Erring on the Side of Caution: Two Failures to Replicate the Derring Effect

It has been claimed that deliberately making errors while studying, even when the correct answers are provided, can enhance memory for the correct answers, a phenomenon termed the derring effect. Such deliberate erring has been shown to outperform other learning techniques, including copying and underlining, elaborative studying with concept mapping, and synonym generation. To date, however, the derring effect has only been demonstrated by a single group of researchers and in a single population of participants. This paper presents two independent, preregistered replication attempts of the derring effect. In Experiment 1, participants studied 36 term-definition concepts in a within-subjects, laboratory study. On error-correction trials, participants were presented with a term-definition concept and were asked to generate an incorrect definition before correcting it. Error-correction trials were compared to copy trials, where participants simply copied the term-definition concepts and underlined the key concepts. Experiment 2 was an online study in which participants studied trivia facts using a similar protocol. Memory for the studied facts was then tested either immediately (Experiments 1 and 2) or after 2 days (Experiment 1). Unlike the original demonstrations of the derring effect, cued-recall performance did not significantly differ between the error-correction and copy conditions, and the Bayes factors provided moderate support for the null hypothesis in both experiments. We discuss potential explanations for our findings and consider them in relation to key theories and the broader literature on the role of errors in learning.