Social Skills and Friendships (and underlying mechanisms) in Autism and ASD


This set of notes begins with further consideration whether high functioning autism and ASD can be differentiated, with the differentiating variables being those of social skill levels (or the extent of social and behavioural problems).

There follows a report of the outcome of a single case study involving interviews with an adolescent diagnosed with Asperger Syndrome and with his mother (with the suggestion that there is no lack of desire for friendships, but that there may be inadequate understanding of how to initiate and maintain relationships, or an inability or reluctance to  apply any understanding that has been gained).

The final section explores the matter of face processing (a basic component in developing social awareness) particularly with reference to the competing hypotheses either that there is some fundamental and neurological anomaly which inhibits successful face processing, or that the anomalous performance arises as a result of limited experience/motivation in this area of functioning. 


M.J.Connor                                                                                                   April 2007



Social Skills and Problems in High-Functioning Autism and Asperger Syndrome


In the introduction to their research concerning whether or not high functioning autism and Asperger Syndrome belong on a single spectrum, Macintosh and Dissanayake (2006b) note the existing evidence for considerable similarities in the characteristics associated with these two conditions, which gave rise to the debate whether these are essentially the same for all practical purposes or might still be seen as distinct diagnostic entities. 


They cite current classification systems which advise that Asperger Syndrome can be identified only in the absence of an intellectual disability.  If high functioning autism also refers to autistic signs and symptoms without an associated intellectual ability, the implication is that some other distinguishing features must be highlighted if Asperger is to have the status of a diagnostic category in its own right.


The existing research has sought similarities and differences in cognitive skills, language and communicative abilities, motor skills, medical and developmental histories, their courses and histories, and core symptomatology. 


These current authors recently reviewed the evidence (2004) and concluded that there is a substantial overlap between the two disorders in many of these areas of functioning (with the implication that they belong on the same spectrum).

The problem is that firm and valid conclusions do not seem appropriate in the light of inconsistent findings between studies, variations in methodology, or simply inadequate empirical data concerning key symptoms … limited research on social behaviours in the two conditions is seen as a case in point.


Their current review of studies on this specific area of functioning leads to the suggestion that both of these clinical groups show gross impairments in social functioning compared to typically-developing peers; and that there are many similarities between these two target groups.  Similarities cited include the extent of eye contact during social interchange, participation in conversation, use of non-verbal means of communication, the demonstration of concern for others, and involvement in shared activities and play.

Where differences have emerged, they have tended to suggest more advanced social skills among individuals with Asperger Syndrome (such as more social responsiveness, more frequent/appropriate means of greeting others, a sharing of interests, etc.). 


Some authors, such as Gillberg (1998), have found that the two groups are comparable in their abilities to make and maintain relationships but that those with Asperger Syndrome have a greater interest in making friends. 

Meanwhile, Ozonoff et al (2000) found a greater social competence among young (4 to 5 year old) Asperger children, but this difference was no longer evident by middle childhood onwards.


In a previous study, these present authors (2006a) analysed the spontaneous social interactions among primary school children with one or other of these target conditions and typically developing peers during break and lunchtimes. 

They found few differences in the social behaviours of the Asperger and the high functioning autistic groups (apart from some greater tendency among the Asperger children to make more social “bids” and to be involved in conversations) and both groups were comparable to the typically developing children in various areas (such as shared activities, social initiations, non-verbal communication etc.). 

However, the overall time spent by members of either target group in social interactions was significantly lower than what was observed among the other children, and they were more often observed to be on their own.


The present authors set out on this further study in order to assess social competences among the Asperger and high-functioning autistic children by means of the Social Skills Rating System (Gresham and Elliott 1990) completed by both parent and teachers.

This rating system looks at cooperation, assertiveness, responsibility, and self control, as well as problem behaviours of an internalising or externalising type plus hyperactivity, which could inhibit the development or use of social skills. 

In using parents and teachers as raters, it was hoped to gain complementary information about the children’s social functioning across a number of settings … school, neighbourhood, and home (with the views of teachers particularly helpful in assessing a child’s performance when interacting with a peer group or series of peer groups). 


Their hypothesis was that children with high-functioning autism would be rated by both parents and teachers as being similar to children with Asperger Syndrome in respect of cooperation, assertiveness, etc.; but, should differences emerge, the greater skills would be observed among the sample of children with Asperger Syndrome. 


The participants included 20 children with high functioning autism, 19 with Asperger Syndrome, and 17 typically developing children.  All the children were boys and were aged between 4.4 and 10+ years.  Apart from three of the typically developing group who were at a pre-school, all the children attended mainstream schools. 

The children were assessed for their cognitive abilities, and parents and teachers completed the SSRS ratings.


In support of their hypothesis, the authors found that both sets of ratings highlighted deficits in the social skills of cooperation, assertion, and self control in both of the target groups compared to the controls, and these were evident across a variety of settings.  There was also agreement that children in both target groups had substantially more problems of a hyperactive and an internalising kind than the typically developing children. 

The authors commented that this consensus was all the more significant given the experience of the SSRS as commonly producing divergent ratings between parents and teachers. 


An analysis of observed scores and the norms for the general population of primary school children (boys) confirmed the relative severity of the social skill deficits and, albeit to a lesser extent, the problem behaviours of the Asperger and high functioning autism groups.  The prevalence of hyperactive and internalising behaviours was high; and it was suggested that the combination of social weaknesses and problem behaviours was the key issue behind their limited interaction with peers in the playground. 


The increased incidence of behaviour problems was held to be consistent with existing evidence indicating that children with Asperger Syndrome or high functioning autism are vulnerable to other conditions such as depression, anxiety, oppositional defiance, and ADHD. 

The implication drawn was that children of this age group, like adolescents and adults, have some awareness of their social difficulties and experience frustration as a result.  An alternative interpretation suggested that inappropriate behaviours may arise as a means of seeking the attention and response of peers in the absence of more conventional initiating behaviour.


Meanwhile, the reported levels of competence in the range of social skills assessed in the SSRS suggested that the two target groups were very similar, and this pattern was considered to be in line with observational data gathered about these children in respect of very little differentiability in the nature or frequency of social activities and social behaviours.  

In other words, once at primary school level, children with Asperger Syndrome and children with high functioning autism show very similar social styles/impairments.


However, contrary to what had been expected, the Asperger sample did not appear to show any greater behavioural problems than the high functioning autism group.  This observation was not in line with previous study evidence (such as that of Kim et al 2000) that both externalising and internalising behaviour problems are indeed more prevalent among children with autism or Asperger syndrome than among typically developing children and young people, but that symptoms of anxiety as well as anti-social or disruptive behaviour are significantly more prevalent among those with Asperger Syndrome than with autism. 


Macintosh and Dissanayake held that this kind of discrepancy could be explained in terms of different sampling procedures in that their own study was limited to children of primary school age, whereas previous work has involved participants drawn from a wider range of age groups (to include adolescents and young adults).

The current and previous data were taken as suggesting that, during middle childhood, the risk for comorbid behavioural problems or emotional difficulties is much the same for children with Asperger Syndrome and with autism but that, with increasing age, the two groups diverge such that adolescents and adults with Asperger Syndrome prove to be at greater risk for these behavioural and emotional difficulties.


The authors summarise by highlighting that there are few differences between children of primary school age who have been diagnosed with high functioning autism or with Asperger Syndrome in terms of social deficits or the frequency of behavioural problems.  There was no evidence that any of the social issues are specific to one or other of the conditions. 

Their suggestion is that one might discontinue the practice of using two separate diagnostic categories (Asperger Syndrome/Disorder and Autistic disorder) and use, instead, one category, namely “ Autism Spectrum Disorder. ”


In practical terms, it is seen as likely that, at least for children of primary school age, similar social skill interventions will be appropriate for children with both Asperger Syndrome and high functioning autism.  Both groups may well need, and benefit from, strategies by which to highlight cooperation and assertiveness and self control, while also being subject to monitoring for signs of comorbid difficulties (with a view to early and specialist intervention as required).


Asperger Syndrome and Friendships 


The notes by Howard et al (2006) start by describing friendships among children as a significant context for social and emotional (and cognitive) development given the opportunities afforded to practise social skills, to express and recognise emotions, and generally to develop an awareness of what is involved in sharing and cooperating.


Autism spectrum disorder (ASD), however, is characterised by atypical social performance and by particular deficits such as impairments in the capacity to use or read non-verbal communications, to initiate and maintain interactions, and to demonstrate any true reciprocity.

Such deficits will impact negatively on the ability or opportunity to develop friendships such that children with ASD are commonly found (see, for example, Bauminger and Shulman 2003) to prefer clearly structured activities with explicit rules and routines and limited need for any social exchange.


On the other hand, one might well speculate that such a preference is simply a reaction to the social difficulties and frustrations, rather than some “primary” quest for being left alone. 

There is converging evidence that children with ASD have no less desire than any other group of children for social relationships and that feelings of isolation are common.  For example, Jones and Meldal (2001) found a quest for interactions and friendship to be a common theme among individuals with Asperger Syndrome. 


The suggestion is that adolescence is a particularly challenging time because of the increasing importance of close relationships, and the increasing anxiety about acceptance or rejection by peers.  The young people with ASD may, at this time, experience a lowering of self esteem as they become more aware of their differences and difficulties with respect of making and maintaining friendships. 

Personal accounts from individuals with ASD have confirmed the desire for friendships, and Howard et al emphasise the importance of hearing the views and sentiments of the individuals concerned as a major step towards understanding their experiences and determining how one might offer support. 


Their own study involved tapping the experience of, and feelings about, friendship on the part of one adolescent boy with Asperger Syndrome, along with the observations of his mother. 

The boy in question (“T”) was nearly 13 years old at the time of the meetings. He lived with his parents and two younger brothers. The diagnosis of Asperger Syndrome was made when he was 7 years old.  After attending a mainstream school for two years, T had been educated at home.


Information and perceptions were gathered by means of two interviews with T, between which he was asked to think about, and take pictures of, people and things that he considered important in his life. One interview was held with his mother. 

The topics covered with T included his preferred activities and companions, the development of relationships with identified friends, and how one defines friendship. 

The mother’s comments focused upon her perceptions of T’s friendships and her own involvement in (promoting) his peer relationships.


From an adapted set of items from a quality of life survey, T selected “ having good friends ” as one of his major goals, raising the question about what are the qualities of a good friend. 
A major aspect of friendship was seen to be the sharing of interests and their spending time together … (although the present writer - MJC - did just wonder whether T’s statement that his friend “ likes a lot of the stuff I like ” could be taken as an assumption that the friend is to adapt to T and his preferred activities rather than a need for T to adapt to the friend).


While T identified seeing someone frequently as one aspect of friendship (“ We became friends because we got to see each other ”) Howard et al felt that T  recognised that friendship involves more than simply regular contact and that there are gradations in friendship. 

On being asked what he does to maintain the relationship with another boy identified as a particular friend. T replied that he liked to be a friend and to help him. When asked what makes this other boy a good friend, T replied that … “ He helps me figure out things that I really like; if I am afraid of something, he helps me get over my fear.” 

In other words, there is an apparent awareness of the principle of mutual support and help as reciprocal elements of friendship … (although, again, one might ponder whether there may be an awareness but the actual behaviours/attitudes tend to be a little lacking in reciprocity … ie it is T’s friend who does the helping??)


T went on to explain that he knows this other boy is a friend because “ he cares about me ”, and that friends are people who respond (with talking on the telephone cited as an example of caring and responding). 


On the issue of reciprocity, T appeared to understand something of the principle by stating that a friend is … “ Someone who looks out for you, and you have to look out the same.”  A given example of reciprocity was the returning of telephone calls, and someone could not be a good friend if they do not call back.

The authors were not sure whether or not T fully understood this issue of reciprocity, but they were encouraged that he was at least able to speak in these terms, and to show some awareness of the principle, since a marked lack of social or emotional reciprocity is a diagnostic marker for ASD. 


Their conclusion was that T did have an understanding of some of the basic components of friendship but that his ability to label or to describe some of these components was not necessarily matched by his putting them into practice during actual interactions.


In the matter of some preoccupation or intense interest, it is noted that one potential danger for relationships is the assumption on the part of the young person with ASD that the (potential) friend shares that preoccupation or interest. 

Limited ability to recognise the feelings or perspectives of another person is one of the deficits associated with ASD, but T was able to say that he may need to put aside his own interests or get interested in what someone else liked to do.


Both T and his mother described the two ways she tries to encourage social relationships … namely by frequently inviting other young people over to meet T, and by offering specific advice or direction about how he might act in a given situation.  An example of the latter was reminding T that the other person may not be interested in what T is saying or doing, and should be asked what (s)he would like to talk about or to do …. (rather akin, one might say, to the way in which parents help along initial interactions between much younger children).


This kind of pattern was described by the authors as common and they refer again to Bauminger et al (op.cit) who reported that parents of children with atypical development often play the role of initiator and supervisor of their children’s budding friendships. 


Howard et al conclude that T’s level of understanding of friendships, and his desire for friendships, should act as a reminder that the observed deficits in social skills and impediments to interactions should not obscure the motivation towards social acceptance and relationships on the part of young people with ASD.  There may be obstacles but no lack of interest in forming friendships among many or most of the young people concerned. 

The critical issue seems to be that of translating this interest and basic understanding of some of the elements of friendship into practical and effective actions, with implications for further studies by which to gain more awareness of how to make this transition, and how the families may be most supportive in the matter of adolescent friendship development.


Atypical Face Processing     


It has been hypothesised that a lack of early social motivation or social interest among children with autism may reduce attention to faces at an early developmental stage; and there is evidence (see, for example, Dawson et al 2002) that autism may be characterised by impairments in face recognition and limited attention to faces ….

(although one would presume that this characteristic does not apply to any significant extent among children eventually identified with high functioning autism or Asperger syndrome).  

The hypothesis continues by suggesting that certain aspects of face processing, such as the identification of emotion conveyed by expressions and the monitoring of the direction of gaze, are critical for the development of theory of mind and social relationships. 


So begins the paper by Webb et al (2006), and their review of studies goes on to highlight the typical pattern whereby infants rapidly begin to show a preference for faces over other visual stimuli, and prefer to look at the mother’s face rather than any other female face.

The typical pattern, revealed by this review, continues with the development at around 4 months of age of the inversion effect where the infant is more efficient at processing upright faces than inverted faces. By 6 months there are differentiable brain activations in response to familiar faces or objects and to unfamiliar ones; and shortly after this, the infants are able to differentiate between a negative and a neutral facial expression.


From about 2 to 4 years, the brain activation pattern shifts in response to faces in that the greater reaction to a familiar face among the younger children in this age range is replaced by an emerging pattern of greater activation in response to novel faces. 


These authors note that the social impairments in autism, such as limited eye contact or difficulty in interpreting emotions conveyed by facial expressions, are linked to the capacity to focus upon and to interpret information from faces, and difficulties in this area could be one of the earliest diagnostic signs of autism.

Osterling and Dawson (1994) carried out a study involving video recordings of children at their first birthday parties and found that a failure to look at other people was the single best discriminator between children who were later diagnosed with autism and those who showed typical development.


Meanwhile, studies involving the discrimination of, and memory for, faces have shown that children with autism perform more poorly than controls matched for age and ability; and one hypothesis holds that the children with autism attempt facial recognition and processing by anomalous strategies.

For example, Joseph and Tanaka (2003) found that, by middle childhood, typically developing children are better at recognising parts of a face when the parts are presented in the context of a whole face, and perform better when seeking to discriminate eyes than mouths.  There also remains the preference, and better memory, for upright faces compared to inverted faces.


However, children with autism are better at recognising isolated facial features and partially concealed faces than their typically developing peers. They also show better memory for the lower half of a face than the upper half; but generally show reduced attention to the major features of the face, like the eyes and nose, relative to their typically developing peers.


Webb et al note the significance of these studies but point out that many used methods which placed a reliance upon the children’s verbal ability and, therefore, the participants were drawn from older and higher functioning children.  However, the use of electrophysiological techniques (to observe and measure brain activation) does not require a verbal response so that they can be employed with younger children and lower functioning children in order to gain some insight into the neural systems involved in face processing among typical and atypical groups of children.


They describe their own previous studies which have demonstrated anomalous brain activation patterns among pre-school children with autism when presented with images of their mothers’ faces compared to images of an unfamiliar face.  No differential patterns of activation were noted; and the activations in response to a familiar and to an unfamiliar toy were also non-differentiated. 

Further, and still unlike typically developing children, the children with autism did not show a different (greater) response to a face expressing fear than to a face with a neutral expression.


This current study by Webb et al set out to explore, via an analysis of event evoked potentials (brain activation responses) whether children with autism have impairments in an early stage of face processing 

The participants were three groups of children between 33 and 54 months.  They included an ASD group (N=63), a group of children with developmental delay (N=37), and a group of typically developing children (N=28).

The face stimuli consisted of two photographs of faces, one familiar to the child

(mother) and the other unfamiliar. The object stimuli paired a familiar item, such as a favourite toy, with an unfamiliar object but matched for size, shape, and type. (The examples were given of a toy fire engine paired with a monster truck, or a backpack paired with a bag.)

Brain activation in terms of evoked potentials was measured following training sessions by which to acclimatise the children to the procedures.


The results showed that, among typically developing children, the rate of response to faces was faster than the rate of response to objects, specifically in the right hemisphere. 

In contrast, the ASD and developmentally-delayed groups showed no face-specific hemispheric response. The DD group also showed no differentiable responses to faces and objects in respect of both speed and magnitude of response; while the ASD group were similar to controls in showing a more negative response to faces than to objects, but showed a faster response to objects than to faces.


The authors summarised the findings by noting that the children with developmental delay (the age and ability matched controls) did not show differential responses to faces and objects, and inferred that children with autism do have an anomalous brain response to faces and objects compared to children with developmental delay and to typically-developing children.


In greater detail, it was noted that the children with autism showed largely typical patterns of brain responses to faces, in respect of extent and speed, in the right hemisphere, but showed abnormalities in the speed of processing of faces in the left, and in the amplitude of responses to objects.

The implication was for abnormal cortical specialisation.


However, it is not clear whether these anomalies among the children with autism are the result of some innate disturbance in face processing capacities, or whether it is a matter of (limited) experience in that there is a history of some lack of attention to faces and greater attention to objects which influences the development of neural circuits involved in face processing.


In other words, one possible explanation refers to some basic impairment in the brain structure that would normally become specialised for the processing of faces or in the mechanisms/circuitry required for effective processing.

The second possible explanation concerns a lack of visual experience with faces from which arises a deficit in face processing.   


While children with autism do not appear to experience any perceptual deprivation, there is a converging view that these children may commonly lack the social motivation for focusing attention upon faces.

Appropriate attention to faces would underlie the development of a face processing skill and the organisation of the necessary neural systems.


The authors suggest that more research data are necessary in order to gain support for one or other of these hypotheses.  Nevertheless, this present work is described as newly defining an early capacity/trend whereby typically-developing children respond preferentially to faces, and as confirming an autistic anomaly in respect of the rate and amplitude of responses to faces and objects (with a need for further work by which to determine the reliability of these current findings.)


                                  *          *          *          *          *          *


M.J.Connor                                                                                                   April 2007                         





Bauminger N. Shulman C., and Agam G.  2003   Peer interaction and loneliness in high functioning children with autism.   Journal of Autism and Developmental Disorders  33(5)  489-506


Dawson G., Carver L., Meltzoff A., Panagiotides H., McPartland J., and Webb S.

2002   Neural correlates of face and object recognition in young children with autism spectrum disorder, developmental delay, and typical development.   Child Development  73  700-717  


Gillberg C.  1989   Asperger Syndrome in 23 Swedish children.   Developmental Medicine and Child Neurology  31  529-531


Gresham F. and Elliott S.  1990   Social Skills Rating System.  Circle Pines, MN; American Guidance Service


Howard B., Cohn E., and Orsmond G.  2006   Understanding and negotiating friendships.  Autism  10(6)  619-627


Jones R. and Meldal T.  2001   Social relationships in Asperger Syndrome.   Journal of Learning Disabilities  5(1)  35-41


Joseph R. and Tanaka J.  2003   Holistic and part-based face recognition in children with autism.   Journal of Child Psychology and Psychiatry  44(4)  529-542


Macintosh K. and Dissanayake C.  2004   The similarities and differences between autistic disorder and Asperger’s disorder.   Journal of Child Psychology and Psychiatry  45  421-434


Macintosh K. and Dissanayake C.  2006a   A comparative study of the spontaneous interactions and social skills of children with high functioning autism and children with Asperger’s disorder.   Autism  10  199-220


Macintosh K. and Dissanayake C.  2006b   Social Skills and problem behaviours in school aged children with high functioning autism and Asperger’s disorder.   Journal of Autism and Developmental Disorders.  36  1065-1076


Osterling J. and Dawson G.  1994   Early recognition of children with autism.   Journal of Autism and Developmental Disorders  24  247-257


Ozonoff S., South M., and Miller J.  2000   DSM-IV defined Asperger Syndrome.  Autism  4  29-46


Webb S., Dawson G, Bernier R., and Panagiotides H.  2006   ERP evidence of atypical face processing in young children with autism.   Journal of Autism and Developmental Disorders  36  881-890                                      

This article is reproduced by kind permission of the author.

© Mike Connor 2007.

Back to NAS Surrey Branch Welcome Page