The forgetting curve is a graph showing how quickly your memory of newly learned information decays over time. First described by German psychologist Hermann Ebbinghaus in 1885, it reveals that forgetting is steepest in the first hours after learning, then gradually levels off. The sharpest drop happens within the first day, and after that, whatever remains tends to stick around much longer.
How Ebbinghaus Discovered the Curve
Ebbinghaus ran his experiments on himself between 1879 and 1880, making him both researcher and sole test subject. He memorized lists of nonsense syllables, meaningless combinations like “DAX” or “BUP” that carried no prior associations. This was deliberate: by stripping away meaning, he could measure raw memory without the boost that comes from understanding or familiarity.
Each experiment involved memorizing eight series of 13 syllables, then relearning those same lists after a set delay. The key measurement was “savings,” meaning how much less effort it took to relearn something compared to learning it the first time. If a list originally took ten minutes to memorize but only four minutes to relearn the next day, that represented 60% savings, suggesting 60% of the memory trace remained even if conscious recall had faded. Over 163 double tests at various time intervals, Ebbinghaus mapped out a consistent pattern of decay.
The Shape of Forgetting
The curve drops steeply at first, then flattens into a long, slow tail. Within the first hour of learning something new, a significant portion is already gone. By 24 hours, roughly two-thirds of the original material has been lost. After a month, only a small fraction remains accessible without review.
What makes the curve so distinctive is its shape: it’s not a straight downward line. Most forgetting is frontloaded. You lose more in the first 20 minutes than you do between day two and day thirty. Ebbinghaus captured this mathematically, and his 1885 equation describes retention as a function of time using a logarithmic relationship. A 2015 replication published in PLOS ONE closely reproduced his original results, confirming that the basic shape of the curve holds up more than a century later.
Why Meaningful Information Fades More Slowly
Ebbinghaus deliberately used meaningless syllables, which means his curve represents something close to a worst-case scenario. Real-world information, the kind that connects to things you already know or care about, decays more slowly.
A large-scale survey of retention studies spanning everything from nonsense syllables to novels found that the type of material changes the shape of forgetting itself. Simple, meaningless items like isolated syllables follow the steep, curving drop Ebbinghaus described. But complex, meaningful material like stories, autobiographical memories, and narratives tends to fade in a more gradual, linear pattern. The memory trace for a novel you read degrades differently than the trace for a random string of letters, because richer material creates more connections in your brain that resist decay. Less complex memory traces degrade quickly, producing that dramatic early plunge that defines the classic forgetting curve.
Sleep Protects Against Forgetting
What happens between learning and forgetting matters, and sleep is one of the strongest protective factors. During sleep, your brain replays and strengthens new memories through a process called consolidation.
Research measuring how much people forget over a 12-hour window found a stark difference depending on whether that window included sleep. People who stayed awake after learning forgot about 33% of what they’d learned within 12 hours. People who slept during that same window forgot only about 17%. In a pilot study of older adults, the gap was even wider: 43% lost while awake versus just 13% after sleep. Sleeping after you learn something doesn’t just passively preserve memories. It actively stabilizes them, roughly cutting the rate of forgetting in half over the near term.
Spaced Repetition Flattens the Curve
The most effective strategy for counteracting the forgetting curve is spaced repetition: reviewing material at gradually increasing intervals rather than cramming it all at once. Each time you revisit something just as you’re about to forget it, you reset the curve and make the next drop less steep.
A large randomized study of over 26,000 physicians tested this directly. Those who used spaced repetition scored 58% on a knowledge assessment, compared to 43% for those who received no repetition, a meaningful gap. Doubling the number of spaced review sessions pushed scores even higher, to 62%. The effect persisted even when tested on material they hadn’t directly reviewed, suggesting spaced repetition doesn’t just help you memorize specific facts but strengthens the underlying knowledge enough to transfer to new situations.
Does Active Recall Beat Passive Review?
A popular claim is that testing yourself (active recall) produces dramatically better long-term retention than passively re-reading material. The reality is more nuanced. A controlled study comparing different review methods for anatomy knowledge, including multiple-choice testing, traditional lectures, and other formats, found no significant differences in retention at three months. Students who took a practice test retained about as much (scoring 53 out of 100) as those who attended a review lecture (55.5 out of 100).
The more important finding was that any form of revisiting the material mattered. Simply encountering the information again, regardless of format, enabled students to recall comparable amounts 14 weeks later. The act of re-exposure itself appears to be the critical ingredient, not whether that re-exposure is “active” or “passive.” This aligns with the core lesson of the forgetting curve: the enemy of memory is time without contact. Any review resets the clock.
Stress and Memory Retrieval
You might expect stress to steepen the forgetting curve, and cortisol (the body’s primary stress hormone) does interact with memory, but not always in the direction you’d assume. Cortisol released during or shortly after learning may actually enhance consolidation, helping lock in new information. The picture for retrieval is less clear: some older studies suggested stress at the time of recall impairs performance, but more recent controlled experiments found no significant difference in memory accuracy between stressed and non-stressed participants during a recognition task, even when cortisol levels were measurably elevated.
In practical terms, being stressed while studying might not hurt your memory formation, and it could even help. Being stressed during a test is less predictable, but the evidence for a large impairing effect is weaker than commonly assumed.
Practical Takeaways
The forgetting curve isn’t a fixed fate. Its steepness depends on what you’re learning, how meaningful it is to you, whether you sleep after studying, and how often you revisit the material. For meaningless or isolated facts, the drop-off is brutal: most is gone within a day. For rich, connected knowledge, the decline is gentler and more gradual.
If you want to retain something long-term, the evidence points to a simple formula: review it before you’ve completely forgotten it, space those reviews out over increasing intervals, and sleep between study sessions. Each review cycle makes the next forgetting curve shallower, until eventually the information stabilizes in long-term memory with minimal maintenance.