Conversational adaptation in children and teens with autism: Differences in talkativeness across contexts.

The study measured only a few language features, mainly word count, so it did not test pitch, grammar, word meaning, or topic changes. The conversations were 3 minutes long, and behavior was scored across the whole conversation, so the study cannot show how adaptation changes during the interaction. The Interested conversation was always first and the Bored conversation second, so order effects cannot be ruled out. The sample was limited to verbally fluent participants aged 7 to 17, so findings do not apply to younger children, adults, or people who are not verbally fluent. Confederates were mainly female undergraduate students and research assistants, not same-aged peers, so partner sex and age may affect the result. The diagnostic groups were not matched on race, and confederates did not report race, so racial and cultural factors are not controlled. The sample of 98 participants limits subgroup analysis, including comparisons by sex and autism symptom severity. The study cannot identify why autistic participants did not adapt, such as cue perception, motivation, skill, anxiety, theory of mind, or receptive language. The method cannot separate the precise causal effect of confederates' verbal and nonverbal behavior on participants' speech.

The supplied text lists research support from the N.I.H., Extramural, but does not state author conflicts of interest.

Do not conclude that autistic children cannot adapt their conversation or that a specific social skills treatment will change word count. The study compared two short conversations with unfamiliar confederates and did not test training, longer interactions, or same-aged partners.