If anyone is interested in the classification of the differing levels of autism, there is an INSAR (International Society for Autism Research) meeting in Rotterdam this month that will discuss this.

The following link is a blog article by Simon Baron Cohen contemplating the possibilities.


  • When Simon Baron-Cohen asks, "Is it Time to Give Up on a Single Diagnostic Label for Autism" ~ I think not. But as for having sub-groups of Autism excluded from the Autistic Spectrum Range in diagnostic terms, giving up on that and instead including subgroups (A.S., P.N.O.S, etcetera) as being part of the 'spectrum', that makes much more sense to me.

  • I would be very interested to hear the proposals and decisions that come out of this meeting! There doesn't seem to be much consensus at the moment on diagnostic terminology and a lot of the proposals mentioned in the blog are interesting - unfortunately harping on about the Hans Asperger 'scandal' isn't the most fair or scientific basis from which to judge the term AS. I would hope that this blog article doesn't too closely reflect the considerations of the INSAR meeting in this respect.  

  • Thanks for drawing attention to this. Many of us were probably aware of the considerations that Baron-Cohen puts forward. Given that autistic people have such a wide variety of strengths and difficulties, it seems like being able to match interventions to particular subtypes would improve outcomes. However, my confidence that psychiatry is able to do this usefully is very low, given a century of abject failures in the area, either institutionalising people or ignoring them. Autism researchers do try to relate various physical and psychological variables to particular 'symptoms', which is valid research, and the typology or classification would come from that data - but I'm not sure there is much in the way of useful data yet.  For example, can autistic 'perseveration' be detected by the Wisconsin Card Sorting Test, or not?

    One of the erroneous psychiatric assumptions seems glaringly obvious when Baron-Cohen writes "the way autism is manifested". That itself makes it sound like a single underlying condition, as if it were kidney failure or a thyroid problem. Isn't it more likely that autism is a 'rag bag' consisting of any sensory, motor or processing difference that isn't covered by other diagnostic definitions? It might shade into Tourette's, but an involuntary tic is a fairly well-defined concept. Hopefully we've moved beyond those days, but some psychiatrists seemed to diagnose based just on if you seem a bit weird and introverted to them. Where a specific biological cause is found, like Rett Syndrome, you may find it excluded from the rump of autism.

    'social and communication difficulties, unusually narrow interests, a strong need for repetition and, often, sensory issues' may well turn out to be a pretty arbitrary construct in my opinion, eliding anything contextual such as bullying or social judgement. Do I have 'unusually narrow interests'? Well, I often find everything including people boring, but I know other autistic people who never get bored. Sensory overload clearly is 'a thing', and the use of the word 'often' means that at least could be the starting point for a taxonomy.  Instead Baron-Cohen revives a simplistic dichotomy "AS and classic autism [Kanner syndrome]", which hasn't served well. At least he rejects the HFA/LFA description, which is based mostly on type of services provided to the individual (eg were they diagnosed through an LD service).

    Having met scores of other people with the diagnosis now (fewer than a score with a learning disability diagnosis), I think a more promising version of this would be from Temple Grandin's Thinking in Pictures, where she suggests a continuum between those with sensory, motor and language problems, and those with social and emotional difficulties.  So on the one hand you have a group who may find speech difficult, but be very emotionally expressive when they can communicate in other ways - this is probably the group most likely to be misdiagnosed as having a learning disability, and shades into neurological problems.  On the other hand, you have the verbally fluent alexithymics like me, where 'autism' shades into 'mood disorders'. In the middle, you could maybe situate the autistic (often female?) who finds it hard to speak up over something emotionally fraught. Some of these behaviours may be as much acquired from upbringing as innate. Then in another dimension, you might have propensity to sensory overload, and many other smaller subsets.

    'only subtype on basis of biology' is probably a long way off, and seems the wrong way to go, if the use of the classification is to identify and meet people's specific needs.  It is good that he criticises 'reliance on IQ tests that frequently underestimate the intelligence of autistic people'. This was the only crude measure available a few decades ago, and wasn't adapted for communication differences. I still find to talk to people with a LD+autism diagnoses, that what they are trying to say may seem off-topic or rambling or ungrammatical, but there's a keen perceptiveness there, and inclusion and listening will improve relevance and connection.

    I downloaded the conference programme from here, and find it a bit depressing. There are various panels on service interventions and no doubt some useful stuff, but the big sponsors seem to suggest research resources are being massively misdirected. At least in the UK, some including Autistica, some in the NAS and the Participatory Autism Research Collective are advocating that research should involve autistic people and centre on quality of life issues. Given all the foregoing, standardisation of conference presentation titles to include the phrase 'with ASD' just makes them look like idiots, which is a shame. I'm afraid I keep thinking of that saying that the 1920s was the first decade in which medicine actually did more good than harm, and wondering when the same will be true of psychiatry.