Diamond and Carey (1986) identify sensitivity to second-order relational information (i.e., spatial relations among features such as the distance between eyes) as a vital part of achieving expertise with face processing. Previous research suggests that 5-month-olds are sensitive to second-order relational information when shown line drawings of faces in which this information has been manipulated to an exaggerated degree. The present series of experiments explored infants sensitivity to second-order relational information using photographs of real faces and with second-order manipulations that were within the normal range of human variability. A discrimination study conducted with adults provided additional evidence that the second-order manipulations were within the normal range. Five- and 7- month-olds exhibited sensitivity to changes in second-order relational information. Moreover, 5-months detected second-order changes in upright but not in inverted faces, thereby exhibiting an inversion effect that has been considered to be a hallmark of second-order relational processing in adulthood. These results suggest that infants as young as 5 months of age are sensitive to second-order relational changes that are within the normal range of human variability. They also indicate that at least rudimentary aspects of face-processing expertise are available early in life.

The experiments in this thesis used behavioural measures and event-related potentials (ERPs) to investigate the influence of race on face processing in the brain. Previous behavioural research has highlighted an own-race effect in face processing, whereby individuals are more accurate at recognizing own-race compared to other-race faces. The current Thesis examined the own-race effect at perceptual and neural levels. Social influences on the own-race effect were also investigated, such as other-race experience, anxiety and implicit social bias, as these may account for differential own- versus other-race face processing. The main aim of the experiments contained in this thesis was to delve deeper into the examination of own and other-race face perception through a series of original experiments. Participants performed a variety of perceptual discrimination and identification tasks, and completed measures of explicit other-race experience and implicit racial bias to record their perceptions of other-race individuals. Chapters 2-4 saw the development of a new paradigm that tested the own-race effect in perception, in contrast to traditional recognition memory investigations. In Chapter 2 the own-race effect was investigated developmentally and found across three age-groups, and was larger in the two older age-groups. Chapters 3 and 4 found that the own-race effect differed across racial groups, and that social variables such as other-race experience influenced the strength of the own-race effect. In the latter experimental chapters, ERPs revealed that the behavioural own-race effect was evident at a neural level. Chapter 7 demonstrated that face-related stages of processing in the brain were sensitive to race of face. In Chapters 8 and 9, the sensitivity of face processing to own and other-race emotional expression processing was also examined. The additional social factor of emotional expression was explored in order to further the investigation of socially relevant information processing from the face. Findings from the last two experimental chapters demonstrated differential emotional face processing for own- versus other-race faces. Confirming the findings of the behavioural experiments, own- versus other-race emotion processing varied across racial groups and was subject to social influences such as other-race experience, intergroup anxiety and implicit racial bias. Overall, behavioural and neural investigations of the own-race effect demonstrated the influence of social variables such as other-race experience, intergroup anxiety and implicit racial bias on the way in which individuals processed own- versus other-race faces in the human brain.

We often think of stigmatized individuals as encountering only one stereotype set at a time. Yet, many individuals belong to multiple stigmatized groups, and stereotypes associated with these groups jointly influence how perceivers evaluate targets. Research suggests that perceivers integrate stereotypes about targets’ obvious identities during impression formation; however, no work has examined whether targets’ obvious (e.g., race) and ambiguous (e.g., sexual orientation) identities jointly influence impressions. Given that gay stereotypes are activated automatically, I expected the co-activation of contradictory Black (e.g., aggressive) and gay (e.g., warm) stereotypes to arouse conflict, weakening activation of negative stereotypes and improving evaluations of Black gay targets compared with Black straight targets. Participants in Study 1 rated faces of White straight men as more likable than White gay men, but rated Black gay men as more likable than Black straight men. Participants in Study 2 performed a race-categorization task designed to make race salient; nevertheless, sexual orientation still influenced impressions, producing a pattern similar to Study 1. Participants in Studies 3A (approach-avoidance task) and 3B (evaluative priming task) formed implicit impressions that converged with the explicit evaluations in Studies 1 and 2. In 3A, participants approaching Whites responded faster to straights than gays, whereas participants approaching Blacks responded faster to gays than straights. In 3B, participants recognized positive words somewhat (but not significantly) faster when primed with White straight (versus White gay) and Black gay (versus Black straight) faces. Studies 4A – C suggest that ambiguous categories modify the activation of obvious stereotypes, but do not make targets’ features look less prototypical of their obvious groups. In 4B, participants were slower to recognize Black-stereotypic words (piloted in 4A) when primed with Black gay (versus Black straight) faces. In 4C, participants rated Black straight and gay faces as similarly prototypical of Black phenotypes. Taken together, this work presents implications for stereotyping in the case of multiply-categorizable targets and for impression formation involving ambiguous categories.

We often think of stigmatized individuals as encountering only one stereotype set at a time. Yet, many individuals belong to multiple stigmatized groups, and stereotypes associated with these groups jointly influence how perceivers evaluate targets. Research suggests that perceivers integrate stereotypes about targets’ obvious identities during impression formation; however, no work has examined whether targets’ obvious (e.g., race) and ambiguous (e.g., sexual orientation) identities jointly influence impressions. Given that gay stereotypes are activated automatically, I expected the co-activation of contradictory Black (e.g., aggressive) and gay (e.g., warm) stereotypes to arouse conflict, weakening activation of negative stereotypes and improving evaluations of Black gay targets compared with Black straight targets. Participants in Study 1 rated faces of White straight men as more likable than White gay men, but rated Black gay men as more likable than Black straight men. Participants in Study 2 performed a race-categorization task designed to make race salient; nevertheless, sexual orientation still influenced impressions, producing a pattern similar to Study 1. Participants in Studies 3A (approach-avoidance task) and 3B (evaluative priming task) formed implicit impressions that converged with the explicit evaluations in Studies 1 and 2. In 3A, participants approaching Whites responded faster to straights than gays, whereas participants approaching Blacks responded faster to gays than straights. In 3B, participants recognized positive words somewhat (but not significantly) faster when primed with White straight (versus White gay) and Black gay (versus Black straight) faces. Studies 4A – C suggest that ambiguous categories modify the activation of obvious stereotypes, but do not make targets’ features look less prototypical of their obvious groups. In 4B, participants were slower to recognize Black-stereotypic words (piloted in 4A) when primed with Black gay (versus Black straight) faces. In 4C, participants rated Black straight and gay faces as similarly prototypical of Black phenotypes. Taken together, this work presents implications for stereotyping in the case of multiply-categorizable targets and for impression formation involving ambiguous categories.

Typescript (Photocopy) / Includes published papers coauthored by the author / Bibliography: leaves 387-428 / xx, 428 leaves : ill. (some col.) ; 30 cm. / Title page, contents and abstract only. The complete thesis in print form is available from the University Library. / Summary: Describes the use of a new measurement technique for the study of the facial expression of emotion. Based on a mathematical model of the face, a microcomputer approach is utilized to quantify facial movement / Thesis (Ph.D.)--University of Adelaide, Dept. of Psychiatry, 1993

Thesis (M.A.)--University of Texas at El Paso, 2007. / Title from title screen. Vita. CD-ROM. Includes bibliographical references. Also available online.

Thesis (Ph. D.)--University of California, San Diego, 2008. / Title from first page of PDF file (viewed July 24, 2008). Available via ProQuest Digital Dissertations. Vita. Includes bibliographical references (p. 123-127).

Thesis (Ph. D.)--University of Nevada, Reno, 2008. / "December, 2008." Includes bibliographical references. Online version available on the World Wide Web.

Thesis (Ph.D.)--University of Melbourne, The Mental Health Research Institute of Victoria and the Dept. of Psychiatry, 2010. / Typescript. Includes bibliographical references (p. 193-229)

Thesis (Ph. D.)--University of Washington, 2005. / Vita. Includes bibliographical references (leaves 157-174).