We have learnt a great deal about the causes of autism in recent decades, but much work remains to be done. We know that it is the result of multiple factors acting together: both genes and environmental influences are important, and there may be differences in the way the brain is wired up during development. We also know that there is absolutely no connection with vaccinations, a fact proven many times over; the original research publication suggesting this link has been retracted (i.e. has proven fraudulent), and the researcher responsible has been removed from the UK medical register.


Autism can sometimes run in families, indicating that genetics plays a key role in the disorder. Early research into the genetics of autism began with studies of twins, both identical and non-identical. Scientists studied the concordance of autism in these sets of twins – the percentage chance that if one twin had autism, the other would too. They found a significantly higher concordance in identical twins, who have largely identical DNA, suggesting that autism has a genetic basis. Under current estimates, there is a 50 to 80 per cent concordance for identical twins; this compares to an overall ~1.5 per cent chance that any two people will have autism. If parents have one child with autism, there is a 5 to 20 per cent chance that their second child will also have autism. 

We now know that as with numerous other traits, including height and intelligence, autism spectrum disorder involves many different genetic variations acting together. No single genetic mutation is enough to produce autism, but the more mutations there are in specific parts of an individual’s DNA, the greater the chances of an autism diagnosis. To identify the exact regions of DNA associated with the disorder, scientists study large numbers of people affected by autism. This approach is called a genome-wide association study (GWAS). QBI Professors Naomi Wray and Peter Visscher are world leaders in such studies.

Environmental factors

As with genetics, no single environmental factor definitively causes autism. Instead, a range of environmental influences seems to exist. For example, an epidemiological study led by QBI Professor John McGrath found that women who are deficient in Vitamin D during pregnancy had children with an increased risk of autism.

A lack of Vitamin D is just one risk factor, but it can also interact with other environmental variables. Associate Professor Thomas Burne, for example, is looking at how Vitamin D deficiency can combine with low levels of alcohol during pregnancy to raise the risk of autism-like behaviours in mice.

Another environmental risk is viral or bacterial infection of the pregnant mother during the first trimester, which has become a common way to study autism in model animals like mice. The laboratory of Professor Darryl Eyles has shown that the presence of Vitamin D protects against autism-like behaviours that are triggered by infections of the mother. The work of Professors McGrath, Eyles and Burne converges in the long-term hope that by supplementing pregnant mothers with Vitamin D, autism rates might be reduced. 

Finally, more research by Professor McGrath has found that older fathers – particularly those over 50 at the time of conception – have a greater chance of having a child with autism, potentially as a result of new genetic mutations that affect brain development. Similarly, experiments in mice have shown that older fathers are more likely to produce offspring with behavioural and genetic changes indicative of autism.

Brain wiring

In the developing brain, connections between and within brain regions are still being formed. Sometimes, though, these connections don’t form in the normal way, which can produce lasting changes in behaviour.

One example relevant to autism is a congenital condition called corpus callosum dysgenesis (CCD). The corpus callosum is a thick bundle of nerve fibres that connects one brain hemisphere to the other. In people with CCD, the corpus callosum is partly or even completely missing. Strikingly, one third of patients with CCD meet the diagnostic criteria for autism. What’s more, the genes involved in CCD and autism are similar. Other studies, however, have found the opposite profile in models of autism – an enlarged corpus callosum.

How can these findings be reconciled? QBI’s Professor Linda Richards is working on an international effort with leading paediatric neurologists, radiologists and clinical psychologists to find out. The international effort is called the International Research Consortium for the Corpus Callosum and Cerebral Connectivity (IRC5) and currently includes investigators from Australia, the USA, Europe and South America. According to Professor Elliott Sherr of the University of California, San Francisco, who is part of this team, one possibility is that autism could result from inefficient communication between brain hemispheres, and that efficiency could be disrupted whether the corpus callosum is too big or too small. 


  What is autism?

  Autism signs and symptoms

  Autism treatment

  Diagnosing autism

Connect with us


Help QBI research