Amended: August 21 (correction to e-mail address below)

When I was training at the Maudsley 30 years ago, anorexic girls were treated as little more than malfunctioning machines. Even when a friend at medical school became terribly thin and suddenly “disappeared” from class one day, no one talked about it.

Very little research had been done on eating disorders and the causes remained a complete mystery to most psychiatrists. The view was that it was an illness that mainly affected middle-class, intelligent, white girls and was little more than an awkward phase of adolescence. Back then services were generally lacking, and treatments were borrowed from other areas of psychiatry and adapted to fit – a catch-all approach that spectacularly missed the incredibly complex causes of anorexia and bulimia.

Ten years later, while studying for my Medical Research Council fellowship, I became fascinated by the idea that brain mechanisms could contribute to a biological understanding of eating disorders. Twenty years on, there has been a huge change in the technologies available in understanding disease. The ability to decode the human genetic sequence means that, if a large enough sample of a particular disease is examined, and the DNA of those it affects is looked at, common risk factors emerge.

What we now realise is that we need to be looking at underlying neural networks in the brain – how patterns of information are processed, how this affects both behaviour and the way an individual reacts to her environment, and why this goes wrong. We need to consider those aspects of how the brain functions that increase the risk of someone falling prey to an eating disorder.

I and others have been working hard to examine such processes and the research so far has produced very interesting findings that we have already been able to put into treatment programmes. We have found, for example, that people with eating disorders find it difficult to change self-set rules and learnt behaviour once fixed in the brain. They also see the world in close-up detail, as if they are looking at life through a zoom lens – but this can be at the cost of having an ability to see and think about self-identity and connections with others without getting lost in the details.

We also discovered that this distorted pattern of processing information has a strong similarity to autistic spectrums. It has even been described as the female form of Asperger’s. Traits that may appear present in childhood, such as obsessive-compulsive disorder or overperfectionism, can often indicate a vulnerability to developing an eating disorder later in adolescence.

Of particular interest are difficulties in “set shifting” – being able to shift back and forth between different tasks or mindsets. Problems may show up in a variety of forms: for example, cognitive inflexibility – an overrigid approach to problem-solving; and response inflexibility – excessively stereotyped behaviour. An example of this is an inability to deal with a last-minute change of plan, such as a meeting. Interestingly, we have found poor set-shifting ability even after recovery, and have found it in the healthy siblings of patients with anorexia.

This trait among healthy sisters implies set shifting or cognitive inflexibility could be an underlying aspect of eating disorders and therefore a possible target for treatment.

As well as poor set shifting, people with anorexia can show weak central coherence, which can be defined as a bias towards the local processing of information rather than placing it into a broader context. For example, if they were in a meeting, they would be distracted by somebody’s fingernail and be unable to switch back to the matter in hand. This is one of the core features in cognition in autism spectrum disorders.

This trait became of great interest in eating disorders after a study a few years ago found the same kind of weak central coherence in people with anorexia as those with autism and Asperger’s. The study also found that more than 20 per cent of the anorexic group could be described as having a disorder from within the autism spectrum.

So how can this translate into therapy? The new Maudsley model of individual treatment includes interventions focused on traits such as perfectionism and rigidity. With this new genetic-based research, it is helpful for us to know whether the markers we determine of underlying brain function run in families – and how much the external environment moderates them.

Examining other family members is very important, especially when there is another young woman in the family. The children of women who themselves have had an eating disorder are particularly interesting in this new area of research, as comparing patterns of the illness across generations can enhance our understanding of environmental as well as genetic factors and how they interact.

But it’s not just about girls. Boys and young men seem to be more protected, perhaps by cultural factors in our society. But looking at the underlying brain function in male cases of eating disorders can improve our overall understanding. The risk factors, though rarer, tend to be more clearly defined.

Work on brain function, of course, does not exclude social and cultural triggers for eating disorders, the kind of thing that generates headlines about size zero models. It is clear that exposure to media images depicting thin women really does reduce body-related self-esteem. And the young models used to promote such images themselves may be at an increased risk of developing eating problems. But much of it happens in the home or playground, far from anything to do with fashion. Teasing and bullying focused on food and weight and body shape, particularly from family members, increases the risk of developing an eating disorder.

Eating disorders still have the highest mortality rate of any psychiatric illness. But there is a trend towards the numbers declining. I personally think a more optimistic outcome is within our grasp as we understand more and more the way the brain works.

Janet Treasure is Professor of Psychiatry at King’s College, London, and head of the Eating Disorders Unit at the South London and Maudsley NHS Trust.

Do you want to take part in research?

Professor Janet Treasure is seeking participants for a variety of research projects into all aspects of eating disorders, including a comprehensive support programme for carers.

If you think you could help, contact Frankie Bishopp on 020-7188 0186 or e-mail f.bishopp@iop.kcl.ac.uk. Further information from the KCL website: www.eatingresearch.com.