This brief paper simply summarises the contents of some articles which are linked together by a particular theme.. namely, support for the belief that there is an emphasis upon the psychological and the physiological levels of assessment at the expense of the instructional level (as well as the simple fact that the present writer gained access to these articles at much the same time!)

Reference is made to the link between Autism and various medical conditions, to the nature of Autism and its link to what is seen as extreme versions of typical male behaviour, and to the conclusions of a study which sought to examine evaluations of existing educational interventions.

M.J.Connor November 1998


In respect of many conditions, it becomes evident that much of the research evidence concerns the source of the problems and the nature of the condition, and there is a relatively little information concerning management or treatment.

A good example of this is Dyslexia where there is a great deal of research devoted to the etiology of the difficulties, including familial history, and to the presenting characteristics. The emphasis is upon a psychological assessment in which one compares the child against characteristics allegedly specific to dyslexia, or upon physiological assessment where one examines where in the central nervous system the problems are located. Only recently has there been significant attention to the instructional level of assessment whereby one begins to evaluate the available programmes of intervention.

It may well be argued that a similar situation relates to Autism and Asperger Syndrome, and a current set of articles reinforced the perception of an emphasis upon describing autistic disorders and upon seeking to explain their source, alongside an acknowledgement that there has been little systematic evaluation of treatment methods.

The articles in question began with a further exploration of the possible link between known medical conditions and autistic disorders.... i.e. the physiological level of assessment.

Barton and Volkmar (1998) describe how there is much research to support the view that neurological or biological features are central in the etiology of Autism. This relationship was first highlighted when it was noted that fits occurred very frequently among people with autistic disorders.

There is also evidence to suggest that genetic features are significantly involved. For example family or twin studies demonstrate that concordance rates are higher for monozygotic rather than dizygotic twins, and that there is an increased risk for Autism in those families where there is already an autistic child.

What is less clear is the significance of other medical conditions in Autism. One school of thought (Rutter et al 1970 ) argues that about ten per cent of autistic people have congenital medical conditions which influence the central nervous system and which may contribute to Autism; and it is noted that the link between autistic disorders and medical conditions appears dependent upon IQ, supporting existing evidence that the more severely retarded children (including those with Autism) are more likely to show some associated medical condition.

Another school of thought (Gillberg 1993) holds that there is a much higher frequency of associations between autistic disorders and medical conditions. It is estimated that around one third of cases of autism may be linked in this way and that the medical conditions concerned include metabolic and immune related diseases.

The problem is that there are insufficient data to determine whether medical disorders are truly significant in the development of Autism or whether they are incidental. Meanwhile it may be true that some medical problems do occur more frequently among persons with Autism but there is no evidence to indicate a causal link. For example, in respect of seizures, it is unclear whether they precede the onset of autistic symptoms or arise as a result of the same neurological injury or deficit which underlies Autism.

It is also possible that different conclusions result from differences in the diagnostic criteria which are used in the first place. It has been found, for example, that the DSM III (revised) criteria tend to over-diagnose Autism particularly among those people with the more severe mental retardation.

This current study set out to examine these issues and used a sample of two hundred and eleven children and adults all diagnosed as autistic or as having pervasive developmental disorders.

Information on medical conditions was obtained and fully recorded and the first outcome was to support the earlier findings of over-diagnosis of Autism when DSM III R is used as opposed to DSM IV or DSM III.

A whole range of medical conditions were observed of which thirty four per cent were linked with individuals diagnosed with autistic disorders. These conditions included seizures, hearing loss, viral illness, foetal alcohol syndrome, microcephaly, maternal rubella, Down's Syndrome, Goldenhar Syndrome, lead poisoning, and hypothyroidism. All the conditions were observed to occur with the highest frequency among those people who had been diagnosed autistic by DSM III R.

In respect of IQ, the results indicated that medical conditions are associated with lower IQ scores regardless of the diagnosis.
However, the overall conclusion was that when one uses strict diagnostic systems there is no difference between a high IQ subjects and a low IQ subjects in respect of the incidence of medical conditions linked to autism. This leads to the suggestion that the more important relationship is between severe mental handicap, rather than Autism per se, and medical conditions.

A range of medical conditions was noted among individuals with autism or mental handicap; but among the autistic sample the medical conditions that seemed most relevant, including rubella, were observed among the more severely handicapped individuals.

The authors concluded that the issue of atypical autism remains an important one for continuing research because it may be among the group of people so-diagnosed, especially those with more severe retardation, that there is most relevance for physiological disorders.


The second article in the current set appeared in the national press and describes how autistic characteristics may be quite common in the population and that among certain professions or occupations they can be of positive benefit. The emphasis, again, is about etiology, physiology, and describing the nature of the condition.

This article by O'Connell points out that autistic spectrum disorders affect men far more frequently than women, and that the most striking characteristic is the failure to understand how other people think or feel.

The author cites the possibility that Theory of Mind has been a critical force behind human evolution. It is argued that the human brain would have increased in size and complexity as a result of the need to determine what are people are thinking or planning and specific areas of the brain have been devoted to social cognition.

It is reported how Baron-Cohen and his team took brain scans of normal subjects and people with Asperger Syndrome while undertaking a test of social intelligence. The test involved guessing what a person was thinking or feeling on the basis of photographs of that person's eyes. It was found that, among the normal subjects, the frontal and temporal regions of the brain were involved in these tasks, along with the amygdala; but among the Asperger sample, this latter part of the brain appeared not to be brought into action. There is, therefore, physiological evidence for altered thinking.

This team also conducted a survey of the fathers and grandfathers of people with Autism and found that they were twice as likely to be engineers than the relatives of non-autistic people; and it is noted that engineering is a predominantly male profession where there is no reliance upon a social intelligence.

A further study involved asking students at Cambridge University whether they had relatives with Autism or with autistic characteristics and it was found that students of science were six times more likely than students of the humanities to have autistic relatives.

The conclusion holds that there is no specific link between science and Autism but that the skills required for being a good scientist are those that might be regarded as extreme versions of typical male characteristics.

It is reported that evidence will shortly be published suggesting that a gene for autism is likely to be carried on the X sex chromosome. Girls inherit one from each parent while boys only have one... from the mother ; and the theory has it that some protective factor on that chromosome may be switched off when inherited from one parent and switched on when inherited from the other. In particular it is argued that it is always switched on when transmitted by a father and switched off when transmitted by the mother and, because girls carry the father's X chromosome, they tend to be protected to some extent from autistic disorders.

In the event the team argue that autistic behaviour may be regarded as an extreme form of male behaviour; is, therefore, not as unusual as might have been thought among the population as a whole; and might well carry some advantages for certain situations or task demands.


The final article in the set is by Jordan and her colleagues who have reviewed the evidence for the effectiveness of educational interventions for children with Autism.

They conclude that there are clear benefits from some form of early intervention and that approaches that involve a degree of integration as part of the total approach will also be likely to have positive outcomes.

It appears that one cannot be more specific in the conclusions... i.e. the instructional level of assessment remains limited... for a number of reasons. For example, there appears to be no consistent and standardised system for assessing the children and for measuring progress. Further, the interventions need to be independently evaluated and to be compared one with another in a way that allows for the control of all the competing variables, such as intensity.

In particular the authors argue that research is needed on the child characteristics that lead to success in different approaches, and on The parental characteristics that lead them to favour a given approach.

Particular mention is made of the Lovaas approach and it is pointed out that positive results may stem from its intensity rather than the actual nature of the treatment, and there are problems in finding a clear scientific basis for the approach given that the original sample were treated with aversives while modem versions of the approach avoid aversives.

The authors finally conclude that there is no clear evidence to suggest that one approach is better than any other, although it is found that early intervention using a programme which involves the parents and which involves direct teaching of essential skills will be the most likely to produce positive outcomes.

M.J.Connor November 1998


Gillberg C. 1993 Autism and related behaviours. Journal of Intellectual Disability Research 37 343-372

Jordan R., Jones G., and Murray D. 1998 Educational interventions for children with

Autism. DfEE Research Report No.77.

O'Connell S. 1998 why men become anoraks. The Independent 13.11.98

Rutter M. 1970 Autistic children: infancy to adulthood. Seminars in psychology 2 435-450.

Barton M. and Volkiriar F. 1998 How commonly are known medical conditions associated with Autism? Journal of Autism and Developmental Disorders 28(4) 273-278.

This article is reproduced by kind permission of the author.

© Mike Connor 1998.

Back to NAS Surrey Branch Welcome Page