What Is the Leitner Box System?

The Leitner box system is a simple spaced repetition method using flashcards and boxes. You start with all cards in Box 1. When you answer correctly, the card moves to the next box, which you review less frequently. When wrong, it goes back to Box 1. Each box has its own review schedule, automatically spacing reviews as you learn.

You don't need an app. You don't need to understand algorithms. Five cardboard boxes and a stack of index cards are enough.

The system was invented in 1972 by Sebastian Leitner, a German science journalist. It's still used today because it works. The method predates every digital spaced repetition app by decades, and the principle behind it is the same one those apps use.

Why You Forget What You Study

You've sat down to revise. You've read through your notes. You've highlighted the important bits. You've reread the chapter. You close the book feeling like you know it.

Then the test arrives, and it's gone.

Not all of it. Some pieces are there. But the stuff you spent the most time on, the concepts you were certain you'd mastered, they've vanished. You can almost see the page where it was written, but the actual information won't come.

You stare at the question. The teacher is saying something about time remaining. The person next to you is writing steadily. You know this. You revised it last night. You read it three times. The page had a diagram in the top right corner. Blue heading. You can picture it exactly.

But the answer itself? Nothing.

So you write something. It's close. Probably not right, but close enough that maybe you'll get partial credit. You move to the next question. Same thing. You've seen this before. You know you've seen it. But what you studied and what you can actually recall when it matters are two different things.

The exam ends. You hand it in. You check the textbook on the way out. There it is. Page 47. Exactly where you remembered. The answer was right there. You read it last night. Why couldn't you remember it an hour ago?

This moment, this exact moment, is where the self-blame starts.

What's wrong with me? Everyone else seemed fine. They were writing. I spent longer on this than most of them probably did. I highlighted. I made notes. I reread the important sections. I did everything you're supposed to do.

Maybe I'm just bad at exams. Maybe I don't have a good memory. Maybe I'm not smart enough for this.

None of that is true. But it feels true when you're standing there with the textbook open, looking at the answer you couldn't produce an hour earlier.

The real problem isn't your memory. It's that the hours you spent felt productive but weren't. You were working hard. You were focused. You were doing all the things that look like studying. But almost none of it was actually making the information stick.

Here's what those hours actually were: you reading the same content over and over, each time feeling more familiar with it, mistaking that familiarity for understanding, and never once testing whether you could recall it without the page in front of you.

Rereading feels like learning. It's not.

When you reread something, it feels familiar. Familiarity registers in your brain as knowing. You think "I've seen this before, so I must know it." Your eyes move across the page and every sentence makes sense. You nod along. Yeah, I get this. I know this. This is fine.

Then the page disappears and you're asked to reproduce it, and there's nothing there.

That gap, that distance between "I recognise this" and "I can recall this," is where most study time gets wasted. Because recognition and recall are not the same process. You can recognise a concept when it's in front of you and still completely fail to recall it when the page isn't there.

Think about the last time someone asked you about a book you read six months ago. Not "did you read it" but "what was it about?" You remember reading it. You remember enjoying it. You might remember a scene or two. But summarise the argument? Explain the main points? Most of it's gone.

Or the article you saved last week. You read it. You found it interesting enough to save. You might have even highlighted a few lines. If someone put it in front of you again, you'd recognise it immediately. But right now, without looking, could you explain what it said? Probably not. And if you can, it's only the broad shape, not the details.

This isn't a focus problem. You weren't distracted. You read the whole thing. You might have read it twice. The issue is that reading something, even reading it carefully, doesn't mean you'll remember it. Your brain files it as "seen" and moves on.

The gap between recognition and recall is where most study time disappears.

You spend three hours reading. You highlight. You make notes in the margin. You reread the tricky sections. At the end of those three hours, you feel like you've learned something. You have, in a sense. You've built recognition. The content is familiar now. If someone showed you that page again tomorrow, you'd recognise it.

But the exam doesn't test recognition. It tests recall. Can you produce the answer when the page isn't there? Can you explain the concept without the textbook open? Can you apply the formula without looking it up?

That's a completely different skill. And rereading doesn't train it.

What happens instead is that you spend hours building recognition, you walk into the exam, and you're tested on recall. Then you're surprised when the answer doesn't come. You saw it. You read it. You highlighted it. Why can't you remember it?

Because recognising something when you see it and recalling it from scratch are two different things. And only one of them was practiced.

This pattern, this cycle of reading and forgetting, doesn't just waste study time. It compounds.

You fail the test, or you scrape through with a lower grade than you expected. That confirms the belief that you're bad at this, or that you don't have a good memory, or that you're not smart enough. So next time, you study even harder. You reread even more. You highlight even more sentences. You make even more notes.

None of it works, because the method hasn't changed. You're still building recognition and hoping it turns into recall. It doesn't. So you do badly again. And the cycle continues.

Or you start avoiding the material entirely. You tell yourself you'll come back to it later. You'll review it when you have more time. You'll figure out a better way to study. But you don't, because the problem isn't time or method. The problem is that rereading feels productive enough that you don't realise it isn't working until the test arrives.

And by then, it's too late.

Worse still, the stuff you forget isn't random. It's usually the stuff you spent the most time on. The difficult concepts. The things you reread three times because they didn't make sense the first time. You put in the hours. You did the work. And a week later, it's gone.

That's the most frustrating part. It's not that you didn't try. You tried hard. You put in more hours than most people. But those hours didn't convert into memory, and now you've got nothing to show for them except the feeling that you're bad at this.

This isn't new information. Hermann Ebbinghaus documented it in 1885. He tested his own memory by learning lists of nonsense syllables, then measuring how much he could recall after different time intervals.

What he found became known as the forgetting curve: without review, you lose about 70% of new material within 24 hours. Within a week, most of it is gone.

That curve has been replicated hundreds of times since. It's one of the most consistent findings in memory research. The shape varies slightly depending on how complex the material is and how well you learned it initially, but the pattern is the same. Memory decays. Fast.

Not because you didn't pay attention. Not because you didn't try hard enough. Because that's what memory does. It discards things that aren't reinforced.

Your brain is not designed to hold onto every piece of information it encounters. It's designed to hold onto information that seems important, and one of the signals it uses to determine importance is how often you retrieve that information.

If you read something once and never think about it again, your brain treats it as irrelevant. It decays. If you reread it, your brain updates the familiarity signal, but that's not the same as a retrieval signal. Familiarity says "I've seen this before." Retrieval says "I needed this and I used it."

Only retrieval convinces your brain the information is worth keeping long-term.

So the standard response to forgetting, which is to study more, reread more, highlight more, make better notes, doesn't fix the problem. It just creates more familiarity. You end up spending even more hours on activities that feel productive but don't actually fight the forgetting curve.

You're running harder on a treadmill, wondering why you're not moving forward.

The real solution isn't more hours. It's a different activity. Instead of rereading, you need to test yourself. Force your brain to recall the answer without the page in front of you. That's called retrieval practice, and it's the only method that consistently beats the forgetting curve.

When you retrieve something successfully, you strengthen the memory trace. Your brain registers that this information was needed and accessed, which is a much stronger signal than "this looked familiar when I saw it again."

When you fail to retrieve something, that failure is also useful. It tells your brain that this information is important but not yet encoded deeply enough. The next time you study it, your brain prioritises it.

Retrieval practice works. The problem is that most people don't do it, because it's harder than rereading. Rereading is comfortable. You sit there with your notes, everything makes sense, you feel like you're learning. Testing yourself is uncomfortable. You have to confront what you don't know. You have to sit with the fact that the answer isn't coming. It feels like failure.

But that discomfort is the signal that real learning is happening.

The Leitner system is retrieval practice with a schedule built in. It forces you to test yourself. It spaces those tests out so you're reviewing at the edge of forgetting. And it does all of this with a structure simple enough that you don't need an app, an algorithm, or anything except five boxes and a stack of cards.

How the Leitner Box System Works

The mechanics are simple. You need five boxes and a stack of flashcards.

Write your questions and answers on index cards. One fact per card. Front side: the question. Back side: the answer. Every card starts in Box 1.

Here's the rule: when you get a card right, it moves to the next box. When you get it wrong, it goes back to Box 1.

Box 1 gets reviewed every day. Box 2 gets reviewed every other day. Box 3 gets reviewed once a week. Box 4 gets reviewed every two weeks. Box 5 gets reviewed once a month.

The boxes themselves are the scheduling system. A card in Box 1 is something you don't know yet, so you see it daily until you get it right. A card in Box 5 is something you've recalled correctly multiple times over weeks, so you only check it monthly to make sure it's still there.

The system is self-correcting. If you think you know something but you get it wrong, it goes straight back to daily review. You can't fool the system by pretending you know something. The only way a card stays in the higher boxes is if you keep getting it right at longer and longer intervals.

Let's walk through an example. You're learning the capital cities of European countries. You write "What is the capital of Slovenia?" on one side of a card and "Ljubljana" on the other. That card starts in Box 1.

Day 1: You review Box 1. You see the Slovenia card. You don't know the answer. You flip it. Ljubljana. You put it back in Box 1.

Day 2: You review Box 1 again. The Slovenia card comes up. This time you remember. Ljubljana. The card moves to Box 2.

Day 4: You review Box 2 (every other day). The Slovenia card is there. You remember. Ljubljana. It moves to Box 3.

Day 11: You review Box 3 (once a week). The Slovenia card is there. You remember. Ljubljana. It moves to Box 4.

Day 25: You review Box 4 (every two weeks). The Slovenia card is there. You remember. Ljubljana. It moves to Box 5.

After that, you only check it once a month. If you ever get it wrong, it goes back to Box 1, and the process starts over.

The beauty of this system is that you don't decide when to review. The boxes decide for you. If you're struggling with a concept, it stays in Box 1 where you see it daily. Once you've got it, it graduates to less frequent reviews. The cards you know well take up almost no time. The cards you're struggling with get all your attention.

Why Moving Cards Between Boxes Helps Memory

The reason this works comes down to a principle called desirable difficulty.

When something is easy to retrieve, practising it again doesn't strengthen the memory much. When something is hard to retrieve, the effort of recalling it produces a much stronger memory trace. This is counterintuitive. You'd think reviewing something you already know well would be useful. It's not. At least not compared to reviewing something you're about to forget.

Dr. Robert Bjork, a psychologist at UCLA, describes this as the difference between storage strength and retrieval strength. Storage strength is how deeply embedded something is in long-term memory. Retrieval strength is how easily you can access it right now.

Here's the key insight: the higher something's current retrieval strength, the smaller the gain in storage strength from studying it again. If something is highly accessible right now, restudying it produces almost no learning. So, in Bjork's phrasing, "forgetting, rather than undoing learning, creates the opportunity to reach additional levels of learning."

Read that sentence again. Forgetting is not the enemy. Forgetting creates the conditions where real learning happens.

This is why cramming doesn't work. When you cram, you're reviewing material that's still highly accessible. You're practising retrieval when retrieval is easy. It feels productive because you get everything right. But you're not strengthening storage. You're just confirming that the material is temporarily in your head.

The Leitner system forces you to review at the edge of forgetting. A card in Box 1 is something you got wrong recently, so retrieval strength is low. Reviewing it takes effort. That effort is what builds storage strength.

A card in Box 5 is something you last reviewed a month ago. Retrieval strength has dropped. When you pull that card out and successfully recall the answer, you're proving to your brain that this information is worth keeping long-term. The effort of recalling something that's partially faded strengthens the memory more than reviewing it when it's fresh.

The spacing itself is critical. Ebbinghaus showed that distributed practice beats massed practice. A 2006 meta-analysis by Cepeda and colleagues synthesised 839 experiments and found an effect size of g = 1.01 for spaced retrieval practice compared to cramming. That's a large effect. It means spacing roughly doubles how much you retain.

The specific intervals Leitner used (daily, every other day, weekly, fortnightly, monthly) aren't magic. They're a reasonable guess at when retrieval strength has dropped enough to make review worthwhile. Modern algorithms like FSRS calculate the optimal interval more precisely, but Leitner's fixed schedule gets you 80% of the benefit with zero computation.

What's the Review Schedule for Each Box?

Leitner's original system used five boxes with these intervals:

You don't have to follow these intervals exactly. Some people use three boxes instead of five (daily, weekly, monthly). Some people double the interval at each box (1 day, 2 days, 4 days, 8 days, 16 days). The principle is the same: each box represents a longer interval than the one before it.

The rule that matters is this: when you get a card right, it moves forward. When you get it wrong, it goes back to Box 1.

That last part is non-negotiable. If a card makes it to Box 5 and you get it wrong, it doesn't go to Box 4. It goes all the way back to Box 1 for daily review. This feels harsh, but it's correct. If you can't recall something after a month, the memory trace is weak. You need to rebuild it from scratch with frequent exposure, not just nudge it back one step.

Some people modify the system to be less aggressive. They send wrong cards back to the previous box instead of Box 1. This makes the system gentler, but it also makes it slower. Cards that you're struggling with stay in the higher boxes longer than they should, and you waste time reviewing things you think you know but actually don't.

Leitner's version is stricter, and that's a feature. It forces honesty. You either know it or you don't. There's no middle ground.

How to Set Up a Leitner Box System

You need five containers and a stack of index cards. The containers don't have to be boxes. You can use envelopes, folders, sections in a drawer, or anything that keeps cards separated.

Label the containers Box 1 through Box 5.

Write your flashcards. One concept per card. Front side: the question or prompt. Back side: the answer. Be specific. "What is the capital of France?" is better than "France?" because the question itself is a retrieval cue.

Put all the cards in Box 1.

Set your review schedule. Box 1 is daily. Box 2 is every other day. Box 3 is weekly. Box 4 is fortnightly. Box 5 is monthly. Write this down and stick it to the boxes so you don't forget.

Now start reviewing.

On Day 1, go through Box 1. Pick up the first card. Read the question. Try to recall the answer. Don't peek. If you can't remember, that's fine. Flip the card. Read the answer. Put the card back in Box 1.

If you do remember, flip the card to check. If you were right, move it to Box 2. If you were wrong (even slightly wrong), it stays in Box 1.

Do this for every card in Box 1. When you're done, Box 1 should have fewer cards than when you started.

On Day 2, go through Box 1 again. Any cards that didn't move yesterday are still there. You'll also go through Box 2 (every other day). Same process: recall, check, move forward if right, send back to Box 1 if wrong.

On Day 3, just Box 1. On Day 4, Box 1 and Box 2. On Day 7, you add Box 3. And so on.

The workload at the start is high because everything is in Box 1. But after a week or two, most cards have moved to higher boxes. Your daily review becomes shorter. Box 1 only contains new cards and the cards you're genuinely struggling with. Everything else is on a longer schedule.

One practical tip: shuffle the cards in each box before each review session. If you review them in the same order every time, you start remembering the sequence rather than the actual content.

Another tip: if a card is ambiguous or poorly written, rewrite it immediately. Don't wait. A bad flashcard is worse than no flashcard because it trains you to recall the wrong thing.

What's the Difference Between Leitner and Modern Apps Like Anki?

Modern spaced repetition apps like Anki use algorithms instead of fixed boxes. The most common algorithm until recently was SM-2, developed in 1987. Newer apps use FSRS, which was released in 2023 and is now the default in Anki.

These algorithms calculate the optimal review interval for each card based on how many times you've reviewed it, how long it's been since the last review, and how difficult the card has been historically. Instead of five fixed boxes, you have thousands of possible intervals, personalised to your performance.

The advantage is efficiency. FSRS can get you to the same retention level with 20-30% fewer reviews than SM-2, and both beat Leitner's fixed schedule. The algorithm also avoids a problem called "ease hell" where cards you struggle with get stuck in a loop of very short intervals. Leitner doesn't have this problem because wrong cards always reset to Box 1, but SM-2 does.

The disadvantage is opacity. You don't see how the algorithm works. You just trust that the app is showing you the right cards at the right time. For some people, that's fine. For others, it's unsettling. The Leitner system is transparent. You can see which box each card is in. You know exactly when you'll review it next. The system is in front of you, not inside an app.

There's also the question of whether you need the extra efficiency. If you're learning 10,000 vocabulary words for a language exam, the difference between 20 reviews and 15 reviews per word adds up. That's 50,000 saved reviews. But if you're learning 200 concepts from a textbook, the difference is 1,000 reviews vs 800 reviews. Meaningful, but not transformative.

Leitner gets you most of the benefit of spaced repetition with none of the setup cost. No account, no algorithm, no subscription. Just boxes and cards.

The other difference is that apps can track more detailed statistics. They know your retention rate, your review speed, your hardest cards, your longest streak. Some people find this motivating. Others find it distracting. Leitner gives you no statistics. You either know the card or you don't.

When Should You Use Physical Boxes vs a Digital App?

Use physical boxes if:

• You're learning a small set of material (under 500 cards)
• You want a system you can see and touch
• You're studying somewhere without reliable internet or you don't want to be on a device
• You're introducing spaced repetition to someone who finds apps overwhelming

Use a digital app if:

• You're learning thousands of items (language vocabulary, medical terms)
• You want to study on multiple devices and sync progress
• You need image occlusion, cloze deletions, or other advanced card types
• You want detailed statistics on your retention

The honest answer is that both work. The best system is the one you'll actually use. Some people need the physicality of cards. Writing them by hand helps them encode the information. Sorting them into boxes is satisfying in a way that swiping on a screen isn't.

Other people need the convenience of an app. They study on their phone during commute time. They want cards synced across devices. They want the algorithm to handle scheduling so they don't have to think about which box to review.

There's no wrong choice. The principle is the same either way: space your reviews, test yourself actively, and focus most of your time on the material you're about to forget.

Frequently Asked Questions

How many cards should I review per day?

Start with 10-20 new cards per day in Box 1. Once those move to higher boxes, add another batch. Your daily workload will stabilise after a few weeks. If Box 1 is overwhelming, reduce the number of new cards you add. Quality beats volume.

What happens if I miss a review day?

Nothing breaks. Just pick up where you left off. If you missed reviewing Box 3 on its scheduled day, do it the next day. The system is forgiving. The intervals are guidelines, not laws. Missing a day delays your progress slightly, but it doesn't ruin the system.

Can I use the Leitner system for subjects other than facts?

Yes, but it works best for discrete knowledge. Capital cities, vocabulary, historical dates, chemical formulas, anatomy terms. It's less effective for concepts that require deep understanding rather than recall. For procedural knowledge (how to do something), you need practice, not flashcards. For essays and synthesis, you need writing, not retrieval.

How long does it take before cards reach Box 5?

If you get a card right every time, it takes about six weeks. Day 1 in Box 1, Day 2 moves to Box 2, Day 4 moves to Box 3, Day 11 moves to Box 4, Day 25 moves to Box 5. In practice, most cards take longer because you'll get them wrong at some point and they'll reset.

Should I remove cards from Box 5 once they're fully learned?

Leitner kept cards in Box 5 indefinitely, reviewing them monthly. Some people remove cards once they reach Box 5 and haven't been wrong in six months. The risk is that you stop reviewing entirely and the memory fades. Monthly review is low effort. Keep the cards unless you're absolutely certain you'll never forget.

Do I need exactly five boxes?

No. Three boxes (daily, weekly, monthly) work fine for smaller sets of material. Seven boxes give you more granular spacing. Five is a good middle ground. The number of boxes is less important than the principle: each box has a longer interval than the previous one, and wrong cards go back to Box 1.

Can I use the Leitner system with an app instead of physical boxes?

Yes. Several apps implement Leitner's fixed-interval system instead of adaptive algorithms. Anki can be configured to use fixed intervals if you prefer. But at that point, you're trading the simplicity of boxes for the complexity of an app without gaining the efficiency of a modern algorithm. If you're going digital, you might as well use FSRS.

What if I don't understand why an answer is right after I flip the card?

Rewrite the card. Add context. Add an example. A flashcard should make sense the moment you flip it. If you're confused, the card is poorly written. Fix it immediately. Don't wait until the next review.

How do I know if I'm making progress?

Watch Box 1 shrink. When you start, every card is in Box 1. After a few weeks, most cards have graduated. If Box 1 stays large, you're either adding new cards too fast, or the existing cards are too hard and need to be broken into smaller pieces.

Is the Leitner system as effective as modern algorithms like FSRS?

Not quite. FSRS is more efficient. But Leitner is simpler, and simplicity has value. It's easier to stick with a system you understand than a system that's optimal but opaque. Leitner gets you 80% of the benefit with 20% of the complexity. For most people, that's enough.

Final Thoughts

The Leitner box system is over 50 years old. It predates computers, smartphones, and every modern spaced repetition app. It's still used because it works.

You don't need an algorithm. You don't need an account. You need five boxes, some index cards, and the discipline to review them on schedule.

The cards you know move forward. The cards you don't move back. The system handles the rest.

If you've been putting off spaced repetition because apps feel overwhelming or you don't trust black-box algorithms, start here. Build the boxes. Write the cards. See what happens.

The knowledge you want to keep is the knowledge you test yourself on. The Leitner system makes that testing automatic.

If you want to go deeper into how memory and spaced repetition work, read about the science of reading retention and how to actually remember what you read.