Investigating the Relation Between Children’s Social Network Properties and the Propensity to Copy or Innovate

• In Event: 3-182 - What are Norms Good For? Outcomes Related to Children’s Early Attention to Social Norms

Integrative Statement

Innovation and social learning are the dual pillars of cultural evolution. Recent theoretical and empirical developments have highlighted the importance of the social environment for both social learning and innovation. Currently, however, we know little about whether individual differences in sociality correlate with individual differences in learning strategy use. This study examined whether, when faced with a novel problem, children’s individual-level social network characteristics influenced their propensity to generate solutions individually or observe others. One hundred and fifty-five seven- to 11-year olds were presented with a novel, tool-use puzzle-box task - the multi-methods box. All participants were asked by an experimenter “Would you like to have a go yourself, or would you to watch me to have a go first?” For those that elected for a demonstration, an adult experimenter performed the same technique four times. Participants then had ten attempts at retrieving sticker rewards from the box. All participants were also presented with an innovation challenge known to be difficult for young children (the Hook Task) and a measure of divergent thinking (the Alternate Uses Task). Social network measures were obtained through ‘social cognitive mapping’, where children named friendship groups from their school class.

Overall, the majority of children elected for social demonstrations, but the tendency to do so decreased with age. Gender differences were also evident, with males more likely than females to elect for no demonstrations. Node-level regressions revealed that network measures did not predict the propensity to elect for social demonstrations or for no demonstrations. However, degree centrality– denoting the number of connections (or ‘friends’) an individual has – was consistently linked to other forms of innovation and exploration. For those who elected for social demonstrations, across their ten subsequent attempts, individuals with many friends (high degree centrality) were more likely to deviate from demonstrated methods than children with lower degree scores. Further, of the children who elected for no demonstrations, those with high degree centrality were less likely to repeat previously used techniques across their ten attempts than children with low degree centrality. Finally, while social network measures showed no relationship with performance on the Hook or Alternate Uses tasks, children who elected for no demonstrations were also more likely than those who opted for social information to both manufacture a tool on the Hook Task, and generated more overall, and more novel, responses for the Alternate Uses task.

This is, to our knowledge, the first study to correlate individual differences in social network measures with children’s learning strategy use and thus sheds light on what may differentiate individual problem solvers from those that look to others. These findings correspond with work from industry settings and nonhuman animal work showing that both companies and individuals with more connections showed greater innovative propensity. Further, these findings provide the first evidence of a relationship between children’s explicit learning strategy choice and their performance on measures of innovative and creative ability. Thus, this study sheds light on what may differentiate individual problem solvers from those that look to others.

Authors

• Bruce Rawlings, Durham University
  Presenting Author

• Rachel Kendal, Durham University
  Non-Presenting Author

• Dan Franks, University of York
  Non-Presenting Author

• Emma Flynn, Durham University
  Non-Presenting Author