David Geffen School of Medicine
Center for Autism Research and Treatment
Advances in Autism 2015
I’d like to go ahead to our next speaker,
Michelle Dean who will be speaking to us
from Cal State University, Channel Islands
on understanding the social behavior of girls with ASD.
So please join me in welcoming Dr. Dean.
Thanks for coming. [Applause]
Hi, “The Art of Camouflage Gender Differences
and the Social Behaviors of Girls and Boys With Autism”
We have a great deal of difficulty identifying and diagnosing
girls with autism without cognitive impairment.
And there may be a male bias in our perception of autism.
Girls with autism have been described as
being better able to camouflage their symptoms of autism
and to use compensatory behaviors
that mask their social challenges.
Camouflage and compensatory behaviors refer to
echoing or mimicking the social behaviors they see other kids do.
But when girls reenact these behaviors,
they’re described as looking superficial,
and seem to lack understanding of the underlying motivations
that inform our social interactions.
So the word “camouflage” highlights the importance of environment.
Understanding the way boys and girls with autism interact with
or blend into the natural social landscape at school
may give us a better understanding of
why it is difficult to detect autism in girls.
We know that children with autism
have a great deal of difficulty in making friends at school.
We know less about gender differences and the social behaviors of autism,
and the actual experiences that children with autism have.
But we do know that children with and without autism
tend to segregate by gender, when they play at school.
And in typically developing populations,
it is widely accepted that there are gender differences
in the way that boys and girls socialize.
Boys are more likely to be playing games,
Girls are more likely to have intimate conversations.
So, in order for girls with autism
to blend into the landscape at school,
their social landscape is different than boys with autism.
So in our research, we ask the following research questions:
To what extent do environmental factors
like gender-related social behaviors and activities
play a role in helping girls with autism to mask their symptoms?
Are girls with autism better at ‘camouflaging’ their symptoms of autism
and using compensatory behaviors to mitigate their social difficulties?
And are the symptoms of autism more obvious and easier to detect in boys?
To examine how children interact with their social environment at school,
we examined the observation data
of 185 children with and without autism.
In… This is a secondary analysis of data
that were collected for a large randomized control trial
of a social skills intervention.
All of these data were baseline data,
are collected before the start of the intervention.
And in the original study,
there were 24 girls of high-functioning autism
or autism without cognitive impairment.
And so we used every girl in the study.
Boys were selected from a larger pool of boys
that were matched to the girls with autism
by age, IQ and city of residence.
The kids in the study lived in
巴尔的摩 安阿伯 洛杉矶和西雅图
Baltimore, Ann Arbor, Los Angeles and Seattle.
We verified the diagnosis of autism using the ADOS.
All the children with autism had an IQ over 70.
And all children with autism were educated in the general education classroom
for more than 80% of the school day.
Now to determine what the social landscape on the playground looks like,
we also use data from typically developing children,
with 69 girls and 68 boys.
These children were selected for the original study
because they were classmates of one of the participants with autism.
In addition, their teacher,
so general education teacher nominated them to participate,
because they had good social skills.
positive prosocial behavior… behaviors.
All children were observed on the playground for 10 to 15 minutes,
using the Playground Observation of Peer Engagement,
or what we call the POPE.
Each line here represents a one-minute interval.
The POPE here yields quantitative and qualitative data.
Quantitative data refers to engagement state,
and qualitative data is a short description of
what the kids were doing during that one minute interval.
The variables to focus for this study are:
独处时 孩子是自己呆着 还是和大人一起
Solitary, the child is alone or with an adult.
Joint Engage, the child is actively socializing with a peer or peers.
Game, the child is actively playing a game
with rules with a peer or peers.
Children were observed by graduate students or research assistants
who were trained to a reliability criteria of at least 0.85
and to maintain coding consistency to coders randomly
and independently coded 20% of the observations.
So here’s an example of what three minutes looks like.
共同合作 和一个女孩一起散步交谈 开心
Joint Engage, walking and talking with one girl, happy.
Joint Engage, walking and talking with one girl.
共同合作 和一个女孩在攀玩架附近交谈 大笑
Joint Engage, talking with one girl on the jungle gym, laughing.
For this analysis, we use concurrent mixed methods,
in which quantitative and qualitative data were collected
at the same time and analyzed concurrently.
Our analytic design was an exploratory case study design
with each group representing one case.
定量 定量分析方面 我们调查
Quantitative, for the quantitative analysis, we examine the frequency
that each group spent some time
in Solitary, Joint Engage or Game.
Then we tested between group differences in the average amount of time
that each group spent in each engagement state.
Qualitatively, we identified the…
we identified each activity
that each participant engaged in, in each engagement state.
And then we identified what the most popular activities
in each engagement state in each group.
Once we’re able to develop a social profile for each participant,
we then went to the qualitative data to select a representative example
of what the social behaviors looked like at school.
The social landscape: typically developing girls.
So this graph represents the proportion of typically developing girls
that had spent some time in each engagement state.
So typically developing girls,
the Joint Engage was the most popular engagement state.
In addition, typically developing girls
spent on average more time than boys in Joint Engage.
The most popular activities for typically developing girls were talking and flitting.
Flitting, we called an activity as flitting
when a participant engaged in three or more activities
within one observation period,
and spent about at equal amounts of time in each activity.
So what typically developing girls looks like were –
Joint Engage: Walking with two girls. Smiling and talking.
Joint Engage: Sitting in a circle pretending to paint each other’s nails
and they did this for two minutes.
Three girls get up.
They hold hands and they start running.
然后追赶一个男孩子 大笑着 非常开心
And then they start chasing a boy, laughing and happy.
The social landscape: typically developing boys.
Typically developing boys, Game was most popular,
followed pretty closely by Joint Engage.
虽然如此 男孩子们 尤其是正常成长的男孩子们
Although, on average, boys, typically developing boys
spent significantly more time in game than girls.
The most popular activities for boys were
first, ball games and second, talking.
So a representative example of what it looks like
to be a typically developing boy on the playground looks like this:
Joint Engage: Getting a game organized.
Going to invite kids to play the game.
Gets out, has a conversation on the bench,
but still watching and paying attention to the game.
Girls with autism, for girls with autism,
Joint Engage was tied with Solitary.
Girls with autism spent their time in Joint Engage or Solitary.
The most popular activities for girls with autism
were talking, like typically developing girls.
And their second most popular activities, so was a tie,
was flitting between solitary activities,
playing on the play structure alone or playing ball alone.
A representative example of what a girl with autism
looked like on the playground was:
Picture a caterpillar painted on a playground
and the girls were hopping from one piece of the caterpillar to the other.
So Joint Engage: Caterpillar activity with 4 girls.
远离女孩群体 走近男孩群体 没讲话
Walked away from the girls. Walked close to boys. No talking.
和一群女孩站得很近 不属于她们的一部分 但是很近
Standing near a group of girls. Not part of the group, but close.
Initiates to the girls to do something else.
They have a short conversation.
Boys with autism: Solitary was the most popular
engagement state for boys with autism.
In fact, a significant interaction effect indicated that
boys with autism spent significantly more time
being solitary than every other group.
So not surprisingly, the most popular activities
for boys were both solitary activities.
And they were talking to an adult and wandering.
So, being a boy with autism,
er… represented an example of what a boy with autism
looked like in this group is:
Sitting alone, eating a snack.
An aide comes up to talk to him.
Another aide comes to talk… comes up to talk to the first aide.
The boys gets up, walks to the playground.
Wandering around with his head down. Alone.
So our findings suggest that
there is some support to the “Camouflage” hypothesis.
The fluidity of female social groups seems
to provide an ideal backdrop to conceal girls with autism,
who tend to be hovering close by.
So when we think about playgrounds,
an adult supervisor, a playground attendant
would be scanning the playground.
But this would be insufficient to identify
the social challenges of girls with autism,
because they look like typically developing girls,
except for typically developing girls are moving
fluidly from group to group,
whereas girls with autism are moving back and forth ,
weaving between Joint Engage and Solitary.
Joint Engage periods were interrupted with Solitary periods,
which suggests that although social challenges might be concealed
by adults on the playground,
they don’t appear to be concealed by other peers.
In contrast, the male landscape seems to make it easier
to detect the social challenges of boys with autism.
Boys primarily played games.
And when you think about games, the structure of games,
there’s one large group, playing a game
for almost all the entire duration of recess.
So it’d be very easy to detect the boy with autism,
because he wasn’t playing the game.
So, simply scanning the playground environment
would be sufficient to identify the social challenges of boys.
So our findings do support the camouflage hypothesis.
And they also suggest that there may be
a male bias in our perception of autism.
If school practitioners or practitioners in general
are looking for social isolation
to detect social challenges at school,
then girls with autism will remain unidentified and overlooked.
Social challenges exist in girls and boys with autism,
but withdrawal and exclusion may be
more nuanced and less obvious in female populations.
So, for future directions,
we need to weigh the social landscape,
when designing social interventions for kids with autism at school.
For the boys with autism in this study,
they would need an intervention that targets…
they’re being in a solitary state…
to help them move from solitary into joint engagement.
We also really need to consider that the male social landscape is
way more physically demanding than the female social landscape.
For girls, we would want to acknowledge
that they’re already in Joint Engage,
but increased the duration of time that they’re in Joint Engage.
How do we help them stay in Joint Engage?
And that may be due to a quality of their interactions.
So perhaps we need to work on
social interactions and improving the quality.
And more work really needs to be done
to examine an intimate inter-social interaction,
so we can understand where the breakdown occurs,
and be even more specific in our interventions.
Finally, perhaps we have underutilized
peers as a source of information
about the social challenges of kids at school.
And future research may wanna consider the peer perspective
to help us gain more understanding
about the social challenges of girls with autism, and boys with autism.
So, also thank you, Connie Kasari,
and Health Resources and Services Administration for funding the study,
以及研究助理 凯特·赖德尔 玛尔塔·威咖
and research assistants, Kate Reidel and Marta Wirga.