A Note on Exploring Rationality in Games

The received concept of strategic (game-theoretic) rationality is attended by a formidable list of paradoxes, anomalies, and empirical failures. This paper reviews three well-known and problematic decision contexts and diagnoses as a common source of difficulty the failure in the received view of rationality to adequately recognize risk-return tradeoffs. This diagnosis is then supported by computational experiments and analysis that demonstrate the effectiveness in strategic contexts of specific and plausible forms of decision making that do better justice to risk-return tradeoffs. The paper suggests that this kind of rationality, called an exploring rationality, be considered as an alternative to the received view.