The Ebbinghaus forgetting curve is a model showing how quickly memories fade after you first learn something. The drop is steep: within the first hour, you lose a significant portion of new information, and after a month, retention can fall to near zero for material you never revisited. The concept comes from experiments conducted by German psychologist Hermann Ebbinghaus in 1880, and it remains one of the most replicated findings in memory research.
How Ebbinghaus Discovered the Curve
Ebbinghaus wanted to study memory in its purest form, stripped of meaning and prior associations. To do this, he invented nonsense syllables, combinations of letters like “DAX” or “BUP” that had no built-in meaning. This was deliberate: real words carry emotional weight, personal connections, and varying levels of familiarity, all of which would muddy the results. He later verified his findings with more natural material like poems, but the nonsense syllables were the core tool.
He was also his own test subject. Ebbinghaus memorized lists of these syllables, then tested himself at various intervals to see how much effort it took to relearn them. His measurement method, called “savings,” was elegantly simple. If a list originally took 20 minutes to memorize but only 12 minutes to relearn the next day, that 40% reduction in effort was his savings score. Higher savings meant more memory had survived; lower savings meant more had been lost. By plotting these savings scores across different time intervals, the forgetting curve emerged.
How Fast You Actually Forget
The curve is not a gentle slope. It drops sharply in the first few hours, then gradually levels off. A 2015 replication study published in PLOS One by Jaap Murre and Joeri Dros closely followed Ebbinghaus’s original method and produced strikingly similar results, even though the replication was conducted in Dutch over a century later.
Here’s what the retention data looks like at each interval, measured as savings (the percentage of relearning effort saved compared to the original):
- 20 minutes: 44% retained
- 1 hour: 47% retained
- 9 hours: 28% retained
- 1 day: 32% retained
- 2 days: 23% retained
- 31 days: 4% retained
A few things stand out. The biggest drop happens between one hour and nine hours. And there’s a curious bump at the 24-hour mark, where retention actually ticks up slightly compared to nine hours. The researchers concluded this jump is a real feature of the curve, not random noise, possibly reflecting the role of sleep in consolidating memories. By 31 days, almost nothing remains of material that was learned once and never reviewed.
Why the Drop Is So Steep
The forgetting curve captures what happens to isolated, meaningless information that you encounter once. That’s the worst-case scenario for memory, and it’s also surprisingly common in daily life. Think of a phone number someone reads to you, a name at a party, or a fact from a lecture you didn’t revisit. Without connections to existing knowledge, emotional significance, or repeated exposure, your brain treats new information as disposable.
Several factors influence how steep or shallow your personal curve is for any given piece of information. Material that connects to things you already know decays more slowly than random facts. The complexity of the material matters too: details tend to fade faster than the overall gist of something. Sleep plays a measurable role in memory consolidation, and disrupted sleep accelerates forgetting. Even what you do immediately after learning matters. If you switch to a distracting task right away, you lose the support of short-term memory before the information has a chance to solidify.
What Flattens the Curve
Ebbinghaus himself identified the most powerful countermeasure: spacing your review sessions over time. This is now known as the spacing effect, and it’s one of the most consistently replicated findings in all of cognitive psychology. When you revisit material at increasing intervals, say one day later, then three days, then a week, each review session resets the curve and makes the next drop less steep. Over several cycles, information that would have vanished in days can stick for months or years.
The reason spacing works is counterintuitive. The slight forgetting that happens between review sessions is actually productive. When you struggle to retrieve something you’ve partially forgotten, the act of retrieval strengthens the memory trace far more than simply rereading it while it’s still fresh. This is why cramming the night before an exam can get you through the test but leaves you with almost nothing a week later. You never gave your brain the chance to forget and recover, which is the cycle that builds durable memory.
Overlearning, continuing to practice material even after you can recall it perfectly, also slows the rate of decay. So does making the material meaningful. Connecting a new fact to a vivid image, a personal experience, or an existing framework of knowledge gives your brain more retrieval paths, which makes forgetting harder. This is why Ebbinghaus used nonsense syllables in the first place: he wanted to isolate pure forgetting from all the tricks your brain naturally uses to hold onto meaningful information.
Does the Curve Hold Up Today?
Remarkably well. The 2015 Murre and Dros replication followed Ebbinghaus’s method as closely as possible and found that, despite being conducted in a different language and more than 130 years later, the resulting curve was very similar to the original. An earlier German replication had found the same. The researchers described the resemblance across all four forgetting curves as “remarkable,” given the expected differences between individual people.
The one notable deviation was at the 31-day mark, where the modern replication showed even lower retention (about 4%) than Ebbinghaus originally recorded. But the overall shape of the curve, the steep initial drop followed by a long, slow tail, was consistent across all attempts.
It’s worth noting that the forgetting curve describes a specific type of memory loss: the decay of newly learned, relatively meaningless material after a single exposure. It doesn’t predict how quickly you’ll forget your wedding day, how to ride a bike, or the plot of your favorite movie. Motor skills, emotionally significant events, and deeply understood concepts follow different retention patterns. The curve is most directly applicable to the kind of learning people do in classrooms, training programs, and language study, where you’re trying to retain large amounts of factual information that doesn’t naturally stick on its own.
Practical Implications for Learning
The forgetting curve isn’t just a piece of psychology trivia. It’s the scientific foundation for spaced repetition systems, which are now built into popular flashcard apps and language learning platforms. These tools automate the spacing effect by showing you information right around the time you’re about to forget it, which is the optimal moment for review.
If you’re studying without software, a simple version of the same principle works. Review new material within the first day, then again after a few days, then after a week. Each review doesn’t need to be long. The goal is repeated retrieval, actually testing yourself rather than passively rereading, spread across increasing intervals. Three short, spaced review sessions will produce dramatically better long-term retention than one long study session of the same total duration.
The core insight of the forgetting curve is that forgetting isn’t a sign of failure. It’s the default. Your brain forgets most of what it encounters, and it does so quickly. The difference between information that vanishes and information that lasts isn’t talent or intelligence. It’s timing.