Learning science
Retrieval practice: why testing yourself beats studying
Being asked to recall something does more for memory than studying it again for the same time. The effect is large, well replicated, and almost universally ignored.
5 min read
Key takeaways
- Students who were tested on material remembered substantially more a week later than students who restudied it for the same length of time.
- The students who restudied felt more confident. Confidence and retention pointed in opposite directions.
- Retrieval has to be effortful to work. Looking at the answer first turns it back into rereading.
- A major review rated practice testing as one of only two high-utility study techniques, and rated rereading and highlighting as low utility.
- Getting it wrong is productive, as long as you then see the right answer. Failed retrieval still improves later learning.
Testing yourself on material does more for long-term memory than studying it again for the same amount of time. In the landmark experiment, students who read a passage once and then took recall tests remembered substantially more a week later than students who read it four times - while the rereading group felt more confident about how well they had learned it.
The experiment
Roediger and Karpicke, 2006. Students studied a short text, then either studied it again or took a test on it with no feedback. Memory was measured after five minutes, two days, or a week.
Two findings, and the second is the important one.
After five minutes, the repeated-study group did slightly better. Cramming works, for about five minutes.
After a week, the tested group did substantially better. The advantage reversed, and it reversed hard.
The students' own predictions went the other way. Those who restudied were more confident they would remember. They were wrong, and they could not tell.
Why confidence misleads
Rereading makes material feel fluent. Your brain reads that ease of processing as evidence of knowing. Retrieval feels difficult - you sit there trying to drag something out and it comes slowly - and that difficulty feels like failure.
This is why students persist with the techniques that work least well. The 2013 Dunlosky review assessed ten techniques and rated only two as high utility: practice testing and distributed practice. Highlighting, rereading and summarising - overwhelmingly the most popular - were rated low.
A technique that feels worse and works better is a hard sell to anyone with an exam in three days.
Why retrieval works
It strengthens the route, not just the file. Remembering is a skill of finding, and each retrieval makes the path easier to travel.
It is diagnostic and immediate. You discover exactly what is missing while there is still time to fix it, rather than during the exam.
It encodes the context of recall. You practise producing the information under conditions resembling the ones where you will need it.
Failing still helps. An unsuccessful attempt followed by the right answer produces better learning than studying alone. The attempt primes the correction.
How to use it
Close the book first. The one rule. Looking at the answer before trying is rereading in a costume.
Blank page recall. Finish a topic, take a blank sheet, write everything you remember. Then compare with your notes and note what was missing. Ten minutes, enormous value.
Use flashcards honestly. Attempt the answer out loud before turning the card. Keep a card in rotation until you have got it right on several separate days, not until you have got it right once.
Practice questions before the worked examples. Attempt, fail, then read the solution. The common order - read the worked example, then do the similar question - is mostly pattern-matching.
Space the retrievals out. Retrieval and spacing compound: the same number of attempts spread across days beats them bunched together. More on spacing here.
What it is not
It is not testing for assessment. There is no mark and nobody sees it. Many students resist because testing has only ever happened to them as judgement, and recasting it as a study tool takes some deliberate reframing.
It is also not a substitute for understanding something in the first place. Retrieval practice makes things durable; it does not make an idea you never grasped suddenly make sense. The right order is understand, then retrieve repeatedly.
Five ways to practise retrieval, ranked
Roughly in order of how much they do per minute spent.
- Blank page recall. Finish a topic, take a blank sheet, write everything you can remember. Compare with your notes and mark what was missing. Ten minutes; nothing else is this efficient.
- Explain it out loud from memory. Self-explanation and retrieval at once. Two to three minutes per concept, and it exposes the joins between ideas rather than the facts.
- Practice questions before the worked example. Attempt, fail, then read the solution. The usual order - read the example, do the similar question - is pattern matching.
- Flashcards, used honestly. Say the answer out loud before turning the card, and keep a card in rotation until you have had it right on several separate days.
- Past papers under time. Closest to the real conditions, and the most expensive in minutes, so save them for when the content is mostly in place.
What is conspicuously absent: rereading, highlighting, copying out notes, watching the lecture again. All four feel productive. All four were rated low utility in the review of the evidence.
Why it feels worse and works better
The sensation of difficulty during retrieval is not a sign that it is going badly. It is the mechanism. Memory is strengthened by the effort of finding something, so a struggle that ends in recall does more than a smooth reread ever could.
This creates a genuine problem for students: the techniques that work feel like failure in the moment and pay off weeks later, while the techniques that do not work feel like progress immediately. Anybody choosing how to spend an evening is choosing against the evidence unless they have been told this explicitly.
For a parent or a teacher
The useful move is to change what a "test" means in your household. If the only tests a child has ever met were judgements, retrieval practice will feel like one. Framed as a study tool - no mark, nobody sees it, the point is finding the gaps while there is time - most students take to it, because it produces an obvious and immediate list of what to do next.
A simple weekly habit: one slot where your child writes or says everything they remember from the week, with nothing in front of them. Then they check. The list of what was missing is the week's revision plan, written by the person who needs it.
Where TruLearn fits
Retrieval is built into how our sessions are shaped rather than offered as a feature. A session opens by asking rather than telling, every concept ends with the student explaining it back from memory, and the specific thing a student got wrong returns days later as a question rather than as a re-explanation.
That design follows directly from this research, which we set out with its limits on our research page. Also worth reading: the explain-it-back method.
Frequently asked questions
- What is retrieval practice?
- Deliberately recalling information from memory rather than reviewing it. Closing the book and writing what you remember, answering questions without looking, using flashcards honestly, or explaining a topic out loud are all retrieval practice. The defining feature is that you produce the material rather than read it.
- How much better is it than rereading?
- In the best-known experiment, students who studied a passage once and then took recall tests remembered considerably more a week later than students who studied it four times. The exact size varies with material and delay, but the direction is consistent: the longer the delay before the final test, the bigger the advantage of retrieval.
- Why does rereading feel better if it works less well?
- Because familiarity is easy to mistake for knowledge. Rereading makes text feel fluent, and fluency feels like understanding. Retrieval feels difficult, and that difficulty is the work being done. Students consistently rate the less effective technique as more effective.
- Does it matter if I get the answer wrong?
- No, as long as you then see the correct answer. Attempting to retrieve and failing still improves subsequent learning, sometimes more than studying would have. The attempt matters more than its success.
- What counts as retrieval practice?
- Flashcards used honestly, practice questions attempted before checking, writing out everything you can remember on a blank page, explaining a topic aloud from memory, past papers under time. What does not count: reading notes, highlighting, watching a lecture again, or looking at a flashcard's answer before trying.
Keep reading
Learning science
Spaced repetition: how to schedule revision so it sticks
The same amount of practice produces far more durable memory when spread over days than when bunched into one session. Here is how to space it without an app.
Oct 3, 20265 min read
Learning science
How to study for AP Biology
AP Biology rewards explaining processes and connecting ideas, not memorising vocabulary. A study plan built around what the exam actually asks.
Oct 3, 20266 min read
Learning science
Why explaining it back is the best study technique nobody uses
Saying an idea in your own words finds the gaps that rereading hides. It is one of the best-evidenced things a student can do, and almost nobody does it.
Oct 3, 20265 min read