Turns out you can’t learn Mandarin Chinese by watching television

It is generally accepted that language exposure in the first year of a baby's life is important. To what certainly must be the dismay of couch potatoes everywhere, researchers at the University of Washington determined that it's particularly important that it come from a live human being. Watching DVDs doesn't have any effect. What was particularly notable to the researchers was that even just five hours of exposure made a significant impact.

Comments (16)
  1. BlakeHandler says:

    Like my Dad always said: "You can’t learn how to drive a car, simply by reading a book"

    Most engineers do need to "get-out" more often! (^_^)

  2. SnappyCrunch says:

    That last sentence is confusing, I think. It should probably read “What was particularly notable to the researchers was that even just five hours of exposure <B>to a person speaking Mandarin</B> made a significant impact.”

    [The last sentence should read “What was particularly notable to the researchers was that even just five hours of exposure of the type the researchers were using, see article for more details, made a significant impact.” But I had assumed that the underlined part was clear from context. -Raymond]
  3. Mark Steward says:

    SnappyCrunch: That last sentence is confusing, I think.  If the person already speaks Mandarin, why do they need exposure? ;-)

  4. Dewi Morgan says:

    Quite clearly, adults CAN learn languages through self-study programmes: many people have, and continue to do so.

    What the study shows is that babies learn language from people, and not measurably from recordings.

    The most obvious conclusion here is not that “nobody can learn from recordings”, but that babies consider recordings to be “background noise” rather than sources of learning.

    I’d be interested in the results using 3D goggles and an interactive 3d avatar, responding to movements and sounds of the baby. The study shows that there may be a real possibility of learning, if the baby’s attention can be kept focused on the teacher, and that babies are pre-wired to focus on people recognised as “human”.

    After all, we’re well past 1984, and it’s about time we started programming babies in Orwellian robotic creches.

    [Turns out you can’t make a funny headline without people nitpicking it. -Raymond]
  5. dislyxec says:

    This sentence in the article struck me as the most telling reason for this:

    "The researchers also noted that the infants who watched the DVD or listened to the audiotape paid significantly less attention than the babies who were in the live Mandarin and English conditions."

    and also to an extent:

    "The babies were mesmerized by the sight and sound of the foreign language speakers."

    I hypothesize that if they can find a way to make the DVD more interactive, and thus more attention retenting, that the results would at least show some learning.

  6. Sheva says:

     But you have to consider the situation in which the person who wanna learn another foreign language probably has little chance of exposure to the live language conditions, and at most times, they have to visualize the scenario in which they are learning from a real speaker or talk with him.

     I have put a lot of my precious time on learning English, so I can tell you it’s absolutely true.


  7. Mark W says:

    I think the title should be "… INFANTS can’t learn Mandarin…" rather than "…YOU can’t learn Mandarin…".  When I was growing up in Hong Kong where Mandarin was not in the school curriculum, most people picked up Mandarin by listening to Mandarin songs and watching Mandrain movies.  There was a <a href=’http://blog.sina.com.cn/m/cailan‘>celebrity</a> who claimed that he learned Japanese by watching one Japanese movie over and over again.

    Mark W

  8. MJP says:

    This is particularly interesting in that it supports suggestions that babies can’t learn *English* by watching television (where English=one’s mother tongue). This suggests that using the box as a babysitter may not be such a good idea.

    The usual reason given why humans are better at teaching babies to speak is that they respond to what the baby says (or babbles), providing positive and negative feedback as appropriate.

  9. db48x says:

    I submit that an adult can learn a language from recordings only because he has already learned how to learn, while the infant has to rely on explicit feedback from another human to learn anything.

    In other words, by the time you’re an adult, you will have (hopefully) learned how to judge for yourself when you have done something correctly, and use that as the feedback that guides your improvement.

    Of course, that doesn’t mean that everybody automatically does that in all situations; you frequently see people doing dumb (or at least things that are only marginally sub-optimal) things over and over because they haven’t applied a little bit of rigor or discipline.

  10. Vince P says:

    Mark: I agree.  When I first saw the title I thought Raymond directed his title directly at me.

    *rolls eyes*

  11. Dewi Morgan says:

    [Turns out you can’t make a funny headline without people nitpicking it. -Raymond]

    Perhaps your “couch potatoes” would be dismayed because they would have to interact with their own baby, rather than because they would not easily be able to learn a language themselves: given the headline, though, I read it the latter way.

    A headline is a synopsis of the article, summarising the “take-home message”, and colours the way you read it.

    To demonstrate, imagine reading about the research, but with two alternative headlines over it: the rational “groundbreaking research in the field of infant education”, and the supermarket-tabloidesque “Scientists subject babies to five-hour Chinese torture ordeal”.

    Even with identical articles, your reading of the author’s intent would be coloured by the headline.

    [This is a blog not a newspaper. I can’t believe I had to write that. -Raymond]
  12. wangd says:


  13. build6 says:

    Dewi – adults that can successfully learn foreign languages etc. may or may not be evidence that falsifies the proposition that exposure to the target language is a necessary element during infancy – we have no idea how linguistically "pure" their auditory environment was when they were children, for one thing.  My bet is that every adult with a fluency in a foreign language, *was* exposed in some way or other to the particular phonetics of the language they learnt, if only as an accent in their main language (perhaps e.g. German immigrants speaking accented English as a neighbour?).  The Eastern-Asian people who *really* cannot tell the difference between "r" and "l" are not pretending to just for amusement or an opportunity to be laughed at.  

  14. Archangel says:

    Dewi Morgan: "After all, we’re well past 1984, and it’s about time we started programming babies in Orwellian robotic creches."

    Are you sure you’re not thinking of Brave New World? I don’t recall any robotic creches (or any robots full stop) in 1984.

  15. Dewi Morgan says:

    Archangel: You’re correct, of course, I get them scrambled as I read them the same time. BNW is set far further ahead in an alternate future, so maybe I shouldn’t be calling for robotic crèches just yet.

    Raymond: Sorry, didn’t mean to sound as if I were telling you what to do. I was trying to acknowledge where I’d misread the body of your post (interpreting "couch potatoes" as "lazy language-learners", rather than "lazy parents"), and to explain why.

  16. Earlier this year, I remember saying

    It is amazing how the number of times you need to bite your…

Comments are closed.