Abstract
Disjunctive reasoning allows us to infer properties of unobserved objects from currently observed objects, formalized by P or Q, not P, therefore Q. We asked whether children and adults can explicitly use disjunctive reasoning to track occluded objects in an attentionally-demanding visual working memory task. Thirty-six 4-7-year-olds and 6 adults viewed animated arrays of 2-3 virtual “cards” depicting images, which were then occluded by opaque “cups”. In the Face-up block, all images were visible during the brief encoding period. In the Disjunctive block, one of the cards was face down during encoding. During the maintenance period, the occluded cards swapped locations (2-3 times). Two images appeared above one of the cups, and participants were asked to select which image was hidden under that cup (chance=.5). In Disjunctive trials, the location of the face-down card was probed on half of the trials, so participants had to use disjunctive reasoning (P or Q, not P, therefore Q) to respond accurately. Adults were above chance on all trials (ps<.001), and there was no difference in their performance on Face-up vs. Disjunctive trials, suggesting they successfully deployed deductive reasoning. However, children performed significantly worse in the Disjunctive versus the Face-up block (p = .033). While children were above chance on all trials in the Face-up block (ps <.006), children’s performance in the Disjunctive block was only above chance on the easiest trial type (2 objects, 2 swaps, p=.006). Children’s performance in Disjunctive trials was higher when the target had been visible during encoding (p=.033), suggesting they failed to use disjunctive reasoning. Children’s use of the disjunctive strategy increased with age (p=.002). Our results suggest disjunctive reasoning is available as a strategy to improve visual working memory performance in adults, but the development of this ability is extremely protracted.