A research workshop at Mannheim University, Germany, brought together a small group of Europe-based experts in system dynamics to discuss basic assumptions about rationality in system dynamics modeling and simulation. As an underlying theme it was found that-although system dynamics is open for various notions of rationality-policies are mostly modeled to be bounded rational and, in this sense, system dynamics models are valid descriptions of real-world situations. By reviewing key points emerging during the debates against the relevant literature, this article identifies a number of important further research issues. Copyright © 2004 John Wiley & Sons, Ltd.Syst. Dyn. Rev. 20, 75-87, (2004) Common use and English dictionaries define rational as reasonable or sensible. A rational decision is understood to be based on somehow intelligent thinking and not purely on emotions. However, in many sciences, particularly in economics, the notion of rationality is a more specific one. In the classical economic view, striving for optimality is a sign of being rational, which is considered to be the "normal" condition of human beings: in this economic sense, men are supposed to exclusively consider their utility (von Neumann and Morgenstern 1944). The basic idea of this theory is that individuals make rational decisions and choose those alternatives of action that maximize their expected utility, giving it the name "theory of expected utility". Under conditions of perfect information and certainty, homo oeconomicus or Economic Man tries to achieve his goals in an optimal manner, only limited by the restrictions of the environment. These goals (the most commonly used is profit maximization) are stable, coherent and explicit. Within an environment of stable goals and perfect foresight, for instance, optimal production programs can be deduced from demand, cost and production functions (see, for instance, Pindyck and Rubinfeld 1995). If the underlying assumptions are accepted, this orthodox view of economics offers great normative power and explanation. Together with the interlinked assumption of perfect competition, perfect rationality as described in this paragraph builds the basis for classical economic theorizing.In reality, individuals usually do not make (and cannot make) rational decisions in the sense of utility maximization-except in artificially simplified settings (Simon 1959). In real-world contexts, people try to achieve some,