Linguistic accommodation made by non-autistic speakers in response to being told the listener is autistic

Many people simplify their speech to accommodate listeners who may face comprehension difficulties, including older adults and non-native speakers, giving rise to registers such as Elderspeak and Foreigner-Directed Speech. Given the communication difficulties autistic people experience and prevalence of negative stereotypes about their communicative competence, they may also be recipients of simplified speech. Indeed, many anecdotal accounts from autistic individuals report this experience, but this has not been systematically investigated. In the current pre-registered study, we investigated whether non-autistic speakers similarly modified their speech in interactions with listeners who disclosed they were autistic. We used the Map Task paradigm; 33 participants (mean age = 36 years) gave verbal instructions to a confederate in two conditions: one in which the confederate disclosed being autistic and the other in which they did not. Participants did not show evidence of linguistic accommodation toward the autistic listener on any of the linguistic measures we coded for, including speech rate, syntactic complexity (e.g., MLU), semantic complexity (e.g., type-token ratio) and discourse features (e.g., repetitions). Although there was no average difference on these measures between the autistic and non-autistic listener conditions, exploratory analyses showed that more positive attitudes toward autism were associated with lower degree of accommodation to the autistic listener on one measure, MLU. No associations were found between level of contact and degree of accommodation. These findings suggest that an autism label alone does not cue linguistic accommodations, and that more positive attitudes toward autism may reduce the likelihood of linguistically accommodating autistic listeners.