Abstract

Humans and other animals need to make inferences about their environment under constraints of limited time, knowledge, and computational capacities. However, most theories of inductive inferences model the human mind as a supercomputer like a Laplacean demon, equipped with unlimited time, knowledge, and computational capacities. In this article I review models of fast and frugal inference, that is, satisfi cing strategies whose task is to infer unknown states of the world (without relying on computationaly expensive procedures such as multiple regression). Fast and frugal inference is a form of bounded rationality (Simon, 1982). I begin by explaining what bounded rationality in human inference is not. 1. Bounded Rationality is Not Irrationality In his chapter in John Kagel and Alvin Roth’s Handbook of Experimental Economics (1995), Colin Camerer explains that “most research on individual decision making has taken normative theories of judgment and choice (typically probability rules and utility theories) as null hypotheses about behavior, ” and has labeled systematic deviations from these norms “cognitive illusions ” (p. 588). Camerer continues, “The most fruitful, popular alternative theories spring from the idea that limits on computational ability force people to use simplifi ed procedures or ‘heuristics ’ that

Citations