The concept of “compensation” makes sense of several autism puzzles

A process involved in neurodevelopmental disorders that we are only just beginning to understand is "compensation" – the way that a deficit can be partially or wholly masked by automatic mental processes and/or deliberate behavioural strategies. For instance, a person with dyslexia may achieve typical levels of reading ability after an earlier diagnosis, not because the disorder has gone away (subtle tests might show continuing problems in phonological processing, for example) but through the use of behavioural strategies, such as reverse-engineering a tricky word from the meaning of words around it. In a new review in Neuroscience and Biobehavioral Reviews Lucy Anne Livingston and Francesca Happé, at the Institute of Psychiatry in London, take us through what compensation might mean for autism.

Compensation in autism could make sense of several puzzles. One is how high-risk individuals – with an autistic sibling, and showing markers for the disorder at the age of one – sometimes beat the odds and go on to seemingly normal development. Similarly, some individuals are diagnosed with autism, but in adulthood achieve what is termed a "good outcome": functional social relationships, work life, and general independence. These cases could be explained by compensation mechanisms kicking in at different times. Meanwhile, people who remain undiagnosed until later life could have been relying to that point on compensation mechanisms that have become ineffective, perhaps because their lives have become more complex.

Moreover, scientists are reassessing the prevalence of autism in girls and women, now understood to stand at around one female for every three males according to one analysis. The reason for the historical under-estimate – even fairly recently standing at a male to female ratio of four to one – and for why autism may present differently in women (who are often diagnosed later in life), may again be compensation; one source of evidence for this is that women with autism report a higher inclination to camouflage their difficulties in order to fit in socially.

A core deficit of autism is an impairment in Theory of Mind: difficulty figuring out what people are thinking in order to make sense of their behaviour. Livingston and Happé asked themselves how this might be compensated. It's possible that what neurotypical people achieve implicitly can also be achieved explicitly, by reasoning through situations to predict behaviour and motives. If true, it's likely higher intelligence would assist this compensatory process. Consistent with this account, intelligence is one of the strongest correlates of "good outcome"; it also tends to be higher in the unaffected siblings of those with autistic spectrum disorders (ASD), but lower in women diagnosed with autism.

Another factor likely to be relevant to autistic compensation is "executive function" – the suite of meta-cognitive functions that includes planning and inhibiting action, flexibility in thinking and switching focus. It's easy to see how being able to prevent undesirable behaviours and plan appropriate social responses ahead of an interaction might prevent visible faux pas and allow an individual to pass as neurotypical. For a long time, executive dysfunction was seen as core to autism, and Livingston and Happé point out this could be because when that dysfunction was present, compensation was lacking, making autism symptoms more obvious.

For autistic people and those around them, understanding the nature of their compensation may be very useful. Compensation may be deep or shallow: Livingston and Happé refer to the development of echolocation by blind people, which genuinely restores some of the functionality of sight (i.e. deep compensation), versus the use of a cane, which serves a much more circumscribed role (shallow compensation). In autism, two individuals might present as equally adjusted to social situations, such as responding to jokes, but the one relying on a shallow strategy (for example, join in whenever other people laugh) is vulnerable to making errors and may struggle in novel situations (e.g. a one-on-one date).

Moreover, compensatory strategies are likely to exact a cost, drawing on resources that would normally be invested into other tasks. A common complaint from those on the spectrum is it's hard work to "act normal" – as recently expressed by TV presenter Chris Packham (see footnote). Diagnosed with Asperger's later in life, Packham says he was "forced to develop" a number of coping mechanisms. Individuals who spent more of their life undiagnosed – and hence without a way to make sense of their difference – are likely to have spent extended periods working to pass invisibly in society, and this may go some way to explain the higher levels of suicidal ideation in this group – at 66 per cent, far higher than in those diagnosed in childhood. Understanding different forms of compensation and their consequences may be an important step in helping autistic people manage their mental health more effectively.

Although it's early days, we're starting to understand how to identify compensation in autism. Work with "good outcome" individuals shows that with sufficiently sensitive tasks, such as anticipating where people direct their gaze in complex social scenarios, it's possible to pick up Theory of Mind deficits that were otherwise undetectable. The next step, argue the authors of this new review, is to develop a gold-standard measure that we agree captures the core, deep marker of Theory of Mind deficits in ASD, just as precise phonological measures do in dyslexia. Through this, we can not only understand the disorder at a scientific, medical level, but we can arrive at a personalised account of the ways individuals live with the disorder, the situations that may challenge them the most, and the psychological toll that they pay to pass in neurotypical cultures.