Are social norms and reciprocity necessary for early helping?

Cortes Barragan and Dweck (1) present evidence that social interactions influence subsequent helping behavior in young children: 2-y-olds are more likely to help an adult when they previously engaged in interactive play, compared with when the child and an adult played in parallel or never interacted. This study shows that helping is a rich social interaction embedded in the context of social experiences more generally. However, the authors also argue that these results challenge the hypothesis that early helping may have a biological basis, and does not emerge as a result of socialization alone (2–4). In particular, Cortes Barragan and Dweck (1) suggest that because this social priming influences helping rates in their study, early helping in young children could result from the values and practices that are subtly communicated in a social situation.

The premise of their argument is that behaviors that are based upon biological predispositions are not malleable or open to social influence. However, the “natural altruism hypothesis” that they aim to refute makes no such claim (2). Rather, this hypothesis states that socialization practices and cultural norms are not foundational for early helping behaviors, even though they nonetheless shape children’s altruism over development. In fact, manipulating the social experience in an experimental test does not address this core question about the developmental factors that initially give rise to a basic competency for helping. The critical test is whether helping emerges in the absence of any relevant socialization experiences and adoption of human social norms. Studies with nonhuman primates provide this critical test, and our discovery that chimpanzees also demonstrate some helping behaviors supports the hypothesis that the adoption of social norms is not a necessary prerequisite for helping behaviors to emerge (2, 3). This second line of evidence is not considered in Cortes Barragan and Dweck’s (1) argument against the natural altruism hypothesis.

A second claim by Cortes Barragan and Dweck is that their experimental manipulation—where the adult either played with the child or only superficially interacted with the child—increases prosocial responses because children apply a “principle of economic reciprocity” (1). This principle is interpreted as evidence against the hypothesis that young children progress from more naïve to more selective cooperators, incorporating reciprocal strategies and social norms into their prosocial actions over development (2). The first direct test of this selectivity hypothesis comes from a study in which 2.5-y-olds were as cooperative toward someone who had cooperated with them as they were toward someone who was selfish. It was not before 3.5 y that children cooperated more with cooperators than defectors (5). The issue is that Cortes Barragan and Dweck (1) use a very broad notion of reciprocity, such that it is difficult to disentangle traditional definitions of reciprocity (exchanging costs and benefits) from social engagement more broadly (passing a toy back and forth). Thus, a more parsimonious account of their finding is that children respond more positively when individuals actually engage with them socially and let them play with the same toys than when the adults have children play by themselves. This result may not reflect “far subtler forms of reciprocity” (1); indeed, it may not be reciprocity at all.