������ý

Uta Frith seems remarkably cheerful and content for someone who’s spent six decades trying and failing to get to grips with her life’s obsession. “Very little has stood the test of time,” she tells me as we sit down in her living room in a leafy estate in Harrow-on-the-Hill, London.

Around us, high-ceilinged walls papered in a luxurious red print are barely visible between rammed bookshelves, several model brains and a collection of abstract art. Frith has been searching for the mechanisms that underpin the enigmatic condition of autism ever since she first met profoundly autistic children in the late 1960s. “We could identify them intuitively, but not really scientifically – and I have to say that this is, unfortunately, still the case.”

Still, Frith’s influence on our ever-shifting understanding of autism has been monumental. She developed two landmark theories about how autistic minds might develop differently to neurotypical minds – and was among the first to test ideas like these using newly available brain scanners in the 1990s.

Since then, the number of autism diagnoses has sharply risen, especially among women and girls – largely because of a softening and broadening of how we define the condition. But Frith thinks that many people at the milder end of the spectrum have little in common with those who are profoundly autistic. “There’s absolutely no overlap,” she says. “That is the sign that the spectrum isn’t holding.”

Over tea and homemade macaroons, we discussed how the evolution of the condition reflects on our present situation. Could the emerging idea that there are a few distinct kinds of autism help us to find a way through the confusion?

Rising diagnoses

I’m no stranger to autism’s recent trajectory. Several years ago, my brother, who had long hidden his struggles, reached out for help, and I supported him through his autism diagnosis. Today, many friends have been diagnosed with autism, ADHD or both, and I have been asking questions about my own possible neurodivergence. So, I was keen to better understand where Frith, with her deep knowledge of the condition, was coming from – even if I might not agree.

Frith never intended to be an experimental psychologist. Instead, she set out to study art history in the early 1960s. “Like so many things in life, it was a complete accident. I just drifted into the psychology department at Saarland University in Germany,” she says. In the university hospital, psychologists gave lectures alongside patients who hallucinated or who spoke in abnormal ways. “I found that absolutely mind-blowing. How would you explain that?” she reflects. “It was really an exciting time for clinical psychology.”

For decades, psychology had been in the grip of psychoanalysis, which saw distress arising from the expression of conflicts in the unconscious mind, often involving repressed memories, desires and fears. Autism, through this lens, was blamed on cold, unloving mothers who made their child withdraw. “The ‘refrigerator mother’ was a dangerous concept,” says Frith. Instead, clinical psychology sought objective explanations based on the cognitive mechanisms that drove behaviours, such as habits and goals. “I was really thirsting to find some facts that we could work with and say, ‘Oh, that really tells us something about our minds,’” she says. “I really go for what is the truth.”

This took Frith to London, where, for the first time, she met autistic children and tried to interact with them. “I was completely baffled and absolutely fascinated by them. How could they have a mind that allowed them to be so oblivious of me?” The same children who struggled so much socially would then complete an extraordinarily complex jigsaw puzzle by themselves, upside down, without being asked. Frith’s affection for and interest in these children led her to dedicate most of her career to trying to figure out the difference between autistic and neurotypical minds.

In 1985, she  with psychologists Simon Baron-Cohen and Alan Leslie that asked: “Does the autistic child have a theory of mind?” They proposed that autistic people can’t understand the knowledge, intentions and emotions of other people, and often struggle to empathise. The idea that autism relates to a lack of empathy is disputed, as , suggesting there is a mutual misunderstanding between autistic and non-autistic people.

Frith emphasises that the lack of a theory of mind can explain only one facet of autism, relating to social behaviour. Another trait Frith uncovered in the 1980s is the impressive ability of autistic people to to . “There’s an over-focus on the detailed components of a big picture, and perhaps this neglect of the big picture – you’re not seeing the wood for the trees,” she says. This also seems to explain the behaviour of autistic savants who have narrow interests along with exceptional skills such as memory, musical or mathematical talents.

Yet to try to demonstrate that these are objective causes of autism, Frith couldn’t rely on cognitive theories and behaviours alone. “You also have the level of the brain where you believe the source for these cognitive mechanisms lies, and underneath that you have the level of evolution and genetics that drives what is actually happening in the brain,” she says.

Imaging the autistic brain

Fortunately, as Frith developed these theories in the 1980s and 90s, brain-scanning technologies such as MRI and fMRI matured, which allowed her to look for the brain circuitry that might underlie autistic traits. “I have to really credit my husband here, because he was right there at the beginning developing the techniques you need to analyse the images you get out of these scans,” she says.

Brain imaging studies like this were only possible because the criteria used in autism diagnosis had recently expanded. To give consent, participants had to understand what the experiments implied, so the studies ended up including many individuals with Asperger’s syndrome – a “high-functioning” form of autism that became formally recognised in the early 90s. The terminology high- or low-functioning is no longer used today, nor is Asperger’s a diagnosable condition.

Initially, Frith welcomed expanding the criteria for autism to include milder cases – she even  to the English-speaking world in a 1991 book. But she now questions whether this line of thinking has, inadvertently, led us astray. “It was almost essential in order to do more research. But now I think perhaps that was a step too far, and certainly what happened afterwards is not what anybody could have foreseen,” she says. In the UK, for instance, between 1998 and 2017, there was an  in autism diagnoses. “It ran away beyond Asperger’s,” she says.

The introduction of Asperger’s syndrome generated cultural depictions of autism, such as Raymond Babbitt in Rain Main. Babbitt is an autistic savant who relies on strict routines and finds social interactions challenging. Many of these popular archetypes have attractive personalities, says Frith, such as Astrid Nielsen, the autistic archivist who helps solve criminal cases in the current Franco-Belgian detective series Astrid et Raphaëlle. “People were presented as very logically minded who, at the same time, seemed completely clueless in social relations,” says Frith. “There was this idea that you could easily classify your friends and say, ‘Well, he’s on the spectrum, isn’t he?’”

Many autistic people and autism researchers welcome how inclusive the condition has become. Broadly speaking, autistic people tend to have difficulties with communicating and socialising, as well as sensory issues and restricted behaviours and interests. But precisely how these facets manifest varies hugely from one autistic individual to another – and the broadening and softening of diagnostic criteria embraces this variation.

Yet Frith is concerned that the traits of a large proportion of autistic people diagnosed since the 90s might be better understood as part of normal personality variation. “We are all individually different,” she says. Autism has become an identity rather than a pathology, she says. “It’s almost like if you find it difficult to make friends, or you have a special interest hobby, or you aren’t always sure what people are thinking, it’s enough to think, ‘Wow, I could be autistic.’”

Those who regard autism as a form of neurodivergence say the condition isn’t just a pathology that causes disability. Autistic traits can also be an immense source of pride and strength. My brother’s knowledge of who scored in every single football match Manchester United has ever played certainly helps him out in the right situations – and I constantly lean on him to jog my childhood memory. While he has his share of troubles, his autism diagnosis helps him to value his unique skills, foster self-understanding and be kinder to himself.

At the same time, it isn’t clear what, if anything, still unites mild autism with profound autism, where people often face severe intellectual disabilities and can’t live independently. Cases of profound autism have increased since the 90s, but not nearly as much as milder forms, which are often identified later in childhood or in adulthood. Frith thinks that these newer groups have no overlap with the autistic children that she first met in the 60s, and that often these conditions are unlikely to be neurodevelopmental, which is a defining feature of autism. “It could be a category which has not yet got a label,” she says. Or it might not be a diagnosable condition at all. “I think a wrong turn was made pathologising what I think are probably extremely useful personality variants that I would call ‘autistic-like’.”

Frith points out that the rate of autism diagnoses has increased most rapidly among women, who are now often diagnosed as teenagers. “They are still labelled autistic, but I wonder whether they would be much better treated and understood if one looked at what their problems actually were,” she says. This trend has brought the historical ratio of male-to-female autism diagnoses down from around 4:1 to an almost equal ratio by the age of 20. But Frith says that including milder forms of autism diagnosed later in life distorts this ratio.

Other researchers, such as psychologist Gina Rippon, say there are good reasons why autistic women and girls are diagnosed later in life – and that rather than omitting these “lost girls” of autism, they should be welcomed into the fold. In this view, social pressures placed on women and girls can lead them to camouflage natural behaviours. Indeed, imaging studies find that brain areas associated with social reward are more active in autistic girls than boys and that they have greater connectivity between social brain networks, which may suggest they are more motivated to adjust their behaviour to fit in.

Masking in this way is linked with high levels of anxiety, depression and  in autistic girls and women. I put these ideas to Frith, but she sticks to her guns. “It creates an extraordinary paradox,” she says. “You can diagnose autism showing symptoms, and you can diagnose autism not showing symptoms, with the assumption that masking is going on. So, how can you ever falsify the diagnosis?” Masking isn’t a scientifically testable idea, says Frith. “We have lots of reasons to doubt subjective experience. As a tool for vigorous science, it’s very difficult.”

I wonder if Frith is biased by her own background. In the 60s, she saw the harm of psychoanalytic approaches entirely rooted in experience, leading to her desire for clear-cut cognitive mechanisms. Even so, isn’t the end goal of autism research to improve the subjective lives of autistic people? “That is questionable; that’s an application of the science,” she responds.

On that front, there seems to be hope. In recent years, a wave of studies has suggested that autism isn’t a single condition, but rather an umbrella term containing multiple distinct subtypes. The grand ambition of this approach is to join the dots all the way from genes through brain circuitry and cognition to behaviours and experiences, something that could then improve support for autistic people. I ask Frith whether subtyping can create order out of the confusion.

“It is feasible to think that this will be achieved eventually, but not very soon,” she says.

Frith reflects on her own excitement when brain imaging tools were introduced in the 80s and 90s. “It was a huge step up, but we didn’t get answers that we expected,” she says. Although they observed differences in the connections between brain regions of autistic people compared with neurotypical people, they struggled to explain these findings.

What’s more, recent subtyping studies often lean on big data to make advances, finding patterns in databases that include thousands of autistic people. Frith is sceptical of these largely bottom-up methods and says that we’ve lost sight of theorising. “I want an approach which goes both top-down and bottom-up,” she says. So far, subtyping studies don’t agree on what the characteristics of each subtype are – or even how many subtypes there are.

“There is an almost desperate need to conserve the idea that it is autism that is being diagnosed, whereas it’s perfectly reasonable to say that it’s not autism, it’s something else that doesn’t have a name,” she says. “I’m astonished that people want to cling to this label when it’s outlived its purpose,” she says. “It’s time to break up this huge spectrum to make it more intelligible.”

If these are distinct conditions, as Frith thinks, using the overarching idea of autism as a guiding light for research might prevent us from reaching the truth. Striving for absolute mechanisms is admirable, but Frith might be setting the bar too high. Relations between the physical brain and our mental worlds are elusive and incredibly complex – and many autistic people care less about mechanisms and more about accessing support that can improve their lives. Only with a diagnosis could my brother ask his workplace to make accommodations that have improved his life.

So, who gets to decide what autism is – autistic people or the scientists studying the condition? Increasingly, researchers are working closely with autistic people, involving them in the design of studies and affirming their experiences of the world. Frith acknowledges this approach is well-meaning, but wants to lay bare the risks that it may entail. “We should start again and invest more in research as opposed to instant treatments which we know very little about,” she says.

“It feels like a time of confusion and contradiction, and it has to be sorted out – not by me, but I’m trying to give some pointers.”