Logical intuitions or matching heuristic? Examining the effect of deduction training on belief-based reasoning judgments

crossref(2022)

引用 0|浏览0
暂无评分
摘要
Are people able to successfully evaluate logic intuitively, without engaging in deliberative thinking? Recent research has revealed that when individuals are asked to simply evaluate the believability of deductively valid or invalid arguments, they often endorse valid arguments at higher rates than invalid ones. This effect of validity on belief judgments – the “logic-belief effect” – is considered to be evidence of intuitive logic or logical intuitions. However, a recent study challenged the intuitive logic interpretation by demonstrating that ‘pseudo-logical’ arguments, such as affirmation of the consequent, where there is a compelling but invalid inference invited, interfere with contrasting belief judgments in the same way as valid logical structures do, and hence, the logic-belief effect may have nothing to do with evaluating logic itself. Instead, this finding suggests that a simple heuristic that relies on the matching of constituent propositions leads to such interference. To further test this matching heuristic account, across two experiments (N = 332), we examined whether training in logic would selectively reduce the endorsement of invalid arguments under logic instructions whilst maintaining the impact of invalid inferences on belief judgments, where a matching cue could drive responses. We instructed participants to evaluate the conclusions of a series of logical and pseudo-logical arguments based on logic or belief, before and after a logic training block. The results showed that whilst both judgment types were impacted by both logical and pseudo-logical structures before training, after training the effect of the latter was indeed minimized in logic judgments but not belief judgments. The results largely support the matching heuristic account, whereby simple considerations of proposition-matching rather than logic assessment underly previous evidence for intuitive logic.
更多
查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要