Retrieval practice
Also known as the testing effect, active recall, test-enhanced learning
Pulling something out of your memory strengthens it far more than putting it in again.
The claim Trying to recall material — and sometimes failing — produces more durable learning than restudying the same material for the same amount of time.
Strong evidence Hundreds of experiments across ages, materials and settings; one of only two techniques Dunlosky's 2013 review rated 'high utility'.
Almost everyone studies by putting information in, and the evidence says learning happens mostly when you take it out. Reading a chapter again, re-listening to a lesson, going back over your highlighted notes — all of these feel like study, and all of them are weaker than closing the book and trying to say what was in it. The act of retrieving changes the memory. It is not a read-out of what you know; it is an edit to it.
The demonstration that convinced the field is Roediger and Karpicke's, in 2006. Students read a short science passage. One group read it four times. Another read it once and then spent the remaining time writing down everything they could remember, with no peeking. Five minutes later the re-readers looked like the clear winners — 83% recall against 71%. A week later they had 40%, and the group that had spent most of its time testing had 61%. The technique that felt worse in the moment left almost half as much again a week on.
The obvious objection is that this must only work for facts you can parrot back. Karpicke and Blunt tested that in 2011 and it did not hold: retrieval beat elaborative concept mapping on questions requiring inference, and beat it even when the final test was itself building a concept map. Whatever retrieval is doing, it is not just drilling surface strings.
The mechanism most researchers favour is that a successful retrieval is a rehearsal of the route, not just the destination. Each time you reconstruct something you strengthen the cues that led you there, so the next reconstruction is easier and can start from further away. Restudying leaves the route untouched — you were handed the answer, so you never travelled.
This is also the technique with the widest gap between what works and what people do. Surveys of study habits find rereading at the top of the list and self-testing well down it, and when students do test themselves they usually stop as soon as they get an item right — which is precisely the point at which further retrieval still pays.
What to do differently
- Close the book first. Before you look anything up, write or say what you already remember; look up only to check and fill gaps.
- Turn headings into questions. A section called 'The causes of the 1848 revolutions' becomes 'What caused the 1848 revolutions?' answered from memory.
- Keep testing past the first success. One correct recall is where most people stop and where the durable gain starts.
- Make the retrieval effortful but reachable — free recall or short answer rather than multiple choice, which lets recognition do the work.
- Space the retrievals out rather than repeating them in one sitting; the two effects compound.
- Get feedback, but not instantly. A short delay before you check the answer is fine and sometimes better.
Limits and trade-offs
- It feels worse than rereading, and it feels worse exactly when it is working. Learners who judge their progress by how fluent study feels will abandon it.
- It is slow to set up. Making good questions is real work, and bad questions drill the wrong thing efficiently.
- It cannot teach you something you never understood. Retrieval strengthens what is there; if the initial explanation did not land, testing yourself on it just rehearses confusion.
- Under high pressure, testing can add anxiety that eats the benefit — the effect is cleanest for low-stakes, self-administered retrieval.
What this means for a language
This is the principle underneath every flashcard app you have ever used, and it explains why passive review of a vocabulary list does so little. It also explains something less obvious: producing the language — saying it, writing it, forcing yourself to recall the word rather than recognising it in a list — is not merely practice for speaking, it is how the vocabulary gets stored in the first place. The one place the finding does not transfer cleanly is pronunciation and listening, where there is no discrete item to retrieve and hours of input do work that recall cannot.
The research
- Roediger & Karpicke (2006), Psychological Science
Repeated testing beat repeated rereading a week later, 61% to 40% — after losing to it, 71% to 83%, five minutes after study.
- Karpicke & Blunt (2011), Science
Retrieval practice beat elaborative concept mapping on inference questions, and beat it even when the final test was a concept map.
- Dunlosky et al. (2013), Psychological Science in the Public Interest
Rated practice testing 'high utility' — one of two techniques out of ten to earn that rating across materials, learners, tasks and settings.
Related
Read next
-
The standard first book. Two of the three authors are the cognitive psychologists behind much of the retrieval-practice literature, so the claims are first-hand rather than second-hand. Heavy on anecdote between the findings, which some readers find padding.
-
The single most useful document here. Ten study techniques rated for utility across materials, learners, tasks and settings. Only practice testing and distributed practice earn 'high'; highlighting, rereading and summarising come out low.
-
The demonstration that made retrieval practice mainstream: rereading won after five minutes and lost badly after a week. The primary source for the 61%-versus-40% figure quoted everywhere since.
-
Closes the obvious objection that retrieval only helps rote memory: it beat concept mapping on inference questions, and beat it even when the final test was itself a concept map.
-
A library of free guides and a well-kept index of the primary papers on retrieval practice, run by a cognitive scientist. Useful precisely because it links the research rather than summarising it away.
-
The best writing anywhere on the part of spaced repetition nobody teaches: what to put on the card. Most flashcard failure is card-writing failure, and this is a craft manual for fixing it.
-
The shortest honest route into the evidence, illustrated throughout. Written by the researchers behind The Learning Scientists, so it doubles as the paper version of their free posters.
-
A cognitive scientist and a classroom teacher writing together, which is why it is unusually specific about what the strategies cost in class time. Aimed at teachers; a self-directed learner has to translate.
-
Thirty-two primary papers summarised and then argued with. The best bridge if you want to move from popular books to reading the literature yourself, and it does not hide where the authors’ own position is contested.
-
Aggressive self-directed projects — an MIT curriculum in a year, a language in three months — turned into nine principles. Strongest on directness and project design; weakest as evidence, since the case studies are selected on success.
-
A cognitive neuroscientist's account, organised around four pillars: attention, active engagement, error feedback and consolidation. The most biologically grounded book on this list, and the most demanding.
-
Takes the same findings as the other books and asks the practical question: what is the smallest change that uses them? Written for university teaching, but the five-minute interventions translate to self-study.
-
The most-enrolled online course ever made, free to audit, about four weeks at a gentle pace. Covers spacing, chunking, procrastination and testing. Its neuroscience is simplified to the edge of metaphor; the study advice is sound.
-
Free, Creative-Commons posters and one-page guides for spacing, retrieval practice, elaboration, interleaving, concrete examples and dual coding. The fastest way to put the evidence on a wall where you will see it.
-
A short free report pairing cognitive-science questions with what the research answers and what it implies for teaching. The second edition adds memory, attention, motivation and misconceptions. Read this if you read nothing else.
-
A long essay on using spaced repetition to read research papers — memory as a tool for understanding rather than a substitute for it. Also the clearest published account of what daily review actually feels like over years.
-
A continuously updated survey of the spacing literature with the arithmetic done: when a card is worth making at all, how review cost compounds, where the published intervals come from. Dense, opinionated, heavily cited.