Welcome to the new version of CaltechAUTHORS. Login is currently restricted to library staff. If you notice any issues, please email coda@library.caltech.edu
Published February 4, 2019 | Supplemental Material
Journal Article Open

Deconstructing Theory-of-Mind Impairment in High-Functioning Adults with Autism


Inferring the beliefs, desires, and intentions of other people ("theory of mind," ToM) requires specialized psychological processes that represent the minds of others as distinct from our own. ToM is engaged ubiquitously in our everyday social behavior and features a specific developmental trajectory that is notably delayed in children with autism spectrum disorder (ASD). In healthy individuals, model-based analyses of social learning and decision-making have successfully elucidated specific computational components of ToM processing. However, the use of this approach to study ToM impairment in ASD has been extremely limited. To better characterize specific ToM impairment in ASD, we developed a novel learning task and applied model-based analyses in high-functioning adults with ASD and matched healthy controls. After completing a charitable donation task, participants performed a "mentalizer" task in which they observed another person (the agent) complete the same charity task. The mentalizer task probed the participants' ability to acquire and use ToM representations. To accurately predict agent behavior, participants needed to dynamically track the agent's beliefs (true or false) about an experimental context that varied over time and use that information to infer the agent's intentions from their actions. ASD participants were specifically impaired at using their estimates of agent belief to learn agent intentions, though their ability to track agent belief was intact and their reasoning about belief and intentions was rational. Furthermore, model parameters correlated with aspects of social functioning, e.g., ADOS severity scores. Together, these results identify novel, and more specific, targets for future research.

Additional Information

© 2019 Elsevier Ltd. Under an Elsevier user license. Received 24 August 2018, Revised 22 November 2018, Accepted 20 December 2018, Available online 24 January 2019. Data and Software Availability: Complete raw data, experiment code (PsychToolBox 3 [33, 34]), materials, and analysis scripts (MATLAB), are available from the authors and can be found online at https://osf.io/ahp5q/. The authors would like to thank Dr. Antonio Rangel for help with initial experimental design; Drs. Daniel McNamee and Jeff Cockburn for help with model-based analyses; and Ghoncheh Ayazi, Marisol Espino, Lynn Paul, Tim Armstrong, Remya Nair, and other members of the Adolphs lab for help with data collection, database management, data analysis, and autism assessment. This work was supported by the National Institute of Mental Health at the National Institutes of Health (K01MH099343 to D.A.S. as well as P50MH094258 and R01MH080721 to R.A.). Author Contributions: D.A.S. and C.A.H. developed the initial experimental concept. I.A.R. and D.A.S. contributed to data collection and analysis. All authors contributed to experimental design and discussion of results. I.A.R., R.A., and D.A.S. contributed extensively to write up of the manuscript. The authors declare no competing interests.

Attached Files

Supplemental Material - 1-s2.0-S0960982218316737-mmc1.pdf

Supplemental Material - 1-s2.0-S0960982218316737-mmc2.pdf


Files (4.9 MB)
Name Size Download all
4.1 MB Preview Download
731.0 kB Preview Download

Additional details

August 22, 2023
October 20, 2023