Towards an integrative theory of conditional reasoning

点击翻译按钮获取中文摘要

Conditional reasoning involves making deductive inferences based on if-then premises. It is a key component of advanced scientific and mathematical reasoning, and understanding its nature and developmental course is an important problem in cognitive psychology, and of real educational importance. The proposed program of research is an extension of an on-going program that has used a modified form of mental model theory as a base. According to this, when people reason, they start by representing the premises using simple mental tokens. Conclusions are verified by a process of counterexample generation; if a counterexample is produced, the conclusion is rejected. Previous results show that this process goes through a sequence from more concrete to more abstract forms, gradually allowing the acquisition of more complete logical reasoning abilities. Other results have shown that people can sometimes use a probabilistic mode of making deductive inferences and that they are often unaware of how logical their reasoning is. The proposed studies will extend these previous results in order to examine developmental patterns in (1) use of a counterexample and a probabilistic mode of inference and (2) the way that the development of reasoning is related to people’s metacognitive awareness of their own reasoning processes. Other studies will examine the way that counterexample and probabilistic ways of reasoning might affect the kinds of inferences that people make when heuristics, such as belief bias, are involved. Finally, the basic theoretical approach that is proposed here suggests that training in divergent thinking might be a practical way to improve people’s logical reasoning, something that will be examined in a series of training studies.