Third-party preferences for imitators depend on the expectation of egalitarian prosociality

• In Event: When it’s okay not to conform: The development of a context-sensitive concern for conformity

Abstract

By one year of age, infants share adults’ preference for social partners who align with their actions and choices (e.g. Mahajan & Wynn, 2012; Meltzoff, 1990; Carpenter et al., 2013). Infants also prefer individuals they observe imitating others, from as early as 4 months of age (Powell & Spelke, 2018; Thomas et al., 2020). But why do infants prefer those they observe imitating others? When participants themselves are imitated, their subsequent preferences are often ascribed to homophily, a liking for similar others (Chartrand & Bargh, 1999). This explanation does not account for third-party preferences, especially given that these preferences do not extend to individuals who are targets of imitation over those who are not (Powell & Spelke, 2018; Thomas et al., 2020).

I suggest, instead, that infants’ preference for imitators is based on the inference that imitators are willing and able to share others’ utility, a central component of helping and cooperation (Hamlin et al., 2013). These attributes are appealing, particularly if the observer expects an imitator’s tendency toward prosocial alignment to extend more broadly, beyond the observed target. Infants and young children hold more egalitarian expectations of benevolence than older children (Ramenzoni & Liszkowski, 2016; DeJesus et al., 2014). Thus, one prediction of this account is that amongst older children, stronger expectations of parochial, rather than egalitarian, behavior will correlate with weaker third-party preferences for imitators (i.e. individuals who align with their social partners), despite showing similar expectations of affiliation between those individuals and the direct targets of their alignment.

We tested this prediction in a two-part experiment with 3-to-11-year-old children (N=81). First, participants watched videos in which a model interacted separately with two social partners. The model initiated each interaction by performing a modified sign language gesture. One partner (the imitator) responded with the same hand gesture, while the other partner (the non-imitator) responded with a different hand gesture in response. Participants were asked to indicate which of the two responders they thought the model liked more, and which of the two responders they liked more themselves. Then, children heard a short story in which one character, Sally, had two cookies that she did not want. Sally had the options of 1) giving both cookies to her best friend, Anne, (the parochial outcome) or 2) giving one cookie to Anne and one cookie to an unfamiliar child, Megan, (the egalitarian outcome). Participants were asked to predict which option she would choose.

The results supported our hypothesis. A logistic regression model replicated the finding that predictions of egalitarian sharing behavior decline with age (b=-0.24, z=2.27, p<.05). Splitting children according to their sharing predictions also revealed that children who expect egalitarianism show a stronger preference for the imitator ( 2=4.04, p<.05; Figure 1). Expectations of the target’s preference did not vary by sharing prediction ( 2=0.21, p>.6). These findings inform our understanding of how children value alignment and conformity, as well as how their social expectations inform their social preferences more broadly.

Author

• Lindsey Powell, University of California - San Diego
  Presenting Author