0:00:01.024,0:00:03.707
I want you to take a look at this baby.

0:00:04.179,0:00:08.975
What you're drawn to are her eyes[br]and the skin you love to touch.

0:00:09.386,0:00:12.526
But today I'm going to talk to you[br]about something you can't see.

0:00:12.550,0:00:15.302
What's going on[br]up in that little brain of hers.

0:00:16.135,0:00:20.056
The modern tools of neuroscience[br]are demonstrating to us

0:00:20.080,0:00:23.975
that what's going on up there[br]is nothing short of rocket science.

0:00:24.736,0:00:28.173
And what we're learning[br]is going to shed some light

0:00:28.197,0:00:34.095
on what the romantic writers and poets[br]described as the "celestial openness"

0:00:34.119,0:00:35.976
of the child's mind.

0:00:37.013,0:00:40.471
What we see here is a mother in India,

0:00:40.495,0:00:44.399
and she's speaking Koro,[br]which is a newly discovered language.

0:00:44.797,0:00:46.604
And she's talking to her baby.

0:00:47.089,0:00:48.245
What this mother --

0:00:48.269,0:00:51.488
and the 800 people who speak[br]Koro in the world --

0:00:51.512,0:00:54.722
understands is that,[br]to preserve this language,

0:00:54.746,0:00:57.332
they need to speak it to the babies.

0:00:57.356,0:01:00.231
And therein lies a critical puzzle.

0:01:00.255,0:01:02.505
Why is it that you can't[br]preserve a language

0:01:02.529,0:01:05.764
by speaking to you and I, to the adults?

0:01:05.788,0:01:07.976
Well, it's got to do with your brain.

0:01:08.538,0:01:12.976
What we see here is that language[br]has a critical period for learning.

0:01:13.000,0:01:16.800
The way to read this slide is to look[br]at your age on the horizontal axis.

0:01:16.824,0:01:19.507
(Laughter)

0:01:19.531,0:01:23.170
And you'll see on the vertical[br]your skill at acquiring a second language.

0:01:23.789,0:01:27.648
The babies and children are geniuses[br]until they turn seven,

0:01:27.672,0:01:29.976
and then there's a systematic decline.

0:01:30.000,0:01:32.497
After puberty, we fall off the map.

0:01:32.934,0:01:35.466
No scientists dispute this curve,

0:01:35.490,0:01:37.445
but laboratories all over the world

0:01:37.469,0:01:39.976
are trying to figure out[br]why it works this way.

0:01:40.518,0:01:44.552
Work in my lab is focused on the first[br]critical period in development,

0:01:44.576,0:01:46.421
and that is the period in which babies

0:01:46.445,0:01:49.810
try to master which sounds[br]are used in their language.

0:01:49.834,0:01:52.324
We think, by studying[br]how the sounds are learned,

0:01:52.348,0:01:54.613
we'll have a model[br]for the rest of language,

0:01:54.637,0:01:57.575
and perhaps for critical periods[br]that may exist in childhood

0:01:57.599,0:02:00.462
for social, emotional[br]and cognitive development.

0:02:01.035,0:02:02.976
So we've been studying the babies

0:02:03.000,0:02:05.503
using a technique[br]that we're using all over the world

0:02:05.527,0:02:07.383
and the sounds of all languages.

0:02:07.407,0:02:09.198
The baby sits on a parent's lap,

0:02:09.222,0:02:12.143
and we train them to turn[br]their heads when a sound changes --

0:02:12.167,0:02:13.625
like from "ah" to "ee."

0:02:13.649,0:02:17.082
If they do so at the appropriate time,[br]the black box lights up

0:02:17.106,0:02:19.090
and a panda bear pounds a drum.

0:02:19.114,0:02:21.532
A six-monther adores the task.

0:02:22.160,0:02:23.321
What have we learned?

0:02:23.345,0:02:25.505
Well, babies all over the world

0:02:25.529,0:02:29.187
are what I like to describe[br]as "citizens of the world."

0:02:29.646,0:02:32.574
They can discriminate[br]all the sounds of all languages,

0:02:32.598,0:02:35.830
no matter what country we're testing[br]and what language we're using,

0:02:35.854,0:02:38.849
and that's remarkable[br]because you and I can't do that.

0:02:38.873,0:02:40.844
We're culture-bound listeners.

0:02:40.868,0:02:43.263
We can discriminate the sounds[br]of our own language,

0:02:43.287,0:02:45.134
but not those of foreign languages.

0:02:45.158,0:02:48.325
So the question arises:[br]When do those citizens of the world

0:02:48.349,0:02:51.233
turn into the language-bound[br]listeners that we are?

0:02:51.257,0:02:54.394
And the answer:[br]before their first birthdays.

0:02:54.418,0:02:57.550
What you see here is performance[br]on that head-turn task

0:02:57.574,0:03:00.037
for babies tested in Tokyo[br]and the United States,

0:03:00.061,0:03:01.499
here in Seattle,

0:03:01.523,0:03:03.272
as they listened to "ra" and "la" --

0:03:03.296,0:03:06.102
sounds important to English,[br]but not to Japanese.

0:03:06.126,0:03:09.504
So at six to eight months,[br]the babies are totally equivalent.

0:03:09.528,0:03:12.082
Two months later,[br]something incredible occurs.

0:03:12.106,0:03:14.786
The babies in the United States[br]are getting a lot better,

0:03:14.810,0:03:16.734
babies in Japan are getting a lot worse,

0:03:16.758,0:03:20.208
but both of those groups of babies[br]are preparing for exactly the language

0:03:20.232,0:03:21.724
that they are going to learn.

0:03:21.748,0:03:26.372
So the question is: What's happening[br]during this critical two-month period?

0:03:26.396,0:03:28.739
This is the critical period[br]for sound development,

0:03:28.763,0:03:30.178
but what's going on up there?

0:03:30.202,0:03:32.181
So there are two things going on.

0:03:32.205,0:03:35.081
The first is that the babies[br]are listening intently to us,

0:03:35.105,0:03:38.838
and they're taking statistics[br]as they listen to us talk --

0:03:38.862,0:03:40.843
they're taking statistics.

0:03:40.867,0:03:43.406
So listen to two mothers[br]speaking motherese --

0:03:43.430,0:03:46.472
the universal language[br]we use when we talk to kids --

0:03:46.496,0:03:48.707
first in English and then in Japanese.

0:03:48.731,0:03:52.170
(Video) Ah, I love your big blue eyes --

0:03:52.194,0:03:55.000
so pretty and nice.

0:03:56.049,0:04:01.976
(Japanese)

0:04:02.452,0:04:05.819
Patricia Kuhl: During the production[br]of speech, when babies listen,

0:04:05.843,0:04:10.612
what they're doing is taking statistics[br]on the language that they hear.

0:04:11.353,0:04:14.263
And those distributions grow.

0:04:14.287,0:04:18.660
And what we've learned is that babies[br]are sensitive to the statistics,

0:04:18.684,0:04:22.512
and the statistics of Japanese[br]and English are very, very different.

0:04:22.536,0:04:25.193
English has a lot of Rs and Ls.

0:04:25.217,0:04:26.976
The distribution shows.

0:04:27.000,0:04:29.840
And the distribution of Japanese[br]is totally different,

0:04:29.864,0:04:32.865
where we see a group[br]of intermediate sounds,

0:04:32.889,0:04:35.304
which is known as the Japanese "R."

0:04:35.328,0:04:39.436
So babies absorb[br]the statistics of the language

0:04:39.460,0:04:41.138
and it changes their brains;

0:04:41.162,0:04:43.313
it changes them[br]from the citizens of the world

0:04:43.337,0:04:46.283
to the culture-bound[br]listeners that we are.

0:04:46.307,0:04:51.020
But we as adults are no longer[br]absorbing those statistics.

0:04:51.044,0:04:53.601
We are governed[br]by the representations in memory

0:04:53.625,0:04:56.174
that were formed early in development.

0:04:56.198,0:04:57.976
So what we're seeing here

0:04:58.000,0:05:01.233
is changing our models[br]of what the critical period is about.

0:05:01.257,0:05:04.026
We're arguing from[br]a mathematical standpoint

0:05:04.050,0:05:07.190
that the learning of language[br]material may slow down

0:05:07.214,0:05:09.594
when our distributions stabilize.

0:05:09.618,0:05:12.480
It's raising lots of questions[br]about bilingual people.

0:05:12.918,0:05:16.949
Bilinguals must keep two sets[br]of statistics in mind at once

0:05:16.973,0:05:19.856
and flip between them,[br]one after the other,

0:05:19.880,0:05:21.652
depending on who they're speaking to.

0:05:21.676,0:05:22.830
So we asked ourselves,

0:05:22.854,0:05:26.261
can the babies take statistics[br]on a brand new language?

0:05:26.285,0:05:29.563
And we tested this[br]by exposing American babies

0:05:29.587,0:05:31.276
who'd never heard a second language

0:05:31.300,0:05:34.016
to Mandarin for the first time[br]during the critical period.

0:05:34.040,0:05:37.577
We knew that, when monolinguals[br]were tested in Taipei and Seattle

0:05:37.601,0:05:40.263
on the Mandarin sounds,[br]they showed the same pattern.

0:05:40.287,0:05:42.758
Six to eight months,[br]they're totally equivalent.

0:05:42.782,0:05:45.318
Two months later,[br]something incredible happens.

0:05:45.342,0:05:48.894
But the Taiwanese babies are getting[br]better, not the American babies.

0:05:48.918,0:05:53.800
What we did was expose American babies,[br]during this period, to Mandarin.

0:05:53.824,0:05:56.832
It was like having Mandarin relatives[br]come and visit for a month

0:05:56.856,0:06:00.653
and move into your house[br]and talk to the babies for 12 sessions.

0:06:00.677,0:06:02.883
Here's what it looked like[br]in the laboratory.

0:06:02.907,0:06:08.463
(Mandarin)

0:06:24.662,0:06:26.908
PK: So what have we done[br]to their little brains?

0:06:26.932,0:06:28.976
(Laughter)

0:06:29.000,0:06:31.671
We had to run a control group to make sure

0:06:31.695,0:06:34.907
that coming into the laboratory[br]didn't improve your Mandarin skills.

0:06:34.931,0:06:37.584
So a group of babies came in[br]and listened to English.

0:06:37.608,0:06:39.010
And we can see from the graph

0:06:39.034,0:06:41.617
that exposure to English[br]didn't improve their Mandarin.

0:06:41.641,0:06:45.236
But look at what happened to the babies[br]exposed to Mandarin for 12 sessions.

0:06:45.260,0:06:47.660
They were as good as the babies in Taiwan

0:06:47.684,0:06:50.660
who'd been listening[br]for 10 and a half months.

0:06:50.684,0:06:54.508
What it demonstrated is that babies[br]take statistics on a new language.

0:06:54.532,0:06:57.976
Whatever you put in front of them,[br]they'll take statistics on.

0:06:58.000,0:06:59.590
But we wondered what role

0:06:59.614,0:07:03.851
the human being played[br]in this learning exercise.

0:07:04.168,0:07:08.484
So we ran another group of babies[br]in which the kids got the same dosage,

0:07:08.508,0:07:10.976
the same 12 sessions,[br]but over a television set.

0:07:11.000,0:07:14.661
And another group of babies[br]who had just audio exposure

0:07:14.685,0:07:16.773
and looked at a teddy bear on the screen.

0:07:16.797,0:07:18.976
What did we do to their brains?

0:07:19.000,0:07:22.375
What you see here is the audio result --

0:07:22.923,0:07:24.661
no learning whatsoever --

0:07:24.685,0:07:26.443
and the video result --

0:07:27.236,0:07:29.081
no learning whatsoever.

0:07:29.105,0:07:33.246
It takes a human being[br]for babies to take their statistics.

0:07:33.622,0:07:35.788
The social brain is controlling

0:07:35.812,0:07:37.917
when the babies[br]are taking their statistics.

0:07:37.941,0:07:41.434
We want to get inside the brain[br]and see this thing happening

0:07:41.458,0:07:45.396
as babies are in front of televisions,[br]as opposed to in front of human beings.

0:07:45.420,0:07:50.117
Thankfully, we have a new machine,[br]magnetoencephalography,

0:07:50.141,0:07:51.434
that allows us to do this.

0:07:51.458,0:07:54.043
It looks like a hair dryer from Mars.

0:07:54.067,0:07:58.550
But it's completely safe,[br]completely noninvasive and silent.

0:07:58.574,0:08:01.017
We're looking at millimeter accuracy

0:08:01.041,0:08:04.357
with regard to spatial[br]and millisecond accuracy

0:08:04.381,0:08:07.003
using 306 SQUIDs --

0:08:07.027,0:08:10.127
these are superconducting[br]quantum interference devices --

0:08:10.151,0:08:13.782
to pick up the magnetic fields[br]that change as we do our thinking.

0:08:14.249,0:08:19.858
We're the first in the world[br]to record babies in an MEG machine

0:08:19.882,0:08:22.273
while they are learning.

0:08:22.297,0:08:23.976
So this is little Emma.

0:08:24.000,0:08:25.908
She's a six-monther.

0:08:25.932,0:08:31.050
And she's listening to various languages[br]in the earphones that are in her ears.

0:08:31.074,0:08:33.298
You can see, she can move around.

0:08:33.322,0:08:37.092
We're tracking her head[br]with little pellets in a cap,

0:08:37.116,0:08:40.091
so she's free to move[br]completely unconstrained.

0:08:40.115,0:08:42.328
It's a technical tour de force.

0:08:42.352,0:08:43.936
What are we seeing?

0:08:43.960,0:08:45.854
We're seeing the baby brain.

0:08:45.878,0:08:51.164
As the baby hears a word in her language,[br]the auditory areas light up,

0:08:51.188,0:08:55.976
and then subsequently areas surrounding it[br]that we think are related to coherence,

0:08:56.000,0:08:59.976
getting the brain coordinated[br]with its different areas, and causality,

0:09:00.000,0:09:02.976
one brain area[br]causing another to activate.

0:09:03.601,0:09:08.985
We are embarking on a grand[br]and golden age of knowledge

0:09:09.009,0:09:11.245
about child's brain development.

0:09:11.269,0:09:13.690
We're going to be able[br]to see a child's brain

0:09:13.714,0:09:17.955
as they experience an emotion,[br]as they learn to speak and read,

0:09:17.979,0:09:21.181
as they solve a math problem,[br]as they have an idea.

0:09:21.515,0:09:24.864
And we're going to be able to invent[br]brain-based interventions

0:09:24.888,0:09:27.158
for children who have difficulty learning.

0:09:27.571,0:09:30.637
Just as the poets and writers described,

0:09:30.661,0:09:34.815
we're going to be able to see, I think,[br]that wondrous openness,

0:09:34.839,0:09:38.263
utter and complete openness,[br]of the mind of a child.

0:09:39.024,0:09:41.700
In investigating the child's brain,

0:09:41.724,0:09:45.720
we're going to uncover deep truths[br]about what it means to be human,

0:09:45.744,0:09:46.899
and in the process,

0:09:46.923,0:09:49.994
we may be able to help keep[br]our own minds open to learning

0:09:50.018,0:09:51.709
for our entire lives.

0:09:51.733,0:09:52.976
Thank you.

0:09:53.000,0:09:56.000
(Applause)