Following a further reminder of the desirability of early diagnosis (and some means of differentiating ASD from other conditions) in order that interventions can also be initiated optimally early, reference is made to two projects designed to increase  shared attention and engagement of children with autism.

A common link is the need to enter the child’s world by seeking to share/imitate his or her activities as opposed, in the first place, to seeking to evoke a particular behaviour.



M.J.Connor                                                                                              January 2005





In referring to interventions, there is raised the implication for necessarily early and accurate identification of the needs in the first place ….. all the more significant in the light of converging evidence for the ways in which disabilities can overlap or become confounded.

There is the associated risk that strategies may not be appropriately targeted such that there may be no observable benefits and even, perhaps, some exacerbation of the signs and symptoms.


This being so, the advice of Perry (2004) is all the more valuable in assisting to differentiate indications of Asperger Syndrome from those relevant to Attention Deficit Hyperactivity Disorder or Obsessive Compulsive Disorder.


The advice is based upon lengthy clinical experience and offers guidelines by which to increase confidence in an Asperger diagnosis.


Reference is made to the frequency with which attentional difficulties and impulsiveness may be observed among samples of children with Asperger Syndrome, with the possibility that a diagnosis of ADHD will be made and the prime need will not be addressed.

However, it is held to be significant that the Asperger children can focus on books or other stimuli which are relevant to their particular or circumscribed interests.  Their initial presentation as young children may reflect a lack of focus, but this picture is likely to change with time and as these interests, which may reach the point of some obsessiveness, are developed and pursued.


The children with ADHD will not demonstrate this pattern, nor will they show the limited or atypical social style and behaviours which are characteristic of Asperger Syndrome.


Meanwhile, the highly focused or circumscribed interests may also be confounded with those applying to children where the prime need relates to Obsessive Compulsive Disorder.

Perry’s experience indicates that, the earlier the start of these interests and preoccupations, the more likely it is that the central issue is about Asperger Syndrome …. cars, trains, maps, computers, and video games, cartoon characters, historical facts, and numbers, are the most common preoccupations in this sample of children.

Further, the intensity of the interest, far exceeding what may be observed in the way of interests among normally-developing children, sets the children apart from peers and may even underlie an alienation, given that the interests are not shared nor do the children show a meaningful social reciprocity.


Further, the Asperger children do not show the compulsive “style” which may involve constant repetitions of actions, going back to the start if some ritualistic action is interrupted, and constant checking.
There may be some repetitive and self-stimulatory behaviours which appear similar to compulsive behaviours but they tend to decrease with age.


Another possible confusion is with Bipolar Disorder given that Asperger Syndrome may be reflected in aggressive behaviours or anxiety or dramatic shifts in emotional state. 

However, such behaviours seem to be a matter of the lack of social perception and failure to recognise the feelings or rights or perspectives of others leading to some invasion of others’ space or demands upon them, while showing anger when anyone else attempts to encroach upon their activities or space. 

Such behaviours are likely to increase in intensity over time, especially as new situations or settings are more frequently presented with implications for anxiety and a tendency towards some avoidance or withdrawal behaviours.

These reactions are all the more likely given the difficulty that the Asperger children have in recognising or describing their own feelings and emotions.


Finally, it is noted that Asperger Syndrome may be associated with sensory and language anomalies, but Perry comments that social problems in the school setting are more likely to reflect memories for earlier confrontations, albeit seemingly trivial to external observers (assuming that they noted or remembered them at all), rather than some specific language-based problem which may be a complication but not the whole cause.


Promoting Attention and Language


Smith et al (2004) describe an approach which seeks to enhance the skills of young children with ASD in respect of sharing social attention and actions thus to foster meaningful communication. 


The initiative was based upon the observed and characteristic trend among children with social communication and autistic difficulties to avoid interactions and to indulge in lone play activities.  In fact, such children may be confused or even intimidated by the actions and approach of others.

A further basis for the initiative was the converging evidence (see, for example, Camaioni and Laicardi 1995) that the processes underlying a sharing of actions, interpreting emotions, and using language to communicate in a  socially meaningful way, are located at the pre-verbal stage of development.


A review of relevant findings completed by Smith et al indicates that the quest for two-way communication and understanding begins at the earliest part of post-natal life with the sharing and mutual imitation of facial expressions and movements. During the early months, the infant and the mother show initial turn-taking in these forms of communication, with the infant gradually recognising the significance in the facial expression, tone of voice, and speech prosody of the mother and of other people.

Meanwhile, the infants begin to recognise the impact of their own utterances and to experiment with variations in their expression.


In this way, the infant and the care-giver gradually develop a system of sharing meaning using movement and sound. There is an intention behind the early communication attempts which is realised through repeated interactions as both partners learn to understand and predict each other’s intentions and behaviours.


Shared and meaningful language fosters the explicit exchange of intention, which, in turn, underlies wider sharing of attention and more advanced communication. 

Words acquire their meaning through their use as a kind of commentary or description of joint activities.


Where young children have a severe delay in their communication skills, as is the case among many children with ASD, there will be difficulties in sharing activities with other people, and the children concerned may be unable to link their play and actions or utterances with any reference to what those other people say or do.

A lack of shared attention has been linked to the social deficits which are expressed as an absent or deficient set of theory of mind skills.


Meanwhile, it has long been recognised (see, for example, Cunningham et al 1981) that interaction patterns with adults become altered because of the non-responsiveness of the infants.  The adult, unable to share play, or lacking overtures from the infant to which to respond, may become more directive of the infant’s actions with an inhibition of the kind of turn-taking and reciprocity which underlie the enjoyment and the positive development of communications.


A practical problem is that children who are eventually diagnosed with significant problems in communication may only be referred for initial assessment as they approach 2 years of age.  By this stage it is too late to focus upon facial attention and basic interactions of the kind relevant to neonate development, and compensatory interventions have to be delivered through shared and structured play contexts.


With this background, Smith et al (op.cit) describe the emphasis in theories of language development upon game formats by which to build the skills necessary for sharing a conversation … the roles of listener and speaker may be modelled through turn-taking games ; and shared play provides the opportunity to develop the understanding of language and social rules.


Further, it is noted that the children with ASD find it difficult to cope with sensory input in more than one channel at a time… they cannot cope readily, for example, with visual and auditory input at the same time.  Accordingly, the intervention described minimises the complexity of new information by using the child’s existing play skills, albeit in a structured and interactive way.


The intervention takes the children’s existing capacity to manipulate objects into a shared and turn-taking game.  This kind of play is intended to foster shared visual attention with the adult, and visual monitoring of the actions within game linked with the facial expressions or bodily movements of the adult partner.

Thus the children learn the meaning of the adult’s reaction as well as learning how to demonstrate their own wishes or intentions by the use of eye contact or gesture or utterances.  The game itself is not important, but what matters is the exchange of turns and the shared attention and mutual imitation.


Reference is made to Bruner (1983) who described 4 stages of building involvement from the role of observer to that of active participant when the child joins in with most or all of the game ; subsequently, the child shows the willingness and capacity to initiate the game with the other person ; and the final stage involves some ability to modify the actions or moves in the game through verbal or non-verbal communicatory means.


The interaction games themselves begin with an emphasis upon shared attention where the child is given access to a range of toys and objects, and the parent or teacher plays with a similar toy to that chosen by the child as a kind of social mirroring.  The child is prompted to look at the adult’s actions. 

Subsequently, with interest aroused, the adult copies the child’s actions with the toys or objects and encourages the child to watch this imitation of his own behaviour. The child is encouraged to repeat or to change the actions made in order to bring about a change in the adult’s behaviour.  The experience is that of intentionally controlling what someone else is doing.


The next stage involves the introduction of a third person so that turn-taking and role shifts become more complex

The final stage involves encouraging the child to take on increasingly the initiator role with access to peers as potential partners, with the (long-term) goal of establishing the social rules and language of a negotiator/organiser.


Evaluation measures, comparing baseline performance against performance after the intervention, used with the sample of children between the ages of 3 and 5 years (all with severe communication delays), included the recordable changes in the level of social engagement in game formats, and speed of progression through the stages of the interactions.  Language measures included raw scores in the WPPSI verbal scale, naming vocabulary test in the BAS, mean length of utterances, and tests from the Scales of Pragmatic Communication Skills (Dewart and Summers 1988).


Observations indicated a clear improvement in the children’s overall level of participation in game formats.  Further, positive changes in all the WPPSI subtests reached a significant level.  Pragmatic skill and utterance length also showed positive changes.


The authors concluded that the rate of progress in game-participation and language use was greater during the intervention period than during the baseline or control phases.


A follow up study involving children between 27 and 42 months of age was completed with children drawn from those referred to the Portage Service and whose language level was 2 SDs below the norm for the age group. An experimental and a control group were established.  The experimental group received input of the above type, designed to promote early interactive conversations, instead of the standard Portage language input.

While the differences did not reach a statistically significant level (presumably because of the relatively short period involved [5 months]), the experimental group achieved higher scores in all the subtests used in the Pre-School Language Scale, with expressive language showing the greatest positive discrepancy in favour of the experimental group.


The overall conclusion suggests the potentially high value of this approach, using the child’s existing repertoire of skills into which the adult seeks admission (as opposed to encouraging the child to join the activities established by the adult).

There is no excessive talk, but a gradual sharing of play and development of communication.

Further, the authors comment that there is not the artificiality that may be said to apply to behavioural approaches, nor the problems in generalising newly acquired skills to different settings.


(The present writer - MJC - would emphasise that this approach, and the approach to be described below, both focus upon entering the world of the child with ASD and joining/imitating his or her activities.  It is not a matter of persuading the child to join the adult’s world or the typically-developing child’s world and to take on their actions … which could well prove meaningless at best and possibly threatening at worst. 

One could describe the general approach as a variant of the SonRise strategy, where the child is accepted as (s)he is, and interventions begin from the child’s existing level and choice of behaviours.)


Fostering Interaction between Children with ASD and Mainstream Peers


In his introduction to a project on shared play between children with ASD and their mainstream peers, Whitaker (2004) describes how it was stimulated by the observed increase in the numbers of children with ASD who are being educated in mainstream schools, with, presumably, the goals of maximising their contact with a wide range of other children and of gaining benefits from the interactions which can be established.


Whitaker expresses the concern that the rhetoric of inclusion is not matched by the reality, at least in terms of the benefits perceived by parents of children with ASD. 

While noting that it is not clear how representative was the sample, reference is made to a survey by the NAS (Barnard et al 2000) which found that parents of children in a specialist ASD school were twice as likely to be “very satisfied” with the education available compared to parents of children attending mainstream schools.


Whitaker recognises how autism-specific settings can provide a curriculum and teaching approaches geared to the specific styles of the children with ASD, but he notes that they may have the disadvantage of affording little or no opportunity for interactions with non-ASD children.

This raises the question of how to take advantage of the potential for interaction in the case of children attending specialist units or classes which are based in mainstream schools.


There is the further problem whereby children with ASD show little capacity for incidental learning, and delayed or disordered social skills are a characteristic feature of ASD. 

This would suggest that simple proximity or incidental contact between children with ASD and their peers are unlikely to bring about meaningful communication or social interactions. 

Experience suggests that children with ASD tend not to make attempts to initiate communication in the integrated school setting ; nor do the mainstream peers commonly make a response to what initiations are made.

Instead, the communication and interaction need to be orchestrated and specifically fostered.


However, there is a risk, as set out by Roeyers (1995), that interventions which involve a directive style on the part of peers can lead to an artificial and intrusive form of interaction with children with ASD.  Further, while there may be some increased level of interaction between the ASD children and the peers who have received some training about initiating interactions, it may be the case that spontaneous interactions with untrained peers are reduced.  


The project described by Whitaker was sited in a village primary school where there is a 10-place unit for children with high levels of ASD and moderate learning difficulty.

The underlying principle was that social engagement and responsiveness can be established and/or increased as a result of imitating the actions taken by the ASD child.

The aims of the project involved increasing the frequency with which the 6 and 7 year old ASD children initiated contact and communication, and exploring the extent to which peers could establish shared play with them.


All the children in the host school had been given some awareness raising sessions about the nature of autism and all had shared classroom contact with children in the ASD unit. However, it was decided to recruit only from the oldest class (Year 6) and to use the teachers’ judgements about the likely commitment and competence by which to select the peer tutors (with parental understanding and agreement gained).


The first step was for the mainstream children to meet their ASD counterparts individually; subsequently, play sessions of around 20-30 minutes took place weekly during lesson time under the supervision of a nurse or LSA from the unit.  


To start with, the peer tutors were simply encouraged to see if they could get their partner child to play with any of the games and activities available.  A baseline video of the extent of interaction was recorded after the initial 3 or 4 sessions.

Further training was then given to the peer tutors with a view to enabling the children to help their partner to enjoy shared play.  The critical principles or steps were …. Get close …. Follow his or her lead …. Talk slow and talk simple …. Make it fun.


Regular play sessions continued for the next two terms, with each pair involved in between 20 and 24 sessions, with the emphasis remaining on the peer tutor joining any activity spontaneously chosen by the ASD partner.  A further video recording was made after a minimum of 20 weeks, with analysis of selected blocks of time to determine the extent of joint attention behaviour, communication, and shared play.


Additional evaluation was completed by means of structured interviews with each tutor just before the end of the project, covering such matters as the level of enjoyment gained, the perceived attitudes of other children and family members, the extent of change in the “buddy”, whether or not time is spent with the buddy outside the scheduled sessions, etc..

The parents of the peer tutors were also interviewed in respect of their perceptions of the advantages and disadvantages of their child’s being involved in the project.


Observations and analysis indicated that the peers proved very competent at following the lead of the ASD child, following his or her focus of attention, and establishing some shared play.  Meanwhile, the target children had established within 3 or 4 weeks a keen anticipation and pleasure about the sessions, and showed no signs of stress from the close presence of the mainstream partner who rapidly learned what were the favourite play activities.


Comparison of the baseline and follow-up video records indicated a significant increase (3-fold) in the frequency with which the ASD children spontaneously made requests of their partners; there was also an increase in the number of comments where the ASD children verbally labelled some aspect of the immediate situation. 


However, it was acknowledged that the change was gained from a very low baseline and also that communications for any other purpose remained very rare.

It was also held to be disappointing that joint attention behaviours remained very limited … behaviours such as showing or giving the partner a toy, or seeking to attract attention verbally or non-verbally.


On the positive side again, with only one exception the tutors showed great enthusiasm for the project and reported gaining much enjoyment.  Some initial anxiety was reported, with the tutors describing their worry lest the partner would not choose to play with them at all. There was also some embarrassment about playing with toys or games suitable for the much younger age group ; but despite this, and despite the repetitious nature of the activities, the sessions were rated positively and as “fairly interesting” or “very interesting”.  

High points commonly involved some spontaneous remark on the part of the ASD children, and low points typically involved the occasions when the tutors felt they were being treated as if they were invisible.


Classmates in the mainstream were also said to be supportive and quite envious of those involved in the project.  Problems of missed work were minimised by scheduling the play sessions during time not used for academic subjects.

The presence of staff was seen as significant in terms of the availability of encouragement and praise when difficulties were met, and being ready to intervene should any signs of challenging behaviour arise, along with advice about engaging the partner, especially in terms of ensuring simplicity of language.


The parents were also positive, believing that the project enabled their children to become more aware of, and to respect, children with differences and disabilities ; and they felt it important to give their children the opportunity to gain experience of helping others and to develop a sense of responsibility for helping other people.


In his summary and conclusion, Whitaker held that the project provided an example of how one could capitalise upon the siting of a specialist ASD unit on a mainstream campus, and upon the goodwill of the mainstream peers, parents, and teachers … even if the interactions evoked tended to be limited.  However, the message gained is that, even with little time necessary for preparation, mainstream peers can engage their ASD counterparts and establish relationships which appear clearly to be enjoyed by those children.


The absence of a meaningful set of controls in this project is recognised by Whitaker, along with the modest outcomes. However, the severity of the level of ASD experienced by the target children is also noted.

Implications include the likely benefits from increasing the contact time … a single session each week would not be expected to bring about major change.

Further, the ongoing project now involves tutors who only take part for one term but who share a greater frequency of sessions in that term, not only reducing missed mainstream opportunities but also maximising the energy and enthusiasm of the tutors with anticipated greater probability of engaging the target children. (There would also be some reduction of a risk lest too great a familiarity with the ASD child’s style and preferences would limit the opportunity or need for the ASD child to initiate communication and shared attention.)


Whitaker’s final point highlights how physical proximity and imitating the lead taken by the child with ASD can establish a foundation of shared play and engagement. Another possible implication is for the likely benefits of further training of peer tutors in order for them to become more proactive once the initial engagement has been established via the reactive (imitative) approach.



                           *          *          *          *          *          *          *


M.J.Connor                                                                                              January 2005







Barnard J., Prior A., and Potter D.  2000   Inclusion and Autism : Is It Working ?

London : National Autistic Society.


Bruner J.  1983   Child’s Talk.   Oxford : Oxford University Press


Camaioni L. and Laicardi C.  1995   Early social games and the acquisition of language.   Journal of Developmental Psychology  3  31-39


Cunningham C., Reuter E., Blackwell J., and Deck J.  1981   Behavioural and linguistic developments in the interaction of normal and retarded children with their mothers.   Child Development  52  52-70


Dewart H. and Summers S.  1988   The Pragmatic Profile of Early Communication Skills.  Windsor : NFER-Nelson


Perry R.  2004   Early diagnosis of Asperger’s disorder : lessons from a large clinical practice.   Journal of the American Academy of Child and Adolescent Psychiatry  43 (11)  1445-1448


Roeyers H.  1995   Peer mediated proximity intervention to facilitate the social interactions of children with a pervasive developmental disorder.  British Journal of Special Education  22(4)  161-163


Smith C., Goddard S., and Fluck M.  2004   A scheme to promote social attention and functional language in young children with communication difficulties and ASD.   Educational Psychology in Practice  20(4)  319-333


Whitaker P.  2004   Fostering communication and shared play between mainstream peers and children with autism.   British Journal of Special Education  31(4)  215-222


This article is reproduced by kind permission of the author.

© Mike Connor 2005.

Back to NAS Surrey Branch Welcome Page