Retrieval practice and rereading can rank differently on immediate and delayed recall

Roediger and Karpicke's 2006 experiments compared repeated study of prose passages with free-recall testing. At a five-minute final test, repeated study produced better recall; at delayed tests after two days or a week, prior testing produced greater retention. Repeated study also increased learners' confidence.

The important distinction is between performance immediately after exposure and retention after a delay. Feeling fluent while looking at material is not itself a delayed memory test.

Illustrative application: after reading a short explanatory passage, close it and write the central ideas from memory. Compare the attempt with the passage, correct omissions, and check recall again later. That added correction step is a practical adaptation, not a condition tested in the cited experiments, which used recall tests without feedback.

This matters when preparing material that must be remembered beyond the current session. Do not assume that the same result applies unchanged to every learner, skill, exam format, or complex professional task. The experiments concerned particular prose-learning conditions, not a universal ranking of educational methods.

Evidence status: source-based synthesis. Examples are illustrative; no new experiment or independent outcome verification is claimed.