Autism : Miscellaneous Thoughts on Nature and Aetiology


These notes begin with a reference to genetic risk for autism and the possible significance of a single X chromosome, which would also provide some explanation for the uneven male-female ratio in diagnosed cases of autism.

Reference is then made to additional evidence or opinion on the MMR issue further indicating the lack of any MMR-autism link.

The final section refers to work which gives a further perspective on the nature of autistic mentalizing, involving the dimensions of systemizing and empathizing, with further implications for the “ extreme male brain ” theory of autism.





The following summaries are linked through their attempts to understand the nature and causes of autism, and the true incidence.

The recent controversy over the possible link between the MMR vaccination and autism may also have intensified efforts to understand the causes of the condition since, while the MMR hypothesis has been consistently rejected, it remains the case that there has been a significant rise over recent years in the reported incidence of children diagnosed with autism.

This has raised the question whether there has been a real rise over the last 25 years or so in the number of children who have autism, or whether it is a matter of a widening of the criteria for diagnosis and the recognition of a spectrum of autistic disorders.



Genetic Risk


Skuse (2003) notes that autism may be as much as 10 times more common among males than among females.  Meanwhile, females with Turner Syndrome (who, like men, have just one X chromosome) are as much as 200 times more likely to have autism than other females.

A logical hypothesis advanced by Skuse is that the most significant risk factor for autism is having the single X chromosome.


A study involving brain imaging, gene mapping, and neuropsychological assessment was then completed on a sample of women with Turner Syndrome and a further sample of women with some genetic abnormality affecting one of their X chromosomes.


As a result it was possible to specify one part of the X chromosome which influences the development of the amygdala which is a part of the brain involved in socio-emotional processing and whose anomalous development has been consistently linked to observed autistic symptoms.  (See “ Autism and the Amygdala ”  MJC  December 2003)

It was further established that those women who lacked the section of genes known to be associated with the development of the amygdala were marked by their difficulty in recognising facial expressions reflecting a particular emotion … like people with autism.


The argument goes, therefore, that it is this particular group of genes whose (non)functioning is critically associated with autism; and that their location on the X chromosome could explain the wide imbalance of the male-female ratio among individuals diagnosed with autism.



Further MMR Evidence


Although there has been a consistent failure to find any reason for maintaining the hypothesis about an MMR-autism link, there has been some speculation whether a particular subgroup of children may prove to have some enhanced vulnerability to autistic symptoms in response to this vaccination.


In response, DeStefano et al (2004) compared the MMR vaccination histories of 624 children with autism and 1824 school-matched non-autistic controls.


It was noted that the children with autism received the first MMR vaccine at similar ages as controls, and the proportions of children vaccinated were similar between the target and control groups, which highlighted the non-association between MMR and autism.

The authors found no such association either in any of the subgroups assessed, including children who showed signs of developmental delay or regression.


Meanwhile, there is further support for the view that thimerosal-containing vaccines have no adverse effect upon neurodevelopmental outcomes. 

In a further paper, DeStefano et al (2003) examined the potential toxicity of such vaccines by looking for any association between developmental disorders and exposure to thimerosal in more than 124,000 infants born between 1992 and 1999 in two districts. 

It  was found that cumulative thimerosal exposure at 3 months showed a significant association with the risk of tics; and such exposure at 3 and 7 months was associated with some non-significant risk of language delay. 

However, there were no associations found between neurodevelopmental effects such as autism or ADHD and thimerosal exposure.


The history of the anxiety over MMR and autism is rehearsed by Skegg (2003)  who begins by commenting how the parents of autistic children usually recognise some anomalous behaviours or development from soon after birth or in early infancy, and how there is clearly a major neurobiologic basis in which a genetic component is significant. 

Skegg goes on to express regret that, even when much evidence and reassurance are available concerning the safety of MMR, there may be a common tendency among the public at large to be suspicious of a cover-up … but reinforces such evidence by citing commentary (Lee et al  1998) to the effect that the original Wakefield et al article of 1998 which discussed a small number of cases with no control group did not adequately establish a temporal association with autism, let alone a causal relationship.

He refers also to the Danish study by Madsen et al  (2002) whose retrospective data about more than 500,000 children indicated no evidence for any link between MMR and autism.


Nevertheless, there has been a rise in the apparent incidence of autism, and, in seeking to understand the observable statistics, Skegg refers to the work of Jick and Kaye (2003) who showed that the marked increase in the incidence rate of recorded autism among boys in British general practices from 1992 to 2000 was balanced closely by a decline in the recorded rate of certain developmental disorders without a diagnosis of autism. 

In other words, the statistics may be explicable at least to an extent in terms of shifting diagnostic criteria and practices.

A direct reading of the Jick and Kaye paper (2003 – op.cit) revealed that it was responding to two main questions that had arisen over the last few years, viz., has the number of children diagnosed with autism progressively increased over the last decade or so, and, if so, can one identify any environmental exposure by which to explain the increase?


Their review paper emphasises the important genetic component, although they acknowledge that it is not known how many genes are involved. Twin studies cited suggest that there is considerable heritability to the liability of autism, but Jick and Kaye hold that it is hardly plausible to explain the large increase in the number of children diagnosed with autism in terms of some change in the genetic characteristics of parents during the recent past.


They then refer to the various risk factors considered as possibly linked to autism noted by the review completed by the UK Medical Research Council (2003).  These included exposure either before or after birth to drugs, vaccines, infections, heavy metals, and physiologic abnormalities affecting the gastrointestinal tract and the immune system.  Reference is also made to cognitive deficits, epilepsy, depression, and anxiety …. albeit with the support for such hypotheses still at a largely anecdotal level.

Jick and Kaye recognise that available data cannot rule out totally the possible aetiological influence of any of these factors in occasional cases of autism, but they also note how converging views and evidence, and their own study of the histories of autistic and non-autistic boys, argue against the possibility that any one of these factors or any permutation thereof could underlie the marked rise in the number of children diagnosed with autism in recent years.  There was little, if any, difference in the frequency of these conditions in boys with autism compared with non-autistic boys, or in the mothers of autistic and non-autistic boys.


The results of their own survey indicated that the pattern of diagnosing autism in children changed during the 1990s in that, during and before the early 1990s, behavioural disorders tended not to lead to a diagnosis of autism.  Subsequently, there was an increasing tendency to diagnose children with such disorders with autism.


The authors concluded that these data suggest that the increased incidence of diagnosed autism is primarily a reflection of changes in diagnostic practices such as improved identification and availability of services.


A Widening of the Scope of Perceived Autistic Characteristics


The work by Baron-Cohen et al (2003) on the “Empathizing-Systemizing” dimension of autism is based upon a recent model of sex differences in cognitive style, and is relevant to the view of some autistic characteristics as exaggerated forms of male behaviour.


Systemizing is seen as a powerful way of understanding and predicting the law- governed world of inanimate machines or objects.  Empathizing is an equally powerful way of understanding and predicting the social and emotional world through identifying another person’s feelings and thoughts, as a result of which one can predict that person’s behaviour and gain insight into how best to respond.

Systemizing seeks to derive the rules which govern the behaviour of a system such that one can predict its actions and gain a degree of control over it.


It is held that there are various types of system, including technical, natural, abstract, social, organizable, and motor; and each may be understood in terms of an input, an operation, and an output. 

The authors use as an example of a technical system the effect of a sail (the input) in response to the angle at which it is set (the operation), and the speed achieved (the output). An example of a natural system is that of a particular flowering plant (input) and variation in the acidic or  alkaline quality of the soil (operation), and the output is observed in terms of the colour of the petals.


It is further argued that, typically, males spontaneously systemize to a greater extent than do females; and females spontaneously empathize to a greater degree than do males.  The two processes are quite different in that systemizing would be largely valueless in predicting ongoing changes in a person’s behaviour, with empathizing  needed if one is to attribute a mental or emotional state to another person and to respond appropriately.  (There is the further complication that the empathizer does not realistically expect there to be a consistent or “governable” relationship between mental state and behaviour, but can only recognise how that mental state may have an influence or a constraining effect upon overt behaviour).

While there are exceptions to a general rule, males are commonly more interested in subjects which require a high level of systemizing, such as physics or engineering; while females are commonly superior at decoding non-verbal information and tuning-in to clues from tone of voice or expression.


So, individuals whose empathizing capacity is the more developed (usually females) may be described as Type E, with those with greater systemizing capacity (usually males) described as Type S.  Many autistic individuals may be perceived as being highly Type S with their fascination with objects but their relative lack of Theory of Mind.


Baron-Cohen et al (op.cit) go on to describe the “Extreme Male Brain Theory of Autism” which was first suggested by Asperger (1944).  The argument is that, in autism, the typical male style or cognitive emphasis is exaggerated to an extreme level.

There is already evidence (Baron-Cohen et al 1999) that on the Faux Pas Test, females are usually better than males at judging what may prove socially insensitive or offensive, and individuals with autism show even lower capacity on this kind of test.  Further, the autistic individual may be limited in respect of empathy but may have islets of ability in areas such as mathematical processes or calendar calculations or music or memory for details … all of which involve factual knowledge or the use of rules, and may be seen as highly systemizable.


The authors suggest that this extreme male brain theory can now be tested by means of this formulation of Type E and Type S individuals, and by the use of self-rating questionnaires which can measure an individual’s place within the systemizing and empathizing spectra and produce a systemizing and an empathizing quotient.

Each questionnaire includes 60 questions, of which some are filler (control) items.


Examples of questions from the Systemizing Scale include ….

“When I listen to a piece of music, I always notice the way it is structured”

“If there was a problem with the electrical wiring in my home, I’d be able to fix it myself“

“I enjoy participating in sport”


Empathizing questions include …..

“I can easily tell if someone else wants to enter a conversation”

“It doesn’t bother me much if I am late meeting a friend”

“I enjoy being at the centre of attention at any social gathering”


Filler questions include …..

(SQ)  “I adhere to common superstitions”

          “I am interested in learning about different religions”

(EQ)  “I prefer animals to humans”

          “I dream most nights”


Questions are to be answered by a choice from a 4-point range : strongly agree - slightly agree - slightly disagree - strongly disagree.


These scales are designed for adults, and the authors used them in assessing the styles of normal adults, comparing a sample of individuals from a range of occupations and a sample of high-achieving students; and a sample of individuals diagnosed with Asperger syndrome or high functioning autism with a history of mainstream schooling and drawn from a range of occupations compared with a control group matched for gender, age, and handedness.


As predicted, among the samples of non-autistic individuals, the males scored significantly higher than the females on the Systemizing Quotient, and females scored higher on the Empathizing Quotient.  A clear and inverse correlation was observed between the SQ and the EQ, further reinforcing the view that the two processes involved are quite different.


In the comparison of individuals with autistic spectrum disorders with normally- functioning individuals, it was found that the autistic group scored significantly higher on the SQ and significantly lower on the EQ …. in line with what would be predicted from the extreme male brain theory of autism.


The authors argue that these findings can be seen as reliable in that, if the autistic group had been somehow disadvantaged overall, this would have been evident on the scores from both questionnaires, whereas the observed pattern was as predicted by the theory.  Further, the groups did not differ in their responses to filler items. In the study of students and non-students, neither age nor level of education led to differentiable results; instead, the observed results would have been best predicted on the basis of gender.


The authors note that the pattern of scores would not have been predicted by existing theories of autism, such as the Executive Dysfunction Theory or the Weak Central Coherence Theory since many aspects of systemizing would involve adequate executive functioning, and a weak central coherence would predict poor scoring on both the systemizing and the empathizing scales.


One possible objection to the Type S and Type E theorising is acknowledged .. viz .. some systemizing skills and empathizing deficits might have been expected as a result of the way in which individuals with autistic spectrum disorders are identified in the first place.

However, this objection could be countered by pointing out that DSM-IV does not use information about systemizing; and, while poor empathy may be a diagnostic pointer, it is not quantified in the formal diagnostic criteria.


The authors conclude that the systemizing-empathizing dichotomy provides a fresh viewpoint for understanding autistic behaviours, although it is it is not clear what are the neurocognitive mechanisms underlying these processes.  The question is raised whether the two processes are independent, or whether they share some neural system or structure so that as one of the capacities improves, the other deteriorates.


The significance of certain brain structures, such as the amygdala or the medial-frontal cortex, is now recognised as a basis for empathizing, but the brain basis for systemizing has yet to be determined. 


                                      *          *          *          *          *        


M.J.Connor                                                                                                March 2004





Asperger H.  1944  Die Autistischen Psychpathen im Kindesalter.   Arch. Psychiatr. Nervenkrank 117  76-113


Baron-Cohen S., O’Riordan M., Jones R., and Plaisted K.  1999   A new test of social sensitivity : detection of faux pas in normal children and children with Asperger syndrome.   Journal of Autism and Developmental Disorders  29  407-418


Baron-Cohen S., Richler J., Bisarya D., Gurunathan N., and Wheelwright S.  2003   The systemizing quotient : an investigation of adults with Asperger syndrome or high functioning autism, and normal sex differences.  Phil.Trans.R.Soc. Lond. 

(Monographs of the Royal Society – published online  DOI 10.1098/rstb.2002.1206)


DeStefano et al  2003   Thimerosal exposure and neurodevelopmental outcome.  Reuters Health Information (November 2003).  Original paper in Pediatrics 112   1039-1048


DeStefano F et al  2004   Interview with Reuters Health, under the byline “ No link observed between MMR vaccination and autism ”   Megan Rauscher :  Medscape, 19 February 2004


Jick H. and Kaye J.  2002   Epidemiology and possible causes of autism.   Pharmacotherapy  23  1524-1530


Lee J., Melgaard B., Clements C., Kane M., Mulholland E., and Olive J.  1998   Inflammatory bowel disease and MMR vaccine.  (Letter)  The Lancet  351 p.905


Madsen K., Hviid A., Vestergaard M., et al  2002   A population-based study of MMR and autism.   New England Journal of Medicine  347  1477-1482


Skegg D.  2003   Autism and MMR vaccine : a challenge for pharmacoepidemiology.

Pharmacotherapy  23(12)  1521-1523


Skuse D.  2003   Presentation to the British Association Festival of Science

(University of Salford : September).  Reported by Jarrett C. “ X-plaining autism ” The Psychologist 16(11)  November p.574

This article is reproduced by kind permission of the author.

© Mike Connor 2004.

Back to NAS Surrey Branch Welcome Page