Spaced Repetition Scheduler
Schedule a flashcard’s next review with the SuperMemo SM-2 algorithm.
New ease factor: 2.50 · Review this card in 15 days.
Spaced repetition delays each review of a flashcard by a growing interval so you restudy it just before forgetting. The SM-2 algorithm sets the next gap from your recall quality. A card at interval 6 days with ease factor 2.5, reviewed well (q = 4), keeps its ease and jumps to round(6 × 2.5) = 15 days.
What the SM-2 algorithm does
SM-2 is the scheduler behind SuperMemo and Anki-style flashcard apps. Each card carries three numbers: the current interval (days until the next review), an ease factor (EF, a difficulty multiplier that starts at 2.5), and the repetition count (n, how many times in a row you have recalled it). After every review you grade your recall on a 0–5 scale, and those three numbers update to place the next review.
The idea is to review each card at the moment memory is about to fade. Get it right and the interval stretches — from 1 day, to 6 days, then multiplied by the ease factor each time. Get it wrong and the card resets to a 1-day interval so you see it again tomorrow. Easy cards drift far apart while stubborn ones stay close, so your study time lands where it is actually needed.
Ease factor is clamped to a minimum of 1.3. The interval multiplies by EF once n ≥ 2.
Worked example
Take a card with interval I = 6 days, ease factor EF = 2.5, and n = 2 prior successes. You review it and grade recall q = 4 (correct after a short hesitation). Because q ≥ 3, the card advances. First the ease updates: EF′ = 2.5 + (0.1 − (5 − 4)(0.08 + (5 − 4)0.02)) = 2.5 + (0.1 − 0.1) = 2.5, so it is unchanged. With n ≥ 2 the new interval is round(6 × 2.5) = 15 days, and n becomes 3.
- 1 Grade your recall. Rate how well you remembered the card on the 0–5 quality scale q, from 5 (perfect) down to 0 (complete blackout).
- 2 Check whether the card passed. If q is 3 or higher the card is a success; if q is below 3 it lapsed and will reset to a 1-day interval.
- 3 Pick the next interval. On a pass: n = 0 gives 1 day, n = 1 gives 6 days, and n ≥ 2 gives round(I × EF). On a lapse the interval becomes 1 day and n resets to 0.
- 4 Update the ease factor. Apply EF′ = EF + (0.1 − (5 − q)(0.08 + (5 − q)0.02)), then clamp it so it never drops below 1.3.
- 5 Store and schedule. Save the new interval, ease factor, and repetition count on the card, and show it again after the interval elapses.
Recall quality and the first intervals
Grades 3–5 advance the card; 0–2 reset it. After a pass, intervals grow 1 day → 6 days → ×EF.
| q | Meaning | Effect |
|---|---|---|
| 5 | Perfect recall | Pass · ease rises by 0.1 |
| 4 | Correct after hesitation | Pass · ease unchanged |
| 3 | Correct with difficulty | Pass · ease falls slightly |
| 2 | Incorrect but familiar | Lapse · reset to 1 day |
| 1 | Incorrect answer | Lapse · reset to 1 day |
| 0 | Complete blackout | Lapse · reset to 1 day |
| — | First interval (n = 0) | 1 day |
| — | Second interval (n = 1) | 6 days |
| — | Later intervals (n ≥ 2) | round(I × EF) |
Why spacing beats cramming
Reviewing at widening gaps forces your brain to retrieve a fact just as it starts to fade, and each effortful recall strengthens the memory far more than rereading it back to back. Cramming packs those retrievals into one session, so the material feels familiar for a day and then collapses. Spaced repetition trades that short-lived familiarity for durable, long-term retention.
SM-2 is the classic engine that made this practical: it powers SuperMemo and the schedulers in Anki-style apps used by medical and language students worldwide. The ease factor is the key tuning knob — cards you find hard earn a lower EF and come back sooner, while easy cards earn a higher EF and drift apart. The 1.3 floor stops a repeatedly-failed card from being scheduled so often that it becomes unmanageable, keeping even your worst cards on a workable rhythm.