Strange Words: Autistic Traits and the Processing of Non-Literal Language.

Austin et al. (2015) asked neurotypical adults to read short stories. Some sentences swapped a familiar noun with a new made-up label.

The team also gave each person the Autism-Spectrum Quotient. They wanted to see if higher autism traits slowed reading when the text turned figurative.

People who scored higher on the AQ took longer to read the odd new words. The delay only showed up when the strange label stood in for something else, not in plain literal lines.

Even in fully neurotypical adults, more autistic-like traits meant measurably slower snap-of-meaning processing.

Palka Bayard de Volo et al. (2021) extends this hint. They tested diagnosed autistic adults and saw the same trouble balloon into large everyday errors on implied meaning. Together the two papers draw a line: subtle slowing in the high-AQ group becomes major comprehension gaps in diagnosed ASD.

Dimitrova et al. (2017) seems to disagree at first glance. Their verbal children with ASD understood pointing and gesture combos as well as language-matched peers. The key difference is modality: gestures are concrete cues, while E et al. looked at abstract, invented metonyms. The studies don’t clash—they show concrete supports can bypass figurative slowdowns.

Perrot et al. (2021) add another piece. Autistic adults also produce fewer mental-state words when talking. Receptive lag (E et al.) plus expressive shortage (Alexandra et al.) point to a shared pragmatic thread worth targeting in therapy.

If a client’s AQ is high or ASD is diagnosed, expect hidden drag when figurative or made-up terms pop up in tasks or social stories. You can pre-teach odd labels, give extra wait time, or swap in concrete cues like gestures or visuals. These tweaks cost nothing and may prevent the bigger comprehension misses C et al. later documented.