Simulating Agency Detection Using Agent-Based Modeling in Python

With this research project we conceptualized and implemented an agent-based model that abstracts the development of religious beliefs, conducted simulations employing this model, and analyzed the collected results. Using established cognitive theories of anthropomorphism and agency detection in the context of religion as our theoretical base, the resulting model was expected to allow us comparing ‘predictive coding’ theory against ‘hyperactive agent detection device’ theory for their explanatory power regarding agency detection. Methodological shortcomings of the study did not allow us to draw revelatory conclusions, but our results do suggest that agent-based modeling is a useful tool for the cognitive science of religion, allowing computer simulations of individual cognitive processes to explore emergent religious phenomena on the group level.