This dissertation puts forward a series of arguments and theoretical proposals about the architecture and development of the human capacity to reason about the internal, psychological causes of behavior, known as “theory of mind” or “mindreading.”

Chapter 1, “Foundations and motivations,” begins by articulating the philosophical underpinnings of contemporary theory-of-mind debates, especially the dispute between empiricists and nativists. I then argue for a nativist approach to theory-of-mind development, and then go on to outline how the subsequent chapters each address specific challenges for this nativist perspective.

Chapter 2, “Pragmatic development and the false-belief task,” addresses the central puzzle of the theory-of-mind development literature: why is it that children below the age of five fail standard false-belief tasks, and yet are able to pass implicit versions of the false-belief task at a far younger age? According to my novel, nativist account, while they possess the concept of BELIEF very early in development, children’s early experiences with the pragmatics of belief discourse initially distort the way they interpret standard false-belief tasks; as children gain the relevant experience from their social and linguistic environment, this distortion eventually dissipates. In the Appendix (co-authored with Peter Carruthers), I expand upon this proposal to show how it can also account for another set of phenomena typically cited as evidence against nativism: the Theory-of-Mind Scale.

Chapter 3, “Spontaneous mindreading: A problem for the two-systems account,” challenges the “two-systems” account of mindreading, which provides a different explanation for the implicit/explicit false-belief task gap, and has implications for the architecture of mature, adult mindreading. Using evidence from adults’ perspective-taking abilities I argue that this account is theoretically and empirically unsound.

Chapter 4, “Character and theory of mind: An integrative approach,” begins by noting that contemporary accounts of mindreading neglect to account for the role of character or personality-trait representations in action-prediction and interpretation. Employing a hierarchical, predictive coding approach, I propose that character-trait representations are rapidly inferred in order to inform and constrain our mental-state attributions.

Because this is a “covering concept” dissertation, each of these chapters (including the Appendix) is written so that it is independent of all of the others; they can be read in any order, and do not presuppose one another.