Active Recall: Why Testing Yourself Beats Rereading (And How to Do It)

By Peter Benson, Cognitive Enhancement Researcher  ·  Updated October 2026  ·  Citations re-verified

Short answer: Active recall means testing yourself, pulling information out of memory instead of rereading it. It’s one of the best-supported study techniques: across hundreds of studies, practice testing beat restudying by a medium effect on later tests, and it helps most when you check your answers.

Most people learn to study by rereading: highlight the important parts, read them again, and again before the exam. Rereading isn’t useless: it builds familiarity and can help on an immediate test. But familiarity isn’t the same as being able to produce the information later, which is what tests, and real life, ask for. Research consistently finds that trying to retrieve information improves later retention more than studying it again, though the size of the advantage depends on the task, how often retrieval succeeds, feedback and the delay before the final test.

This guide to active recall covers what it is, why it beats rereading, how strong the evidence is and where its limits are, how to do it, and how it fits with spaced repetition. It sits in the Memory & Learning hub.

What is active recall?

Active recall is retrieving information from memory without looking at the source. Instead of rereading your notes on, say, the causes of the First World War, you close them and try to reconstruct the main points on a blank page, out loud, or by answering questions. Researchers call the benefit the testing effect: a test doesn’t just measure what you know, it changes how well you’ll remember it.

The key contrast is between recognising and retrieving. Rereading a passage triggers recognition (“I’ve seen this”), which is easy and feels like mastery. Retrieval makes you generate the answer yourself, which is harder. Reviews of the research conclude that retrieval isn’t a neutral readout of memory but strengthens it (Roediger & Butler, 2011).

Why does active recall beat rereading?

A classic 2006 experiment shows the pattern. Students studied short prose passages, then either restudied them or took recall tests on them. Five minutes later, the restudy group did slightly better, which is part of why rereading feels effective. But on tests two days and a week later, the pattern reversed and the retrieval group remembered more. The students who had repeatedly restudied also predicted they’d remember more (Roediger & Karpicke, 2006).

That gap between how learning feels and how well it lasts is sometimes called an illusion of competence. Retrieval feels harder and exposes what you don’t know, which is why people drift back to rereading. Psychologists call this kind of useful effort a “desirable difficulty”. Retrieval also connects to the forgetting curve: material you’ve practised retrieving tends to be forgotten more slowly than material you’ve only reread.

How strong is the evidence for active recall?

Strong, by the standards of education research. A major review rated ten common study techniques; only practice testing and distributed (spaced) practice earned “high utility”, while rereading, highlighting and summarising were rated low (Dunlosky et al., 2013).

Meta-analyses agree. One pooled 272 effect sizes from 188 experiments and found practice testing beat restudying by a medium effect (g = 0.51), and beat filler or no activity by a larger one (g = 0.93) (Adesope et al., 2017). Another found a similar overall effect (g = 0.50), larger when learners got feedback (Rowland, 2014). Repeated retrieval matters even after something has been learned: in one study, continuing to test vocabulary that students could already recall had a large effect on recall a week later, while continuing to restudy it had none, and students’ predictions of their own performance were unrelated to how they actually did (Karpicke & Roediger, 2008). In real classrooms, a review of 222 studies with more than 48,000 students found quizzing raised achievement by a medium amount (g = 0.50) (Yang et al., 2021).

It isn’t limited to facts. In one study, retrieval practice produced better learning than building concept maps, including on questions requiring inference, and even when the final test asked students to draw a concept map (Karpicke & Blunt, 2011).

Where does active recall work less well?

The effect is robust but not unlimited. Some researchers have argued that it shrinks or disappears for very complex material, such as multi-step problem solving, where much of the effort goes into holding many elements in mind at once (van Gog & Sweller, 2015). Others replied that retrieval still helps with complex material and that the null results came from particular study designs (Karpicke & Aue, 2015). The debate isn’t settled. Effect sizes also vary with the material, test format, feedback, prior knowledge and delay, so no single figure from one experiment is the improvement every learner should expect.

Practically: the effect is largest on delayed tests, so it can look worse than rereading if you judge it straight away. It depends on actually retrieving: if you peek or the questions are too easy, it turns back into recognition. And not every difficulty is useful: retrieval should be hard enough to require real recall, but if you repeatedly fail without checking, you mostly practise not knowing. Beginners may do better with worked examples, prompts or guided questions before unaided recall, for example learning the method for a new type of maths problem before testing yourself on it. See also working memory.

Active recall and other study techniques ranked by evidence

🟢 Stronger or replicated evidence  ·  🟡 Limited, mixed or preliminary evidence  ·  🔴 Not established or not supported

How do you practise active recall?

The format matters less than the principle: close the source and make yourself produce the answer. Three formats work well.

The blank page. After studying a topic, put everything away and write down all you can remember. Then check against the source and mark what you missed. It’s demanding, and it shows your gaps clearly.

Flashcards, used properly. Look at the prompt, produce the answer in your head or aloud, then flip to check. The producing step is what matters; reading both sides turns flashcards back into rereading.

Practice questions. End-of-chapter questions, past papers, or questions you write yourself. Attempt the answer before checking, then use the feedback, especially for anything you got wrong or weren’t sure about.

Active recall works best as a loop: try to retrieve the answer without looking, check the source, correct mistakes, and try again later. A failed attempt isn’t wasted, but feedback stops errors from sticking; the aim is accurate, increasingly durable recall, not struggling indefinitely without checking. If retrieval feels effortless, you may be recognising rather than recalling. For the conditions that make effortful study possible, see deep work.

How do active recall and spaced repetition fit together?

They’re often confused, but they’re parts of one system. The forgetting curve is the problem: memories fade over time. Active recall is the action: retrieving a memory strengthens it. Spaced repetition is the schedule: it spreads those retrievals out over time, which the same review rated as the other high-utility technique (Dunlosky et al., 2013).

Put simply, active recall is what you do, and spaced repetition is when you do it. Flashcard apps that schedule reviews combine both. For the biology of how memories are strengthened, see the neuroscience of learning; mixing different types of problem in practice is covered under interleaving.

Frequently asked questions

Scientific references