LENA
Brian MacWhinney
macw at cmu.edu
Sun Feb 24 16:01:27 UTC 2008
Dear Info-CHILDES,
Some people have been asking about the LENA system. There is now an
article about it on www.nytimes.com. It was written by Yudhijit
Bhattacharjee who is a writer on the staff of Science magazine. It
is a pretty informative article and the reporter did a good job. It
may be difficult for people to locate the article and, since this
could be potentially interesting for child language research, I am
making a copy here. However, you should be able to access it also by
going to www.nytimes.com, subscribing and then looking for "LENA".
The following is from the article.
-- Brian MacWhinney
The early days of parenthood are filled with anxiety. Parents fret
over whether their babies are eating enough, growing enough and
sleeping enough. As the children get a little older, parents also
worry if they are talking enough.
But how do you judge a child’s language skills? Infoture, a Boulder-
based company, aims to take the guesswork out of that question by
selling a kind of verbal thermometer. The device, which costs $400, is
called LENA (for “language environment analysis”), and here’s how it
works. A voice recorder tucked into a child’s clothing records all the
sounds in the environment. At the end of each day, special software
evaluates both the amount of exposure the child has had to verbal
stimulation as well as the child’s own utterances. Ultimately, the
device generates percentile rankings that help assess a child’s
language development, just as doctors provide such rankings for a
child’s height, weight and head circumference.
Whatever its merits, LENA represents a radically new way of assessing
language development. Doctors initially judge a child’s skills by
asking parents about what a child can do. Kids with clear difficulties
are referred to a speech pathologist for a more detailed evaluation.
By contrast, Infoture would allow parents to monitor their kids more
precisely and on their own. But is LENA necessary? Some linguists
worry that the technology is more likely to raise false anxieties than
to assuage genuine ones.
The man behind the vision, Infoture’s founder, Terrance Paul, has made
a fortune selling software to assess children’s reading skills. His
current venture was inspired by a well-known 1995 study that found
that professional parents uttered more than three times as many words
to their children as did parents who were on welfare. The children in
the less talkative homes turned out to be less verbal and to have
smaller vocabularies. Other studies have suggested that these gaps
affect later professional success.
One way to close the language gap, Paul reasoned, would be to make
early assessments of a child’s language world. Parents, he figured,
could use the feedback to intervene and enrich their kids’ verbal
environment as needed.
But how to build the ultimate baby monitor? The company’s engineers
soon found that conventional speech-recognition software was not up to
the task. The sounds a baby might encounter — a raspy grandparent, a
TV commercial, a sibling’s chatter — were simply too varied to analyze
successfully. The best solution, it seemed, was to eschew the
identification of particular words and focus on a recording’s acoustic
features. Modeling every conceivable sound in a household, they
designed a system that distinguishes different voices from one
another, gives a rough count of the number of words directed at a
child and counts also the number of conversational “turns” that are
taken as child and interlocutor exchange words. In future versions,
the system may also include a measure called speech entropy, which
represents the increasing complexity of a child’s speech as new
consonants, words and phrases are added to its repertory.
On the basis of recordings from 314 families, Infoture engineers claim
that the number of conversational turns and the entropy measure track
closely with language ability as determined by speech professionals.
Children with diagnosed language delays, for example, have lower
entropy scores than children of a similar age who are developing
normally. But the method has its critics. Tom Roeper, a linguist at
the University of Massachusetts, Amherst, points out that measures of
conversational turns and the like can’t reflect a child’s mastery of
syntax. Learning to speak isn’t just about speaking frequently; it’s
also about knowing how to put the right word in the right place.
If a device like LENA became popular, it might create new benchmarks
for speech development. Mabel Rice, a speech pathologist at the
University of Kansas, speculates that parents might direct repeated
questions to their children in order to score more conversational
turns. The focus on quantity could also reinforce cultural biases
against quieter and perhaps more thoughtful kids; consider Albert
Einstein, who was late to start talking. Even so, Rice says, pressure
on parents to spend more time conversing with their children could
have a positive effect. Partha Niyogi, a computer scientist at the
University of Chicago and an adviser to Infoture, agrees. LENA, for
him, is best understood as an early-warning system. “Suppose you are
talking a lot to your child, but your child is talking very little,”
he says. “It could be a sign of something wrong.”
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups "Info-CHILDES" group.
To post to this group, send email to info-childes at googlegroups.com
To unsubscribe from this group, send email to info-childes-unsubscribe at googlegroups.com
For more options, visit this group at http://groups.google.com/group/info-childes?hl=en
-~----------~----~----~----~------~----~------~--~---
More information about the Info-childes
mailing list