Are Flashcards Effective? What Memory Research Says

Yes, flashcards are effective, but not because of the cards themselves. They work because they make two of the best-supported study techniques easy to do: retrieval practice (pulling an answer out of memory) and spaced practice (reviewing over days and weeks instead of in one session). A large review of learning techniques by Dunlosky and colleagues (2013) rated exactly these two, practice testing and distributed practice, as the only techniques of high utility out of the ten they examined.

So the useful question is not "do flashcards work?" but "am I using them in a way that makes me retrieve, and am I spacing my reviews?" Below is what the research shows, where it stops, and how to apply it.

Retrieval practice: testing yourself is a way of learning

Most students treat a test as a way to measure what they know. Memory research shows that the act of retrieving an answer also changes memory: it makes the answer easier to retrieve the next time.

In a classic experiment, Roediger and Karpicke (2006) had students read short prose passages and then either study them again or take a recall test without feedback. On the delayed tests, prior testing produced substantially greater retention than extra studying, even though re-reading felt more productive at the time.

The effect is not limited to rote facts. Karpicke and Blunt (2011) compared retrieval practice with concept mapping, a popular "deep" study method, on science texts. Practicing retrieval produced greater gains in meaningful learning, including on questions that tested comprehension.

A flashcard is the simplest possible retrieval exercise: a prompt on one side, the answer hidden on the other. The key word is hidden. If you glance at the answer before you have tried to produce it, you have turned retrieval practice back into re-reading.

Spacing: the same reviews, spread out, stick better

The second ingredient is timing. A meta-analysis of 317 experiments on verbal learning by Cepeda and colleagues (2006) found that spreading study over time beats massing it together, and that the best gap between study sessions grows with how long you need to remember the material. A follow-up study by Cepeda and colleagues (2008) put the best gap at roughly 20 to 40 percent of the time until a test one week away, falling to 5 to 10 percent for a test a year away. In practice: for a quiz next week, review every day or two; for a final exam or a licensing exam months away, let the gaps grow to weeks.

Flashcards make spacing easy to see in action. Kornell (2009) had students learn vocabulary with flashcards in one of two ways: a single large stack, where each card came back only after many others, or four small stacks, where each card came back quickly. The large stack, which spaces each card out, was more effective than the small stacks, even though most students believed the opposite.

This is also what spaced repetition software such as Anki automates: it schedules each card for review around the time you are likely to start forgetting it, and pushes cards you know well further into the future.

The most common flashcard mistakes

The research also explains why some students use flashcards for weeks and still feel unprepared.

• Flipping too early. Recognizing an answer when you see it feels like knowing it, but it is not the same as producing it. Say or write your answer before you turn the card.
• Dropping cards as soon as you get them right once. In experiments by Kornell and Bjork (2008), students who were allowed to drop cards they felt they knew learned slightly but consistently less than students who kept studying the whole stack. One correct answer is a start, not proof.
• Cramming the whole stack the night before. It can get you through tomorrow's quiz, but massed review fades quickly. Several short sessions over several days hold up far better on a delayed test.
• Cards that only test isolated words. A card that asks "Which valve sits between the left atrium and left ventricle?" is useful, but exams also ask why and what happens if. Add some cards that ask for causes, comparisons and consequences.

Do you need to make the cards yourself?

There is evidence that making your own cards helps. In six online experiments, Pan and colleagues (2023) found that learners who created their own flashcards learned more than learners who studied premade cards in five of the six.

That does not make premade or AI-generated cards useless. Writing a full deck for a 40-page chapter can take hours that could be spent reviewing it, and a deck you never finish making is a deck you never study. A practical middle path:

1. Generate a first draft of the deck from your lecture slides or textbook chapter.
2. Read through it once and edit: rewrite questions in your own words, delete what your course does not cover, add cards for points your lecturer stressed.
3. Spend the saved time on spaced reviews.

Editing a card forces you to think about what it is really asking, which captures part of the benefit of writing it from scratch.

Multiple-choice flashcards: do they still count as retrieval?

A common worry is that choosing from options is just recognition. It depends on the options. Little, Bjork, Bjork and Angello (2012) showed that properly constructed multiple-choice questions, with alternatives plausible enough to compete with the right answer, can trigger productive retrieval, not just recognition.

The risk with multiple choice is that a wrong option you picked, or nearly picked, sticks in memory. Feedback fixes most of that: in Butler and Roediger (2008), both immediate and delayed feedback after a multiple-choice test increased correct answers and reduced intrusions of wrong options on a later test. So after each answer, read the explanation, especially when you got it wrong.

A useful habit: cover the options, try to answer the question on your own, and only then look at the four choices.

Is there evidence from real courses?

Most of the studies above are controlled experiments with short materials, which is their strength and their limit. Evidence from real courses is thinner and mostly observational. For example, Deng, Gluckstein and Larsen (2015) surveyed 72 medical students at one school and found that the number of unique Anki cards students had studied was associated with higher scores on the USMLE Step 1 licensing exam (Step 1 was still scored then; it has been reported as pass or fail since 2022). Students chose whether and how much to use Anki, so this shows an association, not proof that Anki raised the scores.

Digital or paper?

The studies above do not settle this. Kornell's spacing study used a web program, but it compared stack sizes, not formats. What matters more than the medium is whether you retrieve before you check and whether your reviews are spaced. Software makes spacing automatic and makes it easy to carry thousands of cards; paper costs nothing and keeps you away from your phone. Pick the one you will actually use every day.

A simple routine that applies the research

1. Make or generate cards from the material your exam covers. Use questions, not just terms.
2. Answer every card before you look. Out loud or on paper.
3. Read the explanation when you miss one, and keep that card in rotation.
4. Review a little every day instead of a lot once. Let an app schedule the gaps, or keep a "missed" pile you return to tomorrow.
5. Mix in harder cards over time: why, what if, compare.

If your material is in PDFs, PDFStudy turns a lecture PDF (slides saved as PDF work too) or a photo of notes into multiple-choice flashcards with an explanation for each answer. You can edit any card, practice in the browser, and export the deck to Anki so its scheduler handles the spacing (export formats depend on your plan). Or start with one of our free practice quizzes, such as the anatomy and physiology quiz.