Preschoolers' Persistent Overconfidence in Their Recall Memory

Preschoolers' ability to predict picture recall was examined. They first studied ten pictures, predicted how many they would recall, and then were tested on their memory for the pictures. This study-prediction-test trial was repeated multiple times, with new pictures on each trial. In Experiment 1, children's overconfidence persisted despite repeatedly recalling fewer items than they had predicted across each of three trials. In Experiment 2, the wishful thinking hypothesis was evaluated as an explanation for this persistent overconfidence. Specifically, children were asked to predict their own performance and the performance of another child on the same memory task. Even after five trials, children were overconfident and this overconfidence was just as strong in predictions of another child's recall. Experiment 3 evaluated performance monitoring as yet another explanation for children's persistent overconfidence. The self prediction tasks were performed by two groups, think-back and control. Children in the think-back group made postdictions about their previous recall performance prior to making a prediction for the next trial. Children in the control group did not make a postdiction. Although, children's postdictions were fairly accurate, they did not use this monitoring information in the formulation of their next prediction. Children in both groups were persistently overconfident across five trials. In Experiments 2 and 3, individual differences in vocabulary size, understanding of cardinality, and working memory were also examined as potential explanations for patterns of prediction performance, recall performance, and postdiction accuracy.