Class, Race, and the Face: Social Context Modulates the Cross-Race Effect in Face Recognition

Abstract

The current research investigates the hypothesis that the well-established cross-race effect (CRE; better recognition for same-race than for cross-race faces) is due to social-cognitive mechanisms rather than to differential perceptual expertise with same-race and cross-race faces. Across three experiments, the social context in which faces are presented has a direct influence on the CRE. In the first two experiments, middle-class White perceivers show superior recognition for same-race White faces presented in wealthy but not in impoverished contexts. The second experiment indicates this effect is due to the tendency to categorize White faces in impoverished contexts as outgroup members (e.g., "poor Whites"). In the third experiment, this effect is replicated using different ingroup and outgroup categorizations (university affiliation), with ingroup White faces being recognized better than outgroup White faces. In line with a social-cognitive model of the CRE, context had no influence on recognition for cross-race Black faces across the three experiments.

Interracial contexts debilitate same-race face recognition

“…Insofar as CR faces are likely to elicit strong category activation across contexts, manipulating the task order here should have little effect on CR memory. This is consistent with past research indicating that contextual manipulations have little impact on CR recognition (Shriver, Young, Hugenberg, Bernstein, & Lanter, 2008), and that at baseline, social categorization is stronger for CR than SR targets (Levin, 1996(Levin, , 2000.…”

Removing the own-race bias in face recognition by attentional shift using fixation crosses to diagnostic features: An eye-tracking study

“…In the present study, participants had to fixate upon them for 150 ms in order for the trial to commence. This is not typical practice in other studies: The fixation cross is a cue to the start of the trial and in some cases disappears before the face appears (e.g., Goldinger, et al, 2009) sometimes being replaced with a context image (Shriver, Young, Hugenberg, Bernstein, & Lanter, 2008). Fixation crosses are also typically on screen for longer in face recognition experiments than in the present study (for example, 500 ms in Wiese, Stahl, &Schweinberger, 2009, and2 s in He, Ebner, &Johnson, 2011).…”

On the epistemic costs of implicit bias

“…36 Thanks to Brendan Dill for pointing out the importance of making this distinction, and for suggesting the second formulation. 37 For an interesting discussion of the interaction of race and class in CR deficit, see Shriver et al (2008).…”