Miller, I. D., MacInnis, C. C., & Page-Gould, E. (2017). Cross-group Interaction and Contact. Presented at the Dynamical Systems preconference data blitz at the Society for Personality and Social Psychology. San Antonio, TX, USA. doi:10.6084/m9.figshare.4960589
How can it be that repeated, negative interactions between cross-group members could reliably lead to less negative outcomes over time? There is an apparent contradiction between the intergroup interaction and intergroup contact literature that recent theoretical work has sought to reconcile (MacInnis & Page-Gould, 2015). Using an agent based modeling approach, we test a model based upon MacInnis and Page-Gould’s theoretical predictions by simulating repeated interactions between agents who, over time, become differentially biased on the basis of those interactions.
In this dynamical systems data blitz, I will present results from this study as well as methodological advances that made this computational simulation possible. In particular, I will highlight pplapi (pronounced “people API”), which is a synthetic world population for computational social simulation, n=7,171,922,938. Researchers can submit queries to pplapi using its Application Programming Interface (API) to obtain samples consisting of synthetic agents drawn from this population. Because researchers do not need to host the synthetic population themselves (which is many terabytes in size) pplapi reduces start-up costs associated with using ecologically valid synthetic agents in research. pplapi provides a canonical namespace for Agent Based Social Simulations, permitting comparisons between models implemented with disparate modeling frameworks and programming languages. pplapi is compatible with popular simulation and computation environments, including Netlogo, MASON, Common LISP, R, and others.