This work investigates the development and causes of memory interference effects. Specifically, we measuredproactive and retroactive interference effects in children and adults when learning multiple sets of contingencies, as well asindividuals’ memory binding for the same contingencies. We measured proactive interference by examining memory for asecond set of contingencies after learning a first set, and retroactive interference by examining memory for the first set ofcontingencies after learning the second set. We measured memory binding by presenting participants with partial informationabout each contingency and measuring their accuracy and pattern of errors when asked to identify the completed contingency.Results indicate that both children and adults experienced substantial interference effects, but children were more prone tointerference and substantially worse at memory binding. Additionally, individuals’ memory binding abilities were predictiveof the magnitude of interference effects, suggesting that memory binding is an important mechanism modulating memoryinterference.

Memory and categorization have different goals. The goal of memory is to keep a distinct record of each individualitem. In contrast, categorization aims to treat items as equivalent in some way despite differences. When memory for itemsis the priority, research suggests that adults are able to bind many elements of an experience to form a complex memorystructure, and that such binding may help guard against forgetting due to interference from learning in similar situations. Whencategorization is the priority, however, complex binding structures in memory may impede generalization. In this experimentadults demonstrated robust memory for items after learning one set of categories, but much worse memory for items thatwere categorized differently in a second set. Results suggest that this interference was due to failure to form complex bindingstructures in memory, as a result of selective attention during categorization.

Metacognitive confidence judgments are frequently adopted as a measure of certainty in decision-making tasks, but the mechanisms that underly these judgments have been long debated. In this work, we investigate the effect of the timing of confidence judgments in memory decisions by querying confidence immediately after, with a 3-second delay, or in a separate phase within an associative recognition task. An additional control condition did not probe confidence judgments at all to investigate how metacognitive monitoring may influence the memory decision-making process itself. The results indicate changes in memory performance and response times in conditions where confidence judgments were made, as well as a stronger association between confidence and accuracy when confidence was probed following a 3-second delay. We discuss the implications of these results regarding post- decision processing of metacognitive confidence and the bidirectional relationship between memory and metacognition.