Training, Cognitive Control, and Racial Bias in the Decision to Shoot

本研究探讨了警察射击决策中的种族偏见，重点是训练的影响和训练刺激的复杂性的重要性。 种族偏见影响警官射杀嫌疑犯的意愿和准备。 调查人员使用视频游戏来检查黑人和白色嫌疑人在开枪/不开枪决定中的种族偏见，这些嫌疑人要么是武装的，要么是非武装的。 实验范式涉及到在几个复杂的背景下，以各种姿势站立或蹲伏的人的图像的呈现。参与者必须识别目标手中的一个小物体。这些演习的结果表明，参与者更快地响应目标符合定型观念（武装的黑人，手无寸铁的白人）比反定型目标（武装的白人，手无寸铁的黑人）。此外，这些结果表明，参与者为射击黑色而不是白色目标的决定设置了更宽松的阈值。也就是说，当目标是黑色时，他们更愿意开火。重要的是，PI最近的工作表明，训练最大限度地减少了信号检测阈值中的种族偏见，以至于新手表现出明显的偏见，但训练有素的人（警察“专家”参与者）不会。 有趣的是，训练不会影响反应时间的偏差，因为专家和新手对反刻板印象的反应都比刻板印象的反应慢。因此，专家们似乎显示出一些基于定型观念的处理证据（这延迟了他们对反定型目标的反应），但他们的培训使他们能够克服这些定型观念，最终执行不受种族影响的决定。 本研究旨在研究培训可以减少偏见的过程。 PI认为，专家学会从视野中提取与任务相关的信息，并使用它来指导他们的决策。 研究计划的第一部分测试培训是否可以促进注意力和努力。在第二部分，培训（和它促进的控制）是否可以超越种族陈规定型观念进行审查。最后一组研究试图通过认知负荷、疲劳和恐惧来损害认知控制。通过损害培训运作的机制，这些操纵甚至会加剧专家之间的偏见。 调查结果应提供证据，说明在何种情况下警官的能力最有可能受到损害。 在这项研究中考虑的问题是有趣的和重要的，研究结果可能会在现实世界中产生重大影响。 连同明显的教育影响，这项工作具有重要的意义，在各个领域的培训计划。

This research explores racial bias in police officers' decisions to shoot by focusing on the impact of training and the importance of complexity of training stimuli. Racial biases influence police officers' willingness and readiness to shoot suspects. The investigators employ a videogame to examine racial bias in shoot/don't-shoot decisions with Black and White suspects who are either armed or unarmed. The experimental paradigm involves the presentation of images of men standing or crouching in various positions against several complicated backgrounds. Participants must identify a small object in the target's hand. Results of these exercises show that participants respond more quickly to targets consistent with stereotypes (armed Blacks, unarmed Whites) than to counter-stereotypic targets (armed Whites, and unarmed Blacks). In addition, these results show that participants set a more lenient threshold for the decision to shoot Black rather than White targets. That is, they are more willing to open fire when the target is Black. Importantly, the PIs' recent work shows that training minimizes racial bias in the Signal Detection threshold such that novices show pronounced bias, but highly trained individuals (police & "expert" participants) do not. Interestingly, training does not affect bias in reaction times as experts and novices both respond more slowly to counter-stereotypic than to stereotypic targets. It therefore seems that experts show some evidence of stereotype-based processing (which delays their responses to counter-stereotypic targets), but that their training enables them to override these stereotypes and, ultimately, execute decisions that are not affected by race. This research seeks to examine the processes through which training may reduce bias. The PIs contend that experts learn to extract task-relevant information from the visual field and use it to guide their decisions. The first section of the research program tests whether training can promote attention and effort. In the second section, whether training (and the control it fosters) can override racial stereotypes is examined. The final set of studies attempts to impair cognitive control through cognitive load, fatigue and fear. By compromising the mechanisms through which training operates, these manipulations should exacerbate bias even among experts. Findings should provide evidence of the conditions under which police officers' capacities are most likely to be impaired. The issues under consideration in this research are interesting and important and the findings are likely to have a significant impact in real-world settings. Together with the obvious educational impacts, the work has important implications for training programs in various domains.