Biased Stereotypical Responding in Infants

• In Event: At the Intersection of Social Systems, Perception, and Representation: Going Beyond Perceiving Faces and Languages

Abstract

Infants are primarily exposed to own-race faces and have little exposure to other-race faces. Recent theoretical advances suggest that asymmetrical face-race experiences may lead to biased perceptual and social processing in infancy (the perceptual-social linkage hypothesis, Lee et al., 2017). In accord with this hypothesis, recent studies reported that infants tended to learn from own-race individuals over other-race individuals (e.g., Singh et al., 2019; Xiao et al., 2017). The present study further investigated the early emergence of social biases by studying how infants’ experience with different races of face biases their responding. In a series of studies, infants first learned the reliability of one or several informants (see Figure 1). Here, we indexed reliability by the correctness of signaling the location of an animal with informant gaze. Then, we examined gaze following of a new person of the same race as the informant(s), whose reliability was never revealed to infants. If infants engage in stereotypical (i.e., categorical) responding, their gaze following of the new person will be impacted by the reliability of the prior informants. For example, after learning that the informant is unreliable, infants would not follow the gaze of the new person. Previous child and adult studies suggest a tendency to process other-race faces at a categorical level, but own-race faces at an individual level. Thus, we predicted that infants would exhibit stronger stereotypical responding about other-race individuals compared to own-race individuals. In Experiments 1 & 2, infants learned the reliability of one informant. Asian (Experiment 1, 8-13 months, N=108) and Caucasian infants’ (Experiment 2, 9-14 months, N=49) gaze following of a new person was not affected by the reliability of prior informants if the informants were own race. However, infants learned the reliability of other-race informants, and their tendency to follow the gaze of a new person was significantly affected by the learned reliability (Figure 2A & 2B). We replicated these findings in Experiment 3, where Asian infants learned from four different informants (Figure 2C, 7-11 months, N=156). The convergent findings indicated that infants were likely to engage in stereotypical responding with individuals of an unfamiliar race, but not with own-race individuals. Moreover, Asian infants in the same age range were shown to discriminate between faces of own and other races equally well (Experiment 4, 8-12 months, N=34, Figure 2D), suggesting that the results of Experiments 1-3 reflected biased stereotypical responding, rather than biased face discrimination. Together, the findings suggest that experience with different races of faces influenced how infants generalize learned reliability to other individuals from the same race. Limited experience with other-race faces may lead infants to perceive the behaviour of other-race members categorically leading to greater stereotypical responding. By the same reasoning, infants may perceive the behaviour of own-race members as reflecting that of individuals, thereby limiting their stereotypical responding. In this way, asymmetrical experience with different races may lead infants to respond to other-race members as a collective group (i.e., people) with members of own-race members responded to as persons.

Authors

• Naiqi G Xiao, McMaster University
  Presenting Author

• Shaoying Liu, Zhejiang Sci-tech University

• Paul Quinn

• Kang Lee, University of Toronto