Stay up to date with notifications from The Independent

Notifications can be managed in browser preferences.

Scientists link autism with higher testosterone levels

Science Editor,Steve Connor
Wednesday 12 September 2007 00:00 BST
Comments

Powerful evidence has emerged that may soon lead scientists to discover the causes of autism which, in one form or another, now affects about one in every 100 children in Britain.

Scientists have found that raised levels of the sex hormone testosterone in the womb of pregnant women is a significant risk factor in whether a child develops autistic characteristics.

The researchers emphasised yesterday that although they cannot prove testosterone exposure in the womb causes autism, they strongly believe it may be the smoking gun that eventually leads to the source of the brain disorder. Professor Simon Baron Cohen of Cambridge University said 235 healthy children whose mothers had amniocentesis – a womb test during pregnancy – were closely monitored for eight years and tested for autistic-like behaviour at regular intervals during their development.

The scientists found that high levels of testosterone in the amniotic fluid of the womb were significantly correlated with autistic-like behaviour, such as whether the child tends to be more unsociable or less empathetic than normal.

"It's a significant correlation and it's a correlation that remains significant after you have controlled for a whole set of other factors," Professor Baron Cohen told the British Association's Science Festival at York University.

Previous work on animals has suggested that testosterone in the womb may affect the early development of the brain, which in humans might lead to the sort of extreme behaviour typical of autistic spectrum disorder, including Asperger's syndrome.

"What we knew before this study was that foetal testosterone was showing a correlation with social development at earlier points in childhood," the scientist went on. "But we hadn't been able to look at so-called autistic traits before, so in that respect this is something new.

"The idea that foetal testosterone actually plays a causal role in autism is just a hypothesis. So there is no evidence from any lab in the world that this is actually a causal factor, but this research is certainly consistent with that hypothesis."

Professor Baron Cohen has pioneered the "extreme male brain" theory to explain that autism and its related disorders may be a manifestation of being at the end of a wide spectrum of behaviours seen typically in little boys rather than little girls.

Autistic characteristics for instance includes a fascination with numbers and systems – such as collecting cards – rather than conversational play with other children, which typifies the sort of behaviour seen more commonly in small girls.

"Children with autism seemed to have an exaggeration of the typical male profile because they have a very strong interest in systems, like numbers, but have difficulties with empathy," Professor Baron Cohen added.

Cases of autism – or more accurately autistic spectrum disorder – have increased dramatically over the past 30 years. But this is almost certainly due to better diagnosis and a broadening of the definition to cover other conditions, rather than a real increase.

The ethical dilemma

Professor Simon Baron Cohen is a leading proponent of the theory that autistic spectrum disorder represents the extreme end of the range of behaviours commonly seen in little boys rather than little girls: like many autistic children, boys tend to "systemise" rather than "empathise". The professor calls it the "extreme male brain hypothesis", and the latest study on testosterone levels in the womb is consistent with the idea that when brain development becomes "too male" under the influence of the sex hormone, it results in autistic-like behaviour seen in little boys. If testosterone levels are shown to cause – or partly cause – autism it will raise the ethical issue of what should be done medically. To lower testosterone levels in the womb deliberately may decrease risk of autism, but it would almost certainly interfere with normal foetal development.

Steve Connor

Join our commenting forum

Join thought-provoking conversations, follow other Independent readers and see their replies

Comments

Thank you for registering

Please refresh the page or navigate to another page on the site to be automatically logged inPlease refresh your browser to be logged in