Why Do We Forget What We Studied?
Forgetting is not a personal failing and it is not random. It follows a curve that was measured in the 1880s — and it can be interrupted.
The short answer
Forgetting follows the forgetting curve described by Hermann Ebbinghaus in 1885: retention drops steeply soon after learning and then levels off. The main cause is not decay but retrieval failure — the memory is stored but the route to it weakens without use. Reviewing at increasing intervals, known as spaced repetition, resets the curve each time and makes it flatter, so the same total study time produces far better long-term retention when distributed than when massed into one session.
Transcript
Why do you forget things you definitely studied? It is not a personal defect, and it is not random. Forgetting follows a curve, and it was measured over a century ago.
In 1885 Hermann Ebbinghaus tested his own memory, systematically, over months. He found that retention drops steeply right after learning — you can lose a large share within the first day — and then the decline flattens out. Thats the forgetting curve.
Now the important part. Mostly you are not losing the memory. You are losing access to it. The trace is still stored; the retrieval path has weakened from disuse. Thats why a hint or walking back into the room often brings it straight back.
So what interrupts the curve? Reviewing at the right moment. Each time you successfully retrieve something, the curve resets — and it resets flatter. Forgetting slows down after every review.
This is spaced repetition: review after a day, then three, then a week, then a month. Same total study time, dramatically better retention, because the spacing does the work.
And reviewing just as you are starting to forget is the strongest moment of all.
Test yourself on Study Skills
5 questions, easy to hard. No account needed to try it.
Who first measured the forgetting curve?
What shape does the forgetting curve take?
According to retrieval failure, what has usually gone wrong when you cannot recall something?
What does spaced repetition do to the forgetting curve?
When is a review most effective for long-term retention?
The longer answer
Forgetting feels like a personal weakness, which is why most students respond to it by studying harder rather than differently. The research points the other way. Forgetting is a regular, measurable property of human memory, it follows a known shape, and the shape can be changed by when you review rather than by how much effort you put in.
The foundational work is Hermann Ebbinghaus's, published in 1885. Wanting to study memory without the confound of prior knowledge, he invented nonsense syllables — meaningless three-letter strings — learned lists of them to a criterion, and then tested himself at intervals over hours, days and weeks, using the time needed to relearn a list as his measure of what remained. He was his own sole subject, which is a methodological limitation, but the pattern he found has been replicated repeatedly since with better designs.
The result is the forgetting curve. Retention falls steeply immediately after learning, with a substantial portion lost within the first day, and the decline then flattens: what survives the first few days decays far more slowly. The curve is roughly exponential, not linear. This matters practically, because it means the hours right after studying are when material is most at risk, and it also means that material which has already survived a week is in much less danger than material studied last night.
The next question is what forgetting actually consists of, and here the intuitive model is wrong in a useful way. It is tempting to imagine memories fading like ink, decaying until nothing remains. For most study material, the dominant mechanism is retrieval failure rather than decay: the memory trace is still stored, but the route to it has weakened through disuse. The everyday evidence is familiar — the answer that will not come in the exam but arrives an hour later, the name you cannot produce until someone gives you the first letter, the tip-of-the-tongue state in which you can report the word's length and starting sound while being unable to retrieve the word. None of that is possible if the information is gone. Context effects point the same way: returning to the room where you learned something, or to the same mood or physical state, measurably improves recall.
Interference is the other major contributor. New learning competes with old for the same retrieval cues. Proactive interference is older material obstructing newer; retroactive interference is newer obstructing older. Interference is strongest when the competing material is similar, which is why studying two closely related topics back to back — two languages, two sets of dates, two similar formulas — produces more confusion than studying two unrelated ones.
The intervention that follows from all this is spaced repetition, and its effect is large enough that it is one of the most robust findings in educational psychology. Each successful retrieval does two things: it restores retention toward its original level, and it flattens the subsequent curve so that forgetting proceeds more slowly than it did the first time. Review again and the curve flattens further. This is why the intervals can expand: a first review after about a day, a second after three, then a week, then two weeks, then a month, each spacing chosen so that review arrives before the material is lost but late enough that recall requires effort.
The comparison that makes the point is the spacing effect. Hold total study time constant and distribute it across several sessions rather than massing it into one, and long-term retention improves substantially — often dramatically — at no extra cost in hours. The variable that changes is not effort but scheduling.
The timing principle underneath this is desirable difficulty. Retrieval strengthens memory in proportion to how effortful it is: recalling something that comes instantly does relatively little, while recalling something you nearly could not produce does a great deal. Reviewing five minutes after learning is close to useless for long-term retention because nothing has been forgotten yet and no effort is required. The most valuable moment for review is the point at which you are just beginning to forget — retrieval is difficult but still succeeds. This also explains why the popular practice of rereading notes feels productive and delivers so little: the material is right in front of you, no retrieval occurs, and fluency at reading is mistaken for durable knowledge.
Sleep deserves a mention, because it interacts directly with all of this. Memory consolidation — the process that stabilises new memories and integrates them with existing knowledge — happens substantially during sleep. Losing sleep to extend a study session therefore has a cost on both sides: fewer hours of consolidation, and impaired encoding the following day.
The practical programme is short. Space the study of any given material across days rather than concentrating it. Review just before you expect to have forgotten, not immediately after learning. Make each review an act of retrieval — close the book and produce the answer — rather than re-exposure. Separate similar topics in time to limit interference. And protect sleep, particularly on the nights following heavy learning. None of these require more total hours than the usual approach; they redistribute the same hours into the pattern memory actually responds to.