ASHA journals
Browse

Pronoun comprehension and theory of mind in autism (Abubakare & Snedeker, 2025)

Download (473.08 kB)
online resource
posted on 2025-03-28, 17:31 authored by Oluwatobi Abubakare, Jesse Snedeker

Purpose: Pronouns stitch together discourse by linking referents within and across sentences. Previous research has shown that people often rely on two strategies to interpret pronouns: the subject bias (assuming the pronoun refers to the subject of a prior sentence) and the repeated mention bias (assuming it refers to a person that was mentioned repeatedly). The present study seeks (a) to determine whether autistic adults make use of these strategies to the same degree as non-autistic adults with similar language skills and (b) to assess whether use of these strategies is correlated with theory of mind or vocabulary knowledge.

Method: Native English-speaking autistic and non-autistic adults completed a pronoun comprehension task, as well as a modified version of Happe’s Strange Stories task, the Vocabulary Size Test (VST), and the Autism Spectrum Quotient Test.

Results: Both groups used both strategies most of the time with no reliable differences between the groups. Performance on the VST predicted use of the subject bias strategy.

Conclusions: Autistic adults use some of the same strategies for interpreting ambiguous pronouns as non-autistic adults with similar language skills. Variation in subject bias strategy use is correlated with vocabulary knowledge, a sensitive measure of variation in language skills.

Supplemental Material S1. Instructions for the Pronoun Comprehension tasks, list of stimuli sentences for Pronoun Comprehension tasks, and additional figure to show lines of best fit for each group regarding correlations between Subject Bias and Vocabulary.

Abubakare, O., & Snedeker, J. (2025). Personal pronoun comprehension and theory of mind in autistic adults. Perspectives of the ASHA Special Interest Groups. Advance online publication. https://doi.org/10.1044/2024_PERSP-24-00148

Funding

Research design and data collection was supported by National Institutes of Health Diversity Supplement Grant 3R03HD097629-02S1 awarded to the first author. Data analysis and preparation of this article was supported by the National Science Foundation Graduate Research Fellowship under Grant 1342962 awarded to the first author.

History