An effect that's been measured for over a century
The spacing effect isn't a study hack someone invented recently — it's one of the oldest and most replicated findings in learning research, first documented by Hermann Ebbinghaus in the 1880s. The finding is simple to state: information reviewed multiple times with gaps in between is retained far longer than the same amount of review time spent all at once. Studying something four times in one evening produces weaker long-term retention than studying it four times spread across two weeks, even though the total time invested is identical.
Why cramming feels like it works
Cramming produces a very real, very short-lived improvement — cram the night before an exam, and performance the next morning genuinely is better than no studying at all. The problem is the slope of forgetting afterward. Massed practice (all the review packed together) builds a memory that decays fast; spaced practice builds one that decays slowly. Because the immediate post-cram feeling is "I know this," and the failure only becomes visible weeks later when the material is needed again, cramming keeps getting chosen even though it's the worse long-term strategy. The feedback loop that would normally correct a bad habit is delayed past the point where it changes behavior.
The forgetting curve
Ebbinghaus also mapped out roughly how memory fades over time without review: fast at first, then leveling off. A large fraction of newly learned material is forgotten within the first day or two if it's never revisited. Each time information is successfully retrieved from memory before it's fully forgotten, the curve resets and flattens — the next forgetting curve is shallower than the one before it. This is the actual mechanism behind spaced repetition: it's not that repetition itself is magic, it's that retrieval, spaced to happen right around when the material would otherwise start slipping, strengthens the memory more than review that happens too early (when nothing has been forgotten yet, so there's nothing to relearn) or too late (when so much has faded that it's effectively being learned from scratch again).
Turning this into an actual schedule
The practical version of spaced repetition doesn't require anything complicated:
- Review new material within a day of first learning it. This is the first, most important gap — most forgetting happens in the first 24 hours, so an early review has an outsized effect on what sticks.
- Space subsequent reviews out with growing gaps — a few days, then a week, then a few weeks. The exact numbers matter less than the principle: each successful review earns a longer wait before the next one.
- Let review difficulty guide the schedule, not just a fixed calendar. If retrieval feels effortless, the gap can grow. If it feels shaky, shorten the next gap rather than pushing forward on an artificial timetable.
Flashcard software that implements spaced repetition automates step 3 by tracking how easily each item was recalled and adjusting its next review date accordingly — but the underlying idea works fine with a plain notebook and a rough personal sense of what's getting shaky.
Retrieval practice matters more than re-reading
A closely related finding: spacing helps most when the review itself involves actively retrieving information from memory (answering a question, solving a problem from scratch, explaining a concept without looking) rather than passively re-reading notes or a textbook. Re-reading feels productive because the material feels familiar on the page — but familiarity when the answer is right in front of you says very little about whether it can be retrieved later, unprompted, during an actual exam. Spacing and active retrieval reinforce each other: spaced re-reading is better than massed re-reading, but spaced retrieval practice beats both by a wide margin in the research.
What this means practically
The single highest-leverage change most students can make isn't studying more — it's redistributing the same amount of study time across more, shorter sessions instead of fewer, longer ones, and testing themselves during those sessions instead of just reviewing notes. A subject studied for six hours the night before a test is worse prepared, on average, than the same subject studied for one hour a night across six earlier days — the spacing effect is one of the few places in learning research where "less at once, more spread out" measurably beats "more, all at once," for the same total effort.