Rollins Leslie, Olsen Aubrey, Evans Megan
Department of Psychology, Christopher Newport University, Newport News, VA, USA.
Department of Psychology, Christopher Newport University, Newport News, VA, USA.
Neuropsychologia. 2020 Apr;141:107417. doi: 10.1016/j.neuropsychologia.2020.107417. Epub 2020 Mar 2.
The aim of the present study was to further understanding of how social categorization influences face recognition. According to the categorization-individuation model, face recognition can either be biased toward categorization or individuation. We hypothesized that the face recognition bias associated with a social category (e.g., the own-age bias) would be larger when faces were initially categorized according to that category. To examine this hypothesis, young adults (N = 63) completed a face recognition task after either making age or sex judgments while encoding child and adult faces. Young adults showed the own-age and own-sex biases in face recognition. Consistent with our hypothesis, the magnitude of the own-age bias in face recognition was larger when individuals made age, rather than sex, judgments at encoding. To probe the mechanisms underlying this effect, we examined ERP responses to child and adult faces across the social categorization conditions. Neither the P1 nor the N170 ERP components were modulated by the social categorization task or the social category membership of the face. However, the P2, which is associated with second-order configural processing, was larger to adult faces than child faces only in the age categorization condition. The N250, which is associated with individuation, was larger (i.e., more negative) to adult than child faces and during age categorization than sex categorization. These results are interpreted within the context of the categorization-individuation model and current research on biases in face recognition.
本研究的目的是进一步了解社会分类如何影响人脸识别。根据分类-个体化模型,人脸识别可能偏向于分类或个体化。我们假设,当面孔最初按照某个社会类别进行分类时,与该社会类别相关的人脸识别偏差(例如,年龄偏差)会更大。为了检验这一假设,63名年轻人在对儿童和成人面孔进行编码时,先进行年龄或性别判断,然后完成人脸识别任务。年轻人在人脸识别中表现出年龄偏差和性别偏差。与我们的假设一致,当个体在编码时进行年龄判断而非性别判断时,人脸识别中的年龄偏差幅度更大。为了探究这种效应背后的机制,我们研究了在不同社会分类条件下对儿童和成人面孔的ERP反应。P1和N170这两个ERP成分均未受到社会分类任务或面孔的社会类别归属的调节。然而,与二阶构型加工相关的P2,仅在年龄分类条件下,对成人面孔的反应比对儿童面孔的反应更大。与个体化相关的N250,对成人面孔的反应比对儿童面孔的反应更大(即更负),并且在年龄分类过程中比对性别分类的反应更大。这些结果将在分类-个体化模型以及当前关于人脸识别偏差的研究背景下进行解释。