Pretesting versus posttesting: Comparing the pedagogical benefits of errorful generation and retrieval practice.
Level 2 - randomized trial
Series of five randomized experimental laboratory trials.
PubMed 33793291 · doi:10.1037/xap0000345
What was done
Across five experiments (total n = 1,573), participants studied expository text passages paired with either a pretest (attempted before studying) or a posttest (attempted after studying), alongside a no-test control. Practice tests used multiple-choice (Experiments 1–5) or cued-recall formats (Experiments 2–4), with or without correct-answer feedback (Experiments 3–4). Memory was evaluated using a criterial test administered after a 5-minute or 48-hour retention interval. Experiment 5 specifically examined passage content processing.
What was found
The abstract reports no exact test scores, percentages, or effect sizes. It states that both pretesting and posttesting significantly improved criterial test memory relative to no-test controls, with pretesting producing higher overall retention scores than posttesting. This advantage persisted across both test formats (multiple-choice and cued recall), across retention intervals (5 minutes and 48 hours), and regardless of whether feedback was provided.
Why it matters
While retrieval practice (posttesting) is widely used in pedagogy, these findings demonstrate that errorful pretesting before study can match or exceed posttesting benefits by enhancing subsequent processing of learning materials.
Limits
The abstract provides no descriptive statistics, effect sizes, or confidence intervals. The study evaluates immediate and short-term laboratory memory for expository text passages (up to 48 hours), leaving long-term educational retention in authentic classroom environments unmeasured.