Spaced Repetition: The Science and How to Use It

Picture this: you spend a whole Sunday working through a chapter of biology, a list of Spanish vocabulary, or the principles of contract law. You test yourself at the end and you actually know it. Three weeks later the exam arrives, and most of it is simply gone. You’re relearning from scratch under time pressure.

That isn’t a personal failing. It’s your brain doing what it was built to do: letting go of information it hasn’t been asked to use again. The fix has been known for over 140 years. It’s called spaced repetition: reviewing material at gradually increasing intervals, each review timed for when you’re about to forget. It is one of the most robust findings in cognitive psychology, and most students still don’t use it.

This guide explains what it is, the science behind it, the schedule to follow, and how to start, with nothing more than paper or with an app.

What Is Spaced Repetition?

Spaced repetition is a review method built on two observations about memory:

1. We forget fast, then slower. Most forgetting happens in the first hours and days after learning.
2. Each well-timed review slows forgetting down. After a review, the memory lasts longer than it did before.

So instead of reviewing everything every day, or cramming everything the night before, you review each item just as it starts to fade, and you let the gap grow each time you remember it. A new fact might come back after 1 day, then 3, then 7, then 14, then 30. Items you struggle with come back sooner; items you know well come back rarely.

The result: you spend your study time where it’s needed, on the things you’re about to lose, and almost none on the things you already know.

It works best combined with active recall: at each review you try to answer from memory before checking, rather than re-reading. Spacing decides when you review; retrieval decides how.

The Forgetting Curve: Where It All Started

In the 1880s, the German psychologist Hermann Ebbinghaus did something unusual: he used himself as the only subject, memorized thousands of meaningless syllables (“DAX”, “BUP”, “ZOL”) and measured how quickly he forgot them. Nonsense syllables meant nothing he already knew could help him, so he was measuring memory in its purest form.

The result was the forgetting curve. Ebbinghaus measured how much effort it took to relearn a list compared to learning it the first time. After 20 minutes he had already lost a large part of it; after a day, he needed to relearn about two thirds; after a month, close to 80%. The drop is steep at first and then levels off.

Two details matter in practice:

• Forgetting is front-loaded. The first review, within a day, has the biggest effect.
• The curve flattens with each review. Every time you review before the memory fully fades, the next drop is slower. After a few reviews, material that would have vanished in days stays for months.

Meaningful material is forgotten more slowly than nonsense syllables, so the exact numbers vary by subject and by person. The shape doesn’t.

The Spacing Effect: What 140 Years of Research Say

Ebbinghaus also noticed that spreading his practice across several days produced better retention than the same number of repetitions in one sitting. That observation, the spacing effect, has since become one of the most replicated results in psychology.

• A 2006 meta-analysis by Cepeda and colleagues pooled hundreds of experiments: spaced study beat massed study (cramming) for long-term retention in the large majority of comparisons, across ages, materials and tasks.
• A follow-up study by the same group (2008) looked for the best gap between sessions. The answer depended on how long you need to remember: for a test in a week, a gap of a day or so works; to remember for a year, gaps of several weeks work better. The longer you need the memory, the longer the gaps should be.
• In their 2013 review of study techniques, Dunlosky and colleagues rated distributed practice “high utility”, one of only two techniques to get that rating. The other was practice testing.

The effect holds for vocabulary, medical facts, math procedures, motor skills and more, from children to older adults.

Why It Works: What Happens in Your Brain

The forgetting curve describes behavior. Neuroscience offers part of the explanation, and it changes how you think about studying.

New information is first held with help from the hippocampus, a structure deep in the temporal lobe that binds new experiences together. These early traces are fragile: easily disrupted by new information and quick to decay. Over time, and especially during sleep, memories that keep being used are gradually stabilized in networks across the neocortex, a process called consolidation.

Retrieval plays a central role. When you recall something, the memory becomes briefly changeable and is then stored again, a process researchers call reconsolidation. Each successful retrieval tends to leave the memory more stable and easier to reach.

Spacing helps for at least two reasons:

• Retrieval is harder after a gap, and effortful retrieval strengthens memory more than easy retrieval. Psychologist Robert Bjork calls this a desirable difficulty: the struggle to remember is where learning happens.
• Spaced sessions come with more varied context (different days, moods, places), which gives the memory more cues to be retrieved from later.

This is also why re-reading works so poorly. With the information right in front of you, there’s no retrieval effort, so there’s little strengthening. You build familiarity, not memory.

The Review Schedule

There’s no single magic schedule, but expanding intervals like these are a solid starting point for new material:

Review	When
1st	1 day after learning
2nd	3 days after the 1st
3rd	1 week after the 2nd
4th	2 weeks after the 3rd
5th	1 month after the 4th
Then	2+ months, growing each time

The gaps start short, when the memory is fragile, and grow as it becomes more stable. Two rules adjust the schedule to you:

• If you fail an item, shorten its interval and bring it back soon.
• If an item is easy, stretch its interval further.

For a fixed exam date, work backwards: make sure each topic gets at least three or four reviews before the exam, with the last one a day or two before. The guide on how spaced repetition works for college students shows how to build that calendar around a real exam schedule.

From Paper to Algorithms

The schedule above is easy to follow for a few dozen items. With hundreds, tracking what is due becomes a job in itself. That’s the problem spaced repetition systems solve.

The Leitner system (1970s) uses paper flashcards and a few boxes. All new cards go in box 1, reviewed every day. A card you answer correctly moves up to the next box, reviewed less often (every 2 days, then every 4, and so on). A card you miss goes back to box 1. It’s a simple physical version of expanding intervals, and it still works.

SuperMemo and SM-2 (late 1980s). Piotr Woźniak wrote the first computer algorithm that scheduled each card individually, based on how easily you recalled it. SM-2 became the basis of many apps, including early Anki.

Modern algorithms such as FSRS model each card’s memory with more variables, learned from millions of reviews. They schedule each review for when your probability of recalling the card drops to a target, often around 90%. You choose how much you want to remember; the algorithm works out when to ask.

Whatever the system, you sit down each day to a queue of exactly the cards that are due, and nothing else. We compared the best spaced repetition apps on scheduler, card creation, platforms and price.

Why Cramming Keeps Winning (and Losing)

If the evidence is so clear, why does almost everyone cram? Because cramming works, just not in the way people think.

To pass a test tomorrow, a long session the night before can load a lot into memory, and much of it is still there in the morning. But a week later, most of it is gone, and a month later almost all of it. For a subject you’ll never need again, that trade might be acceptable. For anything cumulative (languages, medicine, law, programming, any course where later chapters build on earlier ones) it means paying the full cost of learning and keeping almost nothing.

The trap is the feedback loop. You cram, you pass, so cramming “works”. The forgetting happens later, when you’ve moved on, so it never counts as evidence against it. Spacing gives the opposite feedback: after a month you notice that material from weeks ago is still there, and that daily sessions get shorter as cards move to longer intervals.

How to Start Using Spaced Repetition Today

You can start with paper or with an app. Both work; apps scale better.

With paper:

1. Turn what you learn into questions: one question per card, answer on the back.
2. Use three to five boxes (or envelopes) for the Leitner system.
3. Review box 1 daily, box 2 every 2-3 days, box 3 weekly, and so on.
4. Correct answers move up a box; wrong answers go back to box 1.

With an app:

1. Create cards or import them, one idea per card.
2. Review the daily queue, answering each card before revealing it.
3. Rate your answer honestly; the rating drives the schedule.
4. Add new cards at a pace your daily reviews can absorb.

The slowest part is usually writing the cards. An AI flashcard generator can draft them from your notes, and if your material is in PDFs you can turn a PDF into flashcards directly. LongTerMemory does both steps: upload your notes, PDFs or slides, and it generates question-and-answer cards and schedules every review for you.

The same scheduling idea works beyond flashcards: speakers use it to memorize presentations, giving each section of a talk its own review interval.

The Habits That Make It Work

The method is only as good as your consistency. A few habits separate people who get results from those who quit after a week:

• Start the first review within a day. It intercepts the steepest part of the forgetting curve.
• Short daily sessions beat long weekly ones. Fifteen to thirty minutes a day keeps cards cycling at the right intervals; a weekly marathon lets them pile up and go stale.
• Don’t skip the hard cards. The ones you keep missing are the ones that haven’t been consolidated yet. They need more contact, not less. If a card keeps failing, rewrite it: it’s often too long or too vague.
• Recall before you reveal. Glancing at a card and thinking “yeah, I know that one” is recognition. Generate the answer, even slowly, before you check.
• Do reviews before new material. If reviews take more than 30-40 minutes a day, add new cards more slowly until the backlog shrinks.

Go Deeper

• Study methods ranked by research evidence
• How it compares to traditional studying
• How to create long-term memory
• When overlearning is worth it
• Applying it to language learning
• How to memorize fast and not forget
• How to study for exams, step by step

FAQ

How does it work, in one sentence?

You review each item just before you’re likely to forget it, and every successful review lets you wait longer before the next one.

How long until I see results?

Within two to three weeks you’ll notice material from the first days is still there with little effort. The real payoff comes over months, as most of your cards move to long intervals and daily sessions get shorter.

What is the best interval between reviews?

Start with about 1, 3, 7, 14 and 30 days, then adjust: shorter for items you miss, longer for items you find easy. Apps do this adjustment automatically for each card.

Does it work for understanding, or only for facts?

It works best for material you can turn into clear questions: facts, definitions, vocabulary, formulas, procedures. For concepts, write cards that ask you to explain, compare or apply ideas, not just define them.

Is it better than cramming?

For anything you need to remember beyond the next few days, yes, clearly. Cramming can get you through tomorrow’s test; spacing keeps the material for the next exam, the next course and the job.

The Compound Interest of Learning

People who study memory sometimes describe spaced repetition as compound interest for knowledge. Each successful review strengthens the memory, so the next review is easier and can come later, so your effort per item keeps shrinking while the memory keeps getting more durable. A card that needs a review every day in its first week may need one every few months a year later.

Compound interest has a flip side too: start late and you miss the best growth period; stop and the balance starts to decay. The students who seem to have exceptional memories usually don’t. They found a method that works with the way memory works, and they used it every day.