Most clinical revision fails for a boring reason: you study something once, feel like you know it, and then forget most of it within a week. Spaced repetition is the most reliable fix we have, and it is not complicated. Here is what it does, and why we did not just use an off-the-shelf version.
The forgetting curve
More than a century ago, Hermann Ebbinghaus showed that memory for new information decays quickly at first, then more slowly. Left alone, a fact you learn today is mostly gone in a few days. The useful part is what happens when you review: each well-timed review flattens the curve, so you forget more slowly the next time. Review at the right moments and the same fact stays accessible for months from only a handful of touches.
The testing effect
Rereading notes feels productive, but it mostly builds familiarity, not recall. Retrieving an answer from memory, the thing a flashcard forces you to do, is what strengthens it. This is the testing effect, and it is one of the most replicated findings in learning science. The struggle to remember is the point. An answer that comes back slowly, with effort, is being learned far more durably than one you simply reread.
The harder you have to work to recall something, within reason, the better you remember it next time.
What an SM-2 style scheduler does
SM-2 is the classic spaced-repetition algorithm that popular flashcard tools are built on. The idea is simple. Every time you see a card you rate how well it came back, from a blank to an easy hit. The scheduler uses that rating to set the next interval:
- Rate a card poorly and it comes back soon, in minutes or the same day.
- Rate it well and the interval grows, from days to weeks to months.
- Each card carries its own ease factor, so genuinely hard cards keep returning more often.
The result is that your limited study time is spent almost entirely on the things you are about to forget, and barely at all on the things you already know cold.
Why we built our own
A generic scheduler treats every card the same. Clinical knowledge is not the same. A dosing threshold you must never get wrong should behave differently from a piece of background context. Cards cluster into topics and sub-topics, and weakness in one often predicts weakness in a neighbour. And the goal is not a tidy review streak, it is being ready for a specific assessment on a specific curriculum.
So our scheduler is tuned for that. It weights what is due against what you are weakest on, it understands the structure of your syllabus rather than treating cards as a flat pile, and it folds the result into a single daily recommendation instead of leaving you to manage decks. You open the app and it tells you the twelve cards that matter most today.
How it shows up in Scripter Academy
In practice you never see the algorithm. You see a short session, sized to the time you have, that surfaces due cards first and then introduces new ones. You rate each card after you flip it. Over days, the topics you find hard rise to the top, the ones you have mastered fade into the background, and your topic analytics show the curve bending the right way. That is the whole trick: small, consistent, well-timed effort, pointed exactly where it counts.