Two-fold nature of autism

The debate in our 'Letters' pages continues.

With regard to the recent correspondence in The Psychologist regarding the diagnosis of autism (Letters, December 2014, March–May 2015), I would like to propose that all of the viewpoints expressed, from the sceptical (Collins) to the orthodox (Jordan) via the confused (Hassall) could be reconciled by a basic change in the way in which autism is conceptualised. The current conceptualisation of autism is best summed up in the phrase of the late Lorna Wing as being a ‘lumpy continuum’, in which autism is regarded as having a two-fold nature, being both a discrete, and hence diagnosable, disorder, albeit one that falls short of the requirements of a syndrome, and simultaneously a set of traits that can be identified independently of both general cognitive ability and other identifiable developmental pathology. However, unlike the wave/particle dual nature of light, there do not appear to be any objective conditions for resolving under what circumstances autism is a category or a trait! However, developments in the field of molecular genetics (Happé & Plomin, 2006) lend support to the notion that autism is probably two genetically distinct traits, namely impaired social communication abilities and repetitive, restrictive and ritualistic behaviour, but which tend to have a high frequency of co-occurrence. It is therefore the degree of co-occurrence that results in autism being erroneously regarded as a distinct clinical entity.

Such a two-fold model of autism currently underpins the current DSM-5 definition of autism, whilst acknowledging the possibility of the separate occurrence of each set of traits, both of which are readily measurable in themselves. In the case of impaired social communication, this is measurable using the tools such as the Social Responsiveness Scale (Constantino et al., 2003), whilst the trait of repetitive, restrictive and ritualistic behaviour is particularly amenable to measurement using instruments such as the Repetitive Behaviour Scale (Moss et al., 2009).

Re-conceptualising autism as two co-occurring traits also enables us to make sense of the continuing difficulty in identifying a specific aetiology or genotype for autism, an issue that is all the more surprising given the effort put into this and the apparent ease with which the aetiologies of much rarer developmental disorders and syndromes have been identified.

On this basis, if we regard autism as a pair of stable human traits that are more or less distributed throughout the population, we can resolve the question posed by Richard Hassall (Letter, May 2015) of whether there is a difference between having autism and having a diagnosis of autism by referring to how autistic someone is, rather in the manner that might refer to how tall a person is. Hence, issues of diagnosis are essentially made redundant whilst still retaining an objective basis for determining clinical, educational and social need.

Dr Dougal Julian Hare CPsychol
School of Psychological Sciences
University of Manchester

Constantino J.N., Davis S.A. Todd R.D. et al. (2003). Validation of a brief quantitative measure of autistic traits: Comparison of the Social Responsiveness Scale with the Autism Diagnostic Interview-Revised. Journal of Autism and Developmental Disorders, 33(4), 427–433.
Happé, F. & Plomin, R. (2006). Time to give up on a single explanation for autism. Nature Neuroscience, 9(10), 1218–1220.
Moss J., Oliver C., Arron K. et al. (2009). The prevalence and phenomenology of repetitive behavior in genetic syndromes. Journal of Autism and Developmental Disorders, 39, 572–588.


I am sorry Richard Hassall (Letters, May 2015) is still confused about autism and even more that I am apparently responsible for some of the confusion. I can see that it appears paradoxical to claim that ‘diagnosis is not a good basis for services’ while still maintaining the value of diagnosis, so let me explain further.

Services should indeed be individually based, but there are conditions (such as congenital blindness, cerebral palsy, autism) where development is atypical and thus where parents and practitioners cannot just ‘read’ individual needs from observing behaviour. In such cases, it is necessary to know the diagnosis to understand what the behaviour might mean in the development of that individual. Treatment for the individual will vary according to that interpretation.

Thus, behavioural observation is vital but not sufficient for determining needs in these cases; the observation has to be through the ‘lens’ of understanding of how development occurs in autism. The ‘behavioural indicators’ of diagnostic systems alert one to the possibility of autism (or another condition) but the diagnostic decision puts those indicators in the context of the developmental history of the individual. The best diagnostic tools (e.g. DISCO: Wing et al., 2002) emphasise careful case histories of development. In autism, one would be looking for signs of early disturbance in instinctive responses to social signals such as those involved in joint attention and response to one’s own name. ‘Complex social deficits’ would be a strong alert for autism, but it would be the nature of those difficulties and their course in the individual’s development that would be key determiners of actual diagnosis.

I am concerned about some of the effects of DSM-5 in making its now duo of diagnostic criteria compulsory for an ASD diagnosis and thus making rigidity in behaviour and thinking a determining criterion. In DSM-IV there was a triad of criteria for ASD with the category PDD-NOS needing to score on only two of the three criteria. Since PDD is now gone from DSM-5, so also has PDD-NOS. This classification has always been problematic (characterised more meaningfully in ICD-10 as ‘atypical autism’), but including these children within the broad category of ASD did allow their possible ‘autistic’ development to be taken account of. Some research has indicated that this group may not develop a full obsessive type of ritual behaviour until late childhood and so they may fail to meet the rigidity criterion in DSM-5 and thus fail to gain an ASD diagnosis.

Richard Hassall assumes (as do many others) that their needs will be met by being included in the new ‘social-communication disorder’ category. However, this is a category set up to cover children with similar social and communication behaviours to autism but with possibly very different developmental histories. It would cover those with pragmatic language disorders where the social difficulties may not be primary, but arise from the communication problems. Where that is the case, such a diagnosis will be misleading when it comes to interpreting the behaviour of those who would be better understood as being on the autism spectrum. Since this category is new, there will also be few staff qualified to understand and work with these children and the outcome of these diagnostic changes needs careful monitoring if many more children are not to ‘slip through the net’ of services and support.

To answer the final question posed by Richard Hassall, a diagnosis of autism provides a signpost to understanding the individual with autism. As we agreed in the NIASA report on diagnosis (2003) where there are behavioural indications of autism but not a definite diagnosis of the condition, it would be the ‘least damaging assumption’ for parents and practitioners to help and support the child as if they had autism, while remembering that it may turn out they do not after all. There is nothing in the good practice recommended for autism that would harm the development of others, whereas neglecting to attend to the autism in a child with the condition is most likely to lead to further problems. The real issue is how to increase understanding of autism in diagnosticians and practitioners so that all children can be included, recognising the importance of this understanding, while basing resources on need determined through that understanding.

Rita Jordan
Emeritus Professor in Autism Studies
University of Birmingham

National Initiative for Autism: Screening and Assessment (NIASA) (2003). National autism plan for children. London: National Autistic Society.
Wing, L., Leekam, S.R., Libby, S.J. et al. (2002). The Diagnostic Interview for Social & Communication Disorders. Journal of Child Psychology and Psychiatry, 43(3), 303–325.

BPS Members can discuss this article

Already a member? Or Create an account

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