The forgetting curve
Memory fades fast right after you learn something, then slower. Every well-timed review resets the drop and makes it shallower. Drag the review days below and watch how much is left on day 60, a month after a day-30 exam.
Drag a numbered handle (or focus it and use ← →). Model: retention = e−t/S; each review multiplies stability S by 2.4. Illustrative, not a prediction for your brain.
What Ebbinghaus actually found
In Über das Gedächtnis (1885), Hermann Ebbinghaus learned lists of nonsense syllables such as "DAX" and "BOK" until he could recite them perfectly, then measured how much less time it took to relearn them later. That "savings" score dropped steeply at first and then levelled off:
| Delay | Savings (Ebbinghaus, 1885) |
|---|---|
| 20 minutes | 58% |
| 1 hour | 44% |
| 9 hours | 36% |
| 1 day | 34% |
| 2 days | 28% |
| 6 days | 25% |
| 31 days | 21% |
A 2015 replication by Murre and Dros at the University of Amsterdam reproduced the curve closely, including a small bump after the one-day delay that they attributed to sleep.
What it means for studying
- The first day matters most. A quick review the day after a lecture catches the material before the steepest part of the drop.
- Each review buys a longer gap. That is why schedules such as 1, 3, 7 and 14 days work, and why the Leitner system doubles the gap each time you get a card right.
- Cramming works, briefly. Reviews on days 28 and 29 give a high score at the day-30 exam and much less by day 60. Compare the two presets above.
- Meaning slows forgetting. Ebbinghaus used nonsense on purpose. Material you understand and connect to other ideas fades more slowly, so combine spacing with the Feynman technique.
To put this into practice, the study planner schedules spaced reviews back from your exam date, and the flashcard maker spaces individual cards for you.
Questions students ask
What is the forgetting curve?
The forgetting curve describes how quickly memory fades after learning if you do not review. Hermann Ebbinghaus measured it on himself in 1885 by learning lists of nonsense syllables and testing how much time he saved relearning them after delays from 20 minutes to 31 days.
How fast do we forget?
In Ebbinghaus’s data, the savings fell to about 58% after 20 minutes, 44% after an hour, 34% after a day and 21% after a month. Meaningful material you understand is forgotten much more slowly than nonsense syllables, but the shape, steep at first then flattening, holds widely.
How do you beat the forgetting curve?
Review before you have forgotten completely, and space the reviews out: each successful review makes the next drop slower, so the gap before the following review can be longer. Reviewing by testing yourself (active recall) rather than rereading makes each review count more.
Is the curve on this page exact?
No. It uses a simple exponential model, retention = e^(−t/S), where the stability S grows each time you review. It illustrates the shape and the effect of spacing; your real forgetting rate depends on the material, how well you understood it and sleep, among other factors.