Culture-Gene Coevolutionary Theory and Children’s Selective Social Learning

Abstract

Human cognition is unique in the degree to which it is shaped by social learning and cumulative cultural evolution. To learn efficiently children cannot just passively absorb all the information others provide, they need to be sensitive to when, if, and who is a good source of information. Several lines of theoretical and empirical inquiry are probing the cognitive mechanisms that underlie humans' flair for cultural learning. Among these, Culture-Gene Coevolutionary (CGC) theory focuses on the evolutionary dynamics our emerging cultural species faced and the learning biases that likely resulted from these selection pressures. CGC theory specifies a suite of hypotheses about which learning biases most effectively extract adaptive information. Here we review the intersection between these and developmental psychologists' recent empirical insights into children's social learning. We focus on two types of biases: 'relative model biases' which help learners identify which models are likely to possess information that is applicable for them given the particular social groups to which they belong; and 'absolute model biases' which help learners select those models most likely to possess objectively better information. In light of recent developmental findings we advocate CGC theory as a useful framework for organizing, understanding, and generating research on children's selective social learning.

Unselective Overimitators: The Evolutionary Implications of Children's Indiscriminate Copying of Successful and Prestigious Models

“…Success (a difference between models in outcomes or payoffs) and prestige (a bias to attend to and imitate models who have received preferential attention from others) have been theoretically predicted to be especially information‐rich cues for young culture learners (Boyd & Richerson, ; Henrich & Gil‐White, ; Henrich & McElreath, ; Richerson & Boyd, ), and these predictions have born out in other domains (e.g., Chudek et al., ; Olson et al., ; Wood et al., ). In contrast, our results here suggest that overimitation is insensitive to success and prestige differences, despite children discriminating between successful and prestigious models when answering explicit questions.…”

“…Though our culture‐pioneering ancestors' patterns of selective learning are the focus of a great deal of evolutionary theorizing (e.g., Boyd & Richerson, ; Cavalli‐Sforza & Feldman, ; Chudek, Muthukrishna, & Henrich, ; Henrich & Henrich, ; Laland & Brown, ; Mesoudi, ; Richerson & Boyd, ), they can no longer be directly observed. However, adaptation claims can be tested by comparing the predictions of evolutionary models to the selective dispositions of contemporary children (for a review, see Chudek, Brosseau‐Liard, Birch, & Henrich, ).…”

Group-level traits can be studied with standard evolutionary theory

“…Humans seem built to learn from one another, and most differences between groups of humans appear to be largely the result of learning rather than genotype (Richerson & Boyd 2005;Richerson et al, in prep). This statement is well supported by the empirical literature on human development (Chudek et al 2013;Harris 2012;Richerson et al, in prep) and cultural psychology (Nisbett et al 2001;Nisbett & Miyamoto 2005). Even the mother-infant attachment relationship, which is for most humans their first participation in a group-level trait and is in many ways a human universal (Grossman & Grossman 2006), is influenced by the culture and past experiences of the mother (van Ijzendoorn et al 2006).…”

Group-level expression encoded in the individual

“…Humans seem built to learn from one another, and most differences between groups of humans appear to be largely the result of learning rather than genotype Baldini et al, under review). This statement is well supported by the empirical literature on human development (Chudek et al 2013;Harris 2012;Baldini et al, under review) and cultural psychology (Nisbett et al 2001;Nisbett & Miyamoto 2005). Even the motherinfant attachment relationship, which is for most humans their first participation in a group-level trait and is in many ways a human universal (Grossman & Grossman 2006), is influenced by the culture and past experiences of the mother (van Ijzendoorn et al 2006).…”