面孔识别一直是社会认知领域的重要研究内容。由于自身经验与认知结构的差异, 人们在识别面孔时会产生特定的社会认知偏向, 如普遍存在自我群体偏向(own-group bias) (Hills, Pake, & Dempsey, 2018)。群体对人类生存和发展具有重要适应价值, 近年来有关面孔识别自我群体偏向的研究得到国内外学者关注, 已取得了一些重要研究成果(Cronin, Craig, & Lipp, 2018; Herliz & Lovén, 2013; Hugenberg, Wilson, See, & Young, 2013; Man & Hills, 2017; Martschuk & Sporer, 2018; Proietti, Macchi Cassia, & Mondloch, 2015; Short, Semplonius, Proietti, & Mondloch, 2014; Tham, Bremner, & Hay, 2015; Wan, Crookes, Reynolds, Irons, & McKone, 2015; Wiese, Komes, & Schweinberger, 2013; Yan et al., 2017; 吕勇, 刘亚平, 罗跃嘉, 2011; 唐卫海, 张晓沛, 唐乐, 李楠, 刘希平, 2017)。然而, 由于面孔识别自我群体偏向涉及到面孔、群体、认知、自我等社会认知领域的多个要素, 关于自我群体偏向的认知加工机制及理论解释等重要问题仍有待探索。
1 自我群体偏向的概念与普遍性
自我群体偏向是指相对于其他群体的面孔, 人们识别和再认自己所属群体(如种族、性别或年龄)面孔的成绩要显著好于其他群体面孔成绩的心理现象(Blandón-Gitlin, Pezdek, Saldivar, & Steelman, 2014; Rule, Ambady, Adams, & Macrae, 2007)。值得注意的是, 面孔识别领域的自我群体偏向现象与内群体偏爱(ingroup bias)的本质不同, 元分析和实证研究均表明这种自我群体偏向与群体态度评价之间没有关联(Bernstein, Sacco, Young, & Hugenberg, 2014), 而只是与对内群体和外群体不同的加工动机(processing motivation)有关。
性别、种族和年龄作为区分自我身份和他人身份的重要“大三” (big three)社会分类维度(佐斌, 温芳芳, 宋静静, 代涛涛, 2019), 在自我群体偏向中受到特别关注。其中, 本族偏向(own-race bias)的研究表明, 相比异族面孔, 人们更容易再认本族面孔(Tanaka, Heptonstall, & Hagen, 2013)。此外, 研究者对同龄偏向(own-age bias)的研究表明, 人们更容易再认和记忆相同年龄范围的人的面孔(Rhodes & Anastasi, 2012; Wiese et al., 2013)。
同性偏向(own-gender bias)也称作跨性别效应(cross-sex effect)或自我性别偏向(own-sex bias), 近年来成为心理学者的一个研究热点(Herliz & Love´n, 2013; Hills et al., 2018)。同性偏向的具体
证明是相比记忆异性面孔, 人们对自己所属性别的面孔能够更好地再认和记忆(Lovén, Herlitz, & Rehnman, 2011)。前人研究通过标准新旧再认范式(standard old/new recognition paradigms) (Hills et al., 2018)、剑桥面孔记忆测试(Cambridge Face Memory Test, CFMT) (McKone et al., 2011)以及面孔倒转任务(Man & Hills, 2017)等多样方法证明了面孔识别时会出现同性偏向现象。例如, 一些研究发现, 女性被试在识别女性面孔时相比男性面孔更准确(Herlitz & Lovén, 2013; Lewin & Herlitz, 2002; Lovén et al., 2011); 同样, 有一些研究表明男性也表现出同性偏向, 男性被试相比女性面孔更好地记忆男性面孔(Hugenberg, Young, Bernstein, & Sacco, 2010; Man & Hills, 2017; Rhodes & Anastasi, 2012; Rule, Garrett, & Ambady, 2010; Sporer, 2001)。
2 自我群体偏向的认知加工过程
当个体在面孔再认任务中对内群体成员的面孔表现出更好的记忆水平时, 这种优势是在认知加工过程的哪一个阶段出现的?注意分配和社会分类等因素对此认知加工过程有怎样的影响?与此同时, 与面孔识别有关的脑区有怎样的活动?自我群体偏向的相关研究聚焦于上述问题进行了较好的探索。
2.1 面孔识别自我群体时的编码过程
目前对面孔识别自我群体偏向的认知加工过程的相关研究尚不多, 概括起来主要聚焦于三个方面：第一, 偏向发生在编码阶段还是后编码阶段?例如, Young, Bernstein和Hugenberg (2010)通过在面孔编码前和面孔编码后分别操作指导语发现, 同性偏向的增强或减弱均只在编码阶段发生, 因此表明偏向主要在面孔编码阶段发生。第二, 面孔编码加工时, 注意起什么作用?例如Lovén等人(2011)比较了全注意和分散注意对同性偏向的影响, 结果发现, 是否分散注意对其作用不明显; 第三, 社会分类对偏向加工会产生怎样的影响?例如, 分类形成的内群体和外群体身份对于自我群体偏向的产生发挥着至关重要的作用( Young et al., 2010)。
2.2 面孔识别自我群体的认知神经机制
前人研究中发现了面孔识别的自我群体偏向, 探索和识别该效应背后的认知神经过程也相应受到重视, 一些新兴技术开始应用于对面孔识别中自我群体偏向的初步探索, 例如眼动、脑电、脑成像等(Wiese & Schweinberger, 2018; Wolff, Kemter, Schweinberger, & Wiese, 2014)。具体来看, 研究者采用眼动追踪技术, 发现被试对于内群体目标存在注意偏向以及更长的注视时间, 更为集中和长时间的注意使得被试在再认内群体目标面孔时有更好的成绩(Kawakami et al., 2014)。有研究者通过脑电研究(ERPs)发现同性面孔引发了女性被试更大的N170波幅(吕勇等, 2011); 其他研究者在对面孔识别的自我种族偏向进行研究时, 发现相比白人面孔, 白人对黑人(外群体)面孔反应时有更长的N170成分(Ofan, Rubin, & Amodio, 2011); 对异族面孔的N170成分潜伏期更长(Wesis, 2012)。吕勇等人(2011)也发现了新旧面孔在女性被试中诱发了不同的P1波幅。但是, P1成分是否真正反映了面孔种族群体差异继而影响面孔识别还存在争议。
3 影响面孔识别自我群体偏向的主要因素
面孔识别自我群体偏向的影响因素可以通过评价者、评价对象的情绪与背景线索和研究任务三个方面加以梳理。评价者方面包括自我参照(self-reference)和个体对互动结果的期待等因素的影响; 评价对象的情绪与背景线索方面包括情绪情感和生态背景等因素的影响; 研究任务包括指导语差异和加工深度等因素的影响。
3.1 评价者因素
3.1.1 自我参照
被试的自我参照会对言语再认记忆产生积极效应。Rogers, Rogers和Kuiper (1977)比较了自我参照加工与其他类型加工(结构、语音、语义)对被试记忆的影响, 结果发现自我参照加工任务中的被试对曾评价过的词语再认记忆成绩最好。基于此, 研究者认为自我参照是一种有力的再认记忆的编码工具。那么, 自我参照是否也会对面孔识别自我群体偏向产生影响呢?
有研究表明, 自我参照会促进面孔识别的自我群体偏向。例如, Ebner (2011)进行了关于面孔识别同龄偏向的研究, 她发现相同年龄目标对象的面孔对于被试来说具有更强的自我相关性, 在面孔识别的相同年龄偏向现象中起到了重要作用。相比其他年龄目标对象的面孔, 被试在编码与自己年龄相近目标对象的面孔时可能涉及到更多的自我参照, 因为这部分面孔与被试更高关联。这种高关联性使得被试对同龄面孔加工的兴趣和动机更强, 因此对其有更加仔细的观察和更长的注视时间, 从而促进了被试对自我群体面孔再认成绩的提高。
3.1.2 个体对互动结果的期待
自我群体偏向的一个典型稳定因素是人们通常期待与内群体成员而不是外群体成员有更频繁和更重要的互动。有研究表明, 个体所期待的人际互动结果在自我群体偏向中发挥着重要作用(van Bavel & Cunningham, 2012)。事实上, 内群体是个体身心支持的稳定来源, 对个体的重要性显而易见, 因此个体对内群体成员的互动期待及关注会更多更细, 认知加工通常是个性化的。而对于外群体成员来说, 尽管个体对外群体成员通常会默认进行类别化对待, 但是当个体看重与外群体成员互动的结果时, 人们对外群体成员可能会进行个性化加工(Wilson, See, Bernstein, & Hugenberg, 2014)。van Bavel和Cunningham (2012)研究中发现, 当告知被试将与其所分配的内群体成员会有更多的互动时, 发现明显的自我群体偏向存在, 但当操作被试会与内外群体成员均具有同等频繁和重要的互动时, 自我群体偏向会随着外群体成员的再认提升而削弱。
3.2 评价对象的情绪与背景线索
3.2.1 情绪情感
以往研究大多采用标准静态面孔作为刺激材料, 但是在日常社会互动中, 人们更多对自然面孔和具有社会背景的面孔进行识别, 随着面孔材料的逐步真实生态化, 面孔识别的自我群体偏向效应也会发生变化。例如, Freeman等人(2013)采用功能性磁共振成像(fMRI)技术对识别典型美国环境之中和典型中国环境之中的面孔的被试进行分析, 发现被试的压后皮质(retrosplenial cortex, RSC)和眼窝前额皮质(orbitofrontal cortex, OFC)在刺激面孔和周围环境显示出一致性时存在高敏感性, 且这种反应会随着典型美国环境中的面孔变得越来越像白人, 以及典型中国环境中的面孔变得越来越像亚洲人而线性增长。
3.3 研究的任务与操作
3.3.1 指导语差异
通过不同的指导语, 被试在学习阶段完成不同的任务, 这会对面孔识别自我群体偏向的大小产生影响。例如, 本族效应的研究表明, 要求被试在学习面孔时尽可能多地注意异族面孔的个性化动机指导语(motivation-individuate instructions)会显著降低本族偏向效应(Young et al., 2010; Young & Hugenberg, 2012)。而在同性偏向的研究中, 研究者主要操纵了被试对目标面孔不同方面的关注程度。例如, 对于男性被试而言, 与评价中性特质(如可爱的)相比, 评价男性特质(如支配性的)时的同性偏向更大(Motta-Mena, Picci, & Scherf, 2016)。要求被试在学习阶段进行吸引力判断或进行独特性判断, 与只是告知被试看面孔也导致认知加工差异, 使被试的面孔再认成绩发生变化(Hills et al., 2018)。相对独特性判断, 进行吸引力判断会让被试倾向于进行更多异性面孔的深加工(Rhodes, Simmons, & Peters, 2005), 从而导致同性别偏向效应减弱。
3.3.2 加工深度
刺激材料的“加工深度”会对自我群体偏向产生影响。人们对面孔进行纯粹的物理特征判断通常只需要进行浅层加工, 而进行人格特质推理则需要进行深层加工。例如, 研究者通过不同的实验任务(例如性别判断、可爱度、诚实度判断等)操纵加工深度, 结果表明对目标面孔在推论特质(如诚实的)上的判断将导致对面孔更深的加工, 进而提升记忆表现, 而对面孔在更浅显属性上的判断(如性别)将导致更浅层次的加工, 因此会降低记忆表现(Bower & Karlin, 1974)。也有研究者进一步认为人们之所以在外群体成员面孔的识别上存在困难, 是因为人们对外群体成员面孔的加工只停留在表面水平(Sporer, 2001)。
自Bower和Karlin (1974)的开拓性研究后, 大量研究采用一系列定向任务对面孔的浅层和深层加工之间的差异进行检验(Mueller, Bailis, & Goldstein, 1979)。总体而言, 大多数研究一致发现两种不同类型取向的指导语会导致被试不同的再认表现。但是也有一些研究结果与此观点相矛盾。例如, Sporer (1991)在研究中比较了8种不同的编码策略, 从而操控了编码深度、精细化加工的程度, 以及自我生成的评价种类, 结果发现当被试采用对面孔做整体加工的策略时, 面孔再认结果大体相同且没有显著好于自然的编码策略, 但当被试持续关注面孔的单一维度时记忆表现是最差的。
4 解释自我群体偏向的理论模型
对于面孔识别自我群体偏向的产生原因, 目前有“知觉经验说” (Perceptual Expertise Hypothesis)与“社会认知论” (Social-Cognitive Approach)两种理论解释。类别化-个性化模型(Categorization- Individuation Model)和双路径模型(Dual-Route Approach to Other-Race Effect)试图从不同的角度将知觉经验论和社会认知论两种解释理论加以整合。
4.1 知觉经验说
知觉经验说认为与不同群体人群的接触经验不同导致了对自我群体和其他群体面孔的识别差异(Michel, Corneille, & Rossion, 2010), 对于他族面孔缺乏经验导致编码面孔时的基本感知不足, 其正确再认低于本族面孔。
不同研究者从不同方面提出了这种感知编码的不足, 包括整体加工的缺乏导致更多依赖于基于局部的加工(Hayward, Crookes, & Rhodes, 2013), 以及面孔空间的低质量表征(Valentine, 1991)。基于经验的解释表明对于不同群体面孔接触的数量或质量会对所观察的偏差产生预测作用。从加工方式角度来看, 面孔加工方式理论认为, 相对其他群体面孔, 人们对自我群体面孔更多采用整体性加工(Michel, Corneille, & Rossion, 2007), 即对整个面孔采用格式塔的方式进行编码, 整体性加工依赖于人们的视觉经验, 通常是无意识自动化加工, 很难用语言表达, 且包含组块, 有利于再认记忆。另一方面, Valentine (1991)的面孔空间模型(face-space model)清晰地描述了经验如何对面孔感知产生影响, 该模型认为每张面孔储存在面孔多重维度的一个空间, 每个维度表征了最常遇到的面孔的面相(McKone, Aitkin, & Edwards, 2005)。当两个常见面孔不容易区分时, 人们可增加面孔空间维度来帮助区分。随着经验的增加, 人们更容易排除或忽略更少相关信息, 产生专门有效的加工方式。经验的增加会导致更少见群体面孔的面孔空间存储在远离面孔空间中心的位置。
本族偏向的研究为基于经验的解释提供了证据, Sporer和Horry (2011)报告住在德国的土耳其人对于德国和土耳其人面孔的再认成绩不存在显著差异, 但是德国被试对德国面孔相比土耳其面孔表现出明显的再认优势。
4.2 社会认知论
与知觉经验说不同, 社会认知论认为人们在识别面孔时会根据类别特征自动划分内外群体成员面孔。人们会自发地对内群体面孔进行个性化加工, 给予他们社会重要性, 进而对内群体成员进行深入地个性化信息加工, 而知觉外群体成员时则关注面孔的群体共同特征而忽视个性化信息, 进而不能准确识别新旧外群体面孔(Hehman, Mania, & Gaertner, 2010; Hugenberg et al., 2010; Pauker et al., 2009; Sporer, 2001)。
一些研究表明增加被试对其他族群面孔个性化的动机会降低甚至消除异族效应(Hehman et al., 2010; Young et al., 2010; Young & Hugenberg, 2012)。这些对动机的操作包括给予让被试个性化加工他族面孔的指导语(Hugenberg et al., 2007; Young et al., 2010; Young & Hugenberg, 2012), 提供引导他族面孔分类为内群体的线索(Hehman et al., 2010), 呈现愤怒面孔表情的面孔(Young & Hugenberg, 2012)等。
同龄偏向的结果为动机解释提供了证据, 例如, 相比更小的儿童, 青春前期儿童具有更大的同龄偏向(Picci, Gotts, & Scherf, 2016), 即更大的儿童可能不愿意与更小的儿童交往, 而更小的儿童通常希望跟更大的儿童交往。
Sporer于2001年提出的面孔加工内外群体模型(In-Group/Out-Group Model of Face Processing)支持了上述两种理论中的社会认知论, 该模型认为：当个体遇到内群体成员的面孔时, 默认、自动的加工将会以构型(configural)编码开始。构型编码即个体将注意力集中于面孔组成元素的组成情况上, 注意局部元素以及元素之间的关系, 这也正是对正常直立面孔的专家化加工的特征。当遇到外群体成员的面孔时, 对于外群体特征线索的感知会在激发其他典型面孔感知策略前激发分类过程。对比人们加工内群体成员面孔和外群体成员面孔过程, 在加工内群体成员面孔时, 人们跳过了分类过程, 而分类过程被假定为加工外群体成员面孔的第一步。由此可见, 个体在识别外群体成员的面孔时要经历更为复杂的认知过程, 因此对外群体成员面孔的识别需要更长的反应时间。
4.3 整合的类别化-个性化模型和双路径模型
在面孔识别自我群体偏向的经典理论模型中, 类别化-个性化模型和双路径模型均试图将知觉经验说和社会认知论整合, 成为用以解释自我群体偏向的综合性模型。但是, 在两种理论模型中, 知觉经验说和社会认知论以不同的方式整合：类别化-个性化模型强调接触经验与社会动机同时对个体的面孔识别起作用, 而双路径模型则认为在识别不同对象的面孔时, 接触经验和社会动机二者之间只有一种起主导作用。
4.3.1 类别化-个性化模型
研究者提出类别化-个性化模型, 试图对知觉经验说和社会认知-动机论加以整合, 来对自我群体偏向加以解释(Hugenberg et al., 2013; Hugenberg et al., 2010)。根据类别化-个性化模型, 一些面孔识别的自我群体偏向在很大程度上(或完全)是由不同水平的个性化动机导致的, 例如团队内部和团队以外的面孔感知差异; 而有些自我群体偏向在很大程度上(或完全)是由不同水平的个性化经验导致的, 例如人类对其他人类的面孔感知与对动物的面孔感知的差异; 还有些内群体偏差同时受到个性化动机和个性化经验的影响, 例如对自己种族面孔的感知和其他种族面孔的感知的差异。需要注意的是, 不同文化或者亚文化中个性化经验和个性化动机可能存在差异。
类别化-个性化模型以“经验”作为横坐标, “动机”作为纵坐标为自我群体偏向提供了直观形象的认知框架, 一方面将知觉经验说和社会认知-动机论进行了有机整合, 为两者的共同作用提供了具体多样的群体样例; 另一方面也为不同类型群体(如性别、种族、性取向、团队等)在类别化-个性化模型的经验-动机坐标体系中所处的位置有更清晰的定位如对“团队”而言, 动机在自我群体偏向中起主导作用, 对“种族”而言, 动机和经验都发挥着重要作用等, 并为群体之间的自我群体偏向特点异同提供了相对的参照比较体系。
4.3.2 双路径模型
Wan等人(2015)试图从不同认知路径的角度将知觉经验说和社会认知论进行一定程度上的整合, 认为对于不同的外群体面孔存在不同的认知路径, 并且对如何改变面孔识别的自我群体偏向, 提高对其他群体的面孔识别成绩提出了可能的方案。当个体看到外群体成员的面孔时, 会有两种路径指向最终的“异族效应”, 即对外群体成员面孔的识别成绩低于自我群体。
路径1：假设当在西方文化中成长的白种人和在东方文化中成长的亚洲人观察对方种族成员的面孔时, 因缺少对目标刺激的经验, 导致较低水平的知觉加工, 进而对对方种族面孔的感知存在缺陷。这种情况下, 提高社会动机对于其他种族成员的面孔再认的提高是没有作用的。想要提高, 可以采用利用计算机程序生成与常模面孔有定量差异的面孔的方法(Sporer, 2001), 生成彼此之间有更大差异的其他种族成员的面孔。
路径2：假设当美国白人观察美国黑人的面孔时, 由于美国黑人更低的社会地位或者其他维度上与美国白人的差异, 美国白人将美国黑人的面孔知觉为“他人”, 因此将注意力集中在类别水平上的信息, 而非个体水平上的信息, 进而对对方种族面孔的感知存在缺陷。这种情况下, 想要提高美国白人对美国黑人的面孔感知是可以通过提高社会动机来实现的。
双路径模型分别从评价者在“缺乏知觉经验”和“缺乏社会动机”不同的情形下, 当看到其他种族成员的脸所产生“异族效应”的两条不同认知路径, 为整合理解知觉经验说和社会认知-动机论提供了系统生态的理论依据。一方面, 双路径模型为自我群体偏向产生中评价者对目标刺激可能存在的动机和经验上的两种不同的个体差异情形进行了清晰的划分归类; 另一方面, 也为在不同情形下的另种不同路径的认知加工过程进行了深入剖析, 为自我群体偏向产生的认知机制具有重要参考借鉴。
5 研究展望
有关面孔识别自我群体偏向的检验、认知加工和神经机制以及理论解释等方面的研究已经取得了比较多的成果。但是, 基于对面孔识别自我群体偏向已有研究的分析, 可以看出以往研究尚存在一些不足, 值得在以下方面进一步探索。
首先, 面孔刺激作为社会认知要素之一, 在面孔识别自我群体偏向中扮演着重要的角色。以往研究大多采用标准正立静态面孔, 很少涉及面孔物理属性的改变, 例如整体/局部特征、正立/倒立、背景颜色的变化等对自我群体偏向产生的影响。特别是在现实生活和社会交往互动中, 人们并非对标准静态面孔进行识别, 而更多面对的是真实自然的面孔, 会考虑面孔所蕴含的丰富信息, 如面孔本身的年龄、种族和情绪状态, 且往往伴随着相应的背景信息, 比如, 有些背景与面孔的性别刻板印象一致/不一致, 有些背景存在不同的群体面孔比例, 还有些背景则折射出内隐的社会群体关系等, 那么这些面孔材料的变化特别是生态化改变又会对自我群体偏向产生怎样的影响?这些面孔识别自我群体偏向的社会生态问题值得深入考察。
其次, 在对影响面孔识别自我群体偏向的认知者和加工情境因素的梳理中可以看出, “自我”在面孔识别自我群体偏向中发挥着核心作用。然而, 目前的研究大多基于“评他”视角, 很少考虑自我的关联信息(如个体生理差异、群体刻板印象、不同社会群体角色身份和社会关系等)对自我群体偏向认知加工机制的影响, 鉴于此, 进一步从“自我”的角度来构建揭示面孔识别自我群体偏向心理机制的整合模型将成为未来研究的重要方向。
再次, 虽然新兴的脑电技术已经被应用于面孔识别自我群体偏向研究中, 但目前自我群体偏向的研究大多仍采用行为指标, 缺乏对自我群体偏向神经生理机制的系统探究, 因此, 采用脑电和脑成像技术、眼动追踪, 以及鼠标追踪技术来为面孔识别自我群体偏向认知加工机制提供更为客观的证据也具有重要的学术价值。
最后, 当前对面孔识别自我群体偏向的研究大多在西方社会文化背景下进行, 而不同文化或者亚文化中个性化经验和个性化动机可能存在差异, 进而对面孔识别产生不同的认知加工机制。现有研究中较少涉及到跨文化比较, 因此文化因素在面孔识别中发挥了怎样的作用还不甚明晰。特别在中国注重群体关系的文化背景下, 自我群体偏向会存在怎样的效应?因此, 未来在对面孔识别自我群体偏向进行本土化研究时, 可以将文化差异作为自变量纳入考虑, 探究身处不同文化或者亚文化中的个体在识别面孔时的心理过程, 在理论层面上对这一领域进行丰富和发展。
参考文献
They all look the same to me (unless they’ re angry): From out-group homogeneity to out-group heterogeneity
The impact of race and inclusionary status on memory for ingroup and outgroup faces
Oxytocin eliminates the own-race bias in face recognition memory
DOI:10.1016/j.brainres.2013.07.015
URL
PMID:23872107
[本文引用: 1]
The neuropeptide Oxytocin influences a number of social behaviors, including processing of faces. We examined whether Oxytocin facilitates the processing of out-group faces and reduce the own-race bias (ORB). The ORB is a robust phenomenon characterized by poor recognition memory of other-race faces compared to the same-race faces. In Experiment 1, participants received intranasal solutions of Oxytocin or placebo prior to viewing White and Black faces. On a subsequent recognition test, whereas in the placebo condition the same-race faces were better recognized than other-race faces, in the Oxytocin condition Black and White faces were equally well recognized, effectively eliminating the ORB. In Experiment 2, Oxytocin was administered after the study phase. The ORB resulted, but Oxytocin did not significantly reduce the effect. This study is the first to show that Oxytocin can enhance face memory of out-group members and underscore the importance of social encoding mechanisms underlying the own-race bias. This article is part of a Special Issue entitled Oxytocin and Social Behav.
Depth of processing pictures of faces and recognition memory
Emotional expressions reduce the own-age bias
The neural basis of contextual influences on face categorization
The other-race effect: Holistic coding differences and beyond
Where the division lies: common ingroup identity moderates the cross-race facial-recognition effect
Sex differences and the own-gender bias in face recognition: A meta-analytic review
Exploring the contribution of motivation and experience in the postpubescent own-gender bias in face recognition
Towards a synthetic model of own group biases in face memory
The categorization-individuation model: An integrative account of the other-race recognition deficit
DOI:10.1037/a0020463
URL
PMID:20822290
[本文引用: 3]
The other-race effect (ORE), or the finding that same-race faces are better recognized than other-race faces, is one of the best replicated phenomena in face recognition. The current article reviews existing evidence and theory and proposes a new theoretical framework for the ORE, which argues that the effect results from a confluence of social categorization, motivated individuation, and perceptual experience. This categorization-individuation model offers not only a parsimonious account of both classic and recent evidence for category-based biases in face recognition but also links the ORE to broader evidence and theory in social cognition and face perception. Finally, the categorization-individuation model makes a series of novel predictions for how the ORE can be exacerbated, attenuated, or even eliminated via perceptual and motivational processes, both by improving other-race recognition and by reducing same-race recognition. The authors propose that this new model for the ORE also leads to applied interventions that differ sharply from other theories of the ORE, while simultaneously providing an integrative theoretical framework for future research on the ORE.
An eye for the I: Preferential attention to the eyes of ingroup members
Encoding of personal information: Self-other differences
Sex differences in face recognition-women's faces make the difference
Women's own-gender bias in face recognition memory
DOI:10.1027/1618-3169/a000100
URL
PMID:21310695
[本文引用: 4]
Women remember more female than male faces, whereas men do not seem to display an own-gender bias in face recognition memory. Why women remember female faces to a greater extent than male faces is unclear; one proposition is that women attend more to and thereby process female faces more effortfully than male faces during encoding. A manipulation that distracts attention and reduces effortful processing may therefore decrease women's own-gender bias by reducing memory for female faces relative to male faces. In three separate experiments, women and men encoded female and male faces for later recognition in full attention and divided attention conditions. Results consistently showed that women, in contrast to men, displayed a reliable own-gender bias. Importantly, the magnitude of women's own-gender bias was not reduced in divided attention conditions, indicating that it is not a result of effortful processing of female faces. We suggest these results reflect that women have greater perceptual expertise for female faces, facilitating recognition memory.
Eye-tracking the own-gender bias in face recognition: Other-gender faces are viewed differently to own-gender faces
Memory for faces in old age: A meta-analysis
Categorical and coordinate relations in faces, or fechner's law and face space instead?
Face ethnicity and measurement reliability affect face recognition performance in developmental prosopagnosia: Evidence from the Cambridge face memory test - Australian
Race categorization modulates holistic face encoding
Race categorization modulates holistic face encoding
Dominance elicits the own-gender bias in males
Dominance elicits the own-gender bias in males
Depth of processing and anxiety in facial recognition
Seeing race: N170 responses to race and their relation to automatic racial attitudes and controlled processing
Not so black and white: Memory for ambiguous group members
A theoretical rut: Revisiting and critically evaluating the generalized under/over-connectivity hypothesis of autism
The own-age face recognition bias is task dependent
Attractiveness and sexual behavior: Does attractiveness enhance mating success?
The own-age bias in face recognition: A meta-analytic and theoretical review
DOI:10.1037/a0025750
URL
PMID:22061689
[本文引用: 2]
A large number of studies have examined the finding that recognition memory for faces of one's own age group is often superior to memory for faces of another age group. We examined this own-age bias (OAB) in the meta-analyses reported. These data showed that hits were reliably greater for same-age relative to other-age faces (g = 0.23) and that false alarms were reliably less likely for same-age compared with other-age faces (g = -0.23). Further meta-analyses of measures of signal detection demonstrated that, although no difference in response criterion was evident (g = -0.01), discriminability was reliably better for same-age compared with other-age faces (g = 0.37). As well, children, younger adults, and older adults exhibited superior discriminability for same-age compared with other-age age faces. Thus, the OAB appears to be a robust effect that influences the accuracy of face recognition. Theoretical accounts of the OAB have generally suggested that it reflects more extensive, recent experiences with one's own age group relative to other-age groups. Additional analyses were supportive of this account as the OAB was present even for groups (e.g., older adults) that had prior experiences as members of another age group. However, the most comprehensive account of the OAB will also likely invoke mechanisms suggested by social-cognitive theories.
Evidence for the self as a cognitive prototype: The "false alarms effect."
Us and them: Memory advantages in perceptually ambiguous groups
Places and faces: Geographic environment influences the ingroup memory advantage
DOI:10.1037/a0018589
URL
PMID:20175617
[本文引用: 1]
The preferential allocation of attention and memory to the ingroup (the ingroup memory advantage) is one of the most replicated effects in the psychological literature. But little is known about what factors may influence such effects. Here the authors investigated a potential influence: category salience as determined by the perceiver's geographic environment. They did so by studying the ingroup memory advantage in perceptually ambiguous groups for whom perceptual cues do not make group membership immediately salient. Individuals in an environment in which a particular group membership was salient (Mormon and non-Mormon men and women living in Salt Lake City, Utah) showed better memory for faces belonging to their ingroup in an incidental encoding paradigm. Majority group participants in an environment where this group membership was not salient (non-Mormon men and women in the northeastern United States), however, showed no ingroup memory advantage whereas minority group participants (Mormons) in the same environment did. But in the same environment, when differences in group membership were made accessible via an unobtrusive priming task, non-Mormons did show an ingroup memory advantage and Mormons' memory for ingroup members increased. Environmental context cues therefore influence the ingroup memory advantage for categories that are not intrinsically salient.
Differential attentional allocation and subsequent recognition for young verus older adult faces
Deep-Deeper-Deepest? Encoding strategies and the recognition of human faces
Recognizing faces of other ethnic groups: An integration of theories
Recognizing faces from ethnic in-groups and out‐groups: Importance of outer face features and effects of retention interval
Perceptual expertise and the plasticity of other-race face recognition
In infancy the timing of emergence of the other-race effect is dependent on face gender
DOI:10.1016/j.infbeh.2015.05.006
URL
PMID:26143499
[本文引用: 1]
Poorer recognition of other-race faces relative to own-race faces is well documented from late infancy to adulthood. Research has revealed an increase in the other-race effect (ORE) during the first year of life, but there is some disagreement regarding the age at which it emerges. Using cropped faces to eliminate discrimination based on external features, visual paired comparison and spontaneous visual preference measures were used to investigate the relationship between ORE and face gender at 3-4 and 8-9 months. Caucasian-White 3- to 4-month-olds' discrimination of Chinese, Malay, and Caucasian-White faces showed an own-race advantage for female faces alone whereas at 8-9 months the own-race advantage was general across gender. This developmental effect is accompanied by a preference for female over male faces at 4 months and no gender preference at 9 months. The pattern of recognition advantage and preference suggests that there is a shift from a female-based own-race recognition advantage to a general own-race recognition advantage, in keeping with a visual and social experience-based account of ORE.
A unified account of the effects of distinctiveness, inversion, and race in face recognition
A social identity approach to person memory: Group membership, collective identification, and social role shape attention and memory
DOI:10.1177/0146167212455829
URL
PMID:22914991
[本文引用: 2]
Evidence indicates that superior memory for own-group versus other-group faces (termed own-group bias) occurs because of social categorization: People are more likely to encode own-group members as individuals. The authors show that aspects of the perceiver's social identity shape social attention and memory over and above mere categorization. In three experiments, participants were assigned to a mixed-race minimal group and showed own-group bias toward this minimal group, regardless of race. Own-group bias was mediated by attention toward own-group faces during encoding (Experiment 1). Furthermore, participants who were highly identified with their minimal group had the largest own-group bias (Experiment 2). However, social affordances attenuated own-group bias-Memory for other-group faces was heightened among participants who were assigned to a role (i.e., spy) that required attention toward other-group members (Experiment 3). This research suggests that social identity may provide novel insights into person memory.
Gender difference in recognition memory for neutral and emotional faces
DOI:10.1080/09658211.2013.771273
URL
PMID:23432017
Research has shown females' advantage in face memory, but few studies have used emotional faces as stimuli. In addition, it is unclear whether gender difference in memory for faces is mediated by individual factors especially associated with emotion. This study examined gender difference in recognition memory for neutral and emotional faces and whether arousal predisposition, emotion reappraisal, and emotion suppression can play a mediation role. Females (N=48) and males (N=45) viewed and memorised neutral, happy, and angry faces, and then took an immediate recognition test. The findings are: (1) Females outperformed males in recognition memory for happy faces, but not for neutral and negative faces. (2) Gender difference in recognition memory (including recollection and familiarity) was not mediated by arousal predisposition and emotion regulation. (3) Females had an own-sex bias in that they outperformed males with regard to recognition memory only for female faces, but not for male faces. The findings suggest that females' advantage is mediated by valence of face emotionality but not by arousal predisposition and emotion regulation, thus extending the literature by demonstrating the boundary condition for occurrence of females' advantage.
A cultural setting where the other-race effect on face recognition has no social-motivational component and derives entirely from lifetime perceptual experience
DOI:10.1016/j.cognition.2015.07.011
URL
PMID:26257000
[本文引用: 2]
Competing approaches to the other-race effect (ORE) see its primary cause as either a lack of motivation to individuate social outgroup members, or a lack of perceptual experience with other-race faces. Here, we argue that the evidence supporting the social-motivational approach derives from a particular cultural setting: a high socio-economic status group (typically US Whites) looking at the faces of a lower status group (US Blacks) with whom observers typically have at least moderate perceptual experience. In contrast, we test motivation-to-individuate instructions across five studies covering an extremely wide range of perceptual experience, in a cultural setting of more equal socio-economic status, namely Asian and Caucasian participants (N = 480) tested on Asian and Caucasian faces. We find no social-motivational component at all to the ORE, specifically: no reduction in the ORE with motivation instructions, including for novel images of the faces, and at all experience levels; no increase in correlation between own- and other-race face recognition, implying no increase in shared processes; and greater (not the predicted less) effort applied to distinguishing other-race faces than own-race faces under normal (
Ageing faces in ageing minds: A review on the own-age bias in face recognition
Inequality between biases in face memory: Event-related potentials reveal dissociable neural correlates of own-race and own-gender biases
DOI:10.1016/j.cortex.2018.01.016
URL
PMID:29475077
[本文引用: 1]
Humans are more accurate at remembering faces from their own relative to a different ethnic group (own-race bias). Moreover, better memory for faces from an observer's own relative to the other-gender (own-gender bias) has also been reported, particularly for female participants. Theoretical explanations for these effects either emphasize differential perceptual expertise or socio-cognitive factors. Importantly, both types of explanations typically assume a single common mechanism for the various biases. The present study examined event-related potentials (ERP) in a combined own-race/own-gender bias experiment. Whereas both male and female participants demonstrated clear own-race biases in memory performance, enhanced memory for own-gender faces was only observed in female participants. Moreover, the own-race bias was accompanied by larger N170 responses for other-race faces, presumably reflecting more effortful perceptual processing of this face category. Neural correlates of the own-gender bias manifested at later processing stages, reflecting the processing of individual faces (N250) and recollection-based memory retrieval (late ERP old/new effect). We conclude that different face memory biases occur at temporally distinct stages of face processing and are therefore based on different mechanisms. This suggestion is at variance with the assumption of a single common mechanism to underlie the various biases in face memory.
Differences in anticipated interaction drive own group biases in face memory
DOI:10.1371/journal.pone.0090668
URL
PMID:24599294
[本文引用: 1]
According to much research, the Own Group Bias (OGB) in face memory occurs as a consequence of social categorization - ingroup members are more likely than outgroup members to be encoded as individuals and remembered well. The current work is an examination of the role of anticipated future interaction in the OGB. We conducted two studies showing that anticipated interaction influences group-based face memory. In Study 1, we provided correlational evidence that beliefs about the amount and importance of future interaction one will have with racial outgroup members is associated with the OGB, such that people expecting more interaction with outgroup members show a reduced OGB. In Study 2, we manipulated expectations about future interactions with lab-created groups and observed that high levels of anticipated future interaction with the outgroup eliminated the OGB. Thus, social group categorization drives face memory biases to the extent that group membership affords the expectation of interpersonal interaction.
What drives social in-group biases in face recognition memory? ERP evidence from the own-gender bias
Own-group face recognition bias: The effects of location and reputation
When do own-group biases in face recognition occur? Encoding versus post-encoding
Individuation motivation and face experience can operate jointly to produce the own-race bias. Social Psychological and
