Reasoning with the THOG problem: a forty-year retrospective

Abstract

Being able to create new information from already existing premises is the essence of human reasoning. This paper focuses on one of the most important experimental tasks that have been used to study how people make inferences: the THOG problem (Wason, 1977, 1978; Wason & Brooks, 1979). It is a hypothetico-deductive reasoning problem in which subjects must formulate and test hypotheses from the comprehension of an exclusive disjunctive statement. Research on this task has shown that it is a difficult problem to solve and few people reach the logically correct answer. This paper presents some of the main theoretical explanations about people’s inferences with this task. From a general perspective, the Dual Process and the Hypothetical Thinking Theories and the Mental Models Theory are found. Some of the more specific proposals have focused on analysing the underlying mechanisms of the cognitive biases such as the Confusion Theory or the Non-Consequential Thinking. Moreover, a review of the empirical investigations on this meta-inference task is presented. Finally, some research on the THOG problem that provides important new clues on broader topics in the study of human reasoning is analyzed