If cognitive development is characterized by a high degree of domainspecificity, so too is the study of cognitive development. Research is frequently conducted in microworlds which are relatively insulated from one another. These research microworlds are, in turn, often focused around a single experimental ‘paradigm task’. This is certainly true of the research to be discussed in this chapter. At the broadest level we are concerned with the nature and development of human reasoning. More specifically, we are concerned with hypothetico-deductive reasoning, and more specifically again, with the use of deductive (and metadeductive) inferences to evaluate the truth status of conditional rules. One experimental task has dominated work in this field for the last quarter of a century: the Wason (1966) selection task.