home about sciencentral contact
sciencentral news
life sciences physical sciences technology full archive
biologygeneticshealthbraineducationanimalspsychology
January 4, 2011
ScienCentral

Live Learning


Post/Bookmark this story:

Search (Archive Only)
 

Hearing Screening (7.17.03) - Researchers are calling for hearing screenings for newborns, because the earlier hearing loss is discovered, the better.

Baby Talk (2.27.03) - How early can babies tell “baby talk” from real language? Our way with words comes a lot earlier than you might expect.

 

How Do Infants Learn Language?

Conversational Mandarin Chinese Online

Do Apes Use Language?



   08.07.03
email to a friend
 
 
Video
cute baby biting toy
(movie will open in a separate window)
Choose your format:
Quicktime
Realmedia

Have you ever wondered how babies learn to talk? As this ScienCentral News video reports, some researchers now think it’s more than just hearing the people around them.

It's Not Greek To Them

If you have trouble learning new languages now, maybe you should have started when you were an infant. Babies are born "citizens of the world," able to distinguish among sound used in all languages, according to Patricia Kuhl co-director of the University of Washington Center for Mind, Brain and Learning and professor of speech and hearing sciences. She found that nine-month-old American babies could pick up Chinese from a live teacher in just a few hours.

Kuhl studied how babies distinguish sounds in language by exposing them to Mandarin Chinese over a four-week period. “We were interested in how babies are taking up information when they’re listening to parents and caregivers talk to them," says Kuhl, who published her research in the Proceedings of the National Academy of Sciences. "We had live Chinese-Mandarin speakers talk to American babies who, of course, have never heard Chinese before, to see what they would learn about the sounds of the language just by listening to it. We exposed the babies by having the Chinese graduate students take turns playing with the babies, using a prescribed set of toys and a prescribed set of books." The nine and ten-month-old babies had twelve of these sessions, each lasting 25 minutes. There was also a control group of American babies looking at the same books and toys, but hearing only English. The total amount of exposure was less than five hours.





baby reading with chinese woman
image: U. of Washington
Then, the babies were tested from two to twelve days after their last exposure to Mandarin. They taught the babies to turn their heads when they could discriminate the sounds of the language. "The test involved both the American babies exposed to English and the American babies exposed to Chinese to see which of the kids can hear this distinction," says Kuhl. "The answer was very clear, that the American babies exposed to the Chinese were excellent at the sound change. Whenever the sound changed they turned their heads. For the kids in the control group who had listened to English, the sound changes did nothing."

So the babies exposed to Mandarin had picked up a lot within a short amount of time. "With five hours of experience, the babies who’d been exposed to live Mandarin speakers learned a tremendous amount," says Kuhl. "They were as good at hearing the sound distinctions of Chinese as babies growing up in Taiwan who had been listening for ten months. So the first experiment convinced us that the baby brain is like a computer, calculating information as they listen to us speak."




The results of the first experiment made Kuhl and her team wonder if the babies would learn just as well if they were watching TV or listening to a tape. They didn't. "In the second experiment, exposure to the DVDs, the television films of the speakers talking—the same speakers, the same material—the babies learned nothing," says Kuhl. "There was no effective exposure either to the audio-visual tapes or to the audiotape alone."

baby in front of TV
Kuhl believes this difference between learning from a person and learning from a machine indicates that social interaction plays an important role in the learning process. "The second experiment is, in a sense, a kind of warning to us that learning is constrained in some interesting respects. You don’t just learn anything. You can’t just expose children to things on DVD and audiotape and expect that their brains are automatically taking up that information. And in evolution, that may have been a pretty important thing. You learn from other human beings who are like yourself. You don’t learn birdcalls and you don’t learn door slams and you don’t learn other irrelevant things either. Perhaps the social brain is a kind of gatekeeper to learning."

Kuhl says that one of the most significant things the research demonstrates is how integrated our brains are. "We sometimes think of our brains as separated into modules or pieces, one of which works on language, one of which works on social, another works on math, and another, if you’re a physicist, works on physics. What these studies tend to show is that the brains, particularly of infants and young children, are very integrated. All parts of your brain are communicating with all other parts. It demonstrates that in learning…infants and children will use every piece of information they can glean from the environment. And learning is very biased by the original conditions under which that learning was forged in evolution."

This research was funded by the National Institute of Child and Human Development, Human Frontiers Science Program, the William P. and Ruth Gerberding Professorship, and Talaris Research Institute.


 
       email to a friend by Karen Lurie
               
     


Science Videos     Terms of Use     Privacy Policy     Site Map      Contact      About
 
ScienCentral News is a production of ScienCentral, Inc. in collaboration with The Center for Science and the Media 248 West 35th St., 17th Fl., NY, NY 10001 USA (212) 244-9577. The contents of these WWW sites © ScienCentral, 2000-2011. All rights reserved. This material is based on work supported by the National Science Foundation under Grant No. ESI-0206184. The views expressed in this website are not necessarily those of The National Science Foundation or any of our other sponsors. Image Credits National Science Foundation