The first section of these notes provides evidence that children and young people diagnosed with Asperger syndrome can be clearly differentiated from high functioning autistic individuals ; and there is a warning that Asperger syndrome, despite placement towards the less extreme end of the autistic spectrum, does carry a significant risk for later emotional or psychiatric disturbance, notably (social) anxiety.

Reference is then made to a study which demonstrates that personal construct approaches can be used effectively with individuals with Asperger syndrome, despite common beliefs that such strategies require a particular level of cognitive and communicative functioning.

The third section continues the theme of the particular perceptual style or perceptual deficits characteristic of autism and these summaries conclude with reference to a practical means of enhancing vocalisations among previously non-verbal children.

M.J. Connor September 1999

Emotional and Behavioural Disturbance

The study of Tonge et al (1999) set out to identify any differences between children or adolescents with high functioning autism or Asperger syndrome in respect of the nature and level of behavioural and emotional difficulty.

They begin by noting that there has been some difficulty in making a clear differentiation between Asperger syndrome and autism as a result of the absence of consistent diagnostic criteria. There may be difficulty in comparing and contrasting results across different studies if the subjects differ widely in respect of their particular needs, ability level, and age. However, reference is made to those diagnostic criteria for Asperger syndrome which include an IQ within the normal range, no significant delay in language development, but clear impairment in social interaction, along with restricted or repetitive patterns of behaviour and activities.

A review of existing studies shows that young children with high functioning autism could be contrasted with children with Asperger syndrome on a range of developmental disorder symptoms, notably those concerned with socialisation and communication, and standard measures of language skill (Szatmari et al 1995). Meanwhile, although motor clumsiness may frequently be associated with Asperger syndrome, a number of studies (such as that of Manjiviona and Prior 1995) have shown that problems in this area do not reliably distinguish between the two groups.

In a subsequent study, Ghaziuddin and Gerstein (1996) found that an Asperger group used correctly formed sentences more frequently than an autism group and their speech was generally more complex. The work of Ozonoff et al (1991) showed that both groups, made up of children or adolescents, showed impaired executive functioning skills but the Asperger group performed significantly better on theory of mind and verbal memory tasks.

There is also evidence that Asperger children are more likely to have disorganised thought processes. Such children are more likely to focus upon their internal experiences and it is suggested that they have more complicated thinking and fantasy lives.

This present study by Tonge et al set out to examine possible differences in the psycho pathology between children and adolescents with Asperger syndrome and those with high functioning autism, and whether they could be differentiated in respect of emotional and behavioural problems.

The subjects chosen included a large sample of those with high functioning autism and those with Asperger syndrome, all of whom met the current American Psychiatric Association criteria for autistic disorder (DSM-1V). The mean age of the former group was 7.4 years and 88 per cent were male. Half the group had a full IQ in the normal range and half in the borderline range. The mean age of the latter group was 9.8 years and a similar percentage were male. 70 per cent had a full IQ in the normal range and the rest had a full IQ in the borderline range. It is underlined that the young people with Asperger syndrome were significantly older than the autistic group.

The Developmental Behaviour Checklist was used to assess behavioural and emotional performance, completed by parents or care givers. The subscales are divided into disruptive, self absorbed, communication disturbance, anxiety, autistic/social relating, and antisocial.

It was found that 65 cent of the autistic group and 85 per cent of the Asperger group met the cut-off criteria for significant levels of behavioural and emotional disturbance. This difference in rates between the two groups was shown to be significant ( p.0.05 ). The Asperger group scored significantly higher on the overall severity of difficulties as measured by the total behaviour problem score of the checklist. This group also scored significantly higher on the disruptive, anxiety, social rating, and anti-social subscales.

It was concluded that children and young people with high functioning autism and Asperger syndrome both show high levels of significant behavioural and emotional disturbance; but the latter group show higher levels of overall problems. Both groups showed a much higher rate of behavioural or emotional disturbance than is found in the general population of young people with intellectual disability, and a vastly greater rate than that found in the general population of children or adolescents.

The implication is that Asperger syndrome and high functioning autism are clearly differentiable. There is also a challenge to the view that Asperger symptoms are at the milder end of the spectrum when actually those young people so-diagnosed have a greater risk of psychiatric disorder than those with high functioning autism.

The authors highlight as particularly relevant the finding of marked symptoms of anxiety and disruptive behaviour in the Asperger group, with the implication that such young people may experience even more difficulty with social interaction than those with high functioning autism even if social impairment is a diagnostic criterion for both disorders.

In seeking to explain the greater number of emotional and behavioural problems experienced by children with Asperger syndrome, the authors suggest that the increased frequency of non-verbal learning difficulty, such as weakness in visual skill, visual memory, or motor competence, can generate anxiety which leads to disruptive behaviour in an attempt to avoid tasks which require these skills.

Alternatively, there is the suggestion that these children are less flexible in the style of information processing and more focused upon their own thoughts or emotions or fantasies. This may be linked to difficulties in using social knowledge to adapt to the changes that constantly occur in any normal social interaction. The difficulty may be increased because the apparently good verbal skills of children with Asperger syndrome might lead other people to expect largely normal social responses and interaction thus putting more pressure upon those young people. As a result, greater anxiety or stress will be evoked, reflected in disruptive or anti-social behaviours.

In summary, there is evidence that many children and adolescents with either high functioning autism or Asperger syndrome will show significant emotional and behavioural disorder. In particular, those individuals with Asperger syndrome may have a high level of psychopathology, with implications for consistent monitoring and support both in the home and school setting.

Psychotherapeutic Work with Individuals with Asperger Syndrome

Neatly following from the above conclusion concerning the potentially high levels of psychopathology among people diagnosed with Asperger syndrome, the work of Hare et al (1999) describes the use of one particular intervention... personal constructs.... with individuals experiencing an autistic spectrum disorder.

They begin up by describing how little has been written about psychotherapy for individuals with Asperger syndrome, but they quote a number of research reports, such as that of Howlin, (1997) which have identified an increased risk of psychological health problems, particularly involving depression or anxiety, as individuals so-diagnosed grow into teenage and adulthood.

Personal construct psychology is based upon the idea that people develop their own idiosyncratic perceptions of reality which may or may not be appropriate and adaptive. These perceptions can be examined by considering how the individual organises his world and what constructs are used in the process. Assessment techniques, which include the use of repertory grids, are practical and can be largely non-verbal. They may, thus, be of particular use with those people who find other forms of psychological assessment difficult. The techniques may apply very well to individuals with autistic spectrum disorders who have an interest in number and in order, prefer things to people, and respond better to relatively formal structures and to external cues.

The study completed by Hare et al involved four young men with a diagnosis of Asperger syndrome, with a mean age of 26 years. All had average measured ability and were considered to be verbal, even if two of these subjects did not spontaneously initiate speech or maintain verbal interaction.

In each case, the initial session involved eliciting from the individual the names of the important people in his life. Subsequently, constructs were sought by using one of the standard methods, such as presenting the individual with two of the named persons (elements) and asking him to say in what ways these two persons are similar and in what ways they are different. Having established a number of constructs to the point where it was clear no more were likely to emerge, the individual was asked to describe the opposite of each one in order to provide a bipolar construct along which to rate each of the elements. For example, one construct might be "kind" and the opposite would be "unkind". Meanwhile care was taken to ensure that there was a shared understanding of what was meant by "kind I unkind".

The next step was to rate the elements along each construct by setting out all the cards containing the important names and asking the individual to put them in order. For example one might ask" Who is the most kind?", "Who would be next?", etc., until all the cards have been placed in a particular order.

One then observed which constructs are particularly significant, and how they cluster. For example, the authors describe one of the subjects as having a recent diagnosis of Asperger syndrome and a previous diagnosis of schizophrenia. A particular feature of this individual was the consistent presence of some internal voices from which he took advice. Analysis of the construct grid suggested that consistency - inconsistency was particularly important for him which was in accord with his self reported need for structure and support in his social environment. It was also found to be important to take note of the voices rather than attempting to ignore them or dismiss them as unimportant or suppress them through medication.

Therefore, later work was designed to include a focus on helping the individual to acknowledge his ownership of these voices by which to organise his life, within which a broadly cognitive behavioural approach was held to be most appropriate.

In the case of one of the other individuals, it was concluded that part of the apparent lack of insight into his difficulties could be interpreted as his failure to regard any of the problems as being his own responsibility. Instead he appeared to find all problems as being due to other people.

This being so, one needed to take account of his probable resistance to any intervention which has, as a starting point, the willingness on the part of the individual to acknowledge difficulties.

A further example is that of one of the individuals who showed that :

He saw flexible people as helpful and tolerant; bossy people as strict and ignorant.

He construed negatively those people outside his own family.

He appeared to see staff as actively working to keep him from his family.

The implication was to work towards long term and trustful relationships with staff, perhaps seeking to replicate a family-like situation. The need was identified for work of a cognitive behavioural type to seek to reduce the very rigid views about the roles of staff and family.

In any event, the authors conclude that personal constructs can provide a practical and useful starting point for therapeutic intervention, and may provide a basis for seeking to encourage new behaviours and new ways of looking at things.

Support was also claimed for The view that this approach has advantages for use with individuals with Asperger syndrome in That it has a kind of intrinsic appeal through its format and through its flexibility ; but also, and importantly, because The individualisation of the approach will demonstrate that the concerns and feelings of people with Asperger syndrome are being taken seriously.

It is recognised that there may still be some difficulty in seeking to move from assessment to ongoing therapeutic work.

There is also likely to be particular difficulty in those aspects of the approach where the individual is required to refer to his own and to others' mental states.

Further, the use of personal constructs may be largely to determine how best to intervene rather than as a formal and individual psychotherapy in its own right.

However, it is important to note that despite the common belief that the use of personal construct techniques requires a minimum level of ability, this current work suggests that intellectual ability is not a major determinant of someone's capacity to work within this approach.

Visual Perception

This next section continues the theme of perceptual process in begun in Current Issues 8 (March 1999) which described the respective significance of central coherence and hierarchisation.

Mottron et al (1999) begin their report with reference to the frequency with which autistic individuals engage in atypical visual behaviours, such as hand-flapping in front of their eyes, gaze avoidance, etc.. They recognise that these behaviours are not limited to autism but they are far more common in this condition than in any other developmental syndrome. Nevertheless, there has been the suggestion from various authors, such as Frith et al (1987), that low level perceptual processes are not impaired in autism because autistic children are usually able to copy simple auditory patterns and to label visual patterns.

What seems to matter is the level of the perceptual process concerned. For example, higher level perceptual processes, such as the recognition) of particular physical characteristics of objects, involve a number of components including selective visual attention, two or three dimensional representation, and the sequencing of local and global cues.

As the complexity of visual perceptual skills has become more evident so there has been greater support for the view that a characteristic of autism is some deficit in perceptual skill. For example, autistic children do not perceive visual illusions in the same way as other children but they can detect hidden figures more easily. Meanwhile, hierarchisation refers to the way that different parts of a visual stimulus may be examined in different ways. Thus, normal subjects examine the overall features or outlines of a figure before internal details; whereas autistic individuals appear independently to process both global and local parts of a stimulus.

This present study set out to examine hierarchical visual processing in a group of individuals with autism. The subjects included high functioning autistic teenagers and young adults, matched with normal controls. The task involved copying a series of black and white drawings, either of familiar items such as a plug or a kettle or designs made up of the basic straight lines or curved lines comprising those objects but reassembled in a different and relatively abstract way.

Observations showed that the autistic individuals could copy object and non-objects at the same speed and with the same accuracy as matched controls. The immediate implication appears to be that differences between the autistic and control subjects is not directly concerned with perceptual or graphic skills but with more general cognitive functioning.

Nevertheless, the autistic group showed a strategy that could be contrasted with that of the control. For both types of visual stimulus, the autistic individuals produced more local features than the control subjects did in the first section of their copying, and less in the latter section. In other words, the autistic individuals appear not to favour the overall global features, unlike the control individuals. This result may be explicable in terms of some impairment in hierarchical processing.

The authors then refer to two models of visual perception where the ventral stream encodes the physical properties of objects whereas the dorsal stream encodes the spatial properties of

objects. The performance observed with the current subjects suggest that autistic individuals have a local bias as a result of impairments in this ventral processing stream.

In a second study, The subjects were asked to copy some impossible figures with the prediction that the autistic subjects would cope more readily with the impossibility effect because they have difficulty in relating the different elements of a figure.

It was indeed found that The normal subjects needed more time to copy impossible figures than possible ones whereas the autistic subjects were less sensitive to the geometric impossibility. It was assumed that the autistic individuals do not integrate a local part of a figure into a perceptual representation of the complete figure to the same extent that control individuals do. That is, they copy each part separately without direct reference to other pans of the figure.

This could be interpreted as a simultagnosia which is a deficit in which two pans of a visual display cannot be seen simultaneously. Alternatively, the explanation may involve long term memory in that the geometric rules which underlie the impossible shape are not activated in memory, or the memory is somehow defective.

One might expect that difficulty in copying an impossible figure would arise from a conflict between what is actually seen and what is remembered and recognised to be possible. The representation in memory of what is possible may not to be activated to the same extent in autistic individuals.

The authors claimed that one could interpret these results in terms of perceptual deficits and that the observed performance offers support to both The hierarchisation deficit view and the (lack of) central coherence view.

One might also adopt an executive interpretation whereby perceptual differences are linked to cognitive planning processes. It is held that an autistic individual may suffer from impairments in what are described as "executive" functions which rely upon the maintenance of several representations in working memory. Reference is made to the work of Leevers and Harris (1998) who asked autistic and control subjects to complete a figure in such a way as to make it impossible. In this task which involved little planning, compared to task where one is asked to draw an entire figure from start to finish, the two groups performed similarly. It may be, therefore, that the lack of emphasis upon global features in the drawing of autistics is related to deficits in graphic planning.

It is further suggested that limitations in the capacity to maintain local and global features of the stimulus in working memory might not allow for the normally observed global primacy affect. Meanwhile, in the second study, the a-typical response to the impossibility effect might be linked to a deficit in the ability to maintain several local parts in working memory during planning. This kind of approach might also account for the good performance of autistic individuals in a test like block design where it is actually helpful if there is some limitation in the number of parts that are processed simultaneously.

In any event, it is possible that autistic individuals are limited in the level of complexity of the operations that they can hold in short term memory.

In conclusion, it is confirmed that individuals with autism use visual information in a characteristic way in that they have a tendency towards a local bias on a range of tasks. This

may be related either to some defect in the processing of the global properties of what is seen or to an impairment in the ability to maintain several representations in spatial working memory.


Enhancing Vocalisation

This last section examines a computer based facility for helping children in respect of (verbal) communication

Bemard-Opitz et al (1999) describe a piece of research to compare the effect of computerised visual feedback with traditional play interaction, that is, computer assisted rather than personal instruction, in promoting vocal imitation among children with autism.

Their report begins by reinforcing the point that some deficit or problem in communication is a core component in the diagnosis and the target for treatment of autistics ; but that computer technology is providing a means towards (verbal) communication.

While many non-verbal children with autism have reduced vocal output and deficits in the skills required to imitate sounds, they may still have adequate or better skills in visual discrimination. The implication, therefore, is for the potential value of visual feedback via a computer to guide the production of vocalisations among these children.

Computer-based speech training has already been used among individuals with neurological impairment with whom, for example, the IBM Speechviewer System provides visual feedback on sound production, loudness, and pitch. Meanwhile, this same system has been used to train pre-school children with phonological deficits.

Existing evidence, such as that of Panyan et al (1984), suggested the potential usefulness of computer instruction upon the attention and performance of children with autism, and upon their social skills. Bernard-Optiz et al (opp.cit) also claim that their previous studies have confirmed higher enthusiasm scores, better compliance, and fewer behaviour problems when children are given computer assisted instruction compared with personal instruction. It is acknowledged, however, that results have been mixed when it comes to learning rates.

This presents study was designed to assess the effect of the Speechviewer on the vocal imitation rate of non-verbal children with autism. Ten children, nine boys and one girl, took part in the study. All of whom were diagnosed using standard criteria; and their ages ranged from three years to seven years, with mental ages ranging from nineteen months to sixty three months. Prior to this study all the children could understand and identify fewer than ten words.

The Speechviewer system converts sound production into a graphic display which can vary in colour or size or movement. If a child correctly imitates some vocal stimulus, there is evoked immediately some visual feedback, such as an image of an expanding balloon, on the screen. In the personal instruction, the adult brings the child's attention to a toy ( such as a ball ), then models a sound or a word ( such as "buh" or "ball" ) and reinforces approximations on the part of the child to the stimulus sound by praise and an opportunity to play with the toy.

The results indicated that the children working in the computer condition compared to those working in the personal instruction condition showed significantly greater vocal imitations.

An enhancement of sounds through the visual feedback was effective for nine out of the ten children. Seven children with vocal imitation rates below twenty per cent at the start of the study increased their performance to eighty per cent or more in the computer condition.

it was acknowledged that the novelty of the computer system may have had some impact upon the effectiveness. However, the authors point out that, for the majority of the children, the extent of imitation showed a consistent increase across time. This trend was not limited simply to those early sessions where novelty could be assumed to play a greater part. Furthermore, while acknowledging the small number of participants, it was pointed out that the positive benefits were observed across a wide range of mental ages.

The authors also recognise the importance of the variable of the particular sounds that the children were given. For example, the adults taking part were instructed to model those sounds that were related to the activities on the screen or to objects used during the play setting, and to select sounds that the child were likely to be able to produce. However, while this may well have had some impact generally upon the ease of sound imitation over time, there is no reason why this would favour the computer condition over the other condition.

Meanwhile, eight out of the ten mothers reported that they believed that enhanced learning via the computer assisted programme had led to an increase of sound production over the sessions; and seven mothers rated the Speechviewer system as a positive influence on motivation and learning.

The authors concluded, therefore, that the enhancing of sound through visual feedback does seem to offer considerable promise for the non-verbal autistic children.

It is held that the visual feedback linked to sounds might be a way for non-verbal children to understand one important feature of communication, namely, that sounds have an effect on the environment.

There is also a general support for the view ( Jordan 1988 ) that computer- based learning has the potential to elicit speech in children with autism through visualising sounds.

M.J. Connor September 1999


Bernard-Opitz V., Siriam N., and Sapian S. 1999 Enhancing vocal imitations in children with autism using the IBM Speechviewer. Autism 3(2) 131-147

Frith U., Baron-Cohen S., and Paul R. 1987 Perception in autistic children. In Cohen and Donellan (Eds) Handbook of Autism and Pervasive Developmental Disorders New York: Wiley

Ghaziuddin M. and Gersten L. 1996 Pedantic speaking style differentiates Asperger syndrome from high functioning autism. Journal of Autism and Developmental Disorders 26(6) 585-595

Hare D., Jones J., and Paine C. 1999 Approaching reality the use of personal construct assessment in working with people with Asperger syndrome. Autism 3(2) 165-176

Howlin P. 1997 Autism : Preparing for Adulthood. London: Routledge

Jordan R. 1988 Computer-assisted learning. In Autism Today and Tomorrow. Report of the International Association Autism-Europe (3rd. European Congress)

Leevers H. and Harris P. 1998 Drawing impossible entities. Journal of Child Psychology and Psychiatry 39 399-410

Manjiviona J. and Prior M. 1995 Comparison of Asperger syndrome and high functioning autism on a test of motor development. Journal of Autism and Developmental Disorders 25(1) 23-29

Mottron L., Belleville S., and Menard E. 1999 Local bias in autistic subjects as evidenced by graphic tasks. Journal of Child Psychology and Psychiatry 40(5) 743-753

Ozonoff S., Rogers S., and Pennington B. 1991 Asperger syndrome : evidence of an empirical distinction from high functioning autism. Journal of Child Psychology and Psychiatry 32 1107-1122

Panyan M., McGregor G., Bennett A., Rysticken N., and Spurr A. 1984 The effect of microcomputer based instruction on the academic and social progress of autistic children. Paper presented to the Council for Exceptional Children - Technology in Special Education Conference. Reno : Nevada

Szatmari P., Archer L., Fisman S., Streiner D., and Wilson F. 1995 Asperger syndrome and autism : differences in behaviour, cognition, and adaptive functioning. Journal of the American Academy of Child and Adolescent Psychiatry 34(12)1662-1671

Tonge B., Brereton A., Gray K., and Einfield S. 1999 Behavioural disturbance in high functioning autism and Asperger syndrome. Autism 3(2) 117-130

This article is reproduced by kind permission of the author.

© Mike Connor 1999.

Back to NAS Surrey Branch Welcome Page