Autism is a hot topic. The Playhouse was packed for this year's Annual Lecture in the Public Understanding of Science. Marcus de Sautoy, Oxford's much-loved Professor of PUS (unfortunate acronym), told us this was the fastest-selling lecture in the series to date.

When the Lecturer, Professor Simon Baron Cohen (Director, Autism Research Centre, Cambridge), started giving us numbers, we could see why: it also appears to be one of the fastest-growing conditions. There were various statistics but perhaps the most dramatic was the increase in the rate of diagnosis of autism in eight-year-olds in America, since he started studying the topic, from 1 in 2,500, to 1 in 48 males, and 1 in 189 females. Of course, both awareness and a broadening definition of the condition play their part in this increased rate of diagnosis. There has been a corresponding increase in the public's interest in autism (widely and recently fuelled by the huge literary and theatrical success of The Curious Incident of the Dog in the Night-Time).

Professor Baron Cohen was a perfect choice for this year's lecture. Not only is the topic of wide interest, but he did a wonderful job of conveying the science used to explore it in an easy, gently paced, accessible way, making full use of simple graphs, charts, photographs and diagrams to provide concise, clear explanations, which enabled him to cover an immense amount of ground in his allotted hour.

After characterising autism as, broadly, the combination of a desire for preciseness and order, a strong interest in patterns and systems, a fear of unpredictability, and difficulties with socialisation and social interactions, he discussed the concept of the "spectrum", a term first coined in 1988, and how to measure autistic traits. It was his own team which came up with the Autistic Spectrum Quotient test (which you can easily find online and take yourself). There are many people on the "spectrum" who share some of these traits. A formal diagnosis is needed only if they interfere with an individual's ability to cope with everyday life.

One of the techniques used to investigate autism is gaze-tracking, with which many Oxford parents will be familiar if their children have participated in BabyLab experiments. This throws up fascinating results. For instance, when watching a face, a typical child will focus on the eye region, whereas an autistic child will focus on the mouth. When a two-year-old is shown simultaneously a human face and a circle containing a geometric design, there is a 100% probability of autism if the child spends more than 70% of the time looking at the design. (This does not work the other way around – not all autistic children will do so.)

One of Professor Baron Cohen's earliest projects was the Theory of Mind: the ability to imagine another person's perspective. Autistic children have a problem with this, finding hide-and-seek games confusing, and having difficulty in reading another person's facial expressions or emotions.

Functional magnetic imaging and autopsies are also used in investigating autism (did you know that Oxford hosts the National Autism Brainbank?). From these, one can identify a number of physical differences in the autistic brain: for instance, the amygdala or "emotion-centre" is larger; there is faster early brain growth from birth to age 5, with a greater amount of grey matter throughout; there are 65% more neurons in the frontal cortex; the corpus callosum (the area of connective tissue linking the two sides of the brain) is smaller; there is, however, more connectivity but the connections are shorter, with less neural "pruning", and a lot more dendritic spines on the neurons. All this indicates that the autistic child is picking up a lot more information – sounds, lights, patterns in the background – leading to the sensory hypersensitivity which is one feature of autism. Autistic children are better at the block design test which is part of a standard IQ test: at taking a big picture apart and looking at its components, spotting patterns quickly, and in this process their brain is superior, doing a job more efficiently.

Professor Baron Cohen's was keen to stress that the autistic brain is not broken, but different – indeed, sometimes, in some ways, better. This was his central message. In the past, autism has been seen purely as a problem – a psychiatric disorder or disease, a medical problem, a disability, something which needs to be prevented, treated, cured. On the contrary, there are numerous instances of autistic people past and present who have achieved great things: from characters like sociopaths Isaac Newton and Albert Einstein, to Derek Paravicini, the blind autistic pianist who can play a new jazz piece after one hearing despite having a "severe learning disability"; Jacob Barnett, the autistic child who blossomed as a university Physics student at age 10; or Alan Gardner, Channel 4's "Autistic Gardener", all of whom have used their abilities to identify and use patterns in different ways with astonishing success.

The emerging view is that autism is simply different. Professor Baron Cohen recommends Steve Silberman's Neurotribes: the Legacy of Autism and the Future of Neurodiversity, which won the Samuel Johnson Prize for Non-Fiction for its history of autism through the ages, arriving at the conclusion that neurodiversity is as essential to life as biodiversity. Autism should not be pathologised; but we should consider the implications of our growing understanding when making decisions about, for example, education and the environment.

The Lecture touched on many aspects of autism – practical, behavioural, psychological and physical – but also left many questions unanswered. For instance, the nature/nurture question is still wide open. On the one hand, it appears to have a genetic basis in that over 400 genes have been associated with it; the probability of autism goes up to 1 in 3 if there is already an autistic person in the family; and autistic children are very often found to have parents or grandparents whose work is centred around systems (eg interest in mechanical systems such as car engines, abstract systems such as mathematics, electronic or natural systems). On the other hand, studies of identical twins show that one can be autistic while the other is not.

The audience also had many questions. Is there is any difference in brain development between Asperger's syndrome and other forms of autism? (A - studies in progress but insufficient data yet). What causes the link between autism and anxiety? (A - tends to occur as a result of being placed in unstructured and unpredictable situations). Given that Hans Asperger claimed that "for success in science a dash of autism is essential", what are the implications of gender differences in terms of comparative levels of empathising and systemising for attempts to get more women into science? (A - in fact, there are already more women than men applying for more empathic science courses such as medicine, but fewer for maths, physics, engineering).

My favourite question was "Can autistic people b etaught to empathise?" Surprisingly, the answer is "yes". Practice improves recognition and understanding of facial and vocal expressions of emotion, if presented appropriately for autistic sensitivities. Not surprisingly, once again it is Professor Baron Cohen's team who developed the relevant educational materials. If you visit http://www.thetransporters.com/ you can find the award-winning videos for 2-8 year olds showing human faces on vehicles travelling predictable roads or tracks. These can be reinforced with entertaining interactive quizzes .It would seem that, unbeknownst to us, Thomas the Tank Engine and his friends have been fulfilling a valuable function over the decades!