https:/.../2019-10-10_psy333d_pt2.mp4

And so, it's very often focused on the
here and now, but it is making--

lacking those little filler words,
'on the table,' just 'put it table,' and

they kind of learn the key aspects,
the language first and then learn

the little filler words and things later.

So, then, they learn the past tense and
the plurals and things, so those all are

a slowly growing, understanding of
the complexities of language and

the particular uses of different
language structures.

Really what we are trying
to understand is how children

learn grammar and how
they learn to communicate.

What's interesting is, are parents really
teaching this or are children just

picking it up like sponges, how is this
happening because there's a lot of

cases where parents are trying to correct
a child's incorrect use of language, but

it is not really, at the time anyway,
showing any kind of correction.

There's some really funny examples
of that in the literature, in the data

that people use to analyze child language.
Here is an example.

The child says, "My teacher holded the
baby rabbits and we petted them."

And the parent corrects this and says,
"Did you say your teacher held

the baby rabbits?" and the child is like,
"Yes."

And the parent says,
"What did you say she did?"

"She holded the baby rabbits
and we petted them."

"Did you say she held them tightly?"
"No, she holded them loosely."

So, they get exactly what the parent
is saying but they are not picking up

on the 'held' versus 'holded'
type of thing. Another example is:

"Want other one spoon, Daddy."
"You mean you want the other spoon?"

"Yes, I want the other one spoon,
please Daddy."

"Can you say 'the other spoon?'"
"Other one spoon."

"Say other."
"Other."

"Spoon."
"Spoon."

"Other spoon."
"Other spoon."

"Now give me other one spoon?"

You know, they are not taking in that kind
of feedback and realizing how to apply it

to the structures that they are trying
to say. So, parents are responding to

the meaning of what the children is saying
ultimately and not necessarily the grammar

of it. They are not stopping the child
and saying "No, that is an incorrect

past form -- past tense form of the verb,
you need to correct this."

But they do model these utterances in
a more correct formula and you can see

that when a child makes an incorrect use
of a verb, for example, the parent is more

likely to rephrase than when they use
it correctly and that is one potential

source of feedback that children
are getting is that, "Oh, if I say it not

quite right, my parent repeats that,
rather than just answering me."

And over time, maybe they learn to make
fewer errors over all.

But it is really interesting that,
at least in the moment, correcting

the child's language, doesn't actually
seem to take very much.

And that has been observed, you know,
widespreadly in the public and the media.

So, in family circus,
"Mommy, Dolly hitted me"

"Dolly HIT me."

"You too?! Boy, she's in trouble!"

Just not picking up on that 'hitted'
versus 'hit' distinction, for example.

Okay, English is a very strange language.

We have a lot of regular verbs
and irregular verbs

and that can make language
challenging for kids to learn

because, in general, you can make
something in the past tense by

adding "ed" but there are enough words
that we use commonly that don't do that.

Ran, went, sang. There are a lot of cases
where the "ed" form is incorrect.

When we do plural, almost always we are
adding an "S," except for a lot of words

that are actually common in children's
books and in kids, where we talk about

mice, geese, teeth, feet, children,
and men for example.

We don't automatically add
the "S" for everything.

What is interesting is children
seem to initially

be correct about this,
they track this okay.

So, if you look at an example of
one child named Abe,

he correctly, initially, said ate and eat,
but then when he turned three

actually kind of seemed
to lose that ability

and started to say
eated for every use of it and it

wasn't until age four or so
that he came out of that

and started properly using ate again.
So you get this initial correct use,

and over-generalization incorrect use,
and then a return to the correct use.

What we think is happening here is,
initially, each word is getting a kind of

single mapping through the system and
not showing any pattern understanding.

But then children form a schema and
realize, "Oh this is a convention that I

can apply to words," and they over
apply it. They apply it to every

word they know and accuracy actually
goes down for these unusual verbs

and unusual past tense.
Then they kind of learn over time,

very slowly, to re-correct and to go
back to actually how they were before

but just for a subset of the words.
As opposed to individually mapping

every single word, they find an efficient
strategy but then have to learn

the cases where that strategy
doesn't apply.

Which is just kind of interesting.
So it creates interesting things

around age three that parents
find very funny.

So, for example, Jill De Villiers was
one of my undergraduate professors

and she's a linguist and
a developmental psychologist

and was really interested in language.

She said, "It's starting to snow again,"
and her grandson said,

"But it already snew!" Jill is like,
"That's perfect! That's my

sentiments exactly!" So he applied,
instead of snowed,

he took the unusual form and
applied that to everything.

So language has a complex hierarchical
structure and children are initially

learning individual instances and then
they are learning how they can

detect patterns in this language stream.

So, in the right conditions, they start
to generalize those strategies really

quickly and it's really fun and magical.
They can produce

infinite variations and surprising thoughts
that people haven't thought

or said before, like snew for snow,
and they are only preschoolers right.

So even though they are not
taught explicitly,

they end up with appropriate
structure that they can produce

an infinite number of sentences from.

Okay, so that's the grammar
understanding-- is that they

start out learning each word individually,
they learn a pattern and then

they learn the exceptions
to that pattern rather slowly.

It's also fun to just think
about how children start to learn

to communicate. You could see the
difference in the kids who have come to

the studio between ages 2 and age 3
and how much more sophisticated that

language has become so quickly. And
they can say some really wonderful things.

Sadie, when I was tucking her,
in was giving me a hard time

at bed time one time, and I said,
"I'm serious Sadie, this is the last time

I'm tucking you in," and she's like,
"Don't be serious Mama, don't be serious."

And in the grocery cart at age 4,
she took her shoes off which was

frustrating her father. He said,
"Why did you take your shoes off?"

She said "I didn't it's just that my
feet were naughty,"

implying that they took them
off themselves.

At age 5 she said, "How can you tell the
poison ivy bee-- the poison bee-- from

the regular bee?" and I said, "I've never
heard of a poison bee.

All bees sting and can hurt
but some people are allergic."

and she said, "It must be true Mommy
because nature is all around us."

This is where they pick up phrases
and sayings, like swear words and things,

from conversations and bring them out
at the most inappropriate time.

So they are also very very curious about
things that are using these languages to

ask questions about things that they are
wondering about. Sometimes also

at inappropriate times. So, we were at a
pizza place and it was pretty quiet and

we were getting ready to leave because
parents with young kids eat super early

and as we are getting up to walk out
at the top volume she says, "Mommy,

what do we do with dead people?"
At the top of her lungs.

And I was just like, "Let's go a outside,
let's go go go," and she's like, "Do we

burn them? Do we bury them?
What do we do,"

and I'm just like,
"Let's leave right now,"

and the whole restaurant
was watching us.

It was crazy.
They are constantly thinking

and learning how to use language
to get attention, which is pretty funny.

Alright so, as another example of
the kind of magical development

that you can see over
a brief period of time,

here is a video clip about kids
who are leaning language.

>> In the course of a year of filming,
we were able to see Avery,

the youngest of three children,
learn to speak.

>> Feel better?

>> At the first Christmas, Avery was
nineteen months old.

>>Want to take a bath now?

She understood a lot of what Felton,
her father, was saying

but still spoke in single words,
when she spoke at all.

>>Go take a bath.

>> Will you help me put these on?

(Avery babbling)

>> With her mother, Anita, she played
at rhythmic vocal games

like nursery rhymes
without words.

>> Pancakes

>>Pancakes.
>>Pancakes.

>> Four months later at breakfast with her
three year old brother Malcolm,

she crams all the meanings
she wants to express into

the single words that are sill the sum
of her language abilities.

>>Pancakes.

>>She says she refuse to eat
my pancakes.

>>Pancakes.

Pancakes.

>> Yes, Pancakes! Are you eating your
pancakes?

>>Is it hot?
>>Yes.

>> Two months later Avery had her
second birthday.

>>My Birthday!

>>Her vocabulary and her ability to put
words together in sentences,

both grew as she imitated her brothers,
five-year old Benjamin

and three-year old Malcolm.

>>It's almost time for you guys to
eat dinner.

>>It's almost time
for you to go to bed.

>> To bed! (Avery imitating)

>> And you're not sleeping here.

>> And tomorrow you're coming back.
>> Coming back. (Avery imitating)

>> Tomorrow.
>>Tomorrow. (Avery imitating)

Where's my muffin, daddy?
>>I ate it.

>> Only four months later, Avery, now
two years and four months,

was speaking in sentences of her own.

>>It's not here, not there.
I ate it.

You kind of abandoned it so I decided
I would eat it.

>> There's more
>> Well, you can eat that, yeah, but--

>> Eat that.

>> Eat that, alright.

>> Now, Avery can play a joke

>> Eat that

>> No I don't want to eat that.
It won't be too good.

>> Eat this.

>> Avery's father, Felton, assessed her
progress.

>> Avery, gone and made a lot of changes.
She has really learned to communicate.

She's gotten very good
at expressing ideas

and expressing who she is
and what she wants to do.

Give me a hint what we are looking for.

>>Yeah.
>>Okay, what we looking for?

>>Hint.
>>Hmm?

Hint.
>>Hmm?

>>Hint.

>>Who?
>>Hint.

>>I'm asking you for a hint.

>> When the next Christmas came around,
Avery was just over two and a half.

>>Oh, come out of there.
>>I want to look at something

>> What are you looking for?
>>Hint.

>>I asked you for a hint.
>>What's the hint?

>> What's the hint.
>>I'll show you

>> You're going to show me,
okay show me show me.

Case in point, my mother in law called
here last week and Avery answered

the phone and when mother in law said,
"Avery go get your mother,"

and Avery said, "No, I don't want to do that.
I want you to talk to me."

Okay, you got the spoon?

>> Yeah

>>You do?

>> Here it is dad.
>>You got it?

>> In learning to speak, Avery has learned
to speak for herself.

>> You ready to stir?

>> Yeah.

>> Language enables her not only to
participate more fully in the life

around her but also to recognize
her own needs and desires

and to say what it is she wants.

>>Keep stirring.

>> With language comes a new self awareness.

A sense of herself as a separate person.

>> Put it on the stove.

>> No, I know how to put it on.
Dad push the chair over.

>> Alright, awesome.

So, you know the extent
to which children can

communicate and hold a conversation
really develops between ages two and five.

Whereas, three-year olds include brief
references to past events, five-year olds

produce narratives and descriptions
of past events that have the

basic structure of a story. Parents can
kind of scaffold that by asking questions

or prompting and asking them to elaborate
that. As children get older

their conversational turns increasingly
relate to what the other person says.

So what's kind of funny is, initially
you are having a conversation but

the conversation is jumping all
over the place on topic.

So you're having back and forth,
but not necessarily sustaining

a conversation. The ability to sustain
that conversation grows so dramatically

over the preschool years and honestly
continues to improve well into adulthood

where it's like, "Can we talk about what we
were talking about and not talk about

something else right now?"

So here's a graph of related turns,
in red, increasing over age, from age two

all the way to adult, here,
and unrelated turns,

meaning you take a turn in the
conversation but you change the topic,

decreasing over that same age-range.

But you can see that it's a slow
improvement after that kind of rapid

scale increase between
ages two and five.

Alright, so we had a great question,
Tehila what was it?

>> Yeah, so the question is:

"How does language production work in
infants who have a non-verbal parent?"

>> Excellent, so I'm hoping we have time
to get to a video that I have on those

children who are learning in a
deaf environment and whether they

babble on things in their manual sign
language but I think the critical point

about this is that children need
language input, it doesn't have to

be an oral language, it doesn't
have to be any language in particular

but they need a rich language environment
in which to learn or they will also have

language difficulties like
impoverished language.

In the case of a non-verbal parent

if the parent is able to communicate via
manual sign language or in other ways

that's really critical, or at least the
child is getting language exposure

from all the other adults in the household
or other caregivers in that child's life.

Because they really-- this window of
language like a sponge is so critical

for cognitive development, language
development and academic success.

Children are really sensitive to
how accurate speakers are,

so that's an interesting dove tail into
the questions that we have here.

Which is if-- Do children trust speakers
who give them incorrect information

or how did they learn about incorrect
information and so

three and four years were shown two
different speakers who were

identifying novel objects and
some familiar objects.

So they start with the object
that the child knows

and one person says, "That's a spoon,"
and the other person says "That's a duck."

For a bottle, "That's a bottle," or
"That's an apple," and so they learn

that one person correctly labels objects
and one person incorrectly labels objects.

Then you ask the child, "Which person
would you like to ask what this novel

object is called?" Or you say, "Here is
a novel object, this person,

that person who was previously accurate,
says it's a 'slod' or this person

who is previously inaccurate
calls it a 'linz.'

Which one do you think it is called?"

And it shows that both three and
four year olds, trusted accurate

over inaccurate speakers on both
types of trials.

So they are really sensitive to
the person speaking to them

and their history and what
they know about the world

and how it matches what they
are expecting and so I think

that's a really critical thing too--

is that goofing around with them is one thing
but being an inaccurate speaker

is something they will learn to assess
and rely on information in a different way.

Okay, so the other kind of major turn
that I kinda wanted to cover in language

and we could talk about
this for a whole semester

and many courses do,

but is the interaction
of language development

with academic success.

If we talk about socio-economic status,
which is also abbreviated as SES.

Children who come from a low
socio-economic status background,

even before they enter kindergarten
are showing a gap, an achievement gap.

That gap continues to grow
throughout elementary school, such that

children who come from
low-income backgrounds are testing less

and fall several years behind by
fifth grade, in many cases.

Also, lose a lot of gains that they make
over the summer,

which is called the
summer slide.

That increases the chance that
they drop out of high school

or don't attend college.

So people are very motivated to try
and understand what is causing

this achievement gap and
how can we address it.

That's where programs like
Head Start and Pre K and

other types of programs are
attempting to work on this.

There is a huge variability in
early vocabulary, right?

So if we look at this graph
These are graphs of children,

over five hundred children, being tracked
from eight months

all the way to 28 months.

You can see that there is not much
variability in the beginning

because there is only
a few words.

But the range and measurements by
the time you are two and a half,

or almost three is really really
massive.

The factors that are influencing
that are of great interest

to try and boost more low producers up.

You can look at the vocabulary
of children and the words they

understand and there is a similar
range, so that by fifteen months,

you know, the bottom ten percent
of the children understand maybe

fifty words but the top ninetieth
percentile of children

understand
three hundred words.

That gap translates into the
number of words they produce.

So that children in the top ninetieth
percentile of words produced are

producing something like two hundred words
by the same age,

whereas the bottom ten percent are maybe
producing two or three.

There is a very famous study in the 90's,
that related to the socio-economic status

of different families and the number of
words they were supposedly speaking to

their children. It became known as the
thirty million word gap,

because they observed at age one,
two, and three

that professional parents or highly
educated parents were speaking

way more words to their child
than children of lower income families.

Based on extrapolation from these
curves, they projected that by the

time these children entered Kindergarten
or Pre-K, they would be

about thirty million words behind.

This just caught fire in
the imagination,

about ways to kind of close the word
gap and, you know,

is this the source of the
achievement gap

between high-income and
low-income individuals.

Obviously, if you are not
hearing as many words

maybe you are learning fewer
words. It had a kind of intuitive sense.

However, you know, it was done
observing only forty two families

in the state of Kansas
and they would go for an hour

and record everything that was said
but just for that one hour

late in the late evening
or late afternoon.

We now have more sophisticated
technology that can track

children over the whole day.
Right?

So it's called the "LENA System,"
which is this little recorder

that you can put in the overalls
of a baby or in a little pocket that they

wear, and it can record all of the
child to child interactions and

the adult to adult interactions
happening in the environment.

And when they did this, they found that
the gap of thirty million was actually

much more like four million, though they
did still note differences in the amount

of child directed speech happening at
different income level households.

But then another study, you know, actually
questioned whether that was true at all.

They went and did a similar recording in
different parts of the country

with different size households and they
found just such a huge range at each

level of income that they didn't find a
consistent gap and felt that it was

actually a non-replication of the
thirty million word gap study.

So this is a very interesting,
controversial kind of finding of,

"Do children of lower income
households really get that

much less of environmental input?"

And it led to a lot of very superficial
type of interventions.

So there was one in
Providence, Rhode Island.

Where it was called "Providence Talks,"
and the solution was merely to focus on

getting parents to talk
more to their children.

And they put five million dollars into
this program, but it's not just about

babbling constantly to your child, right?

It's not just about sheer volume of input.
The quality of the speech really matters.

And how you engage with the
children and the back and forth.

And so, while that program did promote
caregivers to increase their

daily word count by about fifty-five
percent, there's no evidence that just

that key component of number of words
heard really has that big of an impact.

Right? It's not just about words.

It's much more about the give and take
and the communication

and sharing-- not just speaking
at children or around children

but really with them. And giving them
a chance to have a back and forth

even when they aren't
necessarily verbal yet.

And the tone of the language
may matter as well.

And so, what we need is more
research about the wise

level of language use.
And the simplistic level of just,

"Oh, well that's easy, just everyone
talk to their children more."

It might help a little bit, but it's
unlikely to be the magic pill for

closing this achievement gap, right?

A more sophisticated program called
"Thirty Million Words," does a bit better

of a job emphasizing the
quality not just the quantity.

So tune-in. Pay attention to what your
child is communicating to you.

Talk more with your child,
using descriptive words.

And take turns. So they call
that the three "T's."

To try and encourage children to really
have more of a give and take.

But I think this whole thing has been
questioned as of late just because

it really-- focusing on a gap
between incomes and speaking

is a form of deficit thinking. Right?

And deficit thinking is when
you identify a project or a condition

by what is missing.
And in this case, parents talking

to their children in a certain way.
And that deficit often reflects biases of

the researcher, of the institution,
and certain stereotypes.

But we have a question.
What do we have, Tehila?

>> So we have a question about where
the language can come from.

So we've been talking about language
coming from parents. But the question is

what about if there are
older siblings in the house?

Even those in college or high school
who have large vocabularies?

Can that help boost the baby's language?

>> Yeah, perfect. Right, so one drawback
of the original Hartin Risley study of

the 42 families in Kansas is that all of
the low-income families were single

parent, I think, single-child households.

So they also just had fewer people
talking to the child overall.

And so one of the ways that
this has been attempted to be

replicated is to go into more diverse
families and to record,

not just mother-child level of
communication, but household level

of communication and
that is a fantastic point.

Because it doesn't just
have to be one caregiver.

It can absolutely be all sorts of
caregivers or people interacting

with the baby have a chance
to influence their spoken language

and their communication skills.