The development of adaptive conformity in young children: effects of uncertainty and consensus

Abstract

Human culture relies on extensive use of social transmission, which must be integrated with independently acquired (i.e. asocial) information for effective decision‐making. Formal evolutionary theory predicts that natural selection should favor adaptive learning strategies, including a bias to copy when uncertain, and a bias to disproportionately copy the majority (known as ‘conformist transmission’). Although the function and causation of these evolved strategies has been comparatively well studied, little is known of their development. We experimentally investigated the development of the bias to copy‐when‐uncertain and conformist transmission in children from the ages of 3 to 7, testing predictions derived from theoretical models. Children first attempted to solve a binary‐choice quantity discrimination task themselves using asocial information, but were then given the decisions of informants, and an opportunity to revise their answer. We investigated whether children's revised judgments were adaptively contingent on (i) the difficulty of the trial and (ii) the degree of consensus amongst informants. As predicted, older but not younger children copied others more on more difficult trials than on easier trials, even though older children also showed a tendency to stick with their initial, asocial decision. We also found that older children, like adults, were disproportionately receptive to non‐total majorities (i.e. were conformist) whereas younger children were receptive only to total (i.e. unanimous) majorities. We conclude that, whilst the mechanism for incorporating social information into decision‐making is initially very blunt, across the course of early childhood it converges on the adaptive learning mechanisms observed in adults and predicted by cultural evolutionary theory. A video abstract of this article can be viewed at http://youtu.be/Qb6JINGYqVk

The when and who of social learning and conformist transmission

“…These results further support the argument that fewer options underestimate the strength of the conformist transmission bias. In neither experiment did we find any evidence of anti-conformity(Morgan et al, 2014)-negative values.…”

“…This theoretical research provides clear predictions about when individuals, both human and non-human, should rely on their individual or asocial experience and when they should deploy one or more social learning strategies, such as conformist transmission (a tendency to disproportionately copy the majority or plurality). By contrast, relatively little empirical research has sought to directly test these models in the laboratory with human participants, though key exceptions with adult participants include McElreath, et al (2005), Efferson, et al (2008), and Morgan, et al (2012) and with children include Wood, Kendal, and Flynn (2013), Haun, Rekers, and Tomasello (2012), Chudek, Brosseau-Liard, Birch, and Henrich (2013), and Morgan, Laland, and Harris (2014). Here, we aim to advance this research program empirically by testing some novel predictions and implications derived from existing theoretical work, as well as to replicate some prior results in new and more diverse populations.…”

A reappreciation of ‘conformity’

“…Indeed, the disproportionate criterion is typically used in the context of naïve individuals setting out to obtain a useful strategy by means of social learning: the context in which conformist transmission is studied (e.g. Boyd & Richerson, 1985;Morgan, Laland, & Harris, 2014). In the conformist transmission context, when individuals are confronted with a balanced population in which only two possible strategies exist, it is assumed that copiers solely rely on social information and thus have a 50% likelihood of obtaining one or the other strategy.…”

Young children communicate their ignorance and ask questions

“…Adopting this approach, Morgan et al (17) asked how far children would be swayed by varying degrees of consensus among their informants when making numerical judgments. Children ranging from 3 to 7 y of age were asked to say which of two displays, each containing 10-30 dots, was numerically greater.…”

“…More specifically, they appraise potential informants along a variety of dimensions, including their familiarity (9,10), their prior accuracy (9,11,12), apparent group membership (13,14), and degree of consensus (15)(16)(17). Such selectivity is likely to facilitate children's acquisition of those beliefs and norms that are representative of their culture.…”