The brain biology of autism

Debunking the myth that environmental pollutants play a role, asking how testosterone shapes the brain in the womb, and looking into the future, could genetics help screen and diagnose babies? Next we speak with the scientist who came up with the autism systemising hypothesis. Professor Simon Baron-Cohen is Director of the Autism Research Centre at Cambridge University...

Simon -   Autism is a neurodevelopmental condition.  So, it means that it affects the way the brain develops.  The brain isn't developing in the typical fashion.  The symptoms that you observe are that the child is not showing the normal interest in people, but instead they become preoccupied by objects.  So, they have trouble in the peer group in developing social relationships and in communication. 

But the other side of autism is that they become very obsessed with objects or the physical world.  So, they just have interests in the usual way, but they're going to have things much more deeply.  They also have trouble with coping with change and like things to be very predictable.  Autism is a lifelong condition and that it's the way the child is born.  Probably because by prenatal factors, you don't expect the person to kind of grow out of their autism.  It's there for life.

Hannah -   How many people does autism actually affect worldwide?

Simon -   The estimates are that it's about 1% of the population.  For example, in any primary school, usually about 200 kids, there'd one or two kids with the diagnosis of autism.  So, I see that as quite common and those numbers are much higher than they used to be.  If you went back 20 or 30 years ago, the textbooks told us that it was 4 children in every 10,000.  It's become recognised much more and the fact that we're also recognising that much broader spectrum probably means we're picking up more cases.

Hannah -   So, it's that rather than the fact that there's an increase in the number of people that are actually getting the disorder maybe due an environmental factor for example.  So, there has been hypothesis that increased environmental toxins might lead to an increase in the risk of autism?

Simon -   Yeah, I mean, I think it's sort of safer to assume that the increase which is definitely there reflects more about that we're better at picking it up.  And also, that we've broaden the criteria that what counts as autism that maybe a generation ago, a child was just a bit different and maybe obsessed with maths or chess might not have got a diagnosis.  We might have just thought that was an eccentric child, the child is a bit different, or maybe even a bit delicate.  That's the kind of the words that people used to use.  And now, we might give it a psychiatric label. 

But the idea that there's something in the environment, a toxin, I don't think there's any good evidence for that.  The danger of course is that if you start trying to link the increase with some environmental toxin, you can raise a lot of public anxiety.  That's exactly what happened with the MMR debate - the idea that the MMR vaccine, measles, mumps and rubella that's given to children at about 18 months old might cause autism.  Once that idea was floated, it caused a lot of public anxiety about the vaccine programme - public health programme.  But it turned out that there's was no good evidence for it.

Hannah -   It was simply more the case that the age at which symptoms might emerge in children was shortly after the vaccine had been administered.  And so, that's why that link was made in the first place, but it turned out to be nothing more than the coincident and timing?

Simon -   Around 18 months is the age when you'd start to notice.  That's probably the earliest age you'd start to notice that the child wasn't interacting with their peer group or maybe was delayed in language.  So it could be a coincident in timing.  But the original research linking them was actually because they found the measles virus more commonly in the gut of children with autism.  But later, research didn't support it.

Hannah -   So, there are some other symptoms that you've mentioned like for example being fixated on a particular task, getting concerned about change and the structure of your day to day life for example.  Sometimes people with autism may have difficulty reading people's expressions and feeling empathy for other people.  These are in some way things that I relate to.  So, I didn't like change that much and I can't get fixated on certain tasks.  And also, some of the tasks that scientists used to help investigate autism where you try and read a facial expression of an actor, I find it incredibly difficult to tell whether someone is frightened or amazed for example.  So, does that mean that I have a small amount of autism?  Is it a disorder?

Simon -   Right.  Well, on the question of the words 'disorder', I don't intend to use that term in relation to autism.  So, I preferred the terminology autism spectrum conditions because I think 'condition' reflects that it is biomedical event and that this person might need a diagnosis to get support.  So, it's kind of not trivialising the fact these symptoms might be interfering with the person's life, such that they need to go to a clinic and get help.  But calling it a disorder, I think it's a bit harsh to tell somebody, "Your brain is disordered."  I don't think there's any clear evidence for a disorder.  It's just that these are children who are developing differently.  There's a kind of recent concept that's come in called neurodiversity - the idea that in any community, in any population, the brain is going to be developing in different ways in different people.  That there isn't a single route for normal development.  Of course, we've always known that because if you think about left-handedness and right-handedness, that gives you one example of diversity and how the brain gets wired up.  But of course, the brain is much more complex than just about whether the right or the left is dominant.  And you can imagine some children who are more verbal, more or less verbal, some who are more obsessive or less obsessive.  So in that sense, there are many different cognitive or neuro profiles in the population.  Autism may just be one of those profiles.  But if it's interfering with the way that the person can function, they might need a diagnosis.  But it is kind of opening up this whole new way of thinking about autism which is that it's a spectrum of traits that runs right through the population and we all have some autistic traits.

Hannah -   I suppose that maybe at one end of the spectrum, it offers an evolutionary advantage to have some of these traits, these personality traits.

Simon -   Yeah.  I mean, the idea of linking autism with evolution is interesting because autism is partly genetic.  So, although we don't yet know which of their genes that are necessary and sufficient to cause autism, we know that it runs in families.  Potentially, a large number of genes are involved.  So, if you find a trait or a condition that's partly genetic then you immediately start thinking, why is it still in the gene pool?  Has it been adaptive or positively selected in evolution?  As you say, some of the aspects of autism, they're not necessarily disabilities.  They can even be talents.  So, children with autism are very good at spotting details and they love patterns.

Hannah -   So, they'd be incredibly good on 'Where's Wally?' for example, spotting where Wally is within this myriad picture, this very busy picture of lots of different people.

Simon -   Yeah, so the task of Where's Wally is really about attention to detail.  In the modern environment that particular children's game may not be adaptive.  It may just be fun.  But you could imagine millions of years ago, individuals who were good at spotting details might have had some advantage including reproductive advantage in terms of spotting predators or finding resources and food, and kind of learning what's safe and what's harmful in the environment.

Hannah -   Going back to the genetics of autism, it seems as though males are more likely to have autism than females.  Is that something to do with the Y-chromosomes or the chromosomes that only men express?  I've got 2 X-chromosomes and you got an X and a Y.  Is that something to do with that incident difference between genders?

Simon -   The genes for autism, I mentioned, it's not a single gene condition.  There are many genes involved.  Some of them may be on the X-chromosome and the fact that females have two X-chromosomes, males only have one, may mean that the genes for autism vary according to the number of X-chromosomes you have.  The other thing is that autism can occur in a female.  We've been looking at the role of this so-called male hormone testosterone and actually both sexes produce the hormone.  It's just that males produce more of it.  We've been looking at the role of that hormone in shaping brain development in typical children and whether higher levels of prenatal testosterone are linked to more autistic traits.  We've been finding mothers who had amniocentesis during pregnancy where you can take some of the fluid that surrounds the baby and analyse it - in this case, the testosterone - and then wait for the baby to be born.  We've been finding that the children who had higher levels of prenatal testosterone show more autistic traits during postnatal development.

Hannah -   How does this fit in with the view that autism is kind of an extreme male condition?  So for example, I've been listening to the radio recently.  I've heard some women getting in touch saying, "My husband has definitely got autism.  He's just so unempathetic.  He's behaving in a particular way.  He's spending all his time in the tools shed organising his tools and being very fastidious about the work that he does on the house.  He must be autistic."

Simon -   So, the idea that autism might just be an extreme of the typical male brain, that comes from the notion that there are sex differences in the general population in cognition.  Quite a lot of research shows that there are average differences between males and females.  It's not that all males show one pattern and all females show another, but males, even from childhood are more likely to be interested in objects, in machines, constructional tools like Lego.  Females are more interested as a group on average, show more interest in people.  And so, the idea that autism may just be an extreme of the typical male profile, well, it sort of fits that.  that people with autism show even less interest in people.  But the idea that if you show that profile and you a diagnosis, that doesn't necessarily follow.  There are lots of people who enjoy quite solitary activities down in the garden shed, but that's all part of the normal variation in the population.  It doesn't mean they need a diagnosis unless it's causing difficulties at work leading to perhaps losing their job because they're not managing to work with colleagues, causing depression because it's social isolation.  So, we have to kind of distinguish between normal variation which may be linked to sex differences and the point at which it starts to interfere with your ability to cope.

Hannah -   Which then moves on to the diagnosis of autism for this big spectrum where lots of characteristics occur on this continuum of how people behave and what their personalities are made up of?  How do you diagnose autism and are there any biological markers that you can use in order to help with that?

Simon -   So, a diagnosis of autism at the moment rests on behaviour.  I think a lot of clinicians would like it if there was a biological marker to improve diagnosis because otherwise, diagnosis can be a bit subjective.  That's the kind of hope, but we're not there yet.  We're quite a few years away.  I sort of suspect that even if we knew all the genes involved or all the hormones involved, or any other biological marker, we might still need the clinical interview.  So, you'd sort of need to show not only that they've got - as it were- the blueprint for autism, but it's having an impact in their everyday life.

Hannah -   Finally, there's been decades of research, trying to understand autism and yet, we seem so far away from having a full grasp on it.  Why do you think that is?

Simon -   So, autism was first described in 1943.  We've had 70 years of research.  but actually, quite a few decades actually, up until about the year 2000.  There was very little science going on.  If you look at the rates of publishing of journal articles in science on the topic of autism, the line is pretty much flat from 1943 up to 2000 and then it takes off.  That reflects that a lot more funding is going into autism research.  probably, also the results of effective lobbying by parent groups and charities saying, "While we're spending all this funding on other conditions like cancer, why aren't we spending the proportionate amount of money on understanding autism?"  So in some ways, it's kind of early days.  Also, the early research was after we focused on the psychology of autism, we're not so much on the neuroscience or the biology.  So, that's really just opened up since the technology has been available - brain scanning, the new genetics and also, old theories about the causes of autism, and the idea that it was caused by bad parenting kind of got in the way of making scientific progress.

Hannah -   Thanks to Simon Baron-Cohen from Cambridge University.