The other-race effect: Holistic coding differences and beyond

Abstract

We evaluate claims that the other-race effect in face memory reflects stronger holistic coding of own-race than other-race faces. Considering evidence from a range of paradigms, including the inversion effect, part–whole effect, composite effect, and the scrambled/blurred task, we find considerable inconsistency, both between paradigms and between participant ethnicities. At the same time, however, studies that isolate configural and component feature processing consistently show better featural, as well as better configural, processing of own-race faces, for both Caucasian and Asian participants. These results raise the possibility that the key feature of own-race face processing is not stronger holistic processing per se, but rather more effective processing of all types of face information (featural as well as holistic).

Keywords:

• Other-race effect
• Holistic processing
• Featural processing

This research was supported by the Australian Research Council Centre of Excellence in Cognition and Its Disorders (CE110001021), a grant from the Hong Kong Research Grants Council to WGH (HKU744911), an ARC Professorial Fellowship to GR (DP0877379), and an ARC Discovery Outstanding Researcher Award to GR (DP130102300).

This research was supported by the Australian Research Council Centre of Excellence in Cognition and Its Disorders (CE110001021), a grant from the Hong Kong Research Grants Council to WGH (HKU744911), an ARC Professorial Fellowship to GR (DP0877379), and an ARC Discovery Outstanding Researcher Award to GR (DP130102300).

Notes

¹ In this paper we use the term “race” to refer to visually distinct ethnic groups. Most of the studies we use will refer to one of three such groups; Caucasian (also known as White or European), Asian (specifically in this paper referring to East Asian, e.g., Chinese, Japanese, Korean), and Black (including African–American and Afro-European).

² Note that there remains a vigorous theoretical debate as to whether such perceptual processes are truly face-specific, or can also be applied to other, nonface objects in certain situations; see Wong, Palmeri, and Gauthier (2009) and Robbins and McKone (2007) for more information.

³ In this paper, we use “holistic” to refer to integrated processing of the entire face, and “configural” to refer to processing of spatial relations between face features (Maurer, Le Grand, & Mondloch, 2002). However, in most cases, these conceptions will greatly overlap, and, for example, factors that affect the strength of one will almost always affect the other in similar ways.

⁴ However, this lack of correlation could conceivably also be evidence for weak test reliability, so is inconclusive on the issue of other-race experience mediating the strength of the part–whole task for such faces.

⁵ Koreans showed a significantly weaker composite effect in response times for other-race compared to own-race faces. However, by this measure, Caucasians showed no difference in composite effects between own-race and other-race faces.

⁶ Caution is needed, however, because they did not explicitly assess race categorization. Although this might seem unnecessary given that context effects were found, the effects suggested assimilation to the race context (morphs perceived as same race as the context faces), whereas many previous results would lead one to expect a contrast effect (Ng, Boynton, & Fine, 2008; Ng, Ciaramitaro, Anstis, Boynton, & Fine, 2006; Rhodes, Lie, Ewing, Evangelista, & Tanaka, 2010; Webster, Kaping, Mizokami, & Duhamel, 2004).

⁷ Note that although we have organized the composite and part–whole tasks in the same conceptual category, there are many differences between them that might drive different aspects of holistic processing. In particular, the part–whole task changes the stimulus markedly between study and test (at least in the “part” condition), whereas the composite task usually shows the same type of stimulus at study and test. The empirical evidence on this point is mixed; although one recent study has demonstrated a significant relationship between the part–whole task and the composite task (DeGutis, Wilmer, et al., 2013), another found no relationship between the two measures (Wang et al., 2012). Thus, at the very least it seems clear that these tasks are sensitive to nonidentical aspects of holistic processing, and so behavioural dissociations between them should be expected.