Code from: Social learning within groups: Hypernetworks capture effects of within-group dynamics on social contagions

Group size and composition often modulate the outcomes of social processes, including the diffusion of knowledge and behavioral innovations. Incorporating such information within conventional social network analyses is challenging, however, due to their focus on dyadic relationships. Higher-order network approaches such as hypernetworks offer a potential solution by enabling group size and structure to be explicitly incorporated within descriptions of social structure. However, the extent to which this information, relative to dyadic approximations, substantially alters expected outcomes when modeling social processes has not been explored. We develop an agent-based model that simulates the socially transmitted spread of a behavioral innovation. To perform the innovation, knowledgeable individuals must successfully compete over access to limited resources, with competitive interactions modulated by social dominance rank. In our model, we independently vary resource scarcity, dominance skew, individuals’ average number of associates, and whether diffusions are modeled using a dyadic or hypernetwork representation. The R scripts contained here provide functions for running the simulations described above, including user-designated inputs for resource availability, social dominance distribution, and interaction radii. In addition, scripts are included that reproduce the figures in the manuscript that illustrate the relationship between model parameters and a range of diffusion outcomes. No new data were generated as part of this project. Users are free to download, modify, and/or reuse the provided scripts.