Thinking for themselves? The effect of informant independence on children's endorsement of testimony from a consensus

Abstract

Testimony agreement across a number of people can be a reassuring sign of a claim's reliability. However, reliability may be undermined if informants do not respond independently. In this case, social consensus may be a result of indiscriminate copying or conformity and does not necessarily reflect shared knowledge or opinion. We examined children's emerging sensitivity to consensus independence by testing whether it affected their judgements in a social learning context. Children ages 5, 6, and 8-9 years (N 5 92), and 20 adults for comparison received conflicting testimony about an unfamiliar country from two consensual groups of informants: An independent group who responded privately and a nonindependent group who had access to each other's answers. We found increasing levels of trust in independent consensus with age. Adults and 8-9-year olds preferred to accept the claims of the independent consensus, whereas 5-year olds favored the claims of the nonindependent consensus and 6-year olds were mixed. Although previous work has shown that children trust a consensus over a lone dissenter as young as 2 years, the developmental shift in this study indicates that children's reasoning about the nature of consensus and what makes it reliable continues to develop throughout middle childhood.conformity, consensus, knowledge, selective trust, testimony | I NTR OD U CTI ONAs adults, we regularly utilize the knowledge of others to gain information about the world that we have not acquired directly. Although we are aware that people's beliefs may not always be accurate, when a number of sources agree on a particular point we feel more confident that their view is correct and can be trusted

The Illusion of Consensus: A Failure to Distinguish Between True and False Consensus

“…More recent work has investigated how and under what conditions children are susceptible to consensus (Burdett et al, 2016; Hu, Buchsbaum, Griffiths, & Xu, 2013; Hu, Whalen, Buchsbaum, Griffiths, & Xu, 2015). Yet little work has examined sensitivity to the quality of consensus, even though this seems directly relevant to a wide range of psychological and sociological questions (but see Einav, 2014, 2018). Consensus is critical not only as we decide which rumors to take seriously but also as we interpret academic articles, news sources, and virtually any kind of information for which there can be a consensus at all.…”

The role of epistemic and social characteristics in children’s selective trust: Three meta‐analyses

“…Although some social characteristics could be perceived as having both epistemic and social components (i.e., having a native or non‐native accent; Corriveau, Kinzler, & Harris, ; having in‐group or out‐group status; Elashi & Mills, ; belonging to a consensus or dissenting; Corriveau, Fusaro, & Harris, ; Schillaci & Kelemen, ), we classified these studies in the social characteristics category. We did this based on evidence that children make social judgments using these characteristics (e.g., children were more likely to make friends with kids who share the similar accent with them; Souza, Byers‐Heinlein, & Poulin‐Dubois, ) and that the social element of behaviors such as belonging to a consensus is likely to be more salient to 3‐ to 6‐year‐old children than the epistemic element (Einav, ). For inclusion in the third meta‐analysis (interaction of characteristics), children had to be presented with two informants who simultaneously differed along epistemic and social characteristics, where one informant was more knowledgeable but had negative social characteristics, whereas the other informant was less knowledgeable but had positive social characteristics (e.g., accurate/unfamiliar‐inaccurate/familiar; e.g., Corriveau & Harris, ).…”

Of Kids and Unicorns: How Rational Is Children's Trust in Testimonial Knowledge?

“…These findings point to a developmental shift toward a preference for more abstract knowledge, as children strengthen their reasoning capacities and ability to adapt to their learning environment. In a similar way, a recent study by Einav () has demonstrated that in receiving conflicting testimony about an unfamiliar country from two consensual groups of informants, 8‐ and 9‐year‐olds favor the claims of the independent group (who respond privately), while 5‐year‐olds prefer to accept the claims of the non‐independent group (who has access to each other's answers). As older children value the independence of group consensus over conformity to group opinions and, therefore, ensure a greater reliability of consensual decisions, they tend to increase their adaptive fit with respect to solving complex cognitive tasks posed by their environment.…”