What's the point of the 'autism' label?

Professor Rita Jordan and Graham Collin's exchange of letters.

I was interested to read the special edition of The Psychologist ‘Autism: Myth and reality’. Having worked as a clinical psychologist with people with intellectual disability for over 30 years, I have been interested and concerned about the growth of the use of the autism label and its impact on services for people with intellectual disabilities in particular.

The collection of articles all attempted to explain the key concepts of autism, whether it is a continuum or a category, whether there is a way of explaining cognitive differences between people with the label and those without, etc. I was surprised that no author seemed to address a key issue. Surely an important reason to identify and categorise a subgroup of people is the hope that there may be key treatment implications for the members of that group? Yet no articles talked in depth about treatments or management for people with autism.

Yet in a way I am not surprised, because, when you look closely, there are no autism-specific treatment approaches. For example, the NICE Guideline 142 on recognition, referral, diagnosis and management of adults on the autism spectrum used 25 research papers that reached its criteria for acceptable evidence about treatment. Of these 25 papers, 15 were on people with intellectual disability, three were on people with intellectual disability and autistic spectrum disorder, while only seven were with people with autistic spectrum disorder alone. The approaches recommended do not differ from approaches that would be used for anyone with an intellectual disability. Many of the commonly used approaches in autism services such as TEACCH and PECS consist of sound behavioural, cognitive and communication principles that would apply to anyone with an intellectual disability.

Emerson and Baines (2010) estimated that between 20 and 33 per cent of adults known to local councils with social services responsibilities as people with learning disabilities also have autism. There is now a large autism specialist service sector of schools, residential services, day centres, peripatetic supports, etc., which is open to that significant minority of people with intellectual disabilities. If these services really offer nothing substantially different to a good intellectual disability service, we should be questioning whether it is fair that extra help is available to those people with intellectual disabilities who have the autism diagnosis, when those without the diagnosis share similar needs.

The UK autism strategy for England asserts: ‘Diagnosis alone is not enough: the fundamental change we want to see is that diagnosis leads to a person-centred assessment of need, in line with the NHS and Community Care Act 1990’ Department of Health (2010, p.36).  

A final thought: Why do we need the diagnosis in the first place for the person-centred response the guidance seeks?
Graham Collins
Alfreton, Derbyshire

Department of Health (2010). Fulfilling and rewarding lives: The strategy for adults with autism in England. London: Author.
Emerson, E. & Baines, S. (2010). The estimated prevalence of autism among adults with learning disabilities in England. Improving Health and Lives Learning Disabilities Observatory, Public Health England.
National Collaborating Centre for Mental Health (2012). Autism: The NICE guideline on recognition, referral, diagnosis and management of adults on the autism spectrum. National Clinical Guideline Number 142. British Psychological Society and Royal College of Psychiatrists.


It was somewhat ironic (and, perhaps, a necessary curb on my ego) that the issue of The Psychologist that included a ‘One on one’ interview with me (December 2014) also included a letter from Graham Collins that questioned the fundamental premises on which my work in autism has been based. The question he implies is ‘What’s so special about autism?’, and that is the one I hoped that I had answered over a lifetime’s work; seemingly, not.

A letter is rather an inadequate vehicle for once more making my case, but let me address it in outline, at least. The context of Graham Collins’s letter was that of learning difficulties, and the claim made was that there was nothing unique either in characteristics or treatment/educational approaches that would distinguish those with learning difficulties from those with additional autism. Since only around a third of those with autism have additional general learning difficulties, it is not clear what that argument has to say about the majority of those with autism, but I want to address the fundamental point.

I would agree that people with autism do not have unique behaviours but rather share human responses with all other groups of humans; the problem of diagnosis is the current reliance on behavioural characteristics. Diagnosticians have to make a clinical judgement, not relying just on behaviours, but on the psychological processes that underlie those behaviours. In other words, those behaviours have a different meaning when we interpret them through the ‘lens’ of understanding autism. Why is this important and different to all other conditions (and to the typically developing)?

It remains true (in spite of unresolved issues about fundamental biological differences) that the defining feature of autism is that those on the autism spectrum lack instinctive social understanding, not finding social signals salient, and thus having to learn cognitively (and so be taught explicitly) all the things that others (whether or not they have learning difficulties) just acquire naturally. Humans are social animals, born with a very underdeveloped brain. Most of brain development occurs after birth but, unless you have autism, the learning that determines that development is socially guided. Only in autism are infants left on their own to make sense of the world and to learn how to operate within it. It is not unsurprising if their understanding and ways of reacting are idiosyncratic. In that way, individuals on the spectrum are even more different to one another and do not share many (if any) ‘autistic’ behaviours.

This is important in practice, not just because it helps us understand why they behave as they do but because it prevents us from making incorrect and often damaging assumptions about their behaviour. Without understanding autism, professionals and parents naturally assume the social understanding that is there for all other groups. Behaviour that arises from naivety or confusion may then be interpreted as deliberate non-compliance or rudeness.

Understanding autism should also lead to the recognition that the educational curriculum cannot just be about accessing the same curriculum as others but also needs to include ‘therapeutic’ content, addressing the missing social and emotional understanding that others (including those with severe learning difficulty) just develop naturally. Individuals with autism will benefit from some of the same approaches that are used for other groups, but the way they are used may need to differ. Behavioural programmes, for example, may need to focus more on making goals explicit rather than on ‘breaking down’ skills into components; cognitive behavioural therapy may need to devote a lot more time than usual in helping the individual understand what emotions are and helping the development of self-reflection. Teaching communication to those on the spectrum needs to include teaching about communication itself – not just teaching the means for communication. Speech cannot be assumed to be underpinned by understanding of communication, as it can for all other groups.

Working effectively with those on the autism spectrum, and supporting their development, depends on understanding. The biggest myth of all, and the most damaging, is that one can simply train skills, without that understanding. Maybe I need another lifetime to get the message across.
Professor Rita Jordan
Emeritus Professor in Autism Studies
University of Birmingham


In the February edition of the Psychologist, Professor Rita Jordan replies to my letter published in the December issue. This is a response top some of the points she makes.

Firstly, can I say that it was never my intention to disrespect her “lifetime’s work” or that of any of the many hard working researchers in the autism field. I think this research has been very illuminating in investigating social and mental processes. My “lifetime’s work” has been as a Clinical Psychologist with people with intellectual disabilities, which is why I confined my original letter to this group.

My original letter had questioned whether giving someone who already has an Intellectual Disability label the additional label of Autism adds anything to our treatment approaches. I still wait to be convinced.

Professor Jordan’s main point is that people on the autistic spectrum lack “instinctive social understanding”, and that this is qualitatively different to others with intellectual disabilities. Obviously in the constraints of a letter it has not been possible for her to explain this term, but what she says seems to imply that others with intellectual disabilities will “Acquire naturally” social behaviours.  This seems to ignore our personal experience and the research evidence that social deficits are common in people with intellectual disabilities. Here are a few examples.

Many studies have shown social skills deficits for children and adults with ID. For example, Agaliotis and Kalyva (2008) found that children with ID made fewer non-verbal initiations of social interactions than matched non disabled peers. In a different example Pina and colleagues (2013) compared the performance of children with ID with non-disabled peers on a social stories task. This asked the children to choose the best resolution to socially difficult situations. Children with ID were more likely to choose what they termed as ‘dysfunctional’ coping strategies (either aggressive or passive) than their controls.

In another example, Matson, Wilkins, and Ancona (2008) carried out an empirical study to describe symptom patterns in adults with autism and severe intellectual disability. They compared 57 adults with ID who met ICD10 diagnostic criteria for autism, with 57 matched controls with ID only. They DID find that the adults with autism showed a clear and distinct symptom profile compared with matched controls. Significant correlations were found between items on the scales they used and the diagnosis of autism. However, this study raises the interesting general question when we are considering whether the autism diagnosis is helpful when it comes to treatment implications, which is: ‘Does a finding of average group differences actually predict what an individual’s needs are?’ In their study, Matson and colleagues found the five most frequently endorsed symptom items for the people with autism were: Awareness of the rules of play, Interest in social activities, Interest in what other people say, Peer relationships and Limited number of interests. The percentage of the autistic group rated as having problems for these items ranged from 98% to 91%. These were significantly higher than the non-autistic group. However, if we look at the control group who did NOT have an autism diagnosis, the percentage rated as having problems on these items ranged from 68% to 47%. So, a large part of the non-autistic group also had problems in these areas, while a small minority of the autistic group did not. Even when there are statistically significant differences between groups of people labelled autistic and others who are not, this does not imply that we know which problems a specific individual with the label will have.

To me the issue still is whether this categorisation of people actually has treatment implications. I fully accept Professor Jordan’s premise that not everyone learns in the same way, and that we have to understand the individual’s own profile of difficulties, but I think this applies to everyone whether given the autism label or not. There seems to be more heterogeneity within the categories of Autism and Intellectual Disability that there are differences between them.

I do have concerns that by categorising people we risk setting up a two tier service for people with intellectual disabilities. I have seen in practice some people being given extra support worker time or better staffed day centres because they have an autism diagnosis, when others with intellectual disabilities just as great a need have not had access to these services.

A few years ago I said to Lorna Wing at an Autism Conference: “I still don’t understand what giving someone an autism label does for them.” She replied “If I think the label will help the person get better services, I will give it.” Is this right?

Graham Collins

Agaliotis, I. and Kalyva, E. (2008) “Nonverbal social interaction skills of children with learning disabilities.’ Research in Developmental Disabilities 29, 1–10.

Matson, J.L., Wilkins, J. and Ancona, M. (2008).  ‘Autism in adults with severe intellectual disability: An empirical study of symptom presentation.’ Journal of Intellectual and Developmental Disability 33, 1. 36-42.

Pina, F., Flavia, M., Laura, S. and Luana, S. (2013) ‘Learning disabilities and social problem solving skills’ Mediterranean Journal of Clinical Psychology I, 2, 1-23.


Without wishing to monopolise the Psychologist letters pages in any way, I hope I can be allowed one more attempt to clear up misunderstandings over my views on the value of an autism diagnosis. It is unfortunate, if understandable, that diagnostic systems still rely on behavioural ‘symptoms’, but, as I have always maintained, we have to go beyond these in order to see what is uniquely ‘autistic’. It is true that behaviours found in autism (including problematic social behaviours) are also found in other groups with developmental disorders, but I would claim that they are also found in typical development at certain times or under certain circumstances. Thus, there are no such things as ‘autistic’ behaviours; we all share behaviours because we are all humans. What is unique to autism is the origin of these difficulties in an inherent failure to engage socially and emotionally with others.

Those who do not have additional intellectual difficulties can, with effective teaching, learn many of the adaptive behaviours, which the typically developing acquire intuitively or through social guidance. The point is, however, that this adds a significant cognitive load, making them vulnerable to high levels of stress and consequent mental health problems. My plea on behalf of those on the autism spectrum is that we try to understand them and thus learn to support them more effectively. That does not mean they necessarily need more or ‘better’ resources than those with intellectual difficulty alone. I have also always maintained that diagnosis is a poor determiner of services; these should always be based on individual need.

The problem is that, without training, we are liable to be misled in determining those needs when it comes to autism.

Professor Rita Jordan PhD OBE

Emeritus Professor in Autism Studies University of Birmingham


BPS Members can discuss this article

Already a member? Or Create an account

Not a member? Find out about becoming a member or subscriber