The Testing Effect: Why Practice Testing Beats Rereading
You reread the chapter twice, the sentences feel familiar, and you close the book confident. Then the quiz starts and half of it is gone. The testing effect is one of the most replicated findings in memory science, and it explains exactly why: checking what you know beats reviewing it, every time.
Ask most students how they study, and the answer is almost always the same: read the material, then read it again closer to the test. It feels thorough. Every page looks familiar on the second pass, and familiarity is easy to mistake for knowledge.
But familiarity and retrieval are two different skills, and only one of them gets tested on an exam, in a client meeting, or in a conversation in a foreign language. The testing effect — sometimes called the retrieval practice effect — is the research finding that checking what you know, through a quiz or a flashcard, builds a far more durable memory than reviewing the same material passively.
This isn't a minor study tip. It's one of the most heavily replicated results in cognitive psychology, tested across age groups, subjects, and decades of experiments — and it points to a genuinely counterintuitive conclusion: a test is not just a way to measure learning. A test is a way to create it.
Early observations. Psychologists noticed the pattern as far back as the early 1900s: students who recited material from memory outperformed students who simply reread it the same number of times.
The landmark demonstration. In 2006, Henry Roediger and Jeffrey Karpicke published a now-famous experiment: one group of students restudied a passage repeatedly, another studied it once and then tested themselves on it repeatedly. A week later, the group that tested themselves remembered substantially more — even though the testing group had spent less total time with the material in front of them.
A name for the effect. That study, and hundreds that followed it, gave the phenomenon its name: the testing effect, also called retrieval practice or test-enhanced learning depending on the paper.
A finding, not a technique. It's worth separating the two: retrieval practice or active recall is the study technique you use. The testing effect is the body of research that explains why the technique works and how large its benefit tends to be.
What made the Roediger and Karpicke study land so hard in the field wasn't just the result — it was the mismatch between the result and what students expected. Before the delayed test, the restudy group predicted they would remember more than the testing group did. A week later, it was the opposite. People are, on average, not very good at judging how well they've actually learned something, and that miscalibration is a big part of why rereading survives as the default study habit despite the evidence against it.
A ranking of study techniques
A widely cited 2013 review by Dunlosky and colleagues rated ten popular study methods by the strength of evidence behind them. Practice testing and distributed practice earned the top "high utility" rating. Rereading and highlighting — the two most commonly used methods among students — landed at the bottom, rated "low utility."
It holds up across meta-analyses
When researchers pool results across dozens of separate testing-effect experiments, the advantage of practice testing over restudying stays consistently moderate-to-large in size — and it tends to grow, not shrink, as the delay before the final test gets longer.
It works with no feedback at all
Even when a test gives zero feedback — no correct answer shown afterward — testing still beats an equal amount of restudy time in many experiments. Adding feedback afterward makes the benefit larger still, but the retrieval attempt is doing most of the work on its own.
One detail trips people up: rereading and testing often look about the same on a quiz given minutes after studying. That's exactly why the technique feels fine in the moment — short-term memory papers over the difference. The gap between the two only becomes obvious a few days or weeks later, which is precisely when it matters most: at the actual exam, the actual interview, the actual conversation in a language you're learning.
Retrieval effort
Rereading hands you the answer; retrieving forces your brain to search for it. That search itself appears to strengthen the neural pathway leading to the memory — the harder and more successful the search, the more durable the result.
Elaborative retrieval
Trying to answer a question doesn't just activate one memory — it activates related knowledge along the way, as your brain searches multiple possible paths. That extra activity appears to build additional routes back to the same memory, so it's easier to find next time, even from a different angle.
Honest self-monitoring
A test tells you, immediately and unambiguously, what you don't actually know — something rereading almost never reveals. That accurate feedback redirects your remaining study time toward real gaps instead of material you already have down cold.
Rereading & highlighting
- Rated "low utility" in research reviews
- No retrieval, so no strengthening effort
- Inflates confidence without inflating recall
- Gives no signal about what you don't know
- Advantage fades within a few days
Practice testing
- Rated "high utility" in research reviews
- Requires genuine retrieval effort
- Aligns confidence with actual recall
- Pinpoints exactly what still needs work
- Advantage grows over days and weeks
None of this means rereading is completely worthless — a first read-through is how you meet new material in the first place, and there's a place for a quick review right before you start testing yourself. The problem is treating rereading as the study session itself, on repeat, right up to the exam. The research is consistent: swap even one of those extra reread passes for a self-test, and the retention gap starts to open up in your favor.
Middle school science classrooms
A multi-year classroom study quizzed students on some, but not all, of their course material throughout the term. Material that had been quizzed was recalled substantially better on the final exam than material of equal difficulty that had only ever been read.
Medical training
Medical trainees who repeatedly tested their emergency-response knowledge retained it far better months later than trainees who spent the equivalent time simply practicing or reviewing the same material once.
University lecture courses
Courses that add short, low-stakes quizzes throughout the semester tend to see higher scores on the cumulative final exam than sections that rely on optional rereading of slides and notes alone.
Language classrooms
Vocabulary quizzed repeatedly during a course, rather than only reviewed from a list, shows up as better retained on delayed tests weeks or months after the course ends.
Free recall & short answer — strongest
Producing the answer from a blank page, with no options in front of you, demands the deepest retrieval effort and tends to produce the largest retention gains. This is exactly what a good flashcard front-and-back forces you to do.
Cued recall — strong
A partial hint — the first letter, a category, a related image — still requires real retrieval, just with a small nudge. It's slightly easier than pure free recall but still far more effective than simply recognizing an answer among choices.
Multiple choice — moderate
Picking the right option from a list is a shallower form of retrieval, since the correct answer is physically present on the page. It still beats rereading, but produces a smaller benefit than open-ended recall questions.
Recognition alone — weakest form of "testing"
Simply judging whether something looks familiar, without producing or selecting an answer, barely counts as retrieval at all — which is the same reason rereading itself provides so little benefit.
The forward testing effect is the finding that testing yourself on one set of material can improve how well you learn a completely different, new set of material studied right afterward. In other words, the benefit of a quiz isn't limited to the content on the quiz — it appears to carry forward into whatever you study next.
The leading explanation is that retrieval practice helps your brain "close the book" on the material you just tested, freeing up mental resources and attention for new information rather than letting the old material linger and compete for the same working-memory space. Whatever the precise mechanism, the practical takeaway lines up neatly with the rest of the research: a quick self-test between study sessions is rarely wasted time, and it may be actively priming you to learn what comes next.
Turn notes into questions, not summaries
Instead of rewriting a paragraph in your own words, phrase it as a direct question. "What causes X?" builds a retrieval habit; a rewritten summary is still just recognition in disguise.
Answer before you flip the card
Say or think the answer fully before you check it — even a wrong guess primes your brain to encode the correction more deeply than seeing the answer cold.
Grade yourself honestly
"Close enough" is recognition talking, not recall. Mark a card wrong if you hesitated or got only part of it — that's the honest signal that keeps your review schedule accurate.
Space your retests, don't cram them
A single test session helps, but the testing effect compounds when retrieval is spread out over days and weeks — which is exactly what a spaced repetition schedule does automatically.
In Repetit, every review is built around this exact loop: you see the question, retrieve your answer, then grade yourself before the app decides when the card comes back. Cards you struggle with resurface sooner; cards you nail get pushed further out — spaced repetition deciding the timing, the testing effect doing the actual work of building the memory.
Myth: testing only measures what you know
A test isn't a neutral ruler. The act of retrieving changes the memory itself, making it stronger and easier to retrieve next time. Testing is a learning event, not just a measurement of one.
Myth: it needs to be a graded, high-stakes test
Low-stakes, ungraded self-testing — a flashcard deck no one else ever sees — captures most of the benefit. The pressure of a graded exam isn't the active ingredient; the retrieval attempt is.
Myth: it only works for simple facts
Studies extend the effect beyond rote facts to comprehension questions, applied problem-solving, and inference-based material — anywhere you can phrase a genuine question with an answer to retrieve.
Myth: you must see feedback for it to help
Testing without any feedback still produces a measurable retention benefit over restudying. Feedback improves the effect further and corrects mistakes, but it isn't a precondition for testing to work at all.
Myth: one test is enough
A single retrieval attempt helps, but the biggest, most durable gains come from testing the same material repeatedly over spaced-out sessions — one test is a start, not a finish line.
Myth: multiple choice is just as good as free recall
Format matters. Open-ended questions that require producing the answer from scratch generally build stronger memory than picking the right option out of a list, even though both beat rereading.
Exam preparation
Practice questions expose exactly what isn't ready yet, while there's still time to fix it.
Vocabulary & language learning
Recalling a translation from memory, instead of reading it off a list, is what actually transfers to real conversation.
Medical & legal terminology
Dense, high-volume definitions hold up far better when retrieved under pressure than when simply reread.
Programming & certifications
Syntax, commands, and exam objectives translate cleanly into question-and-answer cards for quick, repeated testing.
Professional certifications & onboarding
Corporate training programs that build in short, repeated self-tests see better long-term retention than one-off reading assignments.
Facts, dates & standardized tests
Anything phrased as a clear question with one right answer is ideal material for practice testing.
The testing effect is one of the rare findings in psychology that survives decades of replication, holds up in real classrooms and workplaces, and points to a genuinely simple fix: swap a rereading pass for a self-test. The material doesn't need to be new. The technique does.
Pair it with a spaced schedule and you get the two most evidence-backed study habits reinforcing each other — which is exactly how Repetit's flashcards are built to be reviewed.
You don't need a new course or a new tool to start. Take one topic you already have notes on, turn its key points into a handful of direct questions, and test yourself on them tomorrow instead of reading the notes again. That single swap is where the research says the real difference begins.
FAQ: the testing effect
What is the testing effect in psychology?
The testing effect is the well-documented finding that retrieving information from memory — through a quiz, a flashcard, or a practice question — produces stronger, longer-lasting learning than spending the same amount of time rereading or restudying that same material.
How much does practice testing improve retention compared to rereading?
Pooled results across dozens of experiments show a moderate-to-large advantage for practice testing over restudying, and the gap tends to grow larger the longer the delay before the final test — rereading may look competitive after a few minutes, but practice testing pulls ahead after days or weeks.
Does the testing effect work without feedback?
Yes. Studies show a retention benefit from testing even when learners receive no feedback on whether their answers were right. Feedback still matters — it corrects errors and boosts the effect further — but the retrieval attempt itself is doing most of the work.
What is the forward testing effect?
The forward testing effect is the finding that testing yourself on one set of material can improve how well you learn a second, unrelated set of material studied right afterward — likely because retrieval practice helps you disengage from what you already studied and focus more effectively on what comes next.
Is the testing effect the same as active recall?
They describe the same underlying behavior from two angles. Active recall is the study technique of testing yourself instead of rereading. The testing effect is the research finding that describes why that technique works and how large its benefit is. In practice, using active recall is how you put the testing effect to work.
Does test format change the size of the testing effect?
Yes. Free recall and short-answer questions, which require producing the answer with no options in front of you, tend to produce a larger benefit than multiple-choice questions. Multiple-choice testing still beats rereading, but it is a shallower form of retrieval than open-ended recall.