Author ORCID Identifier

Date of Award

Spring 6-10-2023

Document Type

Thesis (Ph.D.)

Department or Program

Cognitive Neuroscience

First Advisor

Jonathan S. Phillips


This dissertation examines mentalizing abilities, causal reasoning, and the interactions thereof. Minds are so much more than false beliefs, yet much of the existing research on mentalizing has placed a disproportionately large emphasis on this one aspect of mental life. The first aim of this dissertation is to examine whether representing others’ knowledge states relies on more fundamentally basic cognitive processes than representations of their mere beliefs. Using a mixture of behavioral and brain measures across five experiments, I find evidence that we can represent others' knowledge quicker and using fewer neural resources than when representing others’ beliefs. To be considered a representation of knowledge rather than belief, both mentalizer and mentalizee must accept the propositional content being represented as factive (Kiparsky & Kiparsky, 2014; Williamson, 2002). As such, my results suggest that representing the mental states of others may be cognitively easier when the content of which does not need to be decoupled from one’s own existing view of reality.

Our perception of other minds can influence how we attribute causality for certain events. The second aim of this dissertation is to explore how perceptions of agency and prescriptive social norms influence our intuitions of how agents and objects cause events in the world. Using physics simulations and 3D anthropomorphic stimuli, the results of three experiments show that agency, itself, does not make agents more causal to an outcome than objects. Instead, causal judgments about agents and objects differ as a function of the counterfactuals they respectively afford. Furthermore, I show that what distinguishes the counterfactuals we use to make causal attributions to agents and objects are the prescriptions we hold for how they should behave.

Why do we say a fire occurred because of a lightning strike, rather than the necessary presence of oxygen? The answer involves our normative expectations of the probability of lightning strikes and the relative ubiquity of oxygen (Icard et al., 2017). The third aim of this dissertation explores the gradation of causal judgments across multiple contributing events that each vary in their statistical probability. I contribute to ongoing theoretical debates about how humans select causes in experimental philosophy and cognitive science by introducing a publicly available dataset containing 47,970 causal attribution ratings collected from 1,599 adult participants and structured around four novel configurations of causal relationships. By quantitatively manipulating the influence of normality, I systematically explore the continuous space of a causal event’s probability from unlikely to certain. It is my hope that this benchmark dataset may serve as a growing testbed for diverging theoretical models proposing to characterize how humans answer the question: Why?