ASD : More on Social Skills, Interventions, and Outcomes by 9522QzOE


									           ASD : More on Social Skills, Interventions, and Outcomes

The first summary in this set compares and contrasts children with Asperger
syndrome and those with social phobia ( with the finding that children with AS
may possess many social skills but that it is the use of such skills which is
deficient ).
The second summary concerns the use of technology in the teaching or
prompting of social skills, with the provision of appropriate models of
behaviour and feed back via video presentations seen as efficient and easy to
The final summary describes early childhood characteristics predictive of
social skill level in adulthood ( response to joint attention, language
performance, and measured intelligence ).

June 2011

            ASD : More on Social Skills, Interventions, and Outcomes

Social Skills in Asperger Syndrome (AS) and Social Phobia (SP)

The study by Scharfstein et al (2011) opens by citing the converging evidence
that impaired social skills are common to a number of psychological or
developmental disorders which can impact significantly and negatively upon
interpersonal interactions and the probability of making and maintaining
friendships, particularly during childhood, with long term consequences such
as limited achievement at school, behavioural problems/delinquency, and
mental health problems in adulthood.

SP is defined as a pervasive and irrational fear of social or performance
situations where the individual is subject to scrutiny and where (s)he believes
that there is a high probability of embarrassment. SP is first observable
among children of an early age, and is estimated to affect around 5% of
young people.
Typically, the children with SP, compared to controls, respond to social
stimuli/prompts with few words, take longer to make any response, may be
ignored or rejected by peers, and are rated as having limited overall social

AS involves circumscribed interests, some anomalies in language use ( a
pedantic or repetitive style, for example ), and apparent deficits in social skills,
albeit with no history of cognitive or language disorders or delays in the early
years … ( although emerging evidence would suggest that, with hindsight,
some subtle anomalies may be identifiable ). Individual with AS commonly
struggle in respect of establishing relationships, are weak in using or
interpreting non-verbal signals, and show limited reciprocity.

However, the current authors report that there has been no research by which
directly to compare the social skill deficits in these two diagnostic groups.
This is seen as an important issue since clinicians may confound AS and SP (
because of the symptom overlap ) with the risk lest inappropriate interventions
are recommended.

In their own study, samples of children with AS and with SP were compared in
terms of social skills and vocal style. Various “live” scenarios were used by
which directly to observe social behaviours such as the capacity to manage
the topic of conversation and to respond appropriately to another child (
pragmatic social behaviour ); appropriateness of speech production ( eg.
prosody ) ; and non-verbal behaviours
( posture, direction of gaze, and gesture ).
Vocal characteristics explored included variability in voice volume and vocal
intensity, and pitch.

The expectation was that children with either AS and or SP would show social
deficits when compared to typically-developing children, but that the deficits
would be more pronounced among the group with AS.

The sample comprised 23 boys and 7 girls in the SP group ; 26 boys and 4
girls in the AS group : and 22 boys and 8 girls in the control group. The age
range was from 7 to 13 ( with respective group means of 10, 10·5, and 10·7
years ).

All the children were assessed for anxiety using the parent and child forms of
the Anxiety Disorders Interview Schedule.
To confirm the ASD diagnosis, the Autism Disorders Interview (Revised) was
implemented with the parents.
The children completed the Social Phobia and Anxiety Inventory for Children.

The behavioural assessment task was of a role play type. There were 5 brief
scenarios involving interaction with a same-age and typically-developing peer
… starting a conversation, offering help, giving and receiving a compliment,
and responding assertively to inappropriate behaviour.
For example, the researcher would set the scene, telling the child that “ You
have been working hard to remember a poem to recite in class. You finish
reciting the poem and return to your place. The boy/girl sitting next to you
says ….. “
At this point the peer provides the start of the interaction by saying what his
prompt card contains … eg “ You did a great job ”. The response of the
target child is then
noted. The peer goes on to make some further (scripted) comment such as “
You remembered every word and looked so calm and cool ” and again, the
reaction is noted.
This pattern is repeated for each of the 5 scenarios.

Independent observers ( blind to the group to which the observed children
belonged ) rated the signs of social anxiety and the overall social skill of the
children on 4-point scales.
Further, a coding system was used to highlight social skill deficits within three
domains. These were :
     Pragmatic social behaviours ( managing the topic of a conversation,
       appropriateness of response and affect, latency of response, and
       extent of utterances )
     Speech and prosodic social behaviours ( voice volume, appropriate
       inflection )
     “Paralinguistic” social behaviours ( posture, facial orientation while
       speaking or listening, and motor movement ).

The overall results indicated that, in these brief and structured scenarios,
children with SP showed significantly less social skill than children with AS or
typically-developing children. This was evident across all the three domains
of social skills, including speech production and prosody.
 This finding offered strong support for the notion that children with SP have
significant deficits in social skills ; and the implication offered by the authors

was that social skill training programmes for children with SP should include
not only guidelines about what to say but also about how to say it.

Importantly, the evidence indicated that the children with AS may have fewer
of the basic social deficits than children with SP. Clear deficits were noted by
the “blind” observers in the children with SP, but not in the children with AS,
during the scenarios. The level of anxiety reported by the AS children did not
have an impact upon their (observable ) social performance.
The implication was that children with AS may possess the necessary social
skills to cope with brief interactions … raising the question whether they are
willing or able to use these skills in day to day settings.

The authors pondered whether children with AS are likely to function more
effectively in a situation where the need is for minimal and prescribed
( such as simply offering thanks on receiving a compliment ) than in a
situation where more original and spontaneous utterances ( and appropriate
accompanying actions ) are required. This was thought to be all the more
likely in situations requiring the child to take the first step … ie to initiate the

The current findings were seen as out of line with much existing evidence
which indicates clear and consistent social deficits in children with AS … but
these studies have commonly focused upon more complex interactions.
However, the discrepancy supports the interpretation of the present authors
that the level of social skills typically available to children with AS are
adequate for simple, brief, and relatively structured situations, but do not
suffice in situations demanding more extensive or complex or spontaneous
Support was also held to exist for the view that the deficits among children
with AS include the use of what skills they do have and not simply a global
lack of such skills.

Further examination during an unstructured conversation task indicated that
the children with AS showed more appropriate pragmatic competence than
paralinguistic behaviour. This difference was marked, leading the authors to
suggest that the children with AS may know what to say, but need assistance
to develop the skills required for appropriate delivery of the utterances in
terms of posture, voice pitch and intonation, gesture, and facial orientation.

In their concluding summary, the authors highlighted the differences between
children with AS and children with SP during brief interactions. Children with
SP showed less pragmatic, speech and prosodic, or general social skills than
typically developing children ; whereas there was no marked difference
between the children with AS and typically-developing children.
However, the children with AS may have some basic social skills which need
not be covered to any great extent in training programmes, but more subtle
skills ( involved, for example, in speech volume and intensity ) are lacking
which may well lead to ineffective interactions.

What matters is ensuring that children with AS do utilise the skills possessed,
and that research focuses upon the extent to which one can reasonably
expect children with AS to learn and to use more complex social skills
appropriate for longer and unstructured situations.

Technology and Social-Skill Training

The review by DiGennaro Reed et al (2011) has, as its context, the estimated
prevalence of 1 in 110 children with an autistic spectrum disorder. In western
countries, ASD has become a major concern for health and education
departments, and the financial and logistical implications for providing
appropriate services are daunting.

A consistent characteristic of ASD is some delay or disorder in the domain of
social skills and peer interactions. The observed issues include impairments
in the use or understanding of a range of non-verbal behaviours ( direction of
gaze, facial expression, etc. ) ; failure to establish relationships appropriate to
the developmental level ; a lack of attempts to share interests or enjoyments
of activities ; and absent social, and emotional reciprocity.

These social deficits may well have long-term and negative outcomes in terms
of academic performance and of an increase in repetitive and stereotyped
behaviours ( as if to indicate a retreat into a controllable and predictable world
in contrast to the unpredictability of group settings and interactions ).

A logical focus in intervention programmes has, therefore, been upon social
skill development ; and, with a view to managing the pressure upon resources
and staffing, increased attention has been directed towards the use of various
technologies which have the potential for delivering guidelines without high
demands for the active involvement of support professionals.
However, the current authors argue that existing evaluative research is not
very meaningful given that there has been a tendency to focus upon just one
type of intervention, that many studies are purely descriptive, or that there is a
lack of sound methodology and experimental design.

Accordingly, this current review by DiGennaro Reed et al set out to highlight
the number of empirical studies concerned with technological aids for social
skill training, the nature of the skills targeted, the nature and impact of
relevant variables
( demographics, the experimental settings, assessment procedures, etc ), and
the reported reliability of dependent and independent variables.

During June 2010, a systematic search was made of the literature, focusing
upon relevant studies appearing in peer-reviewed journals. 29 studies met
the criteria.
The authors looked at the nature of technologies, the range of social skills
targeted, the demographic data pertaining to participants ( age, gender, etc ),
the settings, the assessment procedures, and reliability of variables.

In respect of what, specifically, the authors were looking for, reference was
made to technologies such as computer programmes, i-pod or smart phone
delivery, virtual reality presentations, video or DVD modelling, tactile prompts
( using a pager, for example ), and audio scripts.

The possible “topography” of targeted social skills included conversational
( initiating, responding, and social conventions ), non-verbal behaviour, play
skills, social problem solving, peer relationships, and emotion ( identification,
regulation, and reciprocity ).
The greatest single category ( around half the studies identified ) focused on
teaching children with autism to initiate conversations. The next largest
category (28%) related to the teaching of play skills.

Settings included school, hospital, home, university, and other centres not

It was found that the participating children in the 29 studies were
predominantly boys ( 83%). The age range was from 2 to 17 years, with a
mean of 7 years.

The overall picture to emerge from the review was that technological aids can
be integrated readily within interventions designed to reduce social skill
The practical significance was the opportunity afforded to replace scarce and
expensive human resources with technological resources ; and the perceived
need now is to conduct further research on the most efficient and effective
procedures, particularly for use in schools where the majority of social skill
programmes are implemented.

The findings indicated that the most commonly used technology involves
video or DVD use. The evidence highlighted success in such methods for
modelling social skills, and, to some extent at least, in providing feedback.
The second most prevalent technology was the use of audio scripts.

It was argued that the advantage of either approach is the ease of their use
and of the training of all concerned in the procedures.
Video presentations, for example, have been held by Ayres and Langone
(2005) to allow the repeated demonstration of a given skill with little need for
direct input from a practitioner. The same would apply to audio scripts.

The authors were surprised at the extent of emphasis upon conversational
skills, and upon play skills. Few studies examined how technology might
address deficits in social problem solving, emotion regulation, or peer
It was their hypothesis that the skills involved in conversation and play are
more readily operationalised and measured.
They also speculated that the use of technology is less suited to the teaching
of advanced and complex social skills ( such as the processes involved in the

making of friends ), albeit acknowledging the importance of play and
conversational skills as a necessary basis for higher level skills

The implication here was for ongoing research to explore how various
technologies could be used to teach the more complex skills.

For practitioners, the implication offered was for ongoing training in social skill
programmes which involve the use of technology and for which there has
grown a body of empirical evidence for effectiveness.
A focus would logically be upon those interventions which are most versatile (
ie modifiable to cover a range of social skills ) and most readily implemented
in natural settings.

The authors suggested that, logically, the first steps should focus upon
video/DVD instruction and audio scripts for social skill development given that
these technologies are readily available and inexpensive and have been
shown to have the potential for success in addressing various elements of
social skill. It appears that they can be modified to meet the particular needs
of given individuals, and modified further as progress is made ( or not ).

As is commonly the case, it appears, according to the current authors, that
evaluative data are limited as a result of the lack of adequate assessments.
There needs to be a clear set of baseline measures if one is to be able to
determine the rate of progress, with follow-up measures adopting a standard
format in order to assist and clarify the monitoring process and in order to
guide the ongoing nature of the intervention.

Predictors of Social Skills in Adulthood

In the introduction to their study, Gillespie-Lynch et al (2011) argue that there
been little research tracking individuals with ASD from childhood into
adulthood, or seeking to relate childhood characteristics ( strengths and
weaknesses ) to the level of social and other functioning in adulthood.

Their own study set out to explore whether joint attention skills in early
childhood are predictive of adult social functioning.

It is further argued that a major problem for identifying predictive factors
across longitudinal studies of autism is the variation in the diagnostic criteria
used, in the sample characteristics ( such as IQ, and age at baseline and
follow up ), the nature of provisions and support provided, and the subjectivity
of outcome evaluations.

Three common measures of adult outcome are the levels of social functioning
( independence and relatedness … whether the individual is employed, has
friends, and can live independently ), adaptive behavioural skills, and
observable autistic symptoms.

Tracking a child, with regular review assessments, into adulthood may offer
insights into the course of the disorder as well as providing a basis for the
planning of support.

Some existing findings have linked adult social functioning to the quantity and
quality of speech before the age of 6 years and to IQ scores. Although
speech and IQ are related, it has been argued ( Rutter et al 1967 ) that each
may account for unique parts of the variance in outcomes.

Language and IQ have also been shown to be predictive of adaptive
functioning although adaptive behaviours may well be lower than expected
according to IQ … particularly in individuals who are not intellectually
For example, Szatmari et al (2009) compared children with autism and
children with Asperger syndrome to find that the AS group produced higher
scores on the Vineland Scales of Adaptive Behaviour across all domains and
across time points, but actual growth in adaptive behaviour was independent
of diagnosis and levelled out during late adolescence. Non-verbal IQ at age 5
was related to the Vineland daily living skills and socialisation but not to
Meanwhile, other evidence has suggested that higher IQ differentiates
between initially non-verbal children who do, and those who do not, go on to
develop language during mid childhood. Positive change in IQ from initial to
follow-up assessments may also be predictive of better social functioning in

Szatmari et al (op.cit) suggested that improvements in autistic symptoms and
in general functioning over time might be linked to a common developmental
precursor … namely, the level of joint attention ( both the initiation of, and
responsiveness to, joint attention ).
Other studies have indicated that relatively frequent initiations seen in children
is predictive of improvements in social and communicative functioning ; and
that both initiations and responsiveness are predictive of expressive

The current authors hypothesised that responsiveness to joint attention (RJA)
with its effects upon language and cognitive development, would predict adult
independence and adaptive skills, albeit with reduced effects when changes
in language and IQ were also included in the analysis.

The participants were those who were already part of a longitudinal study,
beginning when they were aged 3, and followed up previously at ages 11 and
18 years. The original sample comprised 70 individuals with autism. 51 took
part in the second wave of assessments, and 48 in the third.
This current study followed up 20 of the original sample whose mean age was
( SD 3·8 ).

At the first assessment, use had been made of the Early Social
Communication Scales to examine non-verbal communication sills ( including

eye contact, pointing, and showing by which to initiate joint attention ; and the
proportion of prompts, by way of pointing or direction of gaze, needed to elicit
a response to joint attention ). Cognitive ability was assessed either by the
Cattell Scales of Development, the Stanford-Binet Test, or the Mullen Scales
of Early Learning.
At this early childhood stage, language was measured by the Reynell Scales (
or by the CELF in the case of children with good speech )..

The Autism Diagnostic Interview (Revised) was used in the second and third
assessments ; but only current functioning was assessed in this present (4th)
The Vineland Scales were administered in the second, third, and present
assessments. Also for this present assessment, a rating was established to
indicate overall social functioning as reflected in employment, living
circumstances, and friendships. The ratings ( based upon observations,
objective information, and parental responses ) covered a 5-point scale from
very good to very poor where the top rating indicated residential and
occupational independence plus some friendships, and the bottom rating
applied to individuals with minimal independence and living in specialist

The general picture to emerge was described as similar to that produced in
the study by Eaves and Ho (2008) for a sample of individuals with autism born
in the 1970s or 1980s and with a largely comparable level of cognitive ability.
Both studies suggested that there has been a gradual improvement in the
social functioning of people with autism ( observably better than the typical
functioning identified in earlier studies ), most likely to be explicable in terms
of the greater availability of specialist services and interventions.

However, in this current study, it was acknowledged that drop-out rates may
well have been greater for the lower functioning participants, thus inflating the
proportion of individuals with relatively positive outcomes.

Closer inspection indicated that language skills and RJA, but not intellectual
functioning, predicted social functioning in adulthood. It was thought possible
that the limited prognostic significance of intellectual scores may be a function
of the very young age of the children when first assessed, or that such scores
discriminate best among children with poorer outcomes at adulthood.
Meanwhile, some of the predictive power of language ability ( in respect of
later social functioning ) was thought to be attributable to its relationship with
RJA and its scaffolding effect upon developmental quotient and language age.

The general view was that RJA in early childhood may be particularly
predictive of social performance in adulthood. The lack of predictive power of
initiations of joint attention was interpreted as indicating that adult social
outcomes are related to more involuntary non-verbal communicative
behaviours in early childhood.

The observed pattern of developmental trajectories was taken by the authors
to indicate the importance of skills such as RJA in facilitating the process of

learning from others, and of early intervention and parental behaviours which
foster linguistic and cognitive growth.
Caution was also recommended in respect of discussing likely progress
according to observations of performance in very early childhood, given that
the most consistent predictors of adult outcomes were the changes in
language and mental age between the first and third assessments.

The authors recognise limitations in this current study … notably the small
size of the residual sample and the gender bias ( all male ), reliance upon
parental information via telephone interviews, and the lack of information
about available services or the socio-economic status of the families. Further,
individuals may be rated as having “poor” outcomes, but still enjoy happy lives
and acceptance within a community.

Nevertheless, the final conclusion was that RJA and early language are
predictive of a composite measure of adult social functioning and
independence. Language and cognitive scores related to adult adaptive
behaviours, and RJA related to adult non-verbal communication, and social
skills/symptoms. Adaptive behaviours changed with time and development,
but the basic symptoms of autism remained.

General Implications

From the first study, on might emphasise the point that children with autism
and ASD are not necessarily lacking in at least some of the basic social skills
( nor lacking in the motivation to establish interactions with others ), but may
be lacking in the confidence to use any of the skills they do have, or
experience an uncertainty over how to initiate contact, or fail to have
opportunities for contact ( or to recognise them when they arise ).

The concern ( as expressed commonly ) is that simply enabling a child with
ASD to attend a mainstream school is not going, of itself, to bring about social
acceptance and a sense of belonging to a peer group. What may well be
required is the mobilisation of peer resources together with peer-awareness
training about ASD to increase the probability that contacts can be
established ( with the peers making the first move if necessary ).
This is of significance given the still-repeated anxieties ( from the target pupils
themselves or observers ) about break-time or lunch-time during which the
pupils with ASD may be uncertain about where to go or what to do and which
may lead to growing anxiety.

Purely educational integration ( and differentiation of curriculum demands as
appropriate ) is only half of the need, and the social need may well be the
more challenging. There seems to be a tacit assumption that the child with
ASD will be accepted and supported by peers in the mainstream, but this may
be a somewhat over-generous assumption.

From the second study, one would simply highlight the point about the
growing gap between the number of children with ASD who require access to
various support services and the level of resources that are available ( which

are not growing in parallel but possibly diminishing in this period of financial
constraints ).
The need is for the means of providing support which is not wholly dependent
upon 1:1 intervention, so that one would welcome the ongoing research on
technology to provide support which is effective and does not require constant
professional input and supervision.

One can understand why parents or others may still look for that direct input
but the extent of the demands may lead to some conflict … ( rather analogous
to cases where there are arguments about the perceived need for individual
and regular sessions for a child from a speech and language therapist, and
where the responsible authority is offering a language programme devised by
a therapist but delivered by classroom assistants or other adults ).
Direct input from an experienced clinician/practitioner may be the ideal, but
the reality is quite different from that ideal in terms of the time and resources
The need is for evidence that alternative ( and cost-effective ) arrangements
can produce positive outcomes.

From the final study, one notes the information about early predictors of
outcomes. The significance of language and cognitive measures is already
well recognised, but this study highlights further the predictive importance of
responsiveness to joint attention.
The practical implication is for early identification of the likely or definite
presence of autism or ASD, and for supportive intervention by which to foster
both cognitive and communicative performance, within which ( response to )
joint attention appears to be a key element.

                                *      *      *       *      *

June 2011


Ayres K. and Langone J. 2005 Intervention and instruction with video for
students with autism : a review of the literature. Education and Training in
Developmental Disabilities 40 183-196

DiGennaro Reed F., Hyman S., and Hirst J. 2011 Applications of technology
to teach social skills to children with autism. Research in Autism Spectrum
Disorders 5 1003-1010

Eaves L. and Ho H. 2008 Young adult outcome of autism spectrum
disorders. Journal of Autism and Developmental Disorders 38 739-747

Gillespie-Lynch K., Sepeta L., Wang Y., et al 2011 Early child predictors of
the social competence of adults with autism. Journal of Autism and
Developmental Disorders, published on-line DOI : 10.1007/s10803-011-

Rutter M., Greenfield D., and Lockyer L. 1967 A five to fifteen year follow-up
of infantile psychosis II ; social and behavioural outcome. The British Journal
of Psychiatry 113 1169-1182

Scharfstein L., Beidel D., Sims V., and Finnell L. 2011 Social skills deficits
and vocal characteristics of children with social phobia or Asperger’s disorder
: a comparative study. Journal of Abnormal Child Psychology, published on-
DOI : 10.1007/s10802-011-9498-2

Szatmari P., Bryson S., Duku E. et al 2009 Similar developmental
trajectories in autism and Asperger syndrome from early childhood to
adolescence. Journal of Child Psychology and Psychiatry 50(12) 1459-1467


To top