ABSTRACT
This chapter proposes a method for modeling economic systems in which outcomes depend locally on the predictions that agents make of other agents. It develops population games in which each agent adaptively searches for a good model of its environment. The chapter demonstrates that such systems can exhibit persistent dynamics, cyclic oscillations between low and high complexity states, and other complex, yet endogenous, phenomena. It proposes these "adaptively rational" agents as a natural extension of rational expectations, suitable when mutual consistency is not realistic. The chapter discusses the connections between work and the programs of bounded rationality, evolutionary game theory, and models motivated by statistical mechanics. It attempts to formulate a categorization of the natural rationality of agents whose mutual interactions form their world. The chapter compares the results of its analysis and experimentation with the intuitive ideas introduced earlier.
