Out of the empirical box: A mixed-methods study of tool innovation among Congolese BaYaka forager and Bondongo fisher–farmer children.

• Pipe cleaner play did not improve BaYaka and Bondongo children's hook task success. • Children innovated novel uses for pipe cleaners outside the experimental context. • Prior relevant experience may improve children's performance during innovation tasks. • Combining observations and experiments improves our understanding of tool innovation. Tool innovation has played a crucial role in human adaptation. Yet, this capacity seems to arise late in development. Before 8 years of age, many children struggle to solve the hook task, a common measure of tool innovation that requires modification of a straight pipe cleaner into a hook to extract a prize. Whether these findings are generalizable beyond postindustrialized Western children remains unclear. In many small-scale subsistence societies, children engage in daily tool use and modification, experiences that theoretically could enhance innovative capabilities. Although two previous studies found no differences in innovative ability between children from Western and small-scale subsistence societies, these did not account for the latter's inexperience with pipe cleaners. Thus, the current study investigated how familiarity with pipe cleaners affected hook task success in 132 Congolese BaYaka foragers (57 girls) and 59 Bondongo fisher–farmers (23 girls) aged 4–12 years. We contextualized these findings within children's interview responses and naturalistic observations of how pipe cleaners were incorporated into daily activities. Counter to our expectation, prior exposure did not improve children's performance during the hook task. Bondongo children innovated significantly more hooks than BaYaka children, possibly because they participate in hook-and-line fishing. Observations and interviews showed that children imagined and innovated novel uses for pipe cleaners outside the experimental context, including headbands, bracelets, and suspenders. We relate our findings to ongoing debates regarding systematic versus unsystematic tool innovation, the importance of prior experience for the ontogeny of tool innovation, and the external validity of experimental paradigms. [ABSTRACT FROM AUTHOR]