Finding a face in a crowd is a real-world analog to visual search, but extending the visual search method to such complex social stimuli is rife with potential pitfalls. We need look no further than the well-cited notion that angry faces “pop out” of crowds to find evidence that stimulus confounds can lead to incorrect inferences. Indeed, long before the recent replication crisis in social psychology, stimulus confounds led to repeated demonstrations of spurious effects that were misattributed to adaptive cognitive design. We will first discuss how researchers refuted these errors with systematic “face in the crowd” experiments. We will then contend that these more careful studies revealed something that may actually be adaptive, but at the level of the signal: Happy facial expressions seem designed to be detected efficiently. We will close by suggesting that participant-level manipulations can be leveraged to reveal strategic shifts in performance in the visual search for complex stimuli such as faces. Because stimulus-level effects are held constant across such manipulations, the technique affords strong inferences about the psychological underpinnings of searching for a face in the crowd.

中文翻译：

---

在人群中寻找面孔：陷阱和未开发的可能性。

在人群中找到面孔是视觉搜索的现实世界模拟，但是将视觉搜索方法扩展到如此复杂的社交刺激中却充满了潜在的陷阱。我们只需要提一个引人入胜的观点，即愤怒的面孔“冒出”人群，以寻找刺激混杂可能导致错误推论的证据就可以了。的确，在最近的社会心理学复制危机爆发之前，刺激的混乱导致了反复出现的虚假效应，这些虚假效应被误认为是适应性认知设计。我们将首先讨论研究人员如何通过系统的“面对人群”实验来驳斥这些错误。然后我们将争辩说，这些更仔细的研究揭示了实际上可能是自适应的东西，但是在信号的层面上：快乐的面部表情似乎旨在被有效地检测到。最后，我们建议可以利用参与者级别的操作来揭示视觉搜索复杂刺激（例如脸部）的性能的战略性转变。由于在这些操作中刺激级别的效果保持恒定，因此该技术可以强烈推断出在人群中寻找面孔的心理基础。