The Impact of Dynamic Network Structure and Homogeneous Embeddedness on Cultural Polarization

• In Event: 352 - Theory Section Roundtables (one-hour)
  In Refereed Roundtable: Table 05. Knowledge and Information

Abstract

In this study, we use a computer simulation model to investigate the process of cultural polarization in modern society. Numerous studies have used computer simulation models based on the principle of homophily to explain the process of cultural or political polarization and convergence. Most of these assume (1) network structure is fixed and (2) an actor’s ecology, in terms of possible interactions with others, is randomly decided. In contrast to (1), we assume network structure to be dynamic such that it consists in part of a non-trivial amount of non-structuralized, i.e., temporary, weak ties. We take this to be a reasonable assumption because the development of institutional certainty, improved levels of generalized trust, and the development of online communities have made it possible for actors to interact with unanticipated actors without the complicated process of building mutual trust. In contrast to (2), we assume that actors are initially embedded in homophily groups. This is because induced homophily (at the macro level) and repeated choice homophily processes (at the micro level) mean that an actor’s close ties are likely to consist of similar actors. We investigate the effects of these two assumptions by modifying the simulation model developed by Flache and Macy (2011). Our first two experiments show that both dynamic network structures and homogeneous embeddedness have striking effects on cultural polarization. Our second set of experiments investigate reasons for the findings in the first experiments.

Authors

• Yunsub Lee, University of North Carolina-Charlotte

• Joseph M. Whitmeyer, University of North Carolina-Charlotte