Saturday, 25 March 2017

Including the "full intellectual range" in autism vision research

The paper by Alyse Brown and colleagues [1] (open-access available here) is probably not going to gain any significant media headlines (unlike other recent studies - see here and see here) but does cover a rather important question regarding the autism research landscape: how representative is autism research?

Specifically looking at the collected research on visual processing (distinct from physical issues with the eyes that still require greater awareness) with autism in mind, the authors surveyed the research literature to determine "what extent the ASD with-ID [intellectual disabilitypopulation has been excluded from visual research." Intellectual or learning disability is one of the more frequently over-represented comorbidities that can accompany a diagnosis of autism or autism spectrum disorder (ASD). Their answer: "our searches indicate that 80% of the vision research associated with ASD is representative of less than 60% of the appropriate population, i.e., those with ASD without ID while the ASD with ID group who we argue currently represent 42% of the ASD population, have not been adequately considered."

You may well quibble with the "recalculation of ASD prevalence figures, using the criteria of DSM-5" as a means of calculating that '~40% of those with autism have ID too' figure. For me however, the message is quite stark: autism research - specifically related to visual processing issues - is not yet representative of  'all autism'.

"Reluctance to test individuals who are below 80 in IQ is presumably a practical stance as the data collected from these individuals are often hard to obtain, and often close to floor level performance." The authors note however that the presence of ID alongside autism in the area of visual processing is not something that cannot be 'overcome' by researchers with some creative thinking and a few modification(s) to their experimental designs. Indeed, visual processing research lends itself well to quite a few alterations to methods [2]...

How applicable might these results be to other areas of autism research? Well, we just don't know. I daresay that quite a lot of the 'psychology' based autism research in particular might show a bias towards autism without intellectual disability for just those reasons listed above. The problem then of grand, over-arching generalisations to 'all autism' on the basis of results from the more 'cognitively-able' becomes apparent. Of course, in these days of the plural 'autisms' (see here) and the realisation that 'heterogeneity means heterogeneity' when it comes to autism (see here) one could argue that even characterisations based on the presence of ID or not when it comes to autism are equally 'simplistic' and equally 'useless'. How many autisms might well have an ID element to them? Is ID a comorbidity or something rather more central to some of the autisms? These questions and related others are ones that autism research as a whole will eventually have to start looking at and taking into account.

And going back to the issue of eye disorders being potentially over-represented and under-diagnosed in relation to autism, the paper by Mouridsen and colleagues [3] reiterates that intellectual ability when accompanying autism needs more health equality: "The rate of eye disorder was particularly high (24.5%) in those with a co-occurring profound or severe learning disability (IQ < 50)."


[1] Brown AC. et al. Vision Research Literature May Not Represent the Full Intellectual Range of Autism Spectrum Disorder. Front Hum Neurosci. 2017 Feb 14;11:57.

[2] Boot FH. et al. Delayed visual orienting responses in children with developmental and/or intellectual disabilities. J Intellect Disabil Res. 2013 Dec;57(12):1093-103.

[3] Mouridsen SE. et al. Eye Disorders among Adult People Diagnosed with Infantile Autism in Childhood: A Longitudinal Case Control Study. Ophthalmic Epidemiol. 2017 Mar 15:1-4.

---------- Brown AC, Chouinard PA, & Crewther SG (2017). Vision Research Literature May Not Represent the Full Intellectual Range of Autism Spectrum Disorder. Frontiers in human neuroscience, 11 PMID: 28261072