Abilities to Explicitly and Implicitly Infer Intentions from Actions in Adults with Autism Spectrum Disorder.

The sample was small and highly selected: 17 adults with ASD were analysed after four of 21 recruited participants were excluded for control-task performance, and all had IQ scores above 100 with no history of delayed language development or learning difficulties, so it says little about lower-IQ adults or those with broader support needs. The explicit experiment used a smaller analysed sample than the implicit experiment (30 participants, 15 matched pairs), because two adults with ASD could not return, so the two tasks are not compared in exactly the same people. The implicit task was not significant, but it showed a trend, so it cannot be read as proof that implicit intention inference is intact. The videos were brief hand-action clips without faces, so everyday social judgement involving faces, speech, context and conversation may be harder. Three participants with ASD had ADOS scores below the clinical cut-off, although their AQ and SRS scores indicated high traits or social impairment, so the group's clinical status was not uniform.

Funding was from the Economic and Social Research Council. The supplied text does not report author conflicts of interest.

Do not conclude that adults with autism cannot infer intentions in everyday life. The significant deficit was only when participants were explicitly asked to label intentions from hand-action videos, while the implicit partner-choice task was not significantly impaired, and all participants had IQ scores above 100.