UC Davis

The capacity to reason about and project mental states, such as intentions, desires, thoughts, knowledge and emotions, onto others’ minds is fundamental to social cognition, and is the foundation of social behavior. A sophisticated understanding of the mind is referred to as a ‘theory of mind’ (ToM), and is characterized by an understanding of the mind as representational—meaning that it can reflect or hold information independent of the real world—and person-specific—meaning that each individual has their own mind and mental states. Thousands of studies have shed light on the developmental improvements in ToM across childhood, differences in ToM development across neurotypical and neurodivergent populations, and cognitive and environmental factors that support ToM development. However, there remain large unanswered questions that continue to puzzle the field regarding infants’ seeming ability to pass ToM measures when assessed with non-declarative measures, identifying the social brain network and role of the temporo-parietal junction in ToM development, and how to characterize ToM more broadly as either a collection of discrepant mental states or a single social cognitive construct. However, empirically testing these questions is challenging due to limitations in ToM measurement tools. This dissertation illustrates the utility of psychometric examinations in further refining newly-developed ToM measures and in re-evaluating fundamental questions in ToM development, such as how mental states scale in difficulty across childhood. Study 1 (n = 224) evaluates a multi-mental state measure of ToM developed for children ages 3- to 12-years-old (Richardson et al., 2018) within an item response theory framework in order to examine difficulty—how challenging items within the scale are—and discrimination—how well individual items reflect children’s ToM. Results showed that items were generally discriminating, and the measure as a whole was reliable in assessing ToM in a more narrow range of 3- to 7.5-years-old. Findings also expand the developmental trajectory of mental-state reasoning previously described by Wellman and Liu (2004), in which children reasoned about beliefs based on past experience, emotions, morals, and sarcasm generally after understanding false belief. However, this ToM scale included an over-representation of questions about belief reasoning, and did not include equal numbers of tasks from each mental state. Thus, in a second study we developed a novel measure of ToM that was designed with approximately equal numbers of items within each mental-state category. Study 2 (n = 206) develops and evaluates a multi-assessment multi-mental state measure of ToM—the Comprehensive Assessment of Theory of mind (CAT)—in order to create a composite that reflects the range of mental-state reasoning encompassed by ToM, includes mental states not previously included in the Wellman and Liu (2004) scale (e.g., visual perspective taking), in addition to a non-social representational measure (e.g., false sign), and examines the developmental trajectory of different mental states when multiple assessments are administered. Results show that items were highly discriminating, and the measure reliably assessed ToM in 3- to 8-year-old children. Results also suggest a less-straightforward developmental trajectory of ToM when assessed with repeated measures and multiple question-types. These two large-scale studies lay critical groundwork for further psychometric work in ToM research, and illustrate how the inclusion of additional mental states in ToM assessments leads to a richer understanding of how children reason about mental states. We encourage future extensions of these psychometric inquiries and of the Comprehensive Assessment of Theory of mind (CAT).