The forgetting curve: how fast you forget, and how to beat it
In 1885, one psychologist measured exactly how quickly memories fade. His curve still holds today, and so does the way to flatten it.
The forgetting curve shows that we forget most new information within a day, then the loss slows down. Psychologist Hermann Ebbinghaus measured it in 1885, and it has been replicated since. The way to flatten it is to review at increasing intervals and test yourself instead of rereading.
What is the forgetting curve?
The forgetting curve describes how memory of something you just learned declines over time when you make no effort to keep it. It drops steeply in the first hours, keeps falling over the first day or two, then flattens into a long tail. A small part sticks for a long time. Most of it does not.
It was first measured by the German psychologist Hermann Ebbinghaus, who published his results in 1885 in Über das Gedächtnis (Memory: A Contribution to Experimental Psychology). It was one of the first experiments ever done on memory.
An illustration, not a prediction. The curve follows Ebbinghaus' own measurements (about a third left after a day, a fifth after a month), and each review helps more when you have started to forget, as the research on spacing shows. Real memory depends on you and on the material.
Try it: compare cramming three times on the first day with four short reviews spread over two weeks. Almost the same effort, spent at better moments, leaves about twice as much a month later.
How Ebbinghaus measured it
Ebbinghaus was his own test subject. To avoid the effect of meaning and prior knowledge, he memorized lists of nonsense syllables like "DAX" or "BOK" until he could recite them perfectly. Then, after a delay, he learned the same lists again and timed how long it took.
The time he saved the second time was his measure of how much he still remembered. This is called the savings method. If relearning took half as long as the first time, 50% of the memory was still there in some form.
The numbers behind the curve
These are Ebbinghaus' savings scores at each delay. They show how much faster relearning was, not how many syllables he could recite, but the trend is what matters.
| Time after learning | Savings when relearning |
|---|---|
| 20 minutes | 58% |
| 1 hour | 44% |
| About 9 hours | 36% |
| 1 day | 34% |
| 2 days | 28% |
| 6 days | 25% |
| 31 days | 21% |
Two things stand out. First, more than half was lost within an hour. Second, after the first couple of days the decline almost stops: what survives the first week tends to stay a while.
In 2015, Jaap Murre and Joeri Dros at the University of Amsterdam repeated the experiment with the same method and delays. Their subject spent 70 hours learning and relearning lists, and the resulting curve was very close to the 1885 original, with one surprise: a small bump at the 24-hour mark, possibly linked to sleep helping to consolidate memories.
Does it apply to real learning?
Mostly, yes. You forget meaningful material, like the ideas in a book or the content of a course, more slowly than nonsense syllables, because you can connect it to things you already know. But the shape is the same: a steep early drop and a long tail. It explains a familiar experience: finishing a great book, feeling you understood it, and being unable to explain its main argument a month later.
It also explains why cramming works for tomorrow's exam and fails for next semester. Cramming fills the steep part of the curve, then lets it fall.
How to beat the forgetting curve
- Review at increasing intervals. Each time you successfully recall something, the curve that follows is flatter, so you can wait longer before the next review. This is the spacing effect, one of the most replicated results in psychology, and the idea behind spaced repetition.
- Test yourself instead of rereading. Pulling an answer from memory strengthens it far more than looking at it again. Questions beat summaries. Our guide on how to remember what you read covers the research.
- Make it meaningful. Put ideas in your own words and connect them to what you already know or have experienced. Connected memories fade more slowly than isolated facts.
- Review the next day. The first 24 hours are where most forgetting happens, so the first review matters most. A two-minute quiz the next morning does more than an hour of rereading a week later.
- Sleep on it. Sleep helps consolidate new memories. Learning in the evening and reviewing the next day is a simple combination that works.
A simple review schedule
You don't need a perfect algorithm to start. This schedule is a good default for anything you want to keep:
| Review | When | What to do |
|---|---|---|
| 1st | The next day | Answer your questions from memory, then check |
| 2nd | 3 days later | Same questions; note the ones you missed |
| 3rd | 1 week later | Focus on the ones you missed before |
| 4th | 3 weeks later | Quick check of everything |
| 5th | 1–2 months later | Anything you still get right can wait much longer |
If you get something wrong, start it again at a short interval. If something is easy, you can stretch the gap. That adaptive version is exactly what spaced repetition software does.
Want the actual dates? The review schedule calculator turns this schedule into days on your calendar, for the long term or for an exam.
How Prismo uses the forgetting curve
Prismo is the app I built around this curve. You write a few lines about what you learned, or dictate them. Prismo turns them into pieces of knowledge and writes quiz questions from them. Its scheduler, based on the SM-2 spaced repetition algorithm, brings each question back after a short interval, then waits longer every time you answer correctly, and sooner when you don't. You just open the app for a few minutes a day.
Sources
- Ebbinghaus, H. (1885). Über das Gedächtnis: Untersuchungen zur experimentellen Psychologie. Leipzig: Duncker & Humblot.
- Murre, J. M. J., & Dros, J. (2015). Replication and analysis of Ebbinghaus' forgetting curve. PLOS ONE, 10(7), e0120644.
- Dunlosky, J., et al. (2013). Improving students' learning with effective learning techniques. Psychological Science in the Public Interest, 14(1), 4–58.
