The Curious Incident of the Dog in the Night Time and Rain Man both tell the story of people with certain kinds of Autism Spectrum Disorder (ASD) who nevertheless have some remarkable spared abilities. Although almost all children with ASD suffer some type of language impairment, many researchers have argued that grammar is often a spared ability of this type. The idea is that while children with ASD seem to have poor language on the surface, this may be largely due to their impaired social skills. Their knowledge of grammar – the invisible rules under the surface – is actually pretty good (i.e., it’s that they don’t know what to say rather than how to say it). In order to test this idea, we need to get under the surface, and look at what children with ASD know about language, rather than just what they say.
LuCiD Researcher Dr Ben Ambridge (University of Liverpool) recently published a study designed to do just this. Rather than asking children with ASD to produce language, Ben – together with student Georgina Jackson – asked them to rate sentences for how good or silly they sounded, using a 5-point scale. The advantage of this method is that the task does not require children to make use of the social-interaction aspects of language, or – necessarily – to say anything at all (they respond by placing a counter on a scale).
The study found that, compared with typically-developing children matched for IQ, children with high-functioning ASD were less sensitive to the difference between grammatical and ungrammatical sentences (e.g., The cup fell off the shelf vs *Lisa fell the cup off the shelf) used to describe the same scene (e.g., Lisa knocking the cup off the shelf). This was a somewhat surprising finding, as it contradicts the popular view discussed at the beginning of this article: that grammar is generally spared in ASD. Instead, it seems that, when we use a sufficiently sensitive test, we uncover a grammatical impairment in even relatively high-functioning children with ASD. The implication for professionals is that a grammaticality judgment task may be useful for diagnosing subtle impairments that may otherwise go unnoticed, but which may adversely affect language and literacy development (perhaps only in later childhood or early adulthood). This study was published in the Journal of Autism and Developmental Disorders. You can read it here