Research on "the testing effect" routinely conflates direct and forward testing effects: A meta-analysis of testing effects with free recall.
Level 1 - systematic review of randomized trials
Meta-analysis of experimental studies evaluating retrieval practice paradigms.
PubMed 42258276 · doi:10.1037/xlm0001634
What was done
A meta-analysis of free-recall memory studies was conducted to determine how frequently experiments isolate the direct testing effect (improved retention of tested material) versus mixed designs susceptible to both direct and forward testing effects (enhanced learning of subsequent material). The authors evaluated effect size differences between direct and mixed designs, examined mixed subtypes where forward effects theoretically increase, decrease, or moderately alter outcomes, and performed separate moderation analyses.
What was found
The majority of analyzed effects (66%) came from mixed studies that conflate direct and forward testing effects, while only 34% isolated direct testing effects. Mixed studies yielded systematically larger effect sizes than direct studies. Effect sizes also varied across mixed study subtypes depending on whether the forward testing contribution was predicted to increase, decrease, or have an intermediate effect on the outcome. Moderation analyses showed differing moderator patterns between the direct and mixed study populations. Exact numerical effect sizes, variance estimates, and study counts were not reported in the abstract.
Why it matters
Much of the existing literature on retrieval practice does not reflect a single coherent memory mechanism, but rather an unmodeled mixture of direct memory consolidation and forward learning facilitation.
Limits
The abstract does not provide exact effect size values, confidence intervals, or the total number of included studies and participants. The analysis was limited to free-recall paradigms and may not directly generalize to other testing formats (such as cued recall, multiple choice, or recognition).