When I first prepared for a high-stakes exam, I did what had worked in medical school: I took notes, reread them, and did a few questions. Passive study. It had got me through the faculty exams, and it did not work at all for a licensing exam.
Here is what it looked like. I worked a lot and saw little result. My practice scores didn't move from month to month. I did few questions a day and didn't feel confident in the answers I picked. And above all, for every new thing I learned, I felt I was forgetting an older one, as if studying were an endless cycle of learn, forget, relearn.
If that sounds familiar, stop and re-evaluate. The usual cause is studying mostly with passive tools: reading, rereading slides, watching lecture videos without doing anything with them.
Active strategies cost more time and energy per session, and they pay it back many times over in retention and in your ability to apply what you know to a vignette. Making and reviewing flashcards is one of the best-studied of these strategies. It works through active recall: deliberately retrieving a fact from memory instead of re-reading it. Each retrieval strengthens the memory trace, so the next one is easier.
But it only works if the cards are good. Most decks aren't, and a bad deck wastes more time than no deck. Here are six habits that separate a deck that works from one that just makes you feel busy.
A cloze deletion is a sentence with a piece removed and replaced by a blank, which you fill in. It keeps the card to the minimum and is the best format for beginners. The research on it shows that reading and recall both get faster with each repetition.
Example: In STEMI, fibrinolysis is indicated only if transfer to a PCI-capable centre would take longer than […]. Answer: 120 minutes.
Everything you study should be phrased as simply as possible. A complex sentence makes your memory walk a maze; a simple one lights up fewer connections per retrieval, so learning is less tiring and more effective.
One card with two concepts needs more repetitions than two cards with one concept each, because each card can be scheduled at its own pace. Splitting content lowers the total number of reviews and improves retention. Call this information minimalism: each card carries only what is essential. Context can go in a note at the bottom of the card, never in the question.
Example: instead of one card listing the three rules for colonoscopy after diverticulitis, make three cards: one for the contraindication in the acute phase, one for the timing after the episode, one for the exceptions. The answer should be shorter than the question. Keep cloze deletions to one or two per card, unless it's the same term repeated.
A card should trigger the memory in as little time as possible: fewer errors, faster response, better concentration. One sentence per card, no more than two or three clauses.
Example: rather than If peritoneal cytology in a patient with gastric cancer is positive, what stage is assigned?, write Stage of gastric cancer with positive peritoneal cytology? Same content, half the reading.
First Aid and the review books are full of lists. Memorising them as lists is slow, and it usually needs a mnemonic that is itself one more thing to remember. Instead, pick the lists that matter, then reorganise them by a common denominator.
Example: the indications for intubating an asthmatic can be split into clinical criteria (accessory muscle use, altered mental status, inability to speak in full sentences, no response to treatment) and laboratory criteria (falling oxygen saturation, normalising pCO2 or pH). Two short cards, and each one makes you think rather than recite.
When two concepts look alike, they get confused: similar names, several dosages of one drug, two syndromes with overlapping presentations. This interference is one of the commonest sources of errors in flashcard learning, and it's hard to predict. Prevent it by describing each concept unambiguously, keeping to information minimalism, and adding a contextual cue.
Example: neuroleptic malignant syndrome and serotonin syndrome. Both have a psychiatric history, altered consciousness and haemodynamic instability, they sit in different chapters, and they're easily confused in a vignette. The fix is a systematic comparison: a table or an illustration of the distinguishing features (onset, reflexes, clonus, rigidity, the drug involved), turned into cards that ask for the difference, not for each syndrome in isolation.
We evolved to learn by looking. One good illustration can give you ten or twenty cards in a few minutes. Guidelines and review books are full of flowcharts and algorithms you can screenshot; with image occlusion you cover the parts you want to test and reveal them one at a time.
Example: an acute limb ischaemia management algorithm, with each decision node occluded in turn, becomes a set of cards that test the whole pathway.
Paper works, but for the USMLE the standard is Anki: free, and it schedules your cards by spaced repetition, showing new or difficult cards often and mastered ones rarely, which is what the evidence on retention supports. Whether you build your own deck from the questions you get wrong (see what to do with a question you got wrong) or start from a shared deck like the AnKing, the six rules above decide whether the cards do anything for you. The AnKing's YouTube channel covers the software side.
1. Use cloze deletions, from your first card to your last; they make reading and recall faster.
2. Practise information minimalism: when you keep forgetting a card, restructure it, split it, or cut it down.
3. Optimise the wording, the way you would simplify an equation.
4. Don't memorise lists; group the items by what they have in common.
5. Reduce interference with comparison tables, illustrations, examples, or a link to something from your own life.
6. Use image occlusion to turn one illustration into many cards.
Taveira-Gomes, T., Prado-Costa, R., Severo, M. and Ferreira, M.A. (2015). Characterization of medical students recall of factual knowledge using learning objects and repeated testing in a novel e-learning system. BMC Medical Education, [online] 15(1). Available at: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4326410/
Schmidmaier, R., Ebersbach, R., Schiller, M., Hege, I., Holzer, M. and Fischer, M.R. (2011). Using electronic flashcards to promote learning in medical students: retesting versus restudying. Medical Education, [online] 45(11), pp.1101–1110. Available at: https://pubmed.ncbi.nlm.nih.gov/21988625/
Supermemo.com. (2021). Knowledge structuring and representation in learning based on active recall. Available at: https://www.supermemo.com/en/archives1990-2015/english/ol/ks
Ramsay (2020). Are flashcards effective? The top 3 ways they can boost your grades. [online] Brainscape Academy. Available at: https://www.brainscape.com/academy/are-flashcards-effective/
