When Ecphory Fails: Secondary Processes in Eyewitness Identification

目击者在警察列队中辨认错误是无辜者被定罪的主要原因。心理科学家对阵容的改进使法律的系统受益。但是，进一步的改进需要更好地理解次级过程所起的作用。与初级加工（识别记忆）不同，次级加工是缓慢的、费力的、深思熟虑的、可言语化的，而且对罪犯是否在队列中基本上不敏感。这些次级过程包括推论、演绎、排除过程、相对判断和其他形式的推理。假设次要过程参与时，主要过程失败，并可能导致灾难性的肯定识别决策。六个实验（总N=2400）进行，人们认为一个模拟的犯罪，并试图从一个罪犯存在或罪犯删除阵容识别。每个实验都经过精心设计，在一半的条件下诱导识别失败，以研究次级过程的参与。两个实验在记忆获得的水平上操纵识别经验的失败（例如，给予他们中的一些人对所见证的事件的差的看法），两个实验在记忆存储的水平上操纵识别体验的失败（例如，操纵观看和列队之间的时间流逝），并且两个实验在记忆检索的水平上操纵识别体验（例如，粒状的、低分辨率的排列任务）。在排除罪犯的阵容条件下，证人将身份证明转移到填充物（而不是不进行身份证明）的趋势预计在诱导识别失败的条件下最大。此外，参与证人的言语进行了研究，以了解更多关于映射到这些识别错误的二级过程的性质。这些发现可以帮助我们开发更好的列队程序，并在发生错误识别时更好地诊断。

Mistaken eyewitness identification from police lineups is the primary cause of the conviction of innocent people. Improvements to lineups developed by psychological scientists have benefitted the legal system. But, further improvements require a better understanding of the role played by secondary processes. Unlike the primary process (recognition memory) secondary processes are slow, effortful, deliberative, verbalizable, and largely insensitive to whether or not the culprit is in the lineup. These secondary processes include inferences, deductions, elimination processes, relative judgments, and other forms of reasoning. Secondary processes are hypothesized to be engaged when the primary process fails and can lead to disastrous affirmative identification decisions. Six experiments (total N=2400) are conducted in which people view a simulated crime and attempt an identification from either a culprit-present or a culprit-removed lineup. Each experiment is carefully designed to induce a failure of recognition in one-half of the conditions so as to study the engagement of secondary processes. Two experiments manipulate failure of the recognition experience at the level of memory acquisition (e.g., giving some of them a poor view of the witnessed event), two experiments manipulate failure of the recognition experience at the level of memory storage (e.g., manipulating the passage of time between viewing and the lineup), and two experiments manipulate the recognition experience at the level of memory retrieval (e.g., a grainy, low-resolution lineup task). The tendency of witnesses in culprit-removed lineup conditions to shift identifications to fillers (rather than making no identification) is predicted to be greatest in the conditions that induce recognition failure. In addition, the verbalizations of participant witnesses are studied to learn more about the nature of the secondary processes that map into these identification errors. The findings can help us develop better lineup procedures and better diagnose when mistaken identifications occur.