The benefit of retrieval practice over restudy has been demonstrated across a variety of materials and settings. However, past research regarding the efficacy of repeated retrieval practice over repeated restudy has failed to consider participant engagement during passive restudy. Over four rounds, participants studied a list of 76 word-pairs using passive or engaged restudy (answering a semantic yes/no question about each pair). Participants who restudied with semantic engagement performed markedly better on a final cued-recall test than those who used passive restudy. Our findings suggest that the benefit of testing in the current literature may be due in large part to widespread use of an inefficient form of restudy.