Young Researchers Workshop: Kinetic descriptions in theory and applications

Social interactions can drive emergent behavioral diversity and modular social network structure

Christopher Tokita

Princeton University


Complex social systems, including both animal and human societies, can often be described by three key properties: group size, the diversity of individual behavior, and the pattern of interactions among individuals. These properties are of interest not only because they are good descriptors of a society, but also because they can impact other emergent group phenomena, like collective decision-making and information spread. Yet, theoretical explorations of self-organization and emergence in social systems often have a limited scope, focusing on one or two of these system-level descriptors. Division of labor (DOL)—when individuals specialize in certain tasks—is an emergent phenomenon of social systems that requires behavioral diversity, while also being recently shown to have interplay with both group size effects and the structuring of interactions among individuals. Therefore, it is well-suited to be used as a lens to investigate the interplay of these three key properties of social systems. We use individual-based models to examine how social interactions among individuals affect the self-organization of DOL and social network structure. In our model, individuals can influence one another, making others more likely to perform the same task. Additionally, individuals have biased interactions, tending to interact more frequently with individuals performing the same behavior. The combination of these two elements—social influence and biased interactions—results in the non-linear emergence of DOL with increasing group size. Additionally, diversified groups also exhibit the beginnings polarized social networks, where individuals tightly cluster with similar individuals. The findings from this model have broader implications for the underpinnings of societal organization.