Judging guilt and accuracy: highly confident eyewitnesses are discounted when they provide featural justifications
Judging guilt and accuracy: highly confident eyewitnesses are discounted when they provide featural justifications
复制标题
DOI：
10.1080/1068316x.2017.1284220
复制
发表时间：
2017-06-01
影响因子：
1.4
通讯作者：
Dobolyi, David G.
中科院分区：
文献类型：
--
作者：
Dodson, Chad S.;Dobolyi, David G.
陪审员很容易受到自信的目击者的影响。他们是否也受到目击者如何证明他们的信心水平的影响？在这里，我们记录了一个反直觉的效应：当目击者绝对肯定地从一组人中认出一个嫌疑人（“我完全有信心”），并通过提到被告的一个明显特征（“我记得他的鼻子”）来证明他们的信心时，参与者认为嫌疑人有罪的可能性低于目击者绝对肯定地认出一个嫌疑人，但提供了一个不可观察的理由（“我永远不会忘记他”）或根本没有理由。此外，当目击者提供了一个特征证明时，人们认为目击者的识别准确率比目击者提供了一个不可观察的证明或根本没有证明要低近25%。即使目击者的信心水平是明确的，因为他/她已经用数字表达了它（例如，我100%确定），当目击者提供了一个特征的理由时，参与者也会认为目击者不可信（即不准确）。然而，特征证明的效果——仅相对于自信陈述而言——在有一组面孔的情况下是最大的，在有一张面孔的情况下是中等的，在没有面孔的情况下是最小的。结果支持我们的感知诊断说。
Jurors are heavily swayed by confident eyewitnesses. Are they also influenced by how eyewitnesses justify their level of confidence? Here we document a counter-intuitive effect: when eyewitnesses identified a suspect from a lineup with absolute certainty (I am completely confident') and justified their confidence by referring to a visible feature of the accused (I remember his nose'), participants judged the suspect as less likely to be guilty than when eyewitnesses identified a suspect with absolute certainty but offered an unobservable justification (I would never forget him') or no justification at all. Moreover, people perceive an eyewitness's identification as nearly 25% less accurate when the eyewitness has provided a featural justification than an unobservable justification or simply no justification. Even when an eyewitness's level of confidence is clear because s/he has expressed it numerically (e.g. I am 100% certain') participants perceive eyewitnesses as not credible (i.e. inaccurate) when the eyewitness has provided a featural justification. However, the effect of featural justifications - relative to a confidence statement only - is maximal when there is an accompanying lineup of faces, moderate when there is a single face and minimal when there is no face at all. The results support our Perceived-Diagnosticity account.