The effect of familiarity on infants’ social categorization capacity

Abstract

Recent studies indicate that a preference for people from one’s own race emerges early in development. Arguably, one potential process contributing to such a bias has to do with the increased discriminability of own- vs. other-race faces–a process commonly attributed to perceptual narrowing of unfamiliar groups’ faces, and analogous to the conceptual homogenization of out-groups. The present studies addressed two implications of perceptual narrowing of other-race faces for infants’ social categorization capacity. In Experiment 1, White 11-month-olds’ (N = 81) looking time at a Black vs. White face was measured under three between-subjects conditions: a baseline “preference” (i.e., without familiarization), after familiarization to Black faces, or after familiarization to White faces. Compared to infants’ a priori looking preferences as revealed in the baseline condition, only when familiarized to Black faces did infants look longer at the "not-familiarized-category" face at test. According to the standard categorization paradigm used, such longer looking time at the novel (i.e., "not-familiarized-category") exemplar at test, indicated that categorization of the familiarized faces had ensued. This is consistent with the idea that prior to their first birthday, infants already tend to represent own-race faces as individuals and other-race faces as a category. If this is the case, then infants might also be less likely to form subordinate categories within other-race than own-race categories. In Experiment 2, infants (N = 34) distinguished between an arbitrary (shirt-color) based sub-categories only when shirt-wearers were White, but not when they were Black. These findings confirm that perceptual narrowing of other-race faces blurs distinctions among members of unfamiliar categories. Consequently, infants: a) readily categorize other-race faces as being of the same kind, and b) find it hard to distinguish between their sub-categories.

The observation of social interactions helps infants track agents across contexts

“…While infants can form social categories of people (e.g., Ferera et al, 2021;Quinn, Lee, & Pascalis, INFANTS' IDENTIFICATION OF AGENTS ACROSS CONTEXTS 4 2019), people's behaviors are more difficult to predict by their categorical assignment than behaviors and properties of inanimate objects by their kind. One reason for this is that people's behavioral dispositions might be more variable within social categories than object properties within their kinds (Brandone, 2017).…”

Asymmetrical responding to male versus female other‐race categories in 9‐ to 12‐month‐old infants

“…Because infants exhibit greater efficiency in processing female faces, race categorization studies typically have used female faces only, to maximize the likelihood of finding an effect (e.g. Anzures et al, 2010;Quinn et al, 2016Quinn et al, , 2021; for an exception, see Ferera et al, 2021). Hence, data related to whether race and gender interact in infant face processing are scarce.…”

“…In addition, convergent evidence points towards a turning point in other-race face category representation from 9 month of age onwards, where infants begin to group together different types of other-race faces despite clear perceptual differences (Quinn et al, 2016(Quinn et al, , 2018(Quinn et al, , 2020(Quinn et al, , 2021. To what extent this re-organization of category representations has consequences for the perception of face dimensions other than race is unclear, but recent findings suggest that the processing of sub-categories within other-race faces may be especially hindered (Ferera et al, 2021).…”

“…In sum, 6‐month‐olds failed to form discrete categories of own‐ and other‐race faces inasmuch as they formed a category of other‐race faces that excluded own‐race faces on the one hand, but formed a category of own‐race faces that included other‐race faces on the other hand. By 9 months, however, infants formed discrete categories of both own‐ and other‐race faces (Anzures et al, 2010; see also Ferera et al, 2021), with a further re‐organization of other‐race categories into a global outgroup category, lumping together all unfamiliar face race categories (despite clear perceptual differences between them) and excluding own‐race faces (Quinn et al, 2016, 2021). This primitive ingroup–outgroup partitioning of faces has been suggested to constitute a foundation for implicit racial biases later in life (Quinn et al, 2019, 2020).…”

“…These spontaneous preferences disappear in 6‐month‐olds (Fassbender et al, 2016; Liu, Xiao, Xiao, et al, 2015; Quinn et al, 2021), and eventually switch to an other‐race preference in older infants (i.e. 9‐month‐olds: Fassbender et al, 2016; Liu, Xiao, Xiao, et al, 2015; 11‐ to 13‐month‐olds: Ferera et al, 2021). Hence, the similar responding of 6‐month‐olds to White and Asian faces following familiarization with White faces is unlikely to be explained by a preference for own‐race faces that would interfere with the novel category preference, and may find its origins in the categorization process itself.…”