Thinking in patterns : representations in the neural basis of theory of mind

Abstract

Social life depends on understanding other people's behavior: why they do the things they do, and what they are likely to do next. These actions are just observable consequences of an unobservable, internal causal structure: the person's intentions, beliefs, and goals. A cornerstone of the human capacity for social cognition is the ability to reason about these invisible causes; having a "theory of mind". A remarkable body of evidence has demonstrated that social cognition reliably and selectively recruits a specific group of brain regions. Yet, we have little insight into how these neural substrates function at a computational level. This thesis lays the groundwork to address that question, both empirically and theoretically, first by demonstrating that functional neuroimaging can find behaviorally relevant features of mental state representation within the cortical regions that support social cognition, and second by proposing a theoretical framework to interpret activity in these brain regions. In Chapter 1, I review the literature of the last 15 years, and argue that a key next step in understanding the neural basis of social cognition is characterizing the neural representations and computations supported by "social" brain regions. In Chapter 2, I demonstrate in four experiments that functional neuroimaging can be used to find neural representations of distinct features of mental states. Specifically, I show that multivoxel pattern analysis (MVPA) can detect features of mental state representations (e.g., intent), and that these neural patterns are behaviorally relevant, including in autism spectrum disorders. In Chapter 3, I demonstrate that these brain regions contain explicit, abstract representations of another feature of others' mental states: perceptual source. I find that these representations persist in the face of drastic changes in developmental history (congenital blindness), providing evidence that these representations emerge even in the absence of relevant first-person experience. In Chapter 4, I demonstrate that these cortical regions contain representations of epistemic and emotional features of others' beliefs, and that these features are represented along continuous, abstract dimensions. Finally, in Chapter 5, I extend a model from vision and neuroeconomics - predictive coding - and explore its application to the neural basis of social cognition. Together, this work provides a key next step to understanding the neural basis of theory of mind, by demonstrating that it is possible to find abstract, behaviorally relevant features of mental state inferences inside cortical regions that support social cognition, and taking a first step in characterizing their content and format.