Explore versus store: Children strategically trade off reliance on exploration versus working memory during a complex task

How attention and working memory work together in the pursuit of goals: The development of the sampling-remembering trade-off

Most work in the last 50 years on visual working memory and attention has used a classic psychophysical setup: participants are instructed to attend to, or remember, a set of items. This setup sidesteps the role of cognitive control; effort is maximal, tasks are simple, and strategies are limited. While this approach has yielded important insights, it provides no clear path toward an integrative theory (Kristjánsson & Draschkow, 2021) and, like studying a town's walkability by having its college students run the 50-yard dash, it runs the danger of focusing on edge cases. Here, in this theoretical opinion article, we argue for an approach where dynamic relationships between the agent and the environment are understood functionally, in light of an agent's goals. This means a shift in emphasis from the performance of the mechanisms underlying a narrow task ("remember these items!") to their control in pursuit of a naturalistic goal ("make a sandwich!", Land & Hayhoe, 2001). Here, we highlight the sampling-remembering trade-off between exploiting goal-relevant information in the environment versus maintaining it in working memory. We present a dynamic feedback model of this trade-off - where the individual weighs the subjective costs of accessing external information versus those of maintaining it in memory - using insights from existing cognitive control models based on economic principles (Kool & Botvinick, 2018). This trade-off is particularly interesting in children, as the optimal use of internal resources is even more crucial when limited. Our model makes some specific predictions for future research: 1) an individual child strikes a preferred balance between the effort to attend to goal-relevant information in the environment versus the effort to maintain it in working memory, 2) in order to maintain this balance as underlying memory and cognitive control mechanisms improve with age, the child will have to increasingly shift toward remembering, and 3) older children will show greater adaptability to changing task demands.

My Tablet's About to Go Dead! 5- to 6-year-old Children Adjust Their Cognitive Strategies Depending on Whether An External Resource is Reliably Available

There are concerns that reliance on external resources (e.g., information on digital devices) may be harmful to our own internal memory. Here, in a pre-registered study, we investigated how the reliability of an external resource (i.e., whether the information will be available when needed) affects young children's use of it. In our tablet-based Shopping Game, children picked items from a store based on a shopping list. Importantly, the store and the list were not visible simultaneously, but children could toggle between them. In the reliable condition, the list was always available. In the unreliable condition, children were led to believe that the list might disappear. We found that 5-6-year-old children (N = 37) relied more on the list - referring back to it more often and more briefly, and remembering fewer items - when they perceived the list as reliably available (and vice versa, reducing trips to the list by studying it longer, and remembering more, when it was perceived as unreliably available). Nearly all children also identified the reliable condition as easier and preferred. In short, young children not only recognize the opportunity provided by reliably available external resources, but adapt their cognitive effort accordingly.

The Extended Mind in Young Children: Cost-Dependent Trade-Off Between External and Internal Memory

Most work on working memory development has children remember a set of items as well as they can. However, this approach sidesteps the extended mind, the integration of external information with memory. Indeed, adults prefer to use external resources (e.g., lists, models) but will remember more as the cost to access them increases. Here, in our shopping game, we investigated this trade-off in 5- to 8-year-olds. Using a touchscreen, children shopped in a virtual store. Their shopping list and the store were not visible simultaneously but could be toggled. We manipulated access cost by varying a delay (0-4 s) before the list's reappearance. Across three preregistered experiments at two sites (the United States and China, N = 141), a pattern emerged: When it was costlier to do so, children revisited the list less often, studied it longer, and selected more correct items. Also, children recognized the costs, identifying the no-delay condition as easier. Young children showed a cost-dependent trade-off of external-resource use versus working memory.

Don't throw the associative baby out with the Bayesian bathwater: Children are more associative when reasoning retrospectively under information processing demands

Causal reasoning is a fundamental cognitive ability that enables individuals to learn about the complex interactions in the world around them. However, the mechanisms that underpin causal reasoning are not well understood. For example, it remains unresolved whether children's causal inferences are best explained by Bayesian inference or associative learning. The two experiments and computational models reported here were designed to examine whether 5- and 6-year-olds will retrospectively reevaluate objects-that is, adjust their beliefs about the causal status of some objects presented at an earlier point in time based on the observed causal status of other objects presented at a later point in time-when asked to reason about 3 and 4 objects and under varying degrees of information processing demands. Additionally, the experiments and models were designed to determine whether children's retrospective reevaluations were best explained by associative learning, Bayesian inference, or some combination of both. The results indicated that participants retrospectively reevaluated causal inferences under minimal information-processing demands (Experiment 1) but failed to do so under greater information processing demands (Experiment 2) and that their performance was better captured by an associative learning mechanism, with less support for descriptions that rely on Bayesian inference. RESEARCH HIGHLIGHTS: Five- and 6-year-old children engage in retrospective reevaluation under minimal information-processing demands (Experiment 1). Five- and 6-year-old children do not engage in retrospective reevaluation under more extensive information-processing demands (Experiment 2). Across both experiments, children's retrospective reevaluations were better explained by a simple associative learning model, with only minimal support for a simple Bayesian model. These data contribute to our understanding of the cognitive mechanisms by which children make causal judgements.