Loading adults' working memory made them explore like five-year-olds

The manipulation is the number stream. Between conditions the adults' knowledge of the decision game did not change; what changed was how much working memory was left free ^[7]. Within adults, that is an intervention, so the causal reading holds there.

The step to children is a different kind of claim. No one loaded or relieved a five-year-old's working memory in this design, and the comparison runs between an induced adult pattern and a measured child pattern ^[8]. That fits Sloutsky's account, in which the immaturity of the working memory system drives early exploration ^[5]. A direct test would vary working memory load in the children themselves.

The phys.org account calls the two patterns nearly identical and gives no sample size and no effect size for the comparison ^[8][19]. The underlying data is not new. It is work the lab previously reported in the Journal of Experimental Psychology: General, revisited here in a review ^[2][18].

The benefit half of the argument rests on reasoning about environments. "Because children explore, they may not immediately optimize the benefit of the best option, but they also won't lose out if something in the environment changes," Sloutsky said ^[14]. The illustration in the source is fuel prices: you settle on the two cheapest stations you found, keep going back, and miss the stations that cut their prices later ^[16]. The thing this does not tell you is whether the exploring child knows more a week later. The work reports where attention went and which options were chosen, not learning outcomes ^[20].

That gap matters for anyone reading this as guidance about teaching. The children studied were 3 to 8 years old ^[3], the outcome was attentional and choice behavior ^[20], and instruction is not discussed ^[20]. So the idea that designs optimized for selective focus suppress the exploratory mechanism is not tested here in either direction.

The source holds the behavior as two things at once: the output of a system that is not finished ^[5], and, in phys.org's framing, something with an evolutionary purpose that gets young learners to absorb as much about the world as they can ^[21]. Sloutsky notes in the paper that "children tend to explore broadly, and, despite its costs, exploration confers important benefits, especially early in development, when they know little" ^[9]. He hopes studies like this remind parents that kids "are annoying and destructive for a reason" ^[17].