An autistic advocate, podcaster, and author weighs in on the limits of old-school research and services
Recently, developmental psychologist Uta Frith shared with Tes Magazine her opinion that the concept of the autism spectrum “has widened to the point of collapse,” and that the diagnostic criteria for autism should be re-assessed. Frith, an emeritus professor of CogSci at University College London (and who, along with Simon Baron-Cohen, popularized the questionable idea that autistic children lack a “theory of mind“) also recently appeared on the Let’s Talk Neurosense podcast, taking issue with “the dilution of the criteria,” and telling host Naomi Fisher, “It may be that in many cases autism [diagnoses] becomes a label for any form of social exhaustion or anxiety. And then the term just loses its power.”
Frith told Tes reporter Helen Amass she believes there is “very little scientific grounding” for school-based adjustments such as ear defenders and adjusted lighting: “As far as I know,” she said, “there hasn’t been any research into finding out whether these sensory adjustments actually improve lives. I’ve no idea whether it’s good for the child or not. It may feel good to them, but not actually be good.”
(Editor’s note: Autistic researcher Monique Botha-Kite responded to Frith’s stance by noting that, “Not keeping up with the research doesn’t mean the research is broken.” Autistic researcher Steven Kapp points out that Frith is ignoring the history of autistics reporting their sensory issues, as well as related contemporary research.)
While I disagree with Frith, it is unfair that Fisher was hassled for having Frith on the podcast. I’m glad these conversations happened, because they’ve revived a vibrant discussion about the role (and limits) of science in determining how we define ourselves, and how we should provide care and compassion for those we love.
The “Spectrum” Diagnosis and the Debate Around Research
The debate about the autism spectrum goes back to 2013, when experts as the American Psychological Association’s Diagnostic and Statistical Manual (DSM), a central diagnostic tool, introduced the concept of an autism spectrum in its DSM-5, replacing older, separate categories such as Asperger’s that were used by its DSM-IV. The spectrum concept add nuance to the diagnosis, recognizing that people’s support needs change over time. (For an in-depth review of the history, check out Bonnie Evans’s essay How Autism Became Autism, and Steven Kapp’s and Ari Ne’eman’s article Lobbying Autism’s Diagnostic Revision in the DSM-5.)
The spectrum concept has helped to humanize autistic people: instead of a rigid category list, autistics are now understood as a community with shared experiences—and authorship of those experiences. The autistic rights/neurodiversity community has mainly endorsed the spectrum concept because the degree of disability from autism varies so widely throughout the lifespan, and because autistic people can massively relate to each other regardless of their support needs. (Frith disputes the latter, stating “People still hang on to the idea that there is something that unites all the people who are diagnosed as autistic. I don’t believe that any more.”)
In the 10+ years since the spectrum was codified, we’ve also seen a shake-up in autism research. Today, under the rubric of participatory research, non-academic autistics can consult and co-author social science research, being a helpful bridge between autistic research subjects and non-autistic researchers. We are also seeing more research on practical problems, like social isolation and sleep issues, as well as developing communication tools and inclusion models. Which brings us to the question of quantification.
Autism Researchers: Trapped by Data?
Traditionally, autism researchers focused on trying to measure autistic people, tending towards a concept of autism as a staid entity, with palpable subsets. This approach faced an obvious roadblock: that every individual’s brain develops differently according to an endless cascade of factors we will never be able to quantify. The measuring of autism has always been imperfect—and researchers trying to “fix” that are not going to find the clarity they seek.
With new research emphases on quality of life, researchers are spending less time watching mice in mazes, or digging around for an autism gene. But many still remain beholden to scientism—the belief that science can explain almost everything, and that finding the “core” of autism is the most important goal. Scientism imbued a sense of certainty (as well as attracting research dollars), and researchers have over-sold the idea that they can locate and measure autism.
The extra irony here is that most of the science would not pass as rigorous in any other field. As Andrew Whitehouse, Professor of Autism Research at the University of Western Australia, told Spectrum Magazine in 2021, “We accept, and perhaps even expect, lower standards of clinical evidence in the autism field, than in almost any other.” He followed up on my podcast, pointing out, “There are zero other, no other areas of health and medicine that would accept the low standards that we accept in the area of autism.”
Take, for example, Applied Behaviour Analysis, whose foundational “evidence” base (which I’ve documented at length) appears to be little more than a series of poorly executed single-case studies published in industry journals that rarely acknowledge conflicts of interest, or risks. This approach, and others like it, are anything but robust—yet used in thousands of clinical settings that accept a terrible approximation of science as somehow better than no science. (This is just one of many examples, but I’m trying to keep my word count down!)
Practical solutions—and The Power to Name
The solution to the problems in autism research is not just to “science harder.” To the contrary. Autism researchers and professionals would do well to make peace with that fact that autistic people can’t be slotted into a taxonomic flow chart as if we’re a sub-species of the human race. Autism will never fit in the disease model, because it is not a disease; it is part of the human condition. That is the beauty of the spectrum concept. It allows practitioners the opportunity to address specific needs, moving autism research out of the Petrie dish and into the real world.
The fact is, we don’t need a double-blind study to decide if Ben feels better having some ear defenders. We just need to try those noise-canceling headphones out, and see if they work for him. We don’t need more grant dollars squandered on creating new sub-categories of children; we just need to make sure all children are included in social life and school. And if people are feeling anxious, we should ask what they need and help them out.
The debate over classification really belies another conflict—over the power to name and control. We are told that the authority to classify autism is held by experts, even as many of their rubrics crumble into sand upon close examination. We spend thousands of dollars to attain the nod of an expert, in form of a diagnosis that instructs an institution (like school) to include our children. But in the end, all children—all people—are worthy of inclusion. And the power to name one’s experience is far beyond any lab, white paper or conference; it is held by the individual alone.
