Exploring the role of character traits in the iterated prisoner’s dilemma using case-based decision theory

The iterated prisoner’s dilemma (IPD) has become a popular tool to model human interaction in collective action problems. Since the prisoner’s dilemma tournaments conducted by Robert Axelrod in the 1980 s, several researchers have tried to model decision making and best strategies in this game. The literature largely finds that patience, repeated interaction and familiarity with other participants lead to higher rates of sustained cooperation. However, few papers have explored the role of common human character traits such as ambition, envy (wanting to emulate others) and jealousy (wanting others to lose what they have) on long-term decision making in these environments. In this paper, I use agent-based simulation techniques grounded in case-based decision theory (CBDT) introduced by Gilboa and Schmeidler (1995) that allow agents to make decisions based on past experiences to show that jealousy and ambition curtail sustained cooperation in the IPD, while envy promotes cooperation. I show that decision making for these different types of agents is sensitive to how they form their levels of aspiration (i.e., the amount they want to receive after an interaction). I then go on to show that ambitious and jealous agent behavior can affect non-ambitious and non-jealous behavior in mixed environments. Finally, I show that trends in cooperation in an environment where all these traits are randomly distributed align well with what is found in the literature. I hope conclusions in this paper create testable hypotheses for subsequent experimental research analyzing the role of human behavior and cooperation in the IPD.