Three Papers on Beliefs: on their Measurement, their Transmission, and its Implications for Large-scale Change.

Abstract

This dissertation examines the dynamics of widely shared cultural beliefs and the impact that they have on individual-level cognition. It explores how these cultural structures shape cognitive processes like categorization, how the social learning strategies that agents use to learn these structures from one another affect trajectories of cultural transmission, and how our assumptions about how learning occurs across the life-course have implications for large-scale patterns of cultural change. To examine the first element, I collect survey and reaction time data about moral judgments through Prolific. Using sociological approaches to measuring cultural meaning, I show support for the idea that template matching – a specific type of categorization - underpins the attribution of immorality. I then use computational text analysis to test the external validity of my results. To address the second set of questions, I – along with Tom Wolff – build an agent-based simulation where agents use different social learning strategies and place varying weights on the information retrieved with each of these heuristics. We show that processes of cultural transmission vary considerably according to the combination of strategies that agents use, as well as on the topology of the structures where they are embedded. Further, we highlight that there are important parallels between the social learning literature and sociological work on social influence. Our study puts both lines of work in conversation and shows that this intersection is a prolific site of collaboration, especially for sociologists interested in modeling cultural transmission. Lastly, to explore the third set of questions, I build another agent-based simulation that explicitly varies the functional form that the probability of learning takes across an individual’s life-course. I show that the pace and extent of cultural change are both a function of the interplay between the shape of formative periods and the demographic composition of a population. The implication is that if we take seriously the idea that an agent’s probability of learning varies across the life-course, then understanding when and to what extent individuals are open to novel information becomes crucial for explaining how individual-level updating aggregates to large-scale change. Overall, this dissertation hopes to show the productive dialogue that can be established between research within the sociology of culture and growing interdisciplinary interest in culture, its properties and evolution.