Hanfstingl Barbara, Oberleiter Sandra, Pietschnig Jakob, Tran Ulrich S, Voracek Martin
Department of Psychology, University of Klagenfurt, Klagenfurt, Austria.
Department of Developmental and Educational Psychology, University of Vienna, Vienna, Austria.
Front Psychol. 2024 Aug 30;15:1404060. doi: 10.3389/fpsyg.2024.1404060. eCollection 2024.
Over the past few years, more attention has been paid to jingle and jangle fallacies in psychological science. Jingle fallacies arise when two or more distinct psychological phenomena are erroneously labeled with the same term, while jangle fallacies occur when different terms are used to describe the same phenomenon. Jingle and jangle fallacies emerge due to the vague linkage between psychological theories and their practical implementation in empirical studies, compounded by variations in study designs, methodologies, and applying different statistical procedures' algorithms. Despite progress in organizing scientific findings via systematic reviews and meta-analyses, effective strategies to prevent these fallacies are still lacking. This paper explores the integration of several approaches with the potential to identify and mitigate jingle and jangle fallacies within psychological science. Essentially, organizing studies according to their specifications, which include theoretical background, methods, study designs, and results, alongside a combinatorial algorithm and flexible inclusion criteria, may indeed represent a feasible approach. A jingle-fallacy detector arises when identical specifications lead to disparate outcomes, whereas jangle-fallacy indicators could operate on the premise that varying specifications consistently yield overrandomly similar results. We discuss the role of advanced computational technologies, such as Natural Language Processing (NLP), in identifying these fallacies. In conclusion, addressing jingle and jangle fallacies requires a comprehensive approach that considers all levels and phases of psychological science.
在过去几年里，心理科学领域对“叮当”谬误和“混淆”谬误给予了更多关注。当两种或更多不同的心理现象被错误地用同一个术语标记时，就会出现“叮当”谬误；而当用不同的术语来描述同一现象时，则会出现“混淆”谬误。“叮当”谬误和“混淆”谬误的出现，是由于心理学理论与其在实证研究中的实际应用之间存在模糊的联系，再加上研究设计、方法以及应用不同统计程序算法的差异。尽管通过系统评价和荟萃分析在整理科学发现方面取得了进展，但仍缺乏预防这些谬误的有效策略。本文探讨了几种方法的整合，这些方法有可能识别和减轻心理科学中的“叮当”谬误和“混淆”谬误。从本质上讲，根据研究的具体情况进行组织，包括理论背景、方法、研究设计和结果，再结合组合算法和灵活的纳入标准，可能确实是一种可行的方法。当相同的具体情况导致不同的结果时，就会出现“叮当”谬误探测器，而“混淆”谬误指标可以在不同的具体情况始终产生过度随机相似结果的前提下发挥作用。我们讨论了先进的计算技术，如自然语言处理（NLP），在识别这些谬误中的作用。总之，解决“叮当”谬误和“混淆”谬误需要一种全面的方法，该方法要考虑心理科学的所有层面和阶段。
