This latest summary of recently published material includes both theoretical and practical ( in-class ) matters.

Initially, the concepts of central coherence and hierarchization are compared and contrasted ; followed by a comparison of the skills or deficits among children with autism or with language impairment in the processing of information conveyed by faces and voices, with implications both for the validity of the concept of a continuum of disabilities and for the need to avoid assuming among the language group the skills of extracting meaning from what they hear and see.

Reference is then made to a number of studies which have explored the means of enhancing social skills among children with autism, and increasing their peer interaction, with a growing consensus that these targets may best be achieved by directly involving the other children and young people, whether at the early or later stages of schooling, in the intervention.

The final section quotes continuing information and opinion about the possible link between the MMR vaccination and autism, and concludes that the the safety of the vaccine has been further highlighted but that not all parents ( or professionals ) are wholly convinced by the reassurances offered.

Rodgers (2000) introduces her report by noting how studies among individuals with autism have consistently revealed abnormalities in all sensory modalities, such as hypersensitivity, perceptual distortions, or sensory overload. With regard to visual perception, there has been shown to be unexpectedly good performance on certain tasks such as Block Design, along with resistance to visual illusions, or a failure to recognise the impossibility of impossible figures.

There have been two hypotheses by which to explain these findings : one relates to a deficit in Central Coherence ( Frith 1989 ) and the other to a deficit in Hierarchization ( Mottron and Belleville 1993 ).

Simplistically, one could regard Central Coherence as a matter of the way sensory input is organised and processed and interpreted, while Hierarchization appears more to do with the initial perceptual and input mechanisms or idiosyncrasies.

However, most of the research in this area has concerned individuals with autism rather than with Asperger syndrome, but there has been an assumption made that the latter is simply a form of autism without marked learning difficulty so that the nature of perceptual anomalies and the reasons behind them would be the same for individuals diagnosed with autism and those diagnosed with Asperger syndrome. The study by Rodgers set out to examine visual perception in Asperger syndrome in order to test this assumption.

The Central Coherence hypothesis suggests that the individual with autism has difficulty in gaining meaning from what is observed. There is a tendency to examine separate or unconnected stimuli, or parts of the stimulus, rather than perceiving a meaningful whole. There is a preference towards processing individual or local features of the stimulus rather than global features hence the superior performance on tasks such as Block Design or identifying embedded figures. In other words, the typical cognitive style in autism involves " field independence ", indicative of a superior ability to examine elements of a stimulus separately.

An example is that of reading a sentence in which a homophone appears, such as the word "tear", and the normally developing individual identifies how to pronounce and how to interpret that word by rapidly scanning or remembering the rest of the sentence for context and meaning. In the autistic individual, this capacity may be lacking and each word is read separately and not in accordance with the rest of the sentence.

The Hierarchization hypothesis suggests that the individual with autism does not identify those elements of a stimulus which are of most significance .... there is no hierarchy imposed upon the various aspects of the perceptual stimuli.

Among non-autistic people, it is usually found that certain aspects of a stimulus are processed more efficiently than others ; in an autistic sample, there would be a tendency to treat all the aspects similarly, and a lack of capacity to switch from global to local processing according to the nature of the stimulus or of the task. This is illustrated by the typical finding that a person with autism appears unable to identify impossible figures ( such as the well-known picture of a set of stairs which do not go up or down but which maintain a loop ). Again the difficulty is in integrating the various elements of the stimulus into a meaningful whole.

The two hypotheses are actually quite similar in that both predict problems in the day-to-day perceiving of meaningful wholes and global features of images among individuals with autism. However, the two can be clearly differentiated in that the Hierarchization view would suggest that, for autistic subjects, no differences will be found in the rate of identifying local or global features, while non- autistic individuals should be more efficient in processing the global features.

Similarly, autistic and non-autistic individuals can be contrasted in regard to the Central Coherence view in that an autistic group would be predicted to display a preference for the local elements of a stimulus, whereas a control group would display a tendency to process global features more efficiently.

With a sample of individuals with Asperger syndrome, Rodgers noted that the Central Coherence hypothesis would predict that the participants would be superior to a control group in dealing with Block design, and in identifying embedded figures ; but, unlike the control group, they would show no difference in their processing of global or local features in a hierarchical stimuli task. ( This kind of task involves, for example, presenting a series of large letters which are actually made up of small letters, and the participants are asked to read as rapidly as possible either the whole large letters or the small letters .... the prediction is that control subjects will have more difficulty than the autistic group in identifying the small letters because of the interference from the overall large letter ).

The Hierarchization hypothesis would predict that, in the hierarchical stimuli task, the autistic group would also show no difference in speed of detecting local or global features and would make few errors at the local level because of minimal impact from interference, while controls would be more efficient in processing global features. In an impossible figures task, the Asperger group would be predicted to display significantly poorer performance than the control group.

A sample of adults diagnosed with Asperger syndrome, and a control group, matched for age, gender, handedness and IQ, took part in the study involving tests of Block Design, hierarchical stimuli, impossible figures, and embedded figures.

In the Block Design task, no significant differences in the rate of performance were found between the groups. However, participants in the Asperger group failed to complete the task on 21% of occasions compared to only 5% of occasions for the control group.

In the hierarchical task there were, again, no significant differences in processing time between the global and local level for individuals in either the Asperger group or the control group. There was also no difference between the groups in the number of global or local errors made.

When it came to the impossible figures task, the two groups did differ significantly in that the Asperger group showed many more errors than the control group.

With the embedded figures task, there was no difference observed between groups in the time taken to detect the embedded figures. The central coherence hypothesis predicts that the Asperger group should display superior performance. However, although the difference did not quite reach significance ( Mann-Whitney Test : p = 0.06 ) , the findings were in the opposite direction with the control group displaying faster reaction times.

In her discussion, Rodgers restated the aim of the investigation as gaining evidence whether the Central Coherence hypothesis or the Hierarchisationn hypothesis provides sufficient explanation of the visual perceptual style among adults with Asperger syndrome.

The results of the Block Design task, the hierarchical stimuli task, and the embedded figures task, did not support the predictions from the Central Coherence hypothesis. The Asperger group did not display a preference for local stimuli on the hierarchical task nor a superior performance on the embedded figures task.

The results relating to the Hierarchization hypothesis were less clear. This view would predict that the Asperger group would perform equally well on global and local processing, and this was found to be so.

On the other hand, this view also predicts that, among normally developing individuals, global level stimuli would elicit more efficient processing so that one would expect the control group to show a global level bias in completion rates or avoidance of errors. This was not found to be the case as the control group also displayed no significant differences between global and local levels.

Stronger support for this hypothesis emerged from the impossible figures task in that, as predicted, the Asperger group had particular difficulties with this task, and the performance of the control group was superior.

The author recognises the the limited number of participants involved in this study ( 8 subjects in the target group and 8 controls ) but she still makes the tentative suggestion that the Hierarchization model may be a more appropriate explanation of the perceptual functioning in Asperger syndrome than the Central Coherence model.

The next step would involve larger scale studies comparing individuals with Asperger syndrome against individuals with high functioning autism in order further to explore whether either of these models can be shown significantly to predict outcomes in perceptual tasks.

Rodgers holds that, if one can identify differences between autism and Asperger syndrome in perceptual functioning, this would add to the debate concerning whether they are simply to be seen as different levels of impairment along a continuum or can be seen as clearly differentiable diagnostic categories.

Rodgers quotes Schopler (1985) as stating that the term " Asperger syndrome " should not be used unless and until an empirically-based distinction from autism can be demonstrated, and she concludes by suggesting that further investigation of these perceptual functions may provide just that distinction.

Boucher et al (2000) introduce their work by noting how voices and faces are the most important sources of information for infants who are learning about other people, and also the most significant channels for non-verbal communication between infants and their carers.

The processing of this kind of information appears to begin in neonates and develops rapidly, so that, for example, neonates will respond to voices in preference to other stimuli, and can recognise their mothers' voices . At 1 week, babies appear able to recognise the intonation of maternal voices ; and by 6 months, they can respond to the affective content of tone of voice as well as to facial expression.

However, the child with autism does not possess this kind of social awareness or non-verbal communicative school ; and a number of studies have indicated that there are perceptual

encoding anomalies and impairments in the recognition of facial expression or speech. On the other hand, it is recognised that, while children with autism can be differentiated from controls on tests of face recognition, the two groups will overlap.

There is also evidence to suggest that voice processing in children with autism is impaired and a study by Klin (1992) found the young children with autism differed from normally developing children or from children with learning difficulties in that they did not show a preference for listening to their mothers' voices rather than to other speech stimuli.

The authors referred to an earlier study (1998) in which they demonstrated that school age children with autism have significant difficulty compared to a group of language-impaired children in matching familiar voices and faces. The autistic children were also significantly poorer in recognising familiar faces or in recognising familiar voices when compared to the performance of children with specific language difficulty.

The authors suggested that this finding may reflect "simple" impairments in face recognition and voice recognition, but they acknowledged that it could also be a matter of some cross- modal processing deficit. They cite evidence of such an impairment in children with autism such as the study by Bryson (1972) who showed how children with autism performed less well on a task where the spoken word "red" had to be matched to a red object than on a task in which 2 red objects had to be matched.

Their current study therefore set out further to investigate a voice processing in autism; and a series of experiments were implemented. The participants included a group of 19 children with autism, with a mean age of 9-7 ; a group of 19 children with specific language impairment, with a mean age of 9-0 ; while 19 mainstream children, mean age 6-4, took part in experiments 3 and 4.

The first experiment involved the matching of familiar voices and faces but also examined the ability to match objects and sounds. The aim was to assess whether the voice and face matching impairment is specific, and to clarify whether a cross-modal matching impairment may be involved.

The second experiment examined the recognition of familiar voices, with the prediction that children with autism would show impaired recognition of familiar voices compared to control children.

The third experiment assessed the ability to discriminate between unfamiliar voices with the prediction that autistic children would be as competent in this task as control children given the existing evidence that facial discrimination is unimpaired in autism and that there are frequent parallels between voice and face processing skills in normally developing children.

The fourth experiment tested the ability to match the affective content of speech and facial expression, again to investigate the possible role of some cross-modal matching impairment, with the prediction that children with autism would be relatively poor on this task.

The authors assumed that children with a language impairment are not impaired in the processing of socio-emotional stimuli. They went on to predict that children with autism would be more impaired than the language group on a test of matching familiar voices with faces and on a test of familiar voice recognition.

It was further predicted that the autistic group would be poorer than the language group or normal controls in naming vocally expressed emotions or matching vocally and facially expressed emotions.

The final prediction was that the autistic group would not be impaired relative to the other groups on a test of discrimination between unfamiliar voices.

The outcomes supported only 2 predictions, that the autistic group was impaired on the test of matching affect expressed by voice or facial expression, and their discrimination of unfamiliar voices was no different from that observed in the other groups.

However, the children with language impairment performed similarly to, or worse than, the autistic group on the tests of familiar voice and face matching and recognising familiar voices.

The implication was that children with a language impairment have a significant difficulty in interpreting the vocally expressed feelings.

Meanwhile the fact that the children with autism did not appear poorer on naming emotions may have simply reflected the experience of these children in being specifically taught this kind of skill in their schools.

The children with language impairment were also impaired compared to the other groups on a test of matching emotions expressed vocally and facially ; and this may result from simple difficulty in interpreting vocally expressed emotion, but it might also reflect cross-modal processing difficulties which have been demonstrated ( Kamhi et al 1984 ) among children with language impairment. There is also a third possibility that children with language impairment have an impaired ability to process faces.

The authors then refer to a study by Rutter et al (1992) which has linked language impairment in childhood with later social relationship difficulties similar to those seen in mild cases of autism. There is also the common observation that children with language impairments are at risk for behavioural and emotional disorders ; or it is possible that the language impairment and these other problems arise from a common cause such as some neurological immaturity or limited processing capacity.

This latter view is supported by the identification of children who have semantic and pragmatic language difficulty and who may display some of the other behaviours that are typical of autism ( Rapin 1996 ), and there is some question whether to describe these children as autistic or as language impaired.

What matters is that there is support for the view of a continuum of language or communication difficulties with no obvious demarcation between those which include social relation problems and those that do not.

It was acknowledged that the poor performance of the children with language impairment meant that there was no replication of the earlier findings of impaired voice and face matching or familiar voice recognition which could be considered peculiar to, or characteristic of, the children with autism. However, the suggestion is that this is not because the children with autism are performing normally but that the children with language impairment are similar to children with autism in having some problems in voice processing.

The implication drawn from the study is that children with language impairment not only have some weakness in cross-modal processing but additional problems in processing vocally expressed emotion and, possibly, more wide ranging weaknesses in processing social stimuli. In other words, the demarcation between such children and children with autism is not very clear. There is also the suggestion that the autistic children do not have the cross-modal impairment but do have problems in matching emotions expressed vocally and facially ; so the autistic group and the language impaired group both have difficulties which are not at the level of perceptual encoding but at the level of registering meaning.

The authors suggest that there is a need for further studies of this kind since, if language impaired children are confirmed as having significant problems with voice processing ( and, perhaps, face processing ), then the way they are described and how interventions are organised may need to be modified.

Further, there will be no logic in suggesting that a fundamental aspect of, or causal factor in, autism is that of an impaired ability to process social stimuli such as faces or voices, or to recognise expressed emotions, because, by such logic, children with language impairments who also have these social processing difficulties would be described as autistic.

It is recognised that the social processing difficulties in autism may be of a different kind and have a different cause from those which occur in language impairment. It may also be that whatever underlies this processing difficulty in children with autism also brings about the other signs and symptoms of autism. However, it would still be misleading to say that impaired social processing is the cause of autism and instead it should be regarded as a consequence of whatever psychological deficits do underlie autism.

The conclusion from this study is that there is further evidence of the link between language disorder and autism, and one can indeed refer to a kind of seamless continuum from Asperger syndrome, through pragmatic language impairment, to semantic/syntactic/morphological/ phonological language impairment.

Greenway (2000) introduces her discussion of the means of intervening among individuals with autism by referring to a number of assessment instruments and suggesting that the choice of assessment tool will depend in part at least upon what one is seeking to do, and, therefore, the precise nature of the information required.

For example, the Vineland Adaptive Behaviour Scales may assist in analysing functional behaviours, but the focus may not be sufficiently specific for use with Asperger syndrome. To screen for social deficits, the Autistic Continuum is helpful, and theory of mind tasks can help in identifying cognitive deficits. Nevertheless neither of the latter pinpoint particular interventions.

Meanwhile the ABC analysis, the Sue Spence Social Skills Checklist, or personal construct methods, will allow an investigator accurately to describe problem areas and will offer some understanding of the individual's behaviour thus providing useful pointers concerning the intervention.

In respect of social deficits, what little research evidence exists refers to work with children with Asperger syndrome or high functioning autism placed together in specialist settings.

Reference is made to the work of Williams (1989) who, among others, described social skill training with small groups of boys between 8 and 16 years of age in weekly sessions.

Changes of behaviour in order to facilitate mainstream integration were attempted via games, role play, and modelling. Specific targets were achieved by direct instruction, discussion of the consequences of behaviour, and brainstorming to think of new ways to achieve existing social goals. The Sue Spence checklist was a useful starting point to highlight the areas upon which to concentrate, and the intervention materials were found to be a positive means of developing appropriate body language, the interpretation of facial expression and non-verbal cues, and learning how to initiate, maintain, and end a conversation.

Unlike the situation a month normally developing children for whom these materials were devised, it was found that the programme had to move very slowly and one topic per term rather than one topic per weekly session was realistic.

Greenway notes that anyone attempting social skill work with children with autistic spectrum difficulties should remember that the social behaviours taught in one particular setting may not generalise to other settings. It may be necessary to teach the skills in different setting or offer specific advice to the children to use the skills in various contexts.

Reference is then made to the social skill components of the TEACCH programme, and to the work of Mesibov (1984) with adolescent or adult participants in respect of improving interpersonal skills, the development of peer groups, and boosting self esteem.

Small groups met weekly for two terms with each session lasting for one hour but with each child having a further thirty minutes before the session in order to practise those skills to be used in the group. The session was divided into group discussion, listening and talking, role play, and understanding humour. Mezibov is quoted as describing the facilitation of skills in initiating and maintaining conversation, and improved self images ; and as highlighting how all the participants made it clear that they were motivated to be social, but needed specific teaching in order to be able to achieve this goal. The participants appreciated the opportunity to make and keep friends, and the important implication was held to be that these individuals did not lack social interest but social skill.

In respect of behavioural interventions with social skill development, the only study identified is that of Koegel et al (1992) working with boys towards the more severe end of the autistic spectrum.

The aim was to bring about self management of social interactions initiated in various settings. The children were taught to use counters to keep a record of the number of social overtures to which they responded, and a food reward was contingent upon a given number. It is recognised that the programme involved a high level of adult time and organisation, and that the reinforcers were maintained throughout ; and both these issues would make for difficulties in implementation in an ordinary school.

Greenway's review suggests that those interventions which might be described as social constructive this seemed to fit most readily with a mainstream setting and with children with special needs as targets.

Under this heading would be included the Social Story intervention described by Rowe (1999) and the Circle of Friends intervention described by Whitaker et al (1998).

Also included here would be the social skill intervention of Kamps et al (1992) which took place in a class of 7 year olds which contained 3 high functioning autistic pupils as well as 2 children with physical difficult years along with 11 normally developing peers. The teacher was supported by 2 classroom assistants.

The social skill training involved 10 minutes at the beginning of playtime for groups of pupils comprising 1 autistic child and 3 mainstream peers. The focus was on direct teaching of a particular skill, including initiating and responding to interactions, greeting others and conversing on a range of topics, giving and taking compliments, turn taking, seeking and offering help, and including others in activities. Feedback was offered following observations of free play and children were rewarded with stars for social interactions.

Observations over time showed that the autistic children responded as positively as their peers, and the authors are quoted as highlighting how social skill training for children with autism can be effective, and a critical feature may be that the training was conducted with their peers.

( In this respect, it is similar to the Circle of Friends approach which seeks to include vulnerable pupils into a school community by engaging the peer group in offering proactive support ).

The success of Social Stories ( with children with Asperger syndrome ) has been attributed to their visual format, their focus upon specific behaviours relevant to the given child, and their provision of specific advice about social cues and reactions.

Greenway argues that these interventions seem particularly useful and manageable in those mainstream schools which demonstrate a generally positive attitude towards including children with special needs.

The common link is the involvement of support from peers and adults and seeking to alter the social setting at school while focusing upon the specific needs of the children in question.

The overall conclusion would suggest the importance of working on social skills in the social context.

Individual working with children with autistic spectrum disorders, or working with small groups of children all with ASD, may remain abstract ; and, while social skill training groups which involve normal peers may take significant time for organisation, the potential benefits have been demonstrated ...... although there is still scope for research to observe whether the gains are maintained and whether, therefore, mainstream inclusion remains viable in the long term.

Greenway also concludes that work of this type may provide the stimulus or vehicle for multi-professional working, and this could afford the further opportunity to explore how or why success is achieved by these approaches.

One might be able to identify whether it is, for example, a matter of peer modelling or a matter of using simplified language to make the social processes explicit.

Alternatively, if peers are involved in the social skill training and share the same rules, the greater predictability of the situation may reduce the stress upon the children with autism and enhance their learning.

As a footnote to this issue of school experience and social communication, one might refer to the experiences of one able student ( C.) with Asperger syndrome, as reported to Gold (2000).

In respect of the student in question, it was clear well before school age that she had an idiosyncratic style including her own invented language. She went to a small private primary school but found it extremely difficult because of her fear of noises, her emotional blankness, and her failure to interpret body language or day to day chat.

C. became the class loner and the target for teasing and was never chosen to be a partner in any activity. However, she reports her continuing willingness to learn but that there was great difficulty in switching attention from one topic to another or in coping with lengthy verbal explanations.

It is significant that her greatest horrors were playtime, PE, and the lunch queue, which all involved noise, social interaction, physical contact, and incomprehensible rules.

Some teachers were able to recognise her style and set her projects which could utilise her obsessional interests, and allowed her to stay inside alone at playtime to read. Others regarded her as disruptive or lazy.

In retrospect, C. recognises how helpful it would have been if someone had made explicit the various classroom rules that underpin school life while recognising her particular abilities.

In respect of inclusion, one notes the suggestion that this may be appropriate in many cases but the variability among cases of Asperger syndrome is such that no policy will be right for every individual.

She also concludes that, if it were possible to pick up social skills simply by being around normal people, many individuals would not be on the autistic spectrum in the first place. In other words, there is a vindication for the view that inclusion must involve some preparation of the setting and specific support for the individuals with Asperger syndrome in the context of peer interactions.

In the light of the number of very young children who are diagnosed with autistic spectrum disorder, it is increasingly relevant to highlight interventions which are to be implemented in the very early stages of schooling.

The paper by Laushey and Heflin (2000) begins by emphasising how individuals who are diagnosed in this way form a widely heterogeneous group who made differ in many ways, but one of the few common features is a pervasive deficit in social skills. They highlight the evidence that a lack of social skill will inhibit inter-personal relationships and opportunities for interaction but that this kind of skill will not develop without expose to appropriate settings and other people. It is necessary for children with autism to have some exposure to typical peers in order to boost learning of appropriate social behaviours.

Reference is made to Gresham (1984) who also makes the point that simply placing typical peers and children with autism together is not enough to ensure that social skills are developed in the autistic group. For example, such children may have difficulty in imitating appropriate behaviour so that the first step is their being taught how to imitate before they can take advantage of mixing with typical peers. Direct instruction may be required for teaching specific skills because one cannot rely on incidental learning.

There are said to be 2 basics styles for promoting social interaction.

Adult mediated approaches involve prompting or reinforcing particular behaviours so that it is the adult who is seeking to lead the child and, while skills can be elicited, the social behaviours may appear stilted.

In peer mediated approaches, normally developing children are selected and trained to improve the social interaction of children with autism. Peer Tutoring is the most common strategy of this type whereby learners set out to teach each other.

However, there is little research specifically about the training of tutors and what there is tends to be concerned with children much older than 5 years of age ; but, among children with autism, early intervention appears critical so that an logical stratergy would be that of providing training or preparation to young peers without disabilities in order to facilitate learning of children with autism.

A review of research, notably that of Roeyers (1996), has identified 3 types of peer involvement where young children are concerned.

The first is the proximity approach where children with difficulties are placed in a mainstream setting in order to learn by watching and interacting with their non-disabled peers. The intervention is dependent upon some natural transmission of social skills without actual training given to the peers.

The second approach may be described as operant training in which peers are taught to prompt some particular response from the child with autism and then to offer a reinforcement when that response is given.

The third approach is described as a peer-initiated procedure where the non-disabled children are trained how to make social overtures to the target children.

Observations have indicated that all 3 approaches can bring about positive changes in the social behaviours of children with autism, but that the second and third approaches appear more effective with these children.

The study by Laushey and Heflin involved 2 separate kindergarten classes, each containing a child with autism or pervasive developmental disorder. In each class, there were 20-25 children, a teacher, and 2 assistants, one of whom was assigned to work specifically with the child with autism. The 2 target children were 5.6 and 5.8 years old respectively, and both met the DSM-IV M criteria for autism or PDDNOS.

Both boys had adequate spoken language and could read appropriate level stories, but with minimal comprehension. There was some eye contact and some ability to take turns when reminded, but both boys had difficulty recognising social cues or waiting for another child's response. The also had problems in maintaining a conversation.

Baseline data about the amount of interaction and its type were gathered during a period of observation while the target children were in their classes and when no intervention was taking place. In the first treatment phase, a peer tutor training programme took place during play times, and this took the form of a " buddy system ". There followed a period when the children returned to the passive proximity condition, and then there was a second treatment phase where the buddy system was reinstated.

For the buddy system, each child in the class was linked with a daily buddy. A chart was created to display pairs of children's names on cards, and every day the teachers rotated the cards so that each child would have the opportunity to link with a different peer.

During playtime, the children were instructed to pair with their buddies, and the peers working with the autistic boys were reminded of their roles which had been taught during a training phase.

The target behaviour for all children, including the boys with autism, was to stop with, play with, and talk to the buddy. The method was set out by English et al (1996) but in the present study the strategy was modified to allow regular changes of the pairs of children.

One researcher acted as the trainer and the first session described to the children how people can be alike and how people can be different ; and reference was made to how one might choose friends because they appear like ourselves, but how it might also be fun to be friends with others who are not just like ourselves.

The trainer then explained that, in order to have the chance to play with as many children as possible, they were going to begin a buddy system. This would involve their having a particular friend to play with for the day, and that it would be a different buddy each day. This being so, it was pointed out that everyone would have the chance to play with children who are like them in many ways, and others who are different in many ways ..... but they could still enjoy being together and doing many of the same things.

The buddy chart was demonstrated and the children were shown where to find the name of their budddy for the day.

The children were told that being a good buddy meant staying, playing, and talking, with the other child.

Meanwhile, the researchers and early years staff devies a set of 4 social skills deemed as vey important for soacial acceptance for these 5 to 6 year olds. These were :

• Asking for an object and responding according to the answer given ( such as waiting, asking again, asking if (s)he could have it when the other child has finished ).
• Appropriately getting the attention of the other child.
• Waiting for a turn.
• Looking at or towards the person speaking.

Data on these 4 behaviours were taken at regular intervals during play times across the whole of the first treatment phase which lasted for 11 weeks.

The results showed that this buddy system elicited more appropriate social skills in the boys with autism than the passive proximity approach which operated as a baseline condition. In the case of both boys, there was significant regression when the treatment phase ended and then a marked increase once again when the treatment was reinstated. It is important to note that generalisation of the skills occurred naturally because the children with autism had not worked with just 1 or 2 peers, but with the whole group.

During the passive proximity condition, that children with autism engaged in fewer turn- taking interactions, or instances of looking at the person speaking to them, and they displayed lower tolerance for waiting. The treatment phase markedly increased the extent of the desired behaviours while the typically developing children were also shown to benefit by practising their play skills with different friends. Teachers reported that this intervention was helpful for those children who were shy and who had experienced difficulty in making friends.

In their discussion, the authors restate the common view that an immersion approach to skill development will only work if the individuals know what it is they are to observe and interpret and imitate in their more competent peers. The study was held to show that the buddy approach can bring about increased social interaction in the children with autism, while simply placing the children in close proximity to typically developing peers did not bring about significant change.

There is support for the view that training all the children rather than focusing simply upon the children with some disability will produce greater effects. It is also important that this approach avoids singling out the child with a disability.

Meanwhile, observations showed that some of the peers in the study were very social while others were more withdrawn ; but, rather than being seen as a limitation, the variability of peer behaviour was perceived to be more representative of the normal distribution of social behaviour and, therefore, the experience was that much more realistic.

It is recognised that this study has focused upon only 2 children with autism and that both these target children were quite verbal and functioning at the higher end of the autistic spectrum. The researchers would seek, therefore, to replicate the study with more children and with children who display different levels of autistic behaviours.

Nevertheless, they cite evidence that pairing an autistic child with an adult assistant in a mainstream class may actually inhibit social interaction and lead to a kind of dependency in the children with autism ; and restate that social interactions between children with autism and their peers may be best facilitated when all the children are trained in prosocial exchanges and given the necessary structure to practise such skills.

The issue of inclusion, and the changed social environment for many children with special needs, is the theme of the study by Gus (2000) which concerned a year 10 student with autism who was identified as feeling socially rejected at school.

It is pointed out that there can be many benefits from mainstream inclusion, but there are still problems with stereotyping and social integration ; and there is also the risk that peers may have no knowledge about autism even if they are aware that a particular child is autistic. Reference is also made to the possibly damaging impact of additional adult support for the student with autism if this reduces the opportunities to learn with and from peers.

Gus describes Adam as making adequate progress scholastically in his local secondary school where he receives one to one support from a classroom assistant throughout the school day but not at break time or lunch time. Adam was said to have become increasingly unhappy as he became more excluded from peer activities. It was felt that the other young people in Adam's class needed to know that Adam was autistic, and to learn about this condition ; and it was agreed to adapt the Circle of Friends approach. ( The immediate requirement was to reassure Adam's mother about this approach, and agreement was reached that the students in Adams's class were old enough to learn about the autism without his becoming vulnerable to teasing as a result ).

While Adam was engaged in some individual work with his support assistant, the class tutor organised an initial session with the rest of the students in order to explore Adam's circumstances and to work towards a shared understanding about his difficulties. It was acknowledged that it was unusual to discuss a student in this way but that the issues would remain confidential and the motives were positive.

The first step was to ask the students to list positive and negative perceptions about Adam, followed by a presentation by the tutor and educational psychologist about the nature and symptoms of autism, setting out how autistic individuals were similar to non-autistic individuals but highlighting also the particular difficulties linked to the triad of impairments. Comparisons were drawn between the well-established autistic difficulties in respect of language or social relationships or thinking style, and the descriptions set down by the students of where or how they found Adam special.

A week later, there was a feedback session in which the information shared about autism was reviewed and a discussion stimulated about how to help Adam. It is reported that the responses were positive, with a number of pupils acknowledging that previously they did not know about Adam's particular needs but that they were able to be more understanding and patient. Specific suggestions about how to help Adam, such as talking to him more, being more specific in the use of language, making him feel welcome in the class, etc., were shared.

A follow-up was organised at the end of the school year about six months after the initial information sharing and planning. Questionnaires were completed by the class about their feelings towards Adam and their actions towards him, and the findings indicated significant change in attitudes. Students perceived themselves as more understanding or sympathetic or patient, and more aware of how Adam might be feeling. Specific behaviours included increased efforts to talk to Adam, taking care to ensure clarity of what was said, allowing him to join lunchtime activities, intervening if other students were being unpleasant, etc..

In her discussion, Gus highlights the contrast between the experience of rejection, or being ignored by peers, that was evident with Adam when the concern was first raised, and the circumstances following the provision of information and the sessions in which Adam's needs were discussed. It was found that the change in attitude was sustained over a long-term period, and it was argued that the change of attitude and behaviour was linked particularly to allowing the students to express what they had found difficult about Adam together with pointing out the link between his particular behaviours and autism.

It was concluded that sharing specific information about autism with classmates brought about a greater understanding of and attitudes towards the student with autism as a result of which he became a observably happier in school even if his actual social skills were not perceived to have changed as a result of the peer actions.

It was recommended that professionals should be proactive in enabling mainstream peers to be informed about the nature and implications of autism ; and that, ideally, this approach should be completed as early as possible after the child with autism begins attendance at a given school.

Again, the implication is that, if inclusion is to be a positive experience, specific interventions need to involve the peer group rather than focus solely upon the individual with autism. One does not set out to change that individual, one sets out to change the attitudes of those people working alongside him or her, and, thus, to modify the context.

Over the recent months, there has been further emphasis on the part of medical specialists to highlight how the risk associated with being infected with measles or mumps or rubella far outweighs any risk which has been perceived to accompany the MMR injection.

For example, Patja (2000) describes the long term study completed in Finland to examine any cases where there has been some suggestion that the vaccination has led to serious adverse consequences.

According to the report. 173 adverse reactions were claimed out of a total of almost 3,000,000 cases where the vaccination was given. The investigators reported that nearly half of the adverse reactions described proved to be caused or influenced by some other factor so that the incidence of serious adverse events possibly linked to the vaccination was around 5 per 100,000 children.

It was also pointed out that it is still debatable whether the one single fatality, or the onset of diseases such as asthma or diabetes, could be causally related to the vaccination programme. It was concluded that, if the causality were real, then there would be an accumulation of new cases during follow up vaccinations, and this was not found.

This point is reinforced by a further paper ( Feikin et al 2000) which points out that not only is there a risk to the health of children who are exempted from vaccination because of personal beliefs, but that there is an impact upon the whole community.

For example, the overall incidence of measles or pertussis was found to be greater among vaccinated children when the number of non-vaccinated children in the area increased. Schools which experienced pertussis outbreaks had significantly more exemptors than schools where outbreaks were not experienced. Transmission of measles by an exemptor to a vaccinated child was reported in at least 11% of cases.

However, the study does recognise parental anxiety with the implication for spreading information concerning the risks of non-vaccination while constantly highlighting ongoing policies and procedures for monitoring the safety and effectiveness of the vaccines used.

The problem is that not all the public is wholly convinced about the safety of, and need for, the vaccination. There is, in fact, a report ( Ramsay et al 2001) that a significant number of practice nurses believe that there may be a link between the injection and autism or Crohn's disease.

Again the implication is for sharing information among healthcare providers in the first place in order, in the second place, to provide the reassurance to parents that may still be necessary.

In particular, it may be necessary more fully to explore what factors underlie the " genetic fragility "or " genetic predisposition " in respect of autism, as set out by Shattock (2000).

One might summarise his concerns by referring to some continuing question about the methodology in those studies which have sought to demonstrate MMR safety, and to his metabolic thesis of autism concerning vulnerability to damage to enzyme functioning or to breaks in the blood- brain barrier from some external and traumatic factor such as inappropriate diet or some environmental event or toxin, as a result of which certain " rogue peptides " may be able to pass through a damaged (permeable) gut wall and reach the CNS.

It may be that the concerns felt by some parents are unnecessary and that the risk to children is being increased rather than decreased by a seeking to exempt them from the MMR vaccination ; but one would requote the above points to the effect that the concerns should be fully recognised rather than simply dismissed, and more specific information and reassurance given if these concerns are maintained.

M.J.Connor        February 2001

Boucher J., Lewis V., and Collis G. 2000 Voice processing abilities in children with autism, children with specific language impairment, and young typically-developing children. Journal of Child Psychology and Psychiatry 41(7) 847-857

Boucher J., Lewis V., and Collis G. 1998 Familiar face and voice matching and recognition in children with autism. Journal of Child Psychology and Psychiatry 39 171-182

Bryson C. 1972 Short term memory and cross-modal information processing in autistic children. Journal of Learning Disabilities 5 81-91

English K., Goldstein H., Kaczmarek L., and Shafer K. 1996 Buddy skills for pre-schoolers. Teaching Exceptional Children 28(3) 62-66

Feikin D. et al 2000 Subject : Foregoing MMR Vaccination. The Journal of the American Medical Association 284 3145-3150

Frith U. 1989 Autism : Explaining the Enigma. Oxford : Blackwell

Gold K. 2000 Out of this world. Times Educational Supplement : Special Educational Needs. November 3rd.

Greenway C. 2000 Autism and asperger syndrome : strategies to promote prosocial behaviours. Educational Psychology in Practice 16(3) 469-486

Gresham F. 1984 Social skills and self-efficacy for exceptional children. Exceptional Children 51 253-261

Gus L. 2000 Autism : promoting peer understanding. Educational Psychology in Practice 16(3) 461-468

Kamhi A., Catts H., Koenig L., and Lewis B. 1984 Hypothesis testing and non-linguistic symbolic abilities in language impaired children. Journal of Speech and Hearing Disorders 49 169-176

Kamps D., Leonard B., Vernon S., Dugan E., and Delaquadri J. 1992 Teaching social skills to students with autism to increase interaction. Journal of Applied Behaviour Analysis 25 281-288

Klin A. 1992 Listening preferences in regard to speech in 4 children with developmental disabilities. Journal of Child Psychology and Psychiatry 33 736-770

Koegel L., Koegel R., Hurley C., and Frea W. 1992 Improving social skills and disruptive behaviour in children with autism through self management. Journal of Applied Behaviour Analysis 25(2) 341-353

Laushey K. and Heflin L. 2000 Enhancing social skills of kindergarten children with autism through the training of multiple peers as tutors. Journal of Autism and Developmental Disorders 30(3) 183-193

Mesibov G. 1984 Social skill training with verbal autistic adolescents and adults. Journal of Autism and Developmental Disorders 14(4) 395-402

Mottron L. and Belleville S. 1995 A study of perceptual analysis in a high level autistic subject with exceptional graphic abilities. Brain and Cognition 23(2) 279-309

Patja A. 2000 Subject : Serious Events Rarely Related to MMR Vaccination. Pediatric Infectious Disease Journal 19 1127-1134

Ramsay M. et al 2001 Subject : Uncertainty about the MMR Vaccination. British Medical Journal 322 82-85

Rapin I. 1996 Developmental language disorders. Journal of Child Psychology and Psychiatry 37 643-656

Rodgers J. 2000 Visual perception and Asperger syndrome. Autism 4(3) 321-329

Roeyers H. 1996 The influence of non-handicapped peers on the social interactions of children with a pervasive developmental disorder. Journal of Autism and Developmental Disorders 26 303-320

Rowe C. 1999 Do social stories benefit children with autism in mainstream primary schools ? British Journal of Special Education 16(1) 12-14

Rutter M., Mawhood L., and Howlin P. 1992 Language delay and social development. In Fletcher P. and Hall D. (Eds) Speech and Language Impairments in Children. London : Whurr Publishers.

Schopler E. 1985 Convergence of learning disability, higher level autism, and Asperger syndrome. Journal of Autism and Developmental Disorders 32(7) 1107-1122

Shattock P. 2000 Autism : an environmental assault. What Doctors Don't Tell You. 11(6) September 1-4

Whitaker P., Barratt P., Joy H., Potter M., and Thomas G. 1998 Children with autism and peer group support using Circles of Friends. British Journal of Special Education 25(20 60-64

Williams T. 1989 A social skills group for autistic children. Journal of Autism and Developmental Disorders 19(1) 143-155