Reasoning to and from belief: Deduction and induction are still distinct Jonathan St. B T. Evans and David E. Over

In this paper we address a question raised in an earlier discussion of the new paradigm in psychology of reasoning (Evans, 2012). Is there any longer a clear distinction between the psychology of inductive and deductive reasoning? The question arises because it is apparent that developments in the past 10-20 years within the psychology of reasoning have shifted interest away from assumed truth to belief: many contemporary studies focus on the

drawing of conclusions held with degrees of confidence from actual beliefs, rather than deductions from arbitrary assumptions. There has also been increasing interest in pragmatic effects and the influence of context. Correspondingly there has been a shift in the use of normative and computational models based on extensional bivalent logic, dealing only in truth and falsity, to that of Bayesianism, dealing in degrees of belief. Since the study of inductive inference is about uncertainty and the belief system, the two enterprises may now appear very similar.