How Spaced Repetition Actually Works (And Why It Beats Cramming)

The Experiment That Changed How We Study

In 1885, a German psychologist named Hermann Ebbinghaus did something tedious and brilliant. He memorized lists of nonsense syllables (like “WID” and “ZOF”), tested himself at increasing intervals, and recorded exactly how much he forgot over time.

The result was a curve — now called the forgetting curve — that shows memory decaying exponentially after learning. Within an hour, you forget roughly 50% of new information. Within a day, 70%. Within a week, up to 80%.

But Ebbinghaus also discovered something hopeful: each review resets the forgetting curve, and each reset makes the memory decay more slowly. Review at the right moments, and the curve flattens until the memory becomes nearly permanent.

That is spaced repetition. And it is the single most effective study technique ever measured by cognitive science.

What Is Spaced Repetition?

Spaced repetition is a learning technique where you review information at gradually increasing intervals — not on a fixed schedule, but based on how well you remember it each time. If you recall something easily, the next review is scheduled further out. If you struggle, the next review comes sooner.

The goal is to review each piece of information just before you would naturally forget it — the moment of maximum efficiency. Review too early and you waste time. Review too late and you have already forgotten, requiring you to relearn from scratch.

This is different from cramming, where you review everything in a single session. Cramming feels productive but produces almost no long-term retention — the forgetting curve after a cram session looks identical to the curve after a single exposure.

The Science: Ebbinghaus and the Forgetting Curve

Ebbinghaus’s forgetting curve follows a mathematical pattern:

R = e^(-t/S)

R = retention (what you remember)
t = time
S = strength of the memory

Two things to notice:

1. Memory decays fast at first, then slows down. The steepest drop happens in the first hour after learning.
2. Each review increases S (memory strength). After a successful review, the curve resets — but this time it decays more slowly. After several reviews, the curve is nearly flat.

A meta-analysis of spacing effect research published in the journal Psychological Bulletin (Cepeda et al., 2006) reviewed 184 studies and found that spaced practice produced better long-term retention than massed practice (cramming) in every single study — across ages, materials, and retention intervals. The effect is not small: spaced repetition can improve long-term retention by 200-500% compared to cramming.

How Spaced Repetition Algorithms Work

Modern spaced repetition apps do not use Ebbinghaus’s original formula. They use more sophisticated algorithms that track your individual performance on each card and adjust the schedule accordingly.

The SM-2 Algorithm

The most widely used algorithm is SM-2, created by Piotr Wozniak in 1987 (the same algorithm that powers Anki). Here is how it works:

1. You review a flashcard and rate your recall quality on a 0-5 scale:

   • 0 = complete blackout
   • 1 = wrong, but felt familiar
   • 2 = wrong, but easy to recall after seeing the answer
   • 3 = correct, but with significant effort
   • 4 = correct, with some hesitation
   • 5 = perfect recall

2. The algorithm adjusts the interval based on your rating:

   • If you rate 0-2 (failed recall): reset the interval to 1 day (you need to see it again tomorrow)
   • If you rate 3-5 (successful recall): multiply the current interval by an easiness factor (EF)
   • The EF starts at 2.5 and adjusts up or down based on your performance — easy cards get longer intervals, hard cards get shorter ones

3. The easiness factor updates after each review:

   • If you consistently rate 5, the EF increases (intervals grow faster)
   • If you struggle (rating 3), the EF decreases (intervals grow more conservatively)
   • The EF has a floor of 1.3 — it never drops below that

This means the algorithm personalizes the schedule for every single card. A card you always remember easily might have an interval of 6 months after a few reviews. A card you keep forgetting might show up every 2 days.

cnote’s card workstation implements the SM-2 algorithm — and because all data stays on your device, your review history and interval calculations are completely private.

Beyond SM-2: FSRS and Other Modern Algorithms

SM-2 is the industry standard, but newer algorithms have emerged. FSRS (Free Spaced Repetition Scheduler), introduced in 2022, uses machine learning to optimize scheduling based on a larger set of parameters. Some apps offer both SM-2 and FSRS, letting users choose.

For most learners, the difference between SM-2 and FSRS is marginal — both produce dramatically better retention than no spaced repetition at all. The algorithm matters far less than the habit of reviewing consistently.

Active Recall + Spaced Repetition: The Gold Standard

Spaced repetition is powerful on its own, but it becomes dramatically more effective when combined with active recall — the practice of retrieving information from memory rather than passively re-reading it.

Passive review (re-reading notes, watching a lecture again) creates an illusion of competence. The information feels familiar, but familiarity is not the same as recall. When you need to actually use the knowledge — in an exam, a conversation, or real life — passive review fails.

Active recall forces your brain to reconstruct the memory from scratch, which strengthens the neural pathways. This is why flashcards work: the act of trying to remember the answer before flipping the card is where the learning happens.

The combination — active recall + spaced repetition — is sometimes called the “gold standard” of evidence-based learning. Every flashcard app that uses spaced repetition is, by design, implementing both principles.

Practical Applications

Language Learning

Vocabulary is the most natural fit for spaced repetition. Each word is a discrete unit of knowledge, and you either know it or you don’t. Spaced repetition apps can track thousands of words and ensure you review each one at the optimal moment.

Medical School

Medical students are among the heaviest users of spaced repetition. The sheer volume of factual knowledge — anatomy, pharmacology, pathology — makes traditional study methods inadequate. A typical medical student might maintain 10,000+ flashcards reviewed on a daily basis.

Programming

Memorizing syntax, API signatures, and common patterns accelerates the learning curve. While understanding concepts is more important than memorizing syntax, having syntax at your fingertips reduces cognitive load and lets you focus on problem-solving.

Law and Certification Exams

Bar exams, CPA exams, and professional certifications all require retaining large volumes of detailed information. Spaced repetition ensures that material studied early in the prep period is still fresh on exam day.

Common Myths About Spaced Repetition

Myth 1: “Spaced repetition is just flashcards”

Flashcards are the most common implementation, but spaced repetition is a principle. You can apply it to any review schedule — re-reading chapters at increasing intervals, reviewing practice problems on a spaced schedule, etc. Flashcards just happen to be the most efficient format because they force active recall.

Myth 2: “You need to review every day”

No. The algorithm schedules each card individually. Easy cards might have intervals of months. You only review cards that are due on a given day. If you have 1,000 cards but only 30 are due today, you review 30.

Myth 3: “More cards = better”

Quality matters more than quantity. A deck of 500 well-crafted cards is more valuable than 5,000 poorly written ones. Each card should test a single, clear concept — if a card tries to test multiple things, split it.

Myth 4: “Spaced repetition replaces understanding”

Spaced repetition memorizes facts; it does not build understanding. You should understand a concept before you create a flashcard for it. The flashcard ensures you do not forget the details; the understanding ensures you can apply them.

FAQ

How does spaced repetition work?

Spaced repetition works by scheduling reviews at increasing intervals based on your recall performance. Each time you successfully recall information, the interval before the next review gets longer. If you fail to recall, the interval resets. This exploits the psychological spacing effect — reviewing information just before you would forget it produces the strongest long-term memory.

What is the spacing effect?

The spacing effect is the finding that information is better retained when study sessions are spaced out over time rather than massed together in a single session. It was first documented by Hermann Ebbinghaus in 1885 and has been replicated in hundreds of studies across different materials, age groups, and retention intervals.

What is the SM-2 algorithm?

SM-2 is a spaced repetition algorithm created by Piotr Wozniak in 1987. After each review, you rate your recall quality (0-5). Based on your rating, the algorithm calculates the next review interval by multiplying the current interval by an “easiness factor” that adjusts based on your performance. SM-2 is used by Anki, cnote, and many other flashcard apps.

How many flashcards should I review per day?

There is no universal answer — it depends on your goals and available time. A common recommendation is 20-50 new cards per day for casual learning, and 100-200 for intensive study (like medical school). The key metric is consistency: reviewing 30 cards every day is far more effective than reviewing 200 cards once a week.

Is spaced repetition better than cramming?

Yes — dramatically. A 2006 meta-analysis of 184 studies found that spaced practice produced better long-term retention than cramming in every single study, with improvements of 200-500%. Cramming can work for short-term recall (passing a test tomorrow), but for long-term retention, spaced repetition is unequivocally superior.

Can I use spaced repetition without an app?

Yes, but it is tedious. You would need to manually track review intervals for each item — which is why apps like cnote’s card workstation exist. The algorithm handles the scheduling; you just show up and review what is due.

The Bottom Line

Spaced repetition is not a hack or a trick. It is the application of a well-understood principle of human memory — the spacing effect — using an algorithm that personalizes the schedule for each piece of information.

If you are studying anything that requires long-term retention — a language, medical knowledge, programming syntax, legal precedents — and you are not using spaced repetition, you are working harder than necessary and remembering less than you could.

The tools are free. The science is settled. The only question is whether you will start today.

Try cnote’s card workstation — free, privacy-first, powered by the SM-2 algorithm, with all your data stored locally on your device.