N noduly study · spaced repetition
🔥0 · 0/5 All modules

Spaced Repetition Principles

Memory fades on a predictable curve. Reviewing just before you would have forgotten — and spreading those reviews further apart over time — flattens the decay into long-term retention with a fraction of the effort.

The forgetting curve (Ebbinghaus, 1885)

Hermann Ebbinghaus memorized nonsense syllables and re-tested himself at intervals. His data fit a roughly exponential decay: R = e−t/S — retrievability R falls as time t grows, modulated by memory stability S.

Plain English: without review, you lose most of what you learned within hours, and then most of what's left over the next few days. The curve is steepest at the start.

The spacing effect (Hermann Ebbinghaus → modern SRS)

Each successful review makes the next interval longer. Stability roughly doubles with each well-timed retrieval, so reviewing on days 1, 3, 7, 21, 60, 180 gives ~95% retention with about 10 total exposures.

This is the central trade-off of study: massed practice (cramming) feels productive but decays fast; spaced practice feels slower but compounds.

The two engines

R(t) = e−t/S and Snew ≈ Sold × f(grade, R)

R retrievability (0–1, "how likely could I recall it right now?") · S stability (days of half-life before R hits ~50%) · grade how easy this review was for you. Modern algorithms (SuperMemo's SM-2, FSRS, Anki) differ in how they update S from your grade.

Forgetting-curve simulator

Reviews scheduled
Avg retention
Final stability

Three classic algorithms

1972 · paper

Leitner box system

5 physical boxes. New cards start in Box 1 (review every day). Pass → promote one box. Fail → demote to Box 1. Box 2 every 2 days, Box 3 every 4 days, Box 4 every 9 days, Box 5 every 30 days.

No computer needed. Coarse but effective; visualizer below.
1985 · SuperMemo

SM-2 algorithm

The first widely-adopted software algorithm (Piotr Woźniak). Each card has an easiness factor (EF, starts at 2.5). Successful reviews multiply the interval by EF; failures reset the interval and slightly drop EF.

Still the basis of Anki's default scheduler. Walk-through below.
2022 · neural

FSRS

Free Spaced Repetition Scheduler — a 17-parameter model fit to millions of reviews. Tracks both stability and current retrievability; chooses optimal next interval to hit a user-chosen retention target (e.g. 90%).

Default in Anki since v23+ (optional). Outperforms SM-2 in head-to-head tests on the same review logs.

Leitner box visualizer

Start with 25 cards in Box 1. Each "review day" button simulates one pass: every box due that day is reviewed; you can mark a chunk of cards as Got it (promote) or Missed (demote to Box 1).

Day 0 · Cards in long-term Box 5 stay there until missed.

SM-2 walk-through

A card you grade 4 (correct, after hesitation) on 7 successive reviews using SM-2 defaults (EF starts at 2.5, first two intervals fixed at 1 and 6 days):

Review #GradeEasiness factorNew intervalDays from start

Grades 0–5: 0–2 = failure (reset interval, drop EF); 3 = barely; 4 = comfortable; 5 = perfect. Each easy grade pushes the next review further out — that's the spacing.

Connect the dots

Quiz

15 questions on spacing, the forgetting curve, and SRS algorithms.

0
Score
0
Streak
1/15
Question

Flashcards

Tap to flip. Twenty essential terms.

1 / 20
Mastery: —
Space flip · J/ next · K/ prev · 1/2 grade · S save
Daily Spacing Puzzle
A scheduling or retention question every day.
d

Teacher mode

Lesson outline, formula sheet, and a printable worksheet with answer key.

Lesson outline (45 min)

  • 5 min · Hook — Test the class on something they were taught last week without warning. Compare against material from this morning. Demo the forgetting curve in real time.
  • 10 min · Concept — Ebbinghaus and the curve. The spacing effect. Why cramming feels productive but isn't.
  • 10 min · Algorithms — Walk through Leitner (paper-friendly) and SM-2 (the basis of Anki). Show the simulator and Leitner visualizer on the projector.
  • 15 min · Build — Each student picks 10 facts from this week's content and writes them as flashcards. Schedule them on days 1, 3, 7, 21.
  • 5 min · Wrap — Discuss: spaced repetition works only with active recall; segue to that module.

Formula sheet

Forgetting curve
R(t) = e−t/S
R retrievability, t days since last review, S stability (days).
Spacing growth
Snew = Sold × growth
Each successful review multiplies stability; growth ≈ 2–3 in SM-2 / FSRS.
SM-2 next interval
In+1 = In × EF
EF (easiness factor) updated by grade. I1=1d, I2=6d, then ×EF.
EF update (SM-2)
EF += 0.1 − (5−g)·(0.08+(5−g)·0.02)
g ∈ {0..5}. Floor EF at 1.3. Grade ≥3 keeps progress, <3 resets interval.
Time to retention r
t = −S · ln(r)
Days until R drops to r. r=0.9 → t ≈ 0.105 S.
Leitner intervals
Bk ≈ 2k−1 days
Box 1: daily; 2: 2d; 3: 4d; 4: 9d; 5: 30d (common variants).

Worksheet