Definition · Plain-language
Retrieval practice
Retrieval practice is the strategy of deliberately bringing information to mind from memory, which strengthens learning more than restudying it.
The step most authors miss
Doing CRediT right? Don’t stop at the statement.
A CRediT statement credits you inside one paper. The recognition CRediT was built for happens when those roles are tied to you, persistently. Sign in with your ORCID — free — and claim your CRediT contributions on casrai.org, the home of the standard. They become a verified, portable part of your identity, not a line that disappears into one PDF.
Free: claim your contributions, then export a journal-ready CRediT statement, schema.org structured data, JATS XML, CSV or BibTeX — and preview your public profile. A membership publishes that profile publicly and verifies the journals you serve.
How retrieval strengthens memory
Each time you successfully retrieve a piece of information, that memory becomes more durable and easier to retrieve again — a phenomenon supported by extensive research in cognitive psychology. Retrieval practice exploits this by making recall, rather than review, the central study activity. Bringing material to mind without looking is more effortful than re-reading, and that effort is precisely what consolidates the memory. The benefit holds across many kinds of material and is one of the most reliable findings in the science of learning.
Practising it well
Effective retrieval practice means generating answers from memory and then checking them against the correct ones. Useful formats include practice tests and past papers, flashcards, short-answer questions and free recall, where you write down everything you can remember about a topic. Feedback matters: checking your recalled answer corrects errors and confirms what you know. Retrieval practice works best when spaced over time and mixed across topics, combining the testing effect with the benefits of distributed and interleaved practice.
Why it beats re-reading
Re-reading and highlighting are popular but largely passive: they create a sense of familiarity that can be mistaken for mastery, an illusion of competence. Retrieval practice exposes what you genuinely know, because you either can or cannot produce the answer unaided. By surfacing gaps and strengthening what is recalled, it directs further study where it is needed. This is why learning scientists consistently recommend self-testing over re-reading as the more powerful use of the same study time.
Key facts
At a glance
- Definition: deliberately recalling information from memory to strengthen learning
- Underlying principle: the testing effect
- Relation to active recall: essentially the same strategy, deliberately applied
- Common formats: practice tests, flashcards, free recall
- Role of feedback: checking answers corrects errors and confirms knowledge
- Contrast: re-reading and highlighting (passive review)
Common misconceptions
What people often get wrong
Often heard: Retrieval practice is only useful as a way to measure what you know.
Actually: Testing does measure knowledge, but retrieval practice uses it as a learning tool, not just assessment. The act of retrieving strengthens the memory itself, so self-testing improves future recall rather than merely revealing current knowledge. Learning, not measurement, is the goal here.
Often heard: Retrieval practice and active recall are different techniques.
Actually: They describe the same core activity: bringing information to mind from memory to learn it. "Active recall" is the common study-skills name and "retrieval practice" the term favoured in research. The distinction is one of vocabulary, not of method.
Often heard: You should only test yourself once you feel you know the material well.
Actually: Retrieval practice is most useful during learning, not just at the end. Testing yourself early — even with errors, followed by feedback — strengthens memory and reveals gaps to address. Waiting until you already feel confident wastes the strongest benefit of the technique.
Going deeper
