Testing Effect

Even without feedback, test practice enhances delayed performance compared to study practice. However, students often cram by studying just before an exam rather than testing themselves. Cramming appears irrational, but this may depend on whether items are initially retrievable. In this study, an initial test determined item retrievability. Retrievable or non-retrievable items were subsequently presented for repeated study or test. Collapsing across item retrievability, we obtained a crossover interaction of retention interval and practice type. For retrievable items, however, the crossover interaction was quantitatively different with a smaller study benefit for an immediate test and a larger testing benefit after delay. For non-retrievable items, there was a large study benefit only for an immediate test, no testing benefit at all, and even study practice followed by immediate testing failed to produce a testing benefit after delay. These results indicate that for initially nonretrievable material, cramming may be a rational strategy.