can*!)

Confronted with a baby—or puppy—most adults can’t stop themselves from dissolving into baby talk: “WHO’S the cutest? It’s YOU! YES it IS!” We slow down, increase our pitch by nearly an octave, and milk each vowel for all it’s worth. And even if the baby can’t speak yet, we mimic the turn-taking of a conversation.


This story originally appeared on Ars Technica, a trusted source for technology news, tech policy analysis, reviews, and more. Ars is owned by WIRED’s parent company, Condé Nast.

This “parentese” is found across cultures, and babies exposed to more of it at home seem to do better at acquiring their home language. But it’s not all about instinct: a paper published in PNAS this week suggests that parents can be trained to improve their parentese and that this training gives their babies’ language a boost.

Learning to Baby Talk

Why does more parentese go hand in hand with language acquisition? It’s an open question. Recordings from parents and children in their homes show a correlation—the more parentese there is, the more likely the babies are to be a little more advanced with their language abilities. But is the parentese itself actually helping? And if so, how? Or is there another factor at play that boosts them both?

There’s some reason to think the parentese itself is actively helpful. Its simple, exaggerated language could make it easier for babies to grasp what’s being said. But it could also be that its melodic, theatrical qualities grab and hold babies’ attention, while also giving them space to practice conversation by babbling during their “turns.”

A group of researchers at the University of Washington, Seattle wanted to see whether parents could be coached on improving their parentese and whether this would affect their babies’ language development. So they tracked 71 families with young babies over the course of a year, asking the parents to record a full weekend of the family’s conversations when the babies were 6, 10, 14, and 18 months old.

They split the families into two groups, offering coaching to one group but not to the other. The control group still did all the recordings, but the coached group came in to the lab after the researchers had listened to each set of recordings and got personal feedback and pointers.

The coaching helped the parents to identify helpful habits in their own speech, like engaging in back-and-forth interactions with their babies. They were also given suggestions about what kinds of age-appropriate interactions they could have during activities like bathtime or meals.

First Steps

The results were promising: parents in the coaching group showed more use of parentese over time compared to the control group and also engaged in more back-and-forth interactions with their babies. The babies themselves vocalized more, too—if you remove non-linguistic noises like coughing and count prelinguistic noises like babbling, the babies in the coaching group were chattier.

And at the end of the study, babies in the coaching group did better on language assessments than babies in the control group.

The researchers checked that factors like the parents’ level of education weren’t affecting the outcomes. They made sure that this was balanced across the two groups at the start of the experiment and had a look to see whether it was correlated with the children’s outcomes at the end. It wasn’t—babies from across the social class spectrum all seemed to get a boost when their parents received coaching.

But as promising as this research is, it’s just a start, and it does have some important weaknesses. For one thing, the control group didn’t have any intervention at all, while the coached group knew that researchers would be listening closely to their behavior to give them personal feedback. While it’s difficult to keep up an act for a whole weekend, it’s still possible this knowledge could have affected their behavior on the recordings.

And studying babies is messy, difficult, and time-consuming, with a really high drop-out rate among the participants. This, plus limited resources, usually means small samples, and this study is no exception. That doesn’t invalidate the results, but it does mean the data will be noisy, which could mean that the results are exaggerated. So more studies will be needed to confirm these results and understand them better.

Early language ability is linked to advantages later in life, but it’s a messy link that has a lot of different possible explanations. So one crucial question for future research to answer is whether these benefits persist later into the children’s lives—even after the coaching stops.

PNAS, 2018. DOI: 10.1073/pnas.1921653117 (About DOIs).

This story originally appeared on Ars Technica.

More Great WIRED Stories

Read More