An Evolutionary Analysis of Narrative Selection

Abstract: We study the performance of different methods for processing information, incorporating narrative selection within an evolutionary model. All agents update their beliefs according to Bayes' Rule, but some strategically choose the narrative they use in updating according to heterogeneous criteria. We simulate the endogenous composition of the population, considering different laws of motion for the underlying state of the world. We find that conformists -- that is, agents that choose the narrative to conform to the average belief in the population -- have an evolutionary advantage over other agents across all specifications. The survival chances of the remaining types depend on the uncertainty regarding the state of the world. Agents who tend to develop mild beliefs perform better when the uncertainty is high, whereas agents who tend to develop extreme beliefs perform better when the uncertainty is low.