Models and Experiments in Risk and Rationality

Abstract

Decision making under conditions of risk or uncertainty was a popular subject among scientists in the 17th and 18th centuries. At that time, it was primarily developed in connection with parlor games, such as Daniel Bernoulli's famous analysis of the Saint Petersburg Paradox or Waldegrave's solution to de Monmort's game, although it was also applied to such far ranging issues as the sentencing or freeing of court defendants! and even to the very issue of religious faith, as with Pascal's famous wager2. Epistemic uncertainty, indeed, was studied earlier than economic risk, and it took a long time before the latter was recognized as being of a formally distinct nature from the former. When interest in the topic of risk resurfaced among scientists in the 20th century, due to the seminal work of von Neumann and Morgenstern, it initially took a rather different aspect, with an emphasis on economic and strategic behavior. Since then, it has taken on additional and more varied psychological, social and logical aspects, along with its original epistemic one. As a result of this enlarged scope, the field of decision making under risk and uncertainty has become much more complex : not only are the "rules" of social and economic situations more complicated (and ambiguous) than those of simple parlor games, but the underlying criterion of "rationality" comes off as more difficult to model in all but the simplest of situations, and even the notion of "risk" itself emerges as being more subtle than had been thought in the 1950' s. The papers in this volume focus on a selection of these modern- day issues.