A graph that most students have seen
The forgetting curve is a graph that shows memory for something you learned falling over time, steeply at first and then more gradually. You may have seen it in study-skills posters, usually with a line that drops sharply in the first day or two. The idea behind it is simple and, in broad terms, well supported: without review, we lose a good deal of newly learned material quickly, and the loss slows down afterwards.
Where the picture gets shaky is in the numbers. Posters sometimes print exact percentages, such as how much you will remember after a day or a week. Those figures should not be taken at face value. They depend heavily on what was learned, how well, and how memory was tested.
Where it comes from
In the 1880s, the German psychologist Hermann Ebbinghaus studied his own memory. To avoid the effect of prior knowledge, he memorized lists of meaningless syllables and measured how much effort was needed to relearn them after different delays. He found that the savings in effort dropped fast at first and then leveled off. His work was a landmark in experimental psychology, and later researchers have reproduced a similar overall shape using other methods.
There are important caveats. His material was nonsense syllables, which are far harder to remember than meaningful content, and he was the only participant. Real learning tends to involve connected ideas, which decay differently. In practice, the rate of forgetting depends on factors such as:
- Meaningfulness. Material you understand and can link to what you already know tends to last longer.
- Initial learning. Something learned well the first time is generally forgotten more slowly.
- Repetition and recall. Each time you successfully retrieve something, forgetting tends to slow.
- Sleep, stress and attention at the time of learning.
- Interference from similar material learned before or after.
So the curve is a useful model of a general tendency, not a prediction for any individual topic.
What it means for review
If forgetting is fastest soon after learning, then review is most valuable early, and subsequent reviews can be spread further apart. This is the logic behind spaced repetition. A common pattern is:
- A quick review within a day. Even five or ten minutes of recalling what you learned can slow the decline.
- A second review a few days later.
- Later reviews at longer gaps, such as after a week, then a few weeks, depending on how long you need to keep the material.
You will notice this resembles the sample calendar in the article on spaced repetition. The forgetting curve explains why the gaps start short: that is when most of the loss happens. The review itself should be mostly active recall; see retrieval practice for ways to do it.
A small illustration
This is a hypothetical example to show the idea, not a measurement. Suppose you attend a lecture on Monday. Without review, you might find by Wednesday that a good part of it has become vague. If you spend ten minutes on Tuesday writing down what you can remember and checking your notes, you refresh the memory just as it is starting to slip. A second short quiz the following Monday then catches whatever is still shaky. The total time is small compared with relearning everything from scratch before an exam.
Reviewing at the right moment
How do you know when to review? A rough principle is to review when recall is starting to feel difficult but is still possible. Too soon, and you gain little because it is still fresh. Too late, and you have to relearn from scratch. You can find the point with experience: if you test yourself and get most items right with some effort, the gap was about right.
For exams on a fixed date, work backwards from the exam. Place the last review just before it, and earlier ones at widening gaps before that. A weekly revision plan can reserve slots for this. If your time is very limited, a short review soon after learning is usually better value than a long session much later; the article on limited study time offers more ideas.
Slowing the decline
Besides timing, you can influence how steeply memory fades:
- Understand, do not just copy. Explaining ideas in your own words and linking them to what you know gives memory more hooks.
- Organize. Grouping material into meaningful chunks reduces what must be remembered separately.
- Use multiple routes. Combining words and diagrams, as in dual coding, can make recall easier.
- Get enough rest. Evidence suggests sleep helps consolidate memory, though individual effects vary.
- Apply it. Using knowledge in practice problems or projects counts as repeated retrieval.
Misreadings to avoid
- “I will forget most of it anyway, so why bother?” The curve does not say learning is futile. It says review pays off.
- “The numbers are exact.” They are not. Treat any percentages as rough illustrations.
- “Forgotten means gone.” Relearning something you once knew is usually faster than learning it new, which suggests traces remain.
- “More repetition is always better.” Repeating immediately many times in a row gives less long-term benefit than repeating after a gap.
- “Forgetting is purely a failure.” Some forgetting is useful, since it removes details you do not need. The problem is losing what you want to keep.
A note on evidence
Studies of forgetting in meaningful, real-world material are more varied than the classic curve suggests, and reviewers caution against over-precise claims. The broad lesson, that memory fades without use and that well-timed review helps, is dependable. The precise timetable for your own material is something you find by testing yourself and adjusting.