Innovation in children is typically studied by examining their capacity to create novel tools. However, innovation also involves recognising the future utility of a solution. Across two experiments, we examined children's capacity to recognise and construct a tool for future uses. Experiment One presented 3- to 5-year-olds (N=55) with a future-directed problem-solving task. When given a tool construction opportunity in anticipation of returning to the task, only 5-year-olds made the correctly shaped tool above chance levels. Experiment Two assessed 3- to 7-year-olds' (N=92) capacity to build a tool with future, as well as present, utility in mind. Age was positively associated with constructing a tool of greater utility than necessary to solve the present task. Children's propensity to construct longer tools was associated with their capacity to prepare for two alternative possibilities on a secondary task, suggesting performance on our innovation task reflects emerging future-oriented cognition.

Recognising the future utility of a solution is fundamental to our capacity for innovation. However, developmental research has thus far focused on children's capacity to create solutions, rather than recognise existing solutions with ongoing utility. We examined children's capacity to retain and share a solution that would be useful again in the future. Across two rooms, 4- to 9-year-olds (N=83) were given a series of time-based tasks which could be solved by building and using a tool. When given the opportunity to transport a tool between the first and second rooms, children from age 6 onwards retained the tool that would be useful again above chance levels. When subsequently asked to secure a solution for another child, only 8- to 9-year-olds chose this tool above chance. Positive age-partialled correlations between children's retaining and sharing behaviours suggest that these behaviours may reflect a common underlying capacity for recognising future utility.

Across two experiments, we presented children (N = 168; 3 to 6 years) with a memory task in which three targets were hidden sequentially before a search period. In both experiments, younger children were significantly more likely to first search for the last target hidden (in line with the recency effect), whereas older children were significantly more likely to first search for the first target hidden (in line with the primacy effect). In a separate test phase where some but not all targets were were externally marked, younger children were biased towards selecting the marked target first, whereas older children were significantly more likely to search for unmarked targets before marked targets (thus reducing the time spent maintaining the location of the unmarked targets in memory). These results indicate marked shifts in young children's ordered memory recall tendencies, much earlier in development than suggested by previous research.

Humans have increasing opportunities to offload internal cognitive demand, such as by setting reminders to aid future memory performance. Here, we examine how children begin to balance mind and world: weighing up when to offload cognition and when to rely on their unaided capacities. Australian children aged 6 to 9 years (N = 120) were tasked with remembering the locations of 1, 3, 5, and 7 targets hidden under 25 cups. In the critical test phase, children were provided with a limited number of ‘tokens' to distribute across trials, which they could use to mark target locations and assist future performance. Following the final search period, children were invited to evaluate and adjust their initial allocation. Results showed that 8- to 9-year-olds prospectively allocated proportionately more tokens to difficult trials, whereas 6- to 7-year-olds did so only in retrospect. Throughout childhood, humans become increasingly adept at balancing internal and external cognition.