A 138-year-old experiment that still holds up
A Scientist Proved in 1885 That You Forget 66 Percent of Today by Tomorrow. In 2015, We Checked Again.
In 1885, a German psychologist tested his own memory for two years straight and charted exactly how fast it faded. In 2015, scientists ran the same experiment on new volunteers. The curve came back almost unchanged.
Hermann Ebbinghaus did not have a lab, a team, or a grant. He had himself, a stack of homemade flash cards, and a stopwatch. Starting in the early 1880s, he invented thousands of meaningless three-letter syllables (DAX, BUP, ZOL, the kind of thing that carries no story and no association) so that whatever he remembered later could not be explained by anything except memory itself.
He memorized list after list, then tested himself at set intervals: 20 minutes later, an hour later, nine hours later, a day later, two days, six days, a month. He measured not just what he could recall outright, but how much faster he could relearn a list he had already forgotten, a method he called savings. **The result, published in his 1885 book Memory: A Contribution to Experimental Psychology, is one of the most repeated findings in the field: forgetting is fast, and it is fastest right at the start.**
By his numbers, about 56 percent of what he had learned was already gone within an hour. By 24 hours, it was closer to 66 percent. After a week, roughly three-quarters had slipped away. The curve does not fall in a straight line, it drops hard in the first hours and then levels off, so whatever survives the first day tends to stick around much longer than the rest.
1. Modern scientists ran his experiment again. The curve barely moved.
For more than a century, Ebbinghaus's curve was taken mostly on faith, a single 19th-century man testing himself alone in a room. In 2015, Jaap Murre and Joeri Dros at the University of Amsterdam published a formal replication in the journal PLOS ONE. **A modern volunteer went through the same savings method Ebbinghaus used, learning and relearning lists at the same intervals, roughly 70 hours of testing in total.**
The new curve tracked the original one closely enough that the researchers called it a genuine replication, not just a similar shape. They even found the same small, specific quirk Ebbinghaus had reported: a slight uptick in the forgetting rate right around the 24 hour mark, a detail too particular to be a coincidence appearing twice, 130 years apart.
This was not a folk theory that got softer with retesting. A modern, formal replication reproduced Ebbinghaus's curve, down to details he had no way to fake.
2. The one detail Ebbinghaus himself flagged: meaning changes everything
Ebbinghaus was careful to note that his numbers describe the worst case. Nonsense syllables are designed to have no meaning, no story, nothing for the mind to hang on to. In his own writing, he estimated that meaningful material, the kind with a narrative or a connection to something you already know, is forgotten roughly ten times more slowly than a random string of letters.
**That gap is the whole difference between a fact and a memory that means something.** A phone number is closer to a nonsense syllable: it fades fast unless it is repeated. A story about the year your father left for the Army, or the summer your mother learned to drive, is not nonsense. It has people, stakes, a place, an ending. It survives the first brutal day far better. But it still survives inside one thing: whoever is carrying it. Once that person stops repeating it, the story starts on its own curve, the same curve, just delayed.
Meaningful memories fade slower than nonsense syllables, but they are not exempt from the curve. They just get a longer head start.
3. You cannot out-argue the curve. You can get ahead of it.
The forgetting curve is not really about flash cards. It describes something true about every memory, including the ones that live only in a parent's or grandparent's telling: how a grandfather actually talked, the exact joke a mother made every Thanksgiving, the specific reason a family moved the year everything changed. None of that is written down anywhere. It exists only as long as someone is willing and able to say it out loud.
**Ebbinghaus's whole point was that waiting is the enemy, not effort.** The steepest part of his curve is the earliest part. The same is true of a family story: the version told today, in someone's own voice, with their own details, is sharper than the version that gets reconstructed from memory five or ten years from now. A phone call does not stop the curve. It just gets there before it does the most damage.
The best time to record a story is not when it feels urgent. It is before the curve has a chance to work on it.
I always meant to ask my grandmother about coming over from Cuba, the actual details, not the summary I'd heard a hundred times. Ari asked her on a call while I was at work. By the time I listened to it that night, she had told the story better than I'd ever heard it, with names and dates I had never once heard before. I keep thinking about how close I came to never having that.Marisol, 34, recording her grandmother's story
What you get with LifeScribe
- Grounded in a documented psychological finding, Ebbinghaus's 1885 forgetting curve, independently replicated by University of Amsterdam researchers in 2015.
- Ari asks the next question on a call your family is already having, so a story gets captured before the curve starts working on it.
- No special occasion or health scare required: the steepest forgetting happens early, so the best day to record a story is usually today.
- The recording keeps the version with the most detail, the one told in a person's own voice, not the summary that survives after years of retelling.
Start free, with nothing to lose
Try it on your next call this week, free. Nothing to install, nothing to learn. If nothing worth keeping comes up, you have lost nothing but a few minutes.
**Ebbinghaus's data says the sharpest drop happens in the first day, not the first decade.** The same is true of a story nobody has said out loud in a while. The version that exists right now, today, is the sharpest one it will ever be.
The story is at its sharpest today. Record it before the curve gets to it.
Pick a call you already have planned this week. Let Ari ask the question that gets the version worth keeping.
Questions grandparents ask us
What exactly did Ebbinghaus's 1885 study find?
Hermann Ebbinghaus tested his own memory using invented, meaningless three-letter syllables so nothing else could explain what he recalled. Measuring both recall and relearning speed, he found that forgetting is steepest early on: roughly 56 percent of material was lost within an hour and about 66 percent within 24 hours, leveling off after that. He published the findings in his 1885 book Memory: A Contribution to Experimental Psychology.
Has anyone confirmed his findings with modern research?
Yes. In 2015, Jaap Murre and Joeri Dros of the University of Amsterdam published a formal replication in PLOS ONE, using the same savings method on a modern volunteer across roughly 70 hours of testing. The resulting curve closely matched Ebbinghaus's original data, including a small uptick in forgetting rate near the 24 hour mark that Ebbinghaus had also reported.
Does the forgetting curve apply to real memories, not just nonsense syllables?
Ebbinghaus himself noted that meaningful material is forgotten roughly ten times more slowly than nonsense syllables, since it connects to a story or something already known. Meaningful memories, including family stories, are more durable, but they are not exempt from fading. They rely on someone continuing to tell them, and once that stops, the same basic pattern of loss applies.
Why does this matter for family stories specifically?
A story that exists only in someone's memory, never written down or recorded, is subject to the same forces Ebbinghaus measured. The version told today, with full detail and in the person's own words, is sharper than the version pieced back together years later. Capturing it earlier preserves more of it.
How does this connect to LifeScribe?
The forgetting curve is steepest early, which means the best time to capture a story is before it has had years to fade, not after. Ari's job on a call is to ask the question that turns an ordinary conversation into a recording your family keeps in that person's own voice, while the detail is still there.
https://getlifescribeapp.com/blog/you-forget-66-percent-of-today-by-tomorrow