Children Overimitate Adults and Peers More Than Puppets

Abstract

<jats:title>Abstract</jats:title><jats:sec><jats:label /><jats:p>Researchers commonly use puppets in development science. Amongst other things, puppets are employed to reduce social hierarchies between child participants and adult experimenters akin to peer interactions. However, it remains controversial whether children treat puppets like real‐world social partners in these settings. This study investigated children's imitation of causally irrelevant actions (i.e., over‐imitation) performed by puppet, adult, or child models. Seventy‐two German children (<jats:italic>Age<jats:sub>Range</jats:sub></jats:italic> = 4.6–6.5 years; 36 girls) from urban, socioeconomically diverse backgrounds observed a model retrieving stickers from reward containers. The model performed causally irrelevant actions either in contact with the reward container or not. Children were more likely to over‐imitate adults’ and peers’ actions as compared to puppets’ actions. Across models, they copied contact actions more than no‐contact actions. While children imitate causally irrelevant actions from puppet models to some extent, their social learning from puppets does not necessarily match their social learning from real‐world social agents, such as children or adults.</jats:p></jats:sec><jats:sec><jats:title>Research Highlights</jats:title><jats:p><jats:list list-type="bullet"> <jats:list-item><jats:p>We examined children's over‐imitation from adult, child, and puppet models to validate puppetry as an approach to simulate non‐hierarchical interactions.</jats:p></jats:list-item> <jats:list-item><jats:p>Children imitated adults and child models at slightly higher rates than puppets.</jats:p></jats:list-item> <jats:list-item><jats:p>This effect was present regardless of whether the irrelevant actions involved physical contact to the reward container or not.</jats:p></jats:list-item> <jats:list-item><jats:p>In our study children's social learning from puppets does not match their social learning from human models.</jats:p></jats:list-item> </jats:list></jats:p></jats:sec>