Individual Variation and Population Structure in Complex Social Systems: Dynamics and Consequences

Abstract

From ant colonies to human societies, social groups exhibit remarkable collective behavior. Central to the functioning of these self-organizing systems are individual heterogeneity and population structure. This dissertation explores how the interplay between these features influences collective dynamics, and vice versa, in complex social systems. Chapter 1 investigates patterns of behavioral specialization in heterogeneous groups, using a dynamical model based on behavioral response thresholds. Testing the model predictions against experimental data from ant colonies reveals that the simplest form of response thresholds can capture the full range of observed social organization, but only if we consider variation in previously overlooked behavioral parameters. Chapter 2 probes the role of opinion diversity in the coupled dynamics of interindividual cooperation and political polarization. Using a cultural evolution model grounded in evolutionary game theory, we show that increasing interest diversity can improve individual and collective outcomes. But partisanship reduces the dimensionality of opinion space via self-sorting along party lines, potentially yielding greater in-group cooperation at the cost of heightened polarization—an emergent tension between the individual and the collective. Chapter 3 studies the consequences of stereotyping, or generalizing beliefs about social groups, for cooperation. Using a game-theoretic model of indirect reciprocity, we identify conditions under which stereotype use spreads via social imitation. While stereotyping behavior can boost cooperation in some scenarios, group structure in information availability gives rise to in-group favoritism, potentially resulting in an asymmetric improvement in cooperation levels, with individuals cooperating more on average but preferentially with their in-group. Chapter 4 examines mechanisms underlying the emergence of enduring hierarchies. Using an adaptive network model, we prove that feedback between social prestige and individual-level decision-making alone can lead to stratification among otherwise equal competitors. Fitting the model to empirical data, such as hiring patterns among mathematicians, reveals that observed social systems may be near the critical threshold between egalitarianism and hierarchy. Complex social systems—from social media platforms to democracies—shape how we organize our societies. A greater understanding of the connections among diversity, social structure, and collective dynamics may enable us to become better stewards of these systems.