Every student knows the cycle: cram the night before, perform adequately, forget nearly everything within a fortnight. The frustrating part is that the fix has been sitting in the research literature for over a century. Spaced repetition - reviewing material at growing intervals instead of all at once - is arguably the most consistently replicated finding in the science of learning, and almost nobody is taught to use it.
Why forgetting is the raw material of remembering
Memory research going back to Ebbinghaus describes a “forgetting curve”: new information decays rapidly - much of it within days - unless revisited. The counterintuitive discovery is that the most powerful moment to review something is when you are just beginning to forget it. The slight struggle to retrieve a fading memory is precisely what strengthens it; each successful effortful recall flattens the forgetting curve further. Review too soon and the retrieval is effortless and teaches little; too late and you are relearning from scratch. Spacing is the art of hitting that window repeatedly.
The other half: test yourself, do not reread
Spacing pairs with a second heavily-supported principle, retrieval practice: pulling information out of your head (flashcards, practice questions, explaining from memory) builds far stronger memory than pushing it back in (rereading, highlighting). Rereading is popular because it feels fluent - the material seems familiar, so you feel prepared. That feeling is a known illusion; familiarity is not recall. A spaced-repetition system is simply retrieval practice on an intelligent schedule.
How to actually run it
Option one: an app does the scheduling
Anki (free on desktop, the reference tool for medical students worldwide) and similar apps like Quizlet or RemNote track every card and resurface it at the optimal interval: got it right, the gap grows (1 day, 3 days, a week, a month); got it wrong, it shrinks. Your only jobs are writing good cards and showing up daily. Card-writing rules that matter: one fact per card, write your own cards (the writing is half the learning), and prefer questions (“What enzyme unzips DNA?”) to statements.
Option two: paper, with the Leitner system
Five boxes. New cards start in box 1, reviewed daily. Correct answers promote a card one box; each box is reviewed half as often as the last (daily, every 2 days, 4, 8, 16). Any mistake sends the card back to box 1. Simple, satisfying, and the physics of it matches the apps.
Option three: no cards, just a revision calendar
For essay subjects: after learning a topic, schedule active reviews (blank-page brain dumps, past-paper questions - not rereading) at roughly day 1, day 3, day 7, day 14, then monthly. Our 30-day exam preparation guide builds this structure into a full plan.
Why cramming persists (and when it sort-of works)
Cramming does move material into short-term availability, which is why it rescues tomorrow’s exam. What it cannot do is create durable memory - the curve reasserts itself within days - which is catastrophic for cumulative subjects, professional qualifications, and languages, where this term’s material is next term’s foundation. The honest role for cramming: a final-24-hours top-up layered over weeks of spaced work, not a substitute for it.
A realistic starter routine
- Same-day: convert today’s notes into 5-15 flashcards or questions (ten minutes).
- Daily: clear your app queue or box 1 - typically 15-20 minutes once established.
- Weekly: one past-paper question or blank-page dump per subject.
- Trust the schedule on the days material feels “already known” - that feeling is the illusion the system exists to correct.
Twenty minutes a day, every day, quietly outperforms the heroic all-nighter every time it has been tested. See our evidence-based study techniques guide for the companion methods that stack with it.
