I spent two years building flashcard software before realizing most of my users weren't actually using flashcards for half their study time.
A med student told me she'd spend 90 minutes on Anki, then another two hours doing practice questions and teaching concepts to her study partner. A software engineer said he used flashcards for APIs and syntax, but learned system design by drawing architectures from memory on his whiteboard.
They were all doing retrieval practice — the act of pulling information from memory rather than passively reviewing it. Flashcards are one implementation. There are at least six others worth knowing.
TL;DR
Retrieval practice means testing yourself, not re-reading. Flashcards work for discrete facts, but free recall handles concepts, the blank-page method reveals gaps, the Feynman technique catches shallow understanding, teach-back builds fluency, problem sets develop application skills, and low-stakes quizzes simulate exam conditions. Match the technique to what you're learning.
Why retrieval practice works
When you force your brain to retrieve information, you strengthen the neural pathways to that information. This is the testing effect, documented in over 100 years of memory research.
A 2011 study in Science by Karpicke and Blunt compared four study methods: concept mapping, retrieval practice, elaborative studying, and re-reading. Students who practiced retrieval recalled 50% more on a final test one week later than students who used concept maps or re-read the material.
The mechanism: retrieval is effortful. That effort triggers memory consolidation. Passive review feels easier, but it doesn't build durable memory traces.
Flashcards are the most popular retrieval tool because they're structured and repeatable. But they're optimized for atomic facts — vocabulary, formulas, drug names, historical dates. For everything else, you need different retrieval practice techniques.
1. Free recall (brain dump)
Close your notes. Set a timer for 10 minutes. Write down everything you remember about the topic.
No structure, no prompts. Just dump what's in your head onto paper or a blank document.
When to use it: After reading a textbook chapter, watching a lecture, or finishing a study session. Free recall shows you what actually stuck versus what felt familiar while you were reviewing.
Example: You just finished a chapter on the Krebs cycle. Close the book. Write out the steps, substrates, enzymes, and energy yields from memory. When you're done, open the book and check what you missed.
Why it works: Free recall has no retrieval cues. You're generating the structure yourself, which is harder than recognizing the right answer in a multiple-choice question or responding to a flashcard prompt. That difficulty is what strengthens memory.
I use free recall in SmartRecall's "Brain Dump" mode — it gives you a blank text box and tracks how much you wrote compared to previous sessions. No grading, just a forcing function to retrieve.
Limitation: Free recall is exhausting. You can't do it for 90 minutes straight like you can with flashcards. Use it strategically, not as your only method.
2. Blank-page method (structured recall)
Similar to free recall, but with structure. Draw a blank diagram, table, or outline from memory, then fill it in.
When to use it: For visual or hierarchical information — anatomical structures, org charts, decision trees, network topologies, taxonomies.
Example: You're studying cardiac anatomy. Draw the four chambers, valves, and major vessels from memory. Label the flow of oxygenated vs. deoxygenated blood. Then compare your drawing to a reference diagram.
Or you're learning React component lifecycle. Draw the lifecycle diagram from memory, including mounting, updating, and unmounting phases, with method names in each phase.
Why it works: Spatial memory is powerful. When you reconstruct a diagram, you're encoding both the content and its spatial relationships. That dual encoding makes retrieval easier later.
A 2018 study in Applied Cognitive Psychology found that students who drew diagrams from memory outperformed students who copied diagrams or studied them passively.
Limitation: Only works for content that has a visual or structural representation. Not useful for procedural knowledge or abstract concepts.
3. Feynman technique (explain simply)
Named after physicist Richard Feynman. Explain a concept in simple language, as if teaching it to someone with no background knowledge.
When to use it: When you need to understand a concept deeply, not just memorize it. Ideal for foundational topics that underpin everything else — thermodynamics, object-oriented programming, supply and demand, the scientific method.
Example: You're studying how TCP handles congestion control. Explain it out loud or in writing without jargon: "When a network gets crowded, TCP slows down how fast it sends data. It does this by..."
If you can't explain it simply, you don't understand it. The Feynman technique exposes gaps in your mental model.
Why it works: Explaining forces you to organize information into a coherent narrative. You can't fake understanding when you have to generate the explanation yourself.
Limitation: Time-intensive. You can't Feynman-technique 50 vocabulary words. Save it for the 5-10 core concepts that everything else depends on.
4. Teach-back (explain to a person)
Explain what you learned to another person. They ask questions. You answer.
When to use it: When you have a study partner or can recruit a friend, roommate, or family member. Works for any topic where verbal explanation is possible.
Example: You're studying the difference between Type 1 and Type 2 diabetes. Explain it to your roommate. They ask, "Wait, so which one needs insulin?" You answer. They ask, "Why does Type 2 happen?" You answer. If you can't answer, you've found a gap.
Or you're learning SQL joins. Explain INNER JOIN, LEFT JOIN, and RIGHT JOIN to a friend using a real-world analogy (two lists of people, matching by email address). They ask clarifying questions until they get it.
Why it works: Teaching is the highest form of retrieval practice. You're not just recalling facts — you're adapting your explanation to another person's understanding, which requires flexible, deep knowledge.
A 2018 meta-analysis in Educational Psychology Review found that students who expected to teach material learned it more thoroughly than students who expected to take a test.
Limitation: Requires another person. If you don't have a study partner, record yourself explaining the concept and watch it back. Not as effective, but better than nothing.
I've seen SmartRecall users create "teaching decks" — flashcards where the prompt is "Explain X to a beginner" and they record a voice memo as the answer. It's a hybrid approach.
5. Problem sets (application practice)
Solve problems that require you to apply what you learned. No looking at the solution until you've attempted it.
When to use it: For any skill-based domain — math, physics, programming, accounting, statistics, engineering, chemistry.
Example: You're studying integration techniques in calculus. Do 10 integration problems from the textbook. Don't look at the solution manual until you've tried each one.
Or you're learning Python. Write a function that takes a list of dictionaries and returns the dictionary with the highest value for a given key. No Googling until you've attempted it.
Why it works: Problem sets force you to retrieve the relevant concept, recognize which technique applies, and execute it. That's three layers of retrieval in one exercise.
Flashcards can help you memorize the integration rules, but they won't teach you when to use integration by parts vs. u-substitution. Problem sets do.
Limitation: Requires well-designed problems. If the problems are too easy, you're not retrieving. If they're too hard, you're guessing. Look for problems slightly above your current level.
6. Low-stakes quizzes (exam simulation)
Take a practice test under timed, closed-book conditions. Grade yourself immediately after.
When to use it: When preparing for an actual exam — USMLE Step 1, MCAT, NCLEX, AWS certification, JLPT N3, HSK 5, or any high-stakes test.
Example: You're studying for the MCAT. Take a 30-question practice section on biochemistry. Set a timer for 45 minutes. No notes, no phone. When time's up, grade it and review every question you missed.
Or you're preparing for a coding interview. Do a LeetCode medium problem in 25 minutes. No hints, no looking at the solution. When you're done, compare your solution to the optimal one.
Why it works: Low-stakes quizzes simulate the retrieval conditions of the real test. You're practicing not just the content, but the retrieval under time pressure and without external cues.
A 2014 study in Psychological Science found that students who took practice tests outperformed students who re-studied the material, even when both groups spent the same amount of time.
Limitation: Only useful if you have access to high-quality practice questions. Bad practice questions teach you the wrong thing.
SmartRecall integrates with question banks for USMLE, MCAT, and NCLEX, so you can alternate between flashcards and practice questions in the same session.
7. Flashcards (cued recall)
The classic. A prompt on one side, an answer on the other. You see the prompt, retrieve the answer, flip the card, and grade yourself.
When to use it: For discrete, atomic facts — vocabulary, formulas, drug names, historical dates, anatomy labels, API syntax, keyboard shortcuts.
Example: You're learning Spanish. Front of card: "to run." Back of card: "correr." You see "to run," retrieve "correr," flip, and check.
Or you're studying pharmacology. Front: "What is the mechanism of action of metformin?" Back: "Decreases hepatic glucose production, increases insulin sensitivity."
Why it works: Flashcards provide a retrieval cue (the prompt) and immediate feedback (the answer). They're efficient for high-volume memorization.
Spaced repetition algorithms like SM-2 and FSRS optimize flashcard review schedules, showing you cards right before you're about to forget them. That's why tools like Anki and SmartRecall are so effective for long-term retention.
Limitation: Flashcards are terrible for concepts, procedures, and application. If you can't reduce it to a question-answer pair, don't force it into a flashcard.
A common mistake: making flashcards that say "Explain the Krebs cycle." That's not a flashcard — that's a free recall prompt disguised as a flashcard. Use the right tool for the job.
Matching technique to content
Here's how I think about it:
- Discrete facts (vocabulary, dates, formulas): Flashcards
- Concepts (theories, models, principles): Feynman technique, teach-back
- Visual/spatial information (diagrams, maps, structures): Blank-page method
- Procedures (algorithms, protocols, workflows): Problem sets, teach-back
- Application skills (math, coding, clinical reasoning): Problem sets, low-stakes quizzes
- Broad topic review (after a lecture or chapter): Free recall
Most study sessions should use 2-3 techniques. I'll do 30 minutes of flashcards, 20 minutes of problem sets, and 10 minutes of free recall. That hits memorization, application, and synthesis in one hour.
Retrieval practice in SmartRecall
SmartRecall started as a flashcard app, but we've added support for other retrieval practice techniques:
- Brain Dump mode for free recall
- Diagram practice where you reconstruct images from memory
- Problem set integration with question banks for USMLE, MCAT, and NCLEX
- Teaching prompts that ask you to explain concepts in your own words
The spaced repetition algorithm schedules all of these, not just flashcards. If you did a brain dump on the Krebs cycle today, SmartRecall will remind you to do it again in three days, then a week, then two weeks.
Start with one new technique
If you're only using flashcards, pick one technique from this list and try it this week.
If you're studying for an exam, add low-stakes quizzes.
If you're learning a conceptual subject, try the Feynman technique on the three most important ideas.
If you're working through a textbook, do free recall after each chapter.
Retrieval practice works. The question is whether you're using the right retrieval practice for what you're learning.
If this framing is new to you, the foundational comparison is active recall vs passive review—why pulling information out of your head beats re-reading it back in. And once retrieval is a habit, the next lever is when you retrieve: see how the major scheduling algorithms compare in SM-2 vs FSRS vs Leitner vs Anki.
