Department of Psychology.
J Exp Psychol Learn Mem Cogn. 2019 Sep;45(9):1583-1598. doi: 10.1037/xlm0000665. Epub 2018 Oct 8.
It is difficult to recognize the identity of a face presented in negative contrast. This difficulty, however, is substantially reduced when only the eye region is contrast positive in an otherwise negative face image, and recognition of these so-called contrast chimeras approaches performance with full positive faces. This apparently similar accuracy has led researchers to suggest that familiar face representations are built around the eye region. The present study used the N250r, an event-related brain potential correlate of repetition priming, to examine whether chimera recognition is similarly efficient as positive face recognition. In a series of 3 experiments, we found a clear N250r for positive but reduced or even absent repetition effects for negative and chimera faces. This finding held true independent of whether the same basic pictures of familiar faces were used as prime and target stimuli (Experiment 1) or not (Experiments 2 and 3). Similar results were also obtained independent of whether positive, negative or chimera primes preceded full positive targets (Experiments 1 and 2) or targets in the same respective contrast format (Experiment 3). These results indicate that only positive faces contain all information necessary for optimal face recognition and that even though contrast chimeras are recognized highly accurately, the underlying processes work less efficiently as compared with normal face recognition. We conclude that familiar face representations are not built around the eyes but comprise detailed information from other regions of the face. (PsycINFO Database Record (c) 2019 APA, all rights reserved).
呈现的负对比人脸难以识别。然而,当负面孔图像中仅眼部区域为正对比时,这种难度会大大降低,并且对这些所谓的对比嵌合体的识别接近全正面孔的识别水平。这种明显的相似准确性促使研究人员提出,熟悉的面孔表示是围绕眼部区域构建的。本研究使用 N250r(与重复启动相关的事件相关脑电位相关物)来检查嵌合体识别是否与正面孔识别一样有效。在一系列 3 项实验中,我们发现正面孔的 N250r 很明显,而负面孔和嵌合体面孔的重复效应则减少甚至消失。该发现独立于是否使用相同的熟悉面孔基本图片作为启动和目标刺激(实验 1)或不使用(实验 2 和 3)。无论正、负或嵌合启动后是全正目标(实验 1 和 2)还是相同的各自对比格式的目标(实验 3),该结果都成立。这些结果表明,只有正面孔包含最佳面孔识别所需的所有信息,即使对比嵌合体被高度准确地识别,但其潜在过程的效率也低于正常面孔识别。我们得出结论,熟悉的面孔表示不是围绕眼睛构建的,而是包括面部其他区域的详细信息。(PsycINFO 数据库记录(c)2019 APA,保留所有权利)。