Abstract

Cognitive theories of human reasoning can benefit by distinguishing logics from implementations. An semantic analysis of the syllogism reveals what is common between apparently unrelated methods of reasoning and how such an analysis can guide experimental investigation. Existing accounts of syllogistic reasoning oppose rule-based linguistic to model-based methods. Stenning & Oberlander (1995) show that the latter are isomorphic to wellknown graphical methods correctly interpreted. We here extend these results by showing that equivalent sentential implementations exist, thus revealing that all these approaches are members of a family of abstract individual identification algorithms (IIAs) variously implemented in diagrams or sentences. We investigate whether subjects employ the IIA through an experiment using a novel reasoning task. Modelling of this data shows that behaviour is best explained at the abstract level of the IIA with consequences for the semantic and the empiri...

Citations