These notes begin with further thoughts concerning the extreme male brain hypothesis and the possible implications for an understanding of aetiologies.

The next section describes the outcome of a survey comparing progress of children with ASD placed in mainstream and specialist schools, with the result that the school placement does not appear to be the critical factor in determining social or scholastic progress; rather parental coping strategies plus child access to additional facilities such as speech and language therapy are significant.

The final section continues the themes of the non-association between vaccines and the aetiology of SD and other conditions; and the possible value of psycho-stimulant medication for children with ASD and ADHD albeit with caution required in the light of adverse side effects and the limiting of benefits to a sub-sample of this population.


M.J.Connor                                                                                              January 2006


The Extreme Male Brain Hypothesis


The work of Simon Baron-Cohen has included the exploration of the proposition that individuals with autism have an imbalance in two cognitive capacities or styles with the net effect of evoking behaviour which is significantly more characteristic of males than of females.


In an interview (reported by White 2005) Baron-Cohen differentiates systemisers from empathisers.  Males are seen as having a stronger drive towards systemising while females have a stronger drive to empathise, and, in attempts to understand autistic behaviour, it has been noted that individuals on the autistic spectrum show an exaggeration of this male style.  There are more systemisers in this population.


Systemising is a matter of trying to understand the workings of some mechanical or natural or abstract system, be it a machine or a part of the body or some mathematical construct. 

Empathising is about being able to imagine what someone else is thinking or feeling and having an (appropriate) emotional response to that other person’s mental state. 


Empathising is perceived as being more than simply a well developed Theory of Mind in that there are these two components …. not just being able to see another person’s point of view or to recognise his or her feelings, but also having the necessary facility to make a suitable response.  This latter, above and beyond the perception of the other person’s mental state, is the significant part of empathising and what distinguishes it from Theory of Mind capacities per se.


Baron-Cohen goes on to describe how the two capacities were assumed initially to be independent, but longer term observations have indicated a relationship in that individuals in the population at large who are strong in one area tend to be weak in the other.  There is a small but consistent negative correlation, with the possible implication of dependence on a shared mechanism such as some genetic or hormonal influence.


The capacity for empathising is highlighted by assessments which involve the identification of emotions from photographs of faces, and evidence indicates that individuals with autism find this difficult. This pattern relates to sex differences in the general population in that women are more successful than men in this kind of task.

An alternative assessment involves asking someone to listen to a passage of speech and to detect if it contains some hurtful element.  The same patterns of findings emerges in that autistic compared to non-autistic participants, and boys/men compared to girls/women, consistently do more poorly in such a task.


On the other hand, any test that involves systemising tends to produce the better scores among children and adults with autistic spectrum disorder.

Further, Baron-Cohen refers to children with more severe levels of autism (those with little language and scoring low on cognitive measures) as having strong obsessional behaviours and a tendency to act in a repetitive way which could be interpreted as a form of systemising … a pre-occupation with establishing or maintaining a specific and local “system”.

With regard to (measured) ability, there is a developing view that the frequency of finding low levels of general cognitive functioning may be, in some cases at least, an artefact of this systemising-empathising dichotomy.

The stronger the trend towards systemising, the narrower the attention to the environment as a whole; or the drive to focus upon one activity and to repeat it over and over replaces the opportunity to become familiar with other elements of the surroundings, be they objects, ideas, or people.  The net effect would be to depress cognitive scores and identify the child as having learning difficulties.


With regard to “pathways”, Baron-Cohen discusses the concept of  “ assortative mating ” which describes how two individuals are more likely to be mutually attracted and to form a relationship, perhaps leading to long-term commitment and to raising a family, if they share characteristics. 

Individuals tend to choose partners who are seen as similar to themselves, with the possible implication that systemising and empathising are two of the observable characteristics that guide the choice. If the two members of a couple are strong on systemising, they are more likely to produce children with such a style which, in some cases, is associated with an autistic spectrum disorder. 


Evidence cited in support of this supposition emerges, firstly, from a trend in findings from the Embedded Figures Test that both the parents of children with ASD show good attention to detail; and, secondly, that they both show more-than-chance probability of having fathers whose work is within an area requiring good systemising skills  …. ie  good systemising appears commonly to run in both sides of these families, indicative of assortative mating.  (The example is given of Silicon Valley in California where so many people work in the (strongly systemising) computer science industry and where there has been an increase in the incidence of children diagnosed with ASD above the nationally-typical rate …. although it is stressed that the data remain anecdotal and not subject to formal investigation).


This is all further evidence of the strong genetic influences, even if there is a long way to go in identifying which genes are the significant ones and precisely which functions they control.


A related area of study, the significance of foetal testosterone exposure, has offered some further support for this male brain hypothesis in that the higher the level of foetal testosterone (as assessed via an analysis of amniotic fluid) the less eye contact made by the child at age 12 months and the slower to develop language skills.

Following up the same children at age 4 years has shown a link between the foetal testosterone level and the greater level of social difficulty and the narrower range of interests.  The possible implication is that the heightened focus upon this small range of interests  … the systemising style … inhibits social interaction for which a more empathising style would provide a firmer basis.


One area of speculation is the precise role of the X chromosome and the difference in influence according to whether it is inherited from the father or mother.  The X chromosome does have a role in the regulation of testosterone, thus indicating a possible link between genetic and hormonal action.


In sum, the “extreme male brain” could be the end point of a number of pathways, with a need for further studies involving larger samples and the opportunity to make and test predictions about behaviours and capacities.

As a “stop press“ addendum to the above, one notes the item (Baron-Cohen 2006) in which he responds to a suggestion, emerging from a presentation at the British Association Fair  (David Skuse, October 2005), to the effect that the theory of extreme male behaviour is not supported by available data. 

Instead, it was argued that, in typically developing individuals, girls were better than boys at identifying emotions during childhood and early adolescence, but this advantage was no longer evident by late adolescence; and there seemed no sex differences at any stage in respect of remembering faces or detecting eye contact.


Baron-Cohen’s response included the suggestion that one has still got to explain this better emotion-recognition skill among girls, detectable as early as 6 years of age, and that the pattern is in line with his empathising-systemising view of gender difference.


He goes on to argue that one can cite a range of studies to indicate that this female advantage in recognising emotions does persist into adulthood; and that differences between outcomes of his studies and those of Skuse may be an artefact of the precise methods and assessment instruments used.  For example, differences at later ages may have been obscured by ceiling effects within the tests; and the detection of differences requires a test that is sufficiently challenging to give rise to a range of scores.

Baron-Cohen also refers anecdotally to the experience of producing a CD-Rom to illustrate the wide range of human emotions during which there was much more difficulty in identifying male actors than female actors with the capacity convincingly to register complex emotional expressions. It may be, it is suggested, that sex differences include demonstrating emotions through facial expressions as well as reading them.

His final point is that a capacity to recognise faces or to detect eye contact may be as much to do with memory or being able to calculate the direction of gaze as much as any skill in empathising.


Outcomes of mainstream and Special School Placements in Children with ASD


The work by Waddington and Reed (2005) was stimulated by the current emphasis upon mainstream inclusion for children with SENs with its implication that “included” children will enjoy a better quality of life, and achieve better social and educational performance levels.

They point out that, despite this being a critical area for the development of policy and practice, there is very little research data available dealing with the inclusion of children with ASD; and, it might be argued, the policy of inclusion for these children has preceded the gathering of evidence whether it is actually better for their success.


Their own archive study investigated whether children in mainstream placements show enhanced performance relative to those children in specialist placements.


In their introduction, the authors note that “inclusion” or “integration” can mean many things and may apply to children who are in full time attendance at a mainstream school or to those who spend only some limited time in the mainstream setting to pursue particular activities.  This variation is seen as a major inhibitor to making sense of existing studies which have investigated the effects of inclusion or which have compared integrated with segregated provision (with studies frequently failing to define what is meant by inclusion or integration).


However, in their review of existing evidence, they find equivocal results in that, for example, one identified study failed to find language differences either in general level or in rate of progress between children in segregated and integrated provision.


Another identified study indicated that young children appeared to gain greater benefits socially from integrated placements, although a further set of findings indicated that this was a function of the presence or absence of the non-ASD peers. Teachers were observed to interact less with the ASD children when their peers were present.  The target children were also observed to initiate very few interactions in either setting, and it was concluded that physical integration per se was not sufficient to bring about social contact between the children with ASD and peers.


The extent of existing literature was acknowledged to be limited and what evidence existed was seen as inconclusive about whether or not inclusive settings benefit the child.  The ideal of inclusion, therefore, cannot be said to have evidence-based foundations.


Their own study involved “secondary data”, ie information already collected by other investigators often in natural settings and for a range of reasons.  The sources of data for analysis were the LEA archive files about children in their schools from which some elements could offer insight into the performance of children in integrated settings and into the factors associated with positive outcomes.


The sample of children targeted in this current study comprised 73 boys and 11 girls from three boroughs in the South-East of England.  Criteria for inclusion was that they had a diagnosis of ASD and left school no more than 5 years ago.

Measures were gathered about two areas … the child outcomes in terms of national curriculum levels and school placement, and about the interventions that had been made available, such as the amount of speech therapy time and previous access to Portage sessions. 

In addition, parental questionnaires were completed in respect of diagnosis, parenting stress, and developmental/medical history.


The findings showed that, in line with expectation/policy, the children with ASD were more likely to be in mainstream settings.

The severity of the autism appeared not to be a factor in placements, nor the particular characteristics or profiles of the children, with the apparent implication that placement was not linked to precise levels or type of need, but a function of a combination of some other factors such as the level of resources, or governmental policies.


Those children placed in units attached to mainstream schools were receiving more speech therapy than those placed in specialist or mainstream schools, although here was no difference in the number of hours of SLT available to the different settings.


With regard to academic outcomes, no significant differences were found, indicative of a lack of direct impact upon achievement from mainstream schooling.


Significant correlations were found between certain predictive and outcome measures.  For example, social skill group attendance was linked with improved literacy scores

(reading and spelling) suggesting that social skill training may lead to generally improved language skills in children with ASD.

Meanwhile, whether or not the children had been involved in Portage facilities did not influence placement in that those children with a history of Portage and those with no such history were placed equally frequently in mainstream and specialist settings … although the numbers were such as to keep conclusions tentative.

Parental coping abilities were correlated positively with better results in writing and science.  Their access to social and spiritual support was related to better child outcomes in maths, reading, writing, and science; and the willingness of the family to acquire and accept help was linked to better outcomes in English, writing, and science.


In their conclusion, the authors repeat that there is no indication that children in mainstream schools are more successful than those in specialist provision, but acknowledge limitations in terms of the gaps in the data and inconsistencies in the archive files.

The implication is that, if one is to be able to cite truly evidence-based planning, there needs to be access to more data such as fuller NC results, and reports plus assessment data from educational psychologists.


With the addition of questionnaire information, it was possible to highlight how school placement is not the major variable in achievement, but that interventions such as social skill sessions and the maintenance of full attendance are more significant.


Parental factors also play a big part in determining child outcomes in that, for example, the child’s grades will be the poorer as the parental level of stress increases.

The implication here is to identify the level of support that parents need if they are effectively to manage the child with ASD and manage their own potential stress, and to combine LEA and Health Authority resources in identifying and meeting this need.  


Ongoing Debates :  1.Vaccines and ASD   2. ASD, ADHD and Stimulants 


One might have thought that there is sufficient reassurance that vaccines (or the mercury-based preservative in vaccines) are not linked in any causal way with ASD or any other conditions (such as asthma).


As a further study reiterating the same point, one can cite Orenstein (2005) whose review of a range of investigations concludes that research does not implicate thimerasol-containing vaccines or the MMR vaccine in the aetiology of autism or infection.  However, his presentation also included a reference to continuing anxiety, reflected in a refusal among a minority of parents to have their child treated with a generally-recommended vaccine.


Orenstein acknowledges the critical role of the medical profession in ensuring the safety of vaccines, but argues, too, that the science behind all studies should be made explicit.  


For example, he describes one study which appeared to raise parental concerns about thimerasol and autism but which was more ecological than epidemiological in that it did not take account of changes over the recent past in diagnostic criteria and processes, or the increase in the range of services available for children with ASD  which might well have shifted the diagnostic categorisations used.


Further, he was able to cite well-validated studies which have highlighted the non-comparability of the features of children with autism and those of non-autistic children with mercury poisoning.  Studies purporting to show a link between vaccines and ASD commonly appear, he argues, to be short of the epidemiological study quality criteria.

There was also the study recently conducted in Sweden and Denmark which demonstrated that, when thimerasol was removed from vaccines, there was no corresponding change in the rates of autism reported in either country.


Orenstein concludes by expressing the hope that studies can shift away from this area towards more useful research given that trying to find some link between vaccines and morbidity is not leading to productive outcomes.


On the other topic ….. one notes the study by Aman et al (2005) which has shown that around half of children with pervasive developmental disorders and ADHD respond well to methylphenidate. 

However, it is recognised that previous studies examining the effect s of stimulants on children with this kind of comorbidity have yielded mixed results.


Further, in this present study, there were found to be just short of 50% of the sample of children for whom the efficacy of methylphenidate was demonstrated, so that there is a subgroup of this population who are likely to benefit; it is not an intervention that will benefit all cases. 

It was also noted that the beneficial effects were less than those observed among typically-developing children with ADHD and that side effects were more common … such as decreased appetite, interrupted sleep-onset patterns, irritability, and emotional outbursts.


In other words, while the use of psycho-stimulant medication is seen as a potentially useful intervention, it will not be appropriate or free of adverse effects for more than a sub-sample of cases, highlighting again the heterogeneity of this population and the need for individual planning and close monitoring of initial outcomes of interventions.


                                 *          *          *          *          *          *

M.J.Connor                                                                                              January 2006





Aman M. et al  2005    Some children with ASD and ADHD benefit from methylphenidate.   Summary provided in Reuters Health Information (7.11.05)

Original article in Archives of General Psychiatry 62  1266-1274


Baron-Cohen S.  2006   Finding sex differences.  Letter to The Psychologist 19(1) page 14


Orenstein W.  2005   Presentation to the American Academy of Pediatrics National Conference.  Plenary Session.  11.10 05


Waddington E. and Reed P.  (University of Wales, Swansea)  2005   Comparison of the effects of mainstream and special school placements on outcomes in children with ASD.  Report for the SE Regional Special Educational Needs Partnership.


White R.  2005   The neuropsychology of autism and pervasive developmental disorders; an expert interview with Simon Baron-Cohen.  Medscape Psychiatry and Mental Health  10(2)  14th December 2005



This article is reproduced by kind permission of the author.

© Mike Connor 2006.

Back to NAS Surrey Branch Welcome Page