DecksThe Forgetting Curve: What Ebbinghaus Actually Found
The forgetting curve shows that memory fades fast at first and then more slowly. Hermann Ebbinghaus measured it on himself in the 1880s: most of the loss happened within the first day, and after that forgetting slowed down a lot. Spaced repetition schedulers are built around exactly that shape.
What he did
Ebbinghaus learned lists of nonsense syllables until he could recite them perfectly, waited, then relearned them and timed how long it took. The time he saved the second time, compared with the first, measured how much was still in memory. That measure is called savings.
His numbers
| Time since learning | Work saved in relearning |
|---|---|
| 20 minutes | 58.2% |
| 1 hour | 44.2% |
| 8.8 hours | 35.8% |
| 1 day | 33.7% |
| 2 days | 27.8% |
| 6 days | 25.4% |
| 31 days | 21.1% |
From Table 23 of Memory: A Contribution to Experimental Psychology. Note that these are savings, not the share of syllables he could still recite, and they come from one person learning nonsense material. Meaningful material, like words in a language you're learning, is generally forgotten more slowly. The shape is what has held up: steep, then shallow.
What it means for flashcards
- Review early, then less often. The steep start is why a new card comes back within minutes and again the next day.
- Each review flattens the curve. After a successful review, forgetting is slower than before, so the next gap can be longer. That's why intervals grow.
- Timing matters. Reviewing just before you'd forget is the most efficient point: early enough to still recall it, late enough that recalling it takes effort.
How spaced repetition works in Decks shows how those gaps are calculated.
Source: Ebbinghaus, H. (1885/1913). Memory: A Contribution to Experimental Psychology, section 28. psychclassics.yorku.ca/Ebbinghaus.
Published 2026-10-07.