Psychological theories of reasoning have dramatically changed during the last 20 years. Cognitive psychologists once believed that reasoning consisted of the application of context-free rules to content-free symbolic expressions. Syllogisms constituted one of the paradigmatic examples of such reasoning. Researchers investigated the extent to whch the solutions to syllogisms conformed to truth tables, propositional calculus, and predicate calculus. It did not matter what the symbolic expressions meant or what they referred to because the essence of reasoning lay in the proper application of rules that were simple, well-formed, and universally valid. Very few of today's cognitive psychologists would admit to viewing human reasoning in this limited light. Instead, we believe that human reasoning is substantially constrained by the meaning of symbolic expressions and by the entities in the world that these expressions refer to. The application of rules is sensitive to the type of world knowledge that is involved in a problem. Consequently, patterns of reasoning are quite different in the domains of history, law, science and mathematics. Reasoning differs for knowledge that involves causality, spatiality, goals of agents, concept taxonomies, personality attributes, emotions and so on.
Perhaps a Pandora's Box was opened when researchers accepted the view of reasoning that embraced content, context and world knowledge. It produced an explosion of radically different psychological models of reasoning-so many models that a psychological study of reasoning may have lost a coherent shape and definition. Alternatively, perhaps psychologists became emancipated from the systems (or possibly artifacts) of well-formed reasoning that were developed by philosophers, logicians and mathematicians. That is, psychologists may have taken a giant step closer to understanding the normal patterns of human reasoning in naturalistic situations.
The contributors to this special issue represent most of the recent models of human reasoning that have emerged in the cognitive sciences. They collectively explore how reasoning is influenced by domain-specific knowledge, coherent mental models, schemata, specific cases, ill-defined heuristics, discourse, social interaction, context and situated practice. The empirical studies that these researchers report are hardly limited to toy problems in the laboratory. Readers of this special issue will learn about applications of this research to education, the design of computer software, lifespan development, medicine, politics and consumer behaviour.