Personality and social networks: A generative model approach

Social network analysis has produced important insights regarding the causes and consequences of animal social structure. Social structure has been shown to impact longevity, reproductive success, transmission of pathogens and information, and also play important role in the evolution of cooperation. Studies of the determinants of social structure have identified environmental, genetic, and structural factors in a variety of species. At the same time, most studies in the field have been descriptive in approach, statistically identifying patterns in social networks constructed from observed interactions. We argue that there is a need for predictive theory to complement descriptive studies, moving the field from pattern to process. As an example, we provide a simple model of the effect of personality on social network structure and social role differentiation. Our model suggests that variation in behavioral types can result in variation in individual social network traits, and that some patterns found in animal networks in the wild, such as assortativity with respect to personality, may be outcomes of social inheritance and individual variation in it. Our approach and results exemplify the potential of generative models to connect individual-level processes to emergent patterns and advance our understanding of social complexity in nature.