Retrieval practice: testing yourself to learn#

Close the book and try to say it back. That small, slightly uncomfortable act of dragging an answer out of your own head—rather than reading it again on the page—is the single most powerful study move you have. We call it retrieval practice, and the effect is so reliable that researchers gave it its own name: the testing effect. The catch is that it feels worse than rereading while you do it, which is exactly why most people skip it. This page is about why you shouldn’t.

Recalling beats rereading#

Here is the experiment that should change how you study. Roediger and Karpicke gave students a passage to learn, then split them into two groups. One group reread the passage; the other put it away and tried to recall it from memory. Right afterwards, the rereaders felt more confident—and on an immediate test they did slightly better. But a week later, the picture flipped hard: the group that had tested themselves remembered far more (Roediger & Karpicke, 2006). The rereaders had bought a comfortable feeling in the moment and paid for it with forgetting.

This is the trap to watch for. Rereading buys fluency, not memory. When you run your eyes over a highlighted page for the third time, the words feel familiar and your brain whispers “I know this.” That whisper is an illusion of competence—it’s tracking how easy the text is to read, not how well you’d do without it in front of you. Retrieval breaks the illusion, because the moment you try to recall something and can’t, you find out the truth immediately.

The effect isn’t a one-off lab curiosity. A meta-analysis pulling together hundreds of comparisons found a robust advantage for practice testing over restudying, across school subjects, ages, and test formats (Adesope, Trevisan & Sundararajan, 2017). When Dunlosky and colleagues ranked common study techniques by how much they actually help, practice testing came out near the top—while rereading and highlighting, the two things students do most, sat near the bottom (Dunlosky et al., 2013).

Why pulling it out builds the memory#

It helps to know why this works, because it changes how you think about testing. A test isn’t just a measurement of what’s already in your head—the act of retrieving is itself an act of building. Every time you successfully haul a fact up from memory, you strengthen the path back to it, so it comes faster and easier next time. You’re not reading the memory; you’re rewiring it.

A striking study makes the point. Karpicke and Blunt had students learn science material in different ways. One group studied it, then practised retrieving it. Another group built elaborate concept maps—the kind of rich, “deep processing” activity that sounds far more sophisticated. A week later the retrieval-practice group remembered substantially more, including on questions that asked them to draw inferences and connect ideas (Karpicke & Blunt, 2011). The plain act of trying to recall beat the fancy diagram. Effortful retrieval doesn’t just lock in facts; it builds the flexible, connected kind of knowledge you can actually use.

There’s a deeper reason this matters for you as a learner. When recall is a little hard—when you have to strain for it—that very effort is what makes it stick. Easy, fluent study slides off; effortful retrieval sinks in. This is the principle of desirable difficulty: the discomfort isn’t a sign the method is failing, it’s the sign it’s working.

How to actually do it#

Retrieval practice doesn’t need any special tool. It just needs you to close the book and answer from memory before you check. Here are the ways I use most:

Brain dump. Finish a chapter, shut it, and write down everything you can remember on a blank page—no peeking. Then open the book and see what you missed. The gaps you find are pure gold: they’re exactly the things you thought you knew but didn’t.
Free recall. Same idea, out loud. Explain the topic to an empty room, or to a patient friend, as if teaching it. Stumbling mid-sentence tells you precisely where the thin spots are.
Flashcards, used properly. A flashcard only counts as retrieval if you genuinely try to produce the answer before you flip it. Glancing at both sides is just rereading in disguise. (Flashcards pair beautifully with spaced repetition—see the repetition pages.)
Practice questions and past papers. Old exams, end-of-chapter problems, quiz apps. Do them under recall conditions: book closed, answer first, check after.
Close-the-book summaries. Before you reread a page for “revision,” cover it and try to reconstruct its main points first. Read it afterwards, to fill holes—not before, to feel ready.

One rule ties all of these together: always pair retrieval with feedback. Retrieving a wrong answer and never correcting it just rehearses the error. So check your answer afterwards every time—right or wrong. Getting it wrong and then seeing the correct answer is one of the most powerful learning moments there is, far better than never having struggled at all.

The takeaway#

Stop measuring your studying by how smooth it feels and start measuring it by how much you can produce with the book closed. Build a simple habit: after every chunk you learn, shut the material and recall it—brain dump, flashcard, or out-loud explanation—then check what you got. It will feel harder than highlighting. That harder feeling is the sound of memory being built. Testing isn’t the thing you do after learning; for the most part, testing is the learning.