Wojciechowski Bartosz W, Pothos Emmanuel M
Department of Clinical and Forensic Psychology, Institute of Psychology, University of Silesia of Katowice, Katowice, Poland.
Department of Psychology, City, University of London, London, United Kingdom.
Front Psychol. 2018 Apr 5;9:391. doi: 10.3389/fpsyg.2018.00391. eCollection 2018.
Classical probability theory (CPT) has represented the rational standard for decision making in human cognition. Even though CPT has provided many descriptively excellent decision models, there have also been some empirical results persistently problematic for CPT accounts. The tension between the normative prescription of CPT and human behavior is particularly acute in cases where we have higher expectations for rational decisions. One such case concerns legal decision making from legal experts, such as attorneys and prosecutors and, more so, judges. In the present research we explore one of the most influential CPT decision fallacies, the conjunction fallacy (CF), in a legal decision making task, involving assessing evidence that the same suspect had committed two separate crimes. The information for the two crimes was presented consecutively. Each participant was asked to provide individual ratings for the two crimes in some cases and conjunctive probability rating for both crimes in other cases, after all information had been presented. Overall, 360 probability ratings for guilt were collected from 120 participants, comprised of 40 judges, 40 attorneys and prosecutors, and 40 individuals without legal education. Our results provide evidence for a double conjunction fallacy (in this case, a higher probability of committing both crimes than the probability of committing either crime individually), in the group of individuals without legal education. These results are discussed in terms of their applied implications and in relation to a recent framework for understanding such results, quantum probability theory (QPT).
经典概率论(CPT)一直是人类认知中决策的理性标准。尽管CPT提供了许多在描述上非常出色的决策模型,但也有一些实证结果一直给CPT的解释带来问题。在我们对理性决策有更高期望的情况下,CPT的规范规定与人类行为之间的矛盾尤为尖锐。一个这样的例子涉及法律专家(如律师、检察官,尤其是法官)的法律决策。在本研究中,我们在一项法律决策任务中探讨了最具影响力的CPT决策谬误之一,即合取谬误(CF),该任务涉及评估同一嫌疑人犯下两项不同罪行的证据。两项罪行的信息是连续呈现的。在呈现所有信息后,要求每位参与者在某些情况下对两项罪行分别给出个人评分,在其他情况下对两项罪行给出合取概率评分。总体而言,从120名参与者那里收集了360个有罪概率评分,其中包括40名法官、40名律师和检察官以及40名没有法律专业教育背景的人。我们的结果为没有法律专业教育背景的人群中存在双重合取谬误(在这种情况下,犯下两项罪行的概率高于单独犯下任何一项罪行的概率)提供了证据。我们将根据这些结果的应用意义以及与理解此类结果的最新框架——量子概率论(QPT)的关系来讨论这些结果。