Effects of Diversity on Social Dynamics: Theoretical and Experimental Approaches

Abstract: Diversity of individuals is usually discussed and advocated from a demographic point of view. However, *functional* diversity, such as diversity in behavior, background or expertise, would also play an essential role in collective social dynamics. In this talk, we present an overview of our two recent studies on the effects of functional diversity on opinion dynamics and social evolution. The first study investigated how behavioral diversity of individuals might affect informational diversity and connectivity within society, using numerical simulations of adaptive social network models. Results showed that heterogeneity in individual behavioral traits significantly increased both opinion diversity and social network connectivity. This observation was robustly obtained in several distinct models. The second study examined the effects of diversity in individual backgrounds on collaborative design and innovation, using human-subject experiments and machine-learning-based text analysis. We found that spatially clustered collectives with clustered background distributions tended to explore more diverse ideas than in other conditions, whereas collectives with random background distributions consistently generated ideas with the highest utility. Through these interrelated studies, we illustrate how different forms of diversity of social constituents can have different, nontrivial implications for collective social dynamics.