What Is Spaced Repetition? The Science of Remembering More in Less Time

If you have ever read a chapter, felt confident, and then drawn a blank three weeks later, you have met the forgetting curve. Spaced repetition is the most reliable way to beat it — and it is the engine behind every serious flashcard app, from Anki to StudyTab. This is the complete picture: the science, the algorithms, and how to actually run your revision on it.

The forgetting curve

In the 1880s, psychologist Hermann Ebbinghaus tested his own memory and found something uncomfortable: we forget most of what we learn within days. The loss is steep at first, then levels off — in his measurements, retention fell to roughly a third within a day and about a fifth within a month. If that sounds like a Victorian curiosity, it isn't: Murre and Dros replicated the experiment in 2015 with modern controls and got almost exactly the same curve, point for point.

The good news is buried in the same data: every time you successfully recall a fact, the curve gets flatter — you forget more slowly. So the goal isn't to study harder; it's to review at the right moments.

What spaced repetition actually does

Spaced repetition schedules each review for the moment you are about to forget — not sooner (a waste of time) and not later (you've already forgotten). After each successful recall, the next review is pushed further out: 1 day, then 3, then a week, then a month, and so on. A typical card's life looks like this:

Five short reviews, and a fact that would have been gone in a week is still alive months later. Two ingredients make it work — and both are among the best-measured effects in psychology:

1. Active recall. You retrieve the answer from memory instead of re-reading it. In Roediger and Karpicke's (2006) experiments, self-testing beat re-reading 61% to 40% on a one-week test; a meta-analysis of 272 experiments (Adesope et al., 2017) confirms the advantage across ages and subjects. Reading a highlighted page feels productive but builds almost no durable memory; being tested does. (The full evidence.)
2. Spacing. Reviews are spread out over increasing intervals. Across the 317 experiments synthesised by Cepeda et al. (2006), spaced practice produced 47% recall versus 37% for massed practice — same time, ten points more memory. The struggle to recall something you nearly forgot is exactly what strengthens it. (The spacing deep dive.)

Together they compound: spacing makes each retrieval effortful, and effortful retrieval is what builds the memory. In Dunlosky et al.'s (2013) definitive review of ten study techniques, practice testing and distributed practice were the only two rated high-utility — spaced repetition is simply both of them, automated. (How all ten techniques rank.)

The algorithms: SM-2 and FSRS

You don't have to guess the intervals — algorithms do it for you.

• SM-2 is the classic algorithm (from SuperMemo) that Anki used for years. After each card you rate how well you knew it, and it adjusts the next interval with a simple formula.
• FSRS (Free Spaced Repetition Scheduler) is the modern successor. It fits a memory model to your review history per card and targets a retention rate you choose — typically scheduling fewer, better-timed reviews than SM-2 for the same retention.

Which one matters less than using either: the research behind the timing (Cepeda et al.'s 2008 "ridgeline" study of 1,354 learners showed the optimal gap scales with how far away your test is) is exactly what these schedulers implement per fact. Full comparison: FSRS vs SM-2.

How to actually use it

1. Turn material into questions. A definition, a formula, an exception — each becomes a card with a question on one side and the answer on the other. (Hate making cards by hand? An AI flashcard generator can build them from your PDFs, notes or videos in minutes.)
2. Review daily. Spaced repetition only works if you show up most days — the schedule assumes it. Twenty focused minutes daily beats a three-hour weekend session.
3. Be honest when you rate cards. If you guessed, mark it. The algorithm can only help if it knows the truth — and research on the illusion of competence shows your gut overestimates what it knows.
4. Keep decks focused. One card = one fact. Cards that ask for three things at once are hard to schedule and harder to recall.
5. Start early, not perfectly. The system's power compounds with time. A rough deck started in July beats a beautiful deck started in December.

Do this in StudyTab

StudyTab is built end-to-end around this research:

• Both algorithms, per deck. Schedule with FSRS or SM-2, switchable per deck, with retention presets — choose 90% retention for core subjects, relax to 85% for breadth material, and the scheduler adjusts every interval.
• See the cost before exam season. The workload forecast projects your daily review load months ahead, so you know in September what February mornings will look like.
• Cards without the card-writing. Generate question cards from PDFs, YouTube lectures, web pages and even handwritten notes — cloze deletions for formulas, MCQs in NEET/JEE format, picture quizzes for diagrams.
• Retention you can see. The knowledge graph tracks concept-level mastery from green to red, so "is my revision working?" has an answer that isn't a feeling.

The bottom line

Spaced repetition isn't a hack — it's the two highest-utility findings in learning research (testing and spacing) turned into a daily system and run by an algorithm. Test yourself, space the reviews, and let the scheduler pick the timing. You'll remember more of your syllabus, for longer, in less time than you spend re-reading today.