This is a collection of essays on reasoning – deductive, inductive, abductive, belief-revision, defeasible, cross-cultural, conversational, argumentative – oriented toward contemporary empirical research. The study of reasoning is, of course, a major focus of investigation in both philosophy and psychology, and this volume attempts to combine these perspectives. Philosophy, especially logic and philosophy of science, inspired psychology's interest in deduction, induction, and other forms of inference. In the past thirty years, the focus of common interest has greatly enlarged to include causal reasoning, cross-cultural comparisons, moral reasoning, and argumentation. There is increasing attention to reasoning as a response to new information or change of belief (belief-revision or defeasible reasoning), which is a far-reaching concern in epistemology. These advances, along with the more than twenty-year debate on rationality, have made reasoning a focal point of cognitive science research and reflections.

Three additional paths also are now converging on reasoning as major research interests. The first is a focus on fallacious reasoning and its reduction through education, which usually goes under the banner of informal logic or critical thinking. Second, the development of pragmatics, particularly following Grice's foundational contribution, has given rise to fruitful applications to traditional reasoning tasks. A number of commentators in the rationality debate have suggested connections between performance on those tasks and conversational expectations. These suggestions are coming in for experimental testing and some corroboration.

The idea that formal logic bears a close relationship to human reasoning is extremely controversial within cognitive science. For example, Wason and Johnson-Laird (1972: 245) concluded their influential study of reasoning by stating that “only gradually did we realize first that there was no existing formal calculus which correctly modeled our subjects' inferences, and second that no purely formal system would succeed.” This kind of opposition is based on evidence that subjects' judgments about an argument sometimes depart from the answer that the experimenter derived by translating the argument into some system of logic and assessing its correctness within that system. The strength of this evidence, however, clearly depends on the system the investigator uses. If the investigator's conception is too narrow, what is classified as nonlogical behavior may turn out logical after all. To evaluate the evidence, we need some clear idea of what logic has to offer.

Most cognitive theories that contain a logic-like component (e.g., Braine 1978; Johnson-Laird 1975; Osherson 1974, 1975, 1976; Rips 1983) are based on the notion of proof, particularly the “natural deduction” proofs originally devised by Gentzen (1935/1969) and Jaskowski (1934). The basic notion is familiar to anyone who has taken a high school mathematics course: If you want to know whether a particular argument is deductively correct, you can find out by taking its premises as given and then trying to derive its conclusion by applying a specified set of rules.