Exceptional Skills in Autism Are Common but Poorly Measured

What the review found

Between 10 and 50 percent of autistic people are reported to have savant syndrome. A spread that wide is a measurement problem, and the review that assembled these figures calls the prevalence estimates inconsistent.

The review is Hyun Ok Park’s, at Baekseok University in Korea, published in 2023 in the Journal of the Korean Academy of Child and Adolescent Psychiatry. Forty articles from 2008 to 2022 were synthesised, half of them literature reviews or position papers and half surveys, with participants ranging in age from 7 to 90 and most studies concentrated on people from their late teens to about forty. Roughly half of all people with savant syndrome are autistic; how many autistic people have savant syndrome is the question nobody can answer.

Savant syndrome has no official diagnostic criteria. The term describes an ability standing far above what the rest of a person’s functioning would predict. Neither the DSM nor the ICD, the manuals that define recognised conditions, includes it, so researchers borrow six items from the Autism Diagnostic Interview-Revised and use those as a proxy. Screening questionnaires have been written, though none has been validated or standardised, and reported sex ratios vary between six males to every female and four to one.

A registry of 319 savants records 79 percent male and 90 percent with a congenital rather than an acquired skill, autism being the underlying disability in 75 percent of those. Music was the most frequent principal skill, with art and memory following.

Among 137 autistic adults of average age 24 studied by Howlin and colleagues, 39, or 28.5 percent, met criteria for a savant skill or an exceptional cognitive skill; a third of the men and 19 percent of the women showed some form of outstanding ability. No participant with a non-verbal IQ below 50 met criteria for a savant skill, and the savant group’s overall cognitive function exceeded the non-savant group’s, averaging a non-verbal IQ near 80. The familiar picture of extraordinary talent emerging from severe impairment has thin support here.

A screening questionnaire found that 42 percent of respondents had a child with one or more skills outstanding given the child’s overall functioning, and those skills were associated with strong working memory and highly focused attention; obsessionality showed no such association. Four cognitive theories compete to explain the phenomenon, one of which, detail-focused processing, describes as a capability what an older account called weak central coherence. Park concludes that the four overlap, and that no single one accounts for savant syndrome. What the field has, after fifteen years of study, is something it can describe and cannot yet define.

What it means for you and your child

A parent waiting for that definition to arrive will wait a long time. There are no validated instruments and no agreed threshold, and Park notes that research into whether savant skills can be taught or strengthened has scarcely been attempted, which makes any promise about developing a child’s talent a promise the evidence does not cover.

One study asked parents directly instead of applying a threshold. Almost half of the children, 46 percent, had at least one talent their parents could name, and a further 23 percent with no extraordinary talent had at least one personal strength. Those children also performed differently on standardised tests of visuospatial ability, drawing, computation, reading and memory than children without a skill in the domain concerned, which is the review’s argument for assessing strengths area by area rather than treating a child’s ability as one quantity.

Assessed that way, the question changes from whether a child qualifies for a rare designation to what this particular child does easily and in which domain. An aptitude for music, for drawing, for calculation, for recalling routes or dates, is worth recording as an aptitude whether or not it clears a bar that nobody has agreed on. The skills that do appear generally advance rather than fade; the one famous case of loss, a child called Nadia whose drawing vanished after intensive language instruction, is described as exceedingly rare.

None of this rests on a diagnosis of savant syndrome, and given the state of the criteria, none of it should. Schools can be asked about a specific subject rather than about a general profile, and a strength recorded in one area does not have to survive being averaged against a weakness in another to count. The review’s finding that such skills accompanied focused attention and strong working memory, and not obsessive preoccupation, gives a parent plainer language for where the ability comes from.

What remains uncertain in this literature is how to count exceptional skill and what to call it. Forty studies do not dispute that autistic children have abilities worth finding, and in the one study that simply asked, nearly half of the parents could already name their child’s. Recognising what your child does well has never required the field’s permission.