Spaced repetition for board exams: the science behind iatroX adaptive learning

Key takeaways

The forgetting curve is real and exponential. Without active reinforcement, you lose approximately 50% of newly learned information within 24 hours and up to 80% within a week. This has been demonstrated by Ebbinghaus (1885) and replicated in modern research.
Spaced repetition counteracts the forgetting curve by scheduling reviews at increasing intervals — just before you would forget. Each review strengthens the memory trace and extends the retention period. Published evidence shows students using spaced repetition software score 6.2–10.7% higher on standardised medical exams compared to traditional study methods.
The testing effect (active recall) amplifies spaced repetition. Testing yourself — answering questions, not re-reading notes — is a far more effective learning method than passive review. The combination of spaced repetition + active recall is the most powerful evidence-based study strategy available.
Anki is the most widely used spaced repetition tool in medical education (86%+ adoption among US medical students), but it requires manual flashcard creation and offers no curriculum mapping, no adaptive difficulty, and no clinical context.
iatroX applies spaced repetition and adaptive learning principles to curriculum-mapped, exam-specific question banks — for UKMLA, MRCGP AKT, MRCP, USMLE Step 2 CK/Step 3, MCCQE, AMC, ABIM, ABFM, and more. The system identifies your weak areas, adjusts question difficulty, and schedules reviews at optimal intervals — all within a clinical context, not a flashcard.

The science: why you forget (and how to stop)

The Ebbinghaus forgetting curve

In 1885, German psychologist Hermann Ebbinghaus published Memory: A Contribution to Experimental Psychology, establishing the foundational principles of human memory that still govern how we learn today. His key finding: memory decays exponentially after initial learning, following a predictable curve.

The practical numbers are sobering. Without any review, you retain approximately:

100% immediately after learning
~50% after 24 hours
~30% after 48 hours
~20% after one week
~10% after one month

This means that a medical student who spends an entire day studying cardiology and does not revisit the material will have lost roughly half of what they learned by the next morning. By the time they sit the exam four weeks later, they may retain only 10–20% of the original material.

The forgetting curve was replicated by Murre and Dros in 2015, confirming that Ebbinghaus's findings hold in modern experimental conditions. The exponential decay of memory is not a metaphor — it is a measurable, predictable phenomenon.

Spaced repetition: bending the curve

Ebbinghaus also discovered the solution. By reviewing material at specific intervals along the forgetting curve, you can "reset" the decay — and each review extends the retention period further.

The optimal review schedule, based on Ebbinghaus's work and subsequent research:

First review: Within 1 hour of initial learning
Second review: Within 24 hours
Third review: After approximately 1 week
Fourth review: After approximately 1 month
Subsequent reviews: At increasing intervals (months, then years)

With each well-timed review, the forgetting curve flattens. By the fourth or fifth review, retention approaches 90–95% with minimal effort per session. The total time investment is modest — perhaps 30 minutes spread across a month — but the retention gain is enormous compared to cramming.

The testing effect: why questions beat re-reading

Spaced repetition is powerful on its own. But it becomes dramatically more effective when combined with active recall — the practice of testing yourself rather than passively re-reading material.

The testing effect, extensively documented by Roediger and Butler (2011) and replicated across hundreds of studies, shows that the act of retrieving information from memory strengthens the memory itself. When you answer a question (even if you get it wrong), you engage neural pathways that passive reading does not activate. The effort of retrieval — the "desirable difficulty" — is precisely what makes the learning durable.

This is why question banks are more effective than textbooks for exam preparation, and why adaptive question banks with spaced repetition are the optimal combination — they test you (active recall), at the right time (spaced repetition), on the right material (adaptive difficulty).

The evidence in medical education

The research in medical education specifically is now substantial:

Gilbert et al. found that medical students using spaced-repetition software like Anki scored 6.2–10.7% higher on standardised exams compared to those using traditional study methods.
Durrani et al. reported a statistically significant improvement in test performance among medical students using flashcards, with a mean increase of 2.93 points (p < 0.01).
Kerfoot et al. (2007) demonstrated in a randomised controlled trial that spaced education improved the retention of clinical knowledge by medical students compared to massed learning — and the effect persisted at long-term follow-up.
A 2025 Frontiers in Medicine study found that students in an intervention group using spaced-repetition flashcards achieved significantly higher post-test scores (16.24 ± 2.37) than a control group (11.89 ± 2.94), with p < 0.0001.

The evidence is not subtle. Spaced repetition works, and it works particularly well for the kind of knowledge medical exams test: large volumes of factual, conceptual, and applied clinical information.

The tools: Anki vs AMBOSS vs Passmedicine vs iatroX

Anki: the manual powerhouse

Anki is the most widely used spaced repetition tool among medical students globally, with an estimated 86%+ adoption rate among US medical students. It is free on desktop and Android, and its spaced repetition algorithm (SM-2) is well-validated.

Strengths: Completely free (desktop/Android). Massive library of pre-made decks (AnKing, Zanki, etc.). Total user control over card content and scheduling. Huge community.

Limitations: Requires manual flashcard creation or downloading pre-made decks that may not align with your specific exam blueprint. No curriculum mapping. No adaptive difficulty — the algorithm adjusts timing but not the content based on your performance patterns. No clinical context — flashcards are decontextualised facts, not clinical scenarios. No exam-specific question formats (SBAs, EMQs).

Best for: Students who want total control and are willing to invest time in card creation and deck management.

AMBOSS

AMBOSS offers a comprehensive QBank integrated with a medical library, plus a "Study Plan" feature that incorporates spaced repetition principles. The platform is popular globally, particularly for USMLE and European medical exams.

Strengths: High-quality, clinically contextualised questions. Integrated medical library for deep reading. "Study Plan" with spaced elements. Strong for USMLE.

Limitations: Paid subscription ($29–$39/month during clinical rotations). Less UK-specific than dedicated UK platforms. The spaced repetition is a feature within the study plan, not the core architecture of the question engine.

Best for: US medical students and those preparing for USMLE who want an integrated QBank + library.

Passmedicine

The dominant UK QBank for postgraduate exams (MRCGP AKT, MRCP). Includes a "Knowledge Tutor" feature with spaced repetition elements.

Strengths: Massive UK-specific question bank. Well-established among UK trainees. Knowledge Tutor supports retention.

Limitations: Paid subscription. The spaced repetition is a secondary feature, not the core adaptive engine. Less integration of clinical reasoning or AI-driven personalisation.

Best for: UK trainees preparing for MRCGP AKT or MRCP who want a proven, high-volume QBank.

iatroX: adaptive, curriculum-mapped, AI-powered spaced repetition

iatroX Quiz is built from the ground up around the principles of spaced repetition and adaptive learning, applied to curriculum-mapped, exam-specific question banks across multiple countries and exams.

How the iatroX adaptive engine works:

Curriculum mapping: Every question is tagged by topic, difficulty level, and exam blueprint — aligned with the specific requirements of UKMLA, MRCGP AKT, MRCP, USMLE Step 2 CK/Step 3, MCCQE, AMC, ABIM, ABFM, MRCEM, PSA, and PANE.
Performance tracking: The system monitors your accuracy, speed, and confidence across every topic and difficulty level.
Adaptive difficulty: Questions are selected based on your performance profile — targeting your weakest areas with appropriately challenging questions. If you consistently answer cardiology questions correctly but struggle with endocrinology, the system prioritises endocrinology.
Spaced repetition scheduling: Questions you answered incorrectly (or slowly, or with low confidence) are re-surfaced at optimal intervals — just before the predicted forgetting point. Questions you answered correctly and confidently are scheduled at longer intervals.
Clinical context: Unlike flashcards, iatroX questions are clinical vignettes — the same format you will encounter on the actual exam. You are practising recall in the context of clinical decision-making, not decontextualised fact retrieval.

What makes this different from Anki: You do not create cards. You do not manage decks. You do not download third-party content of uncertain quality. You sit down, choose your exam, and the system presents you with the right question, at the right difficulty, at the right time. The cognitive science is embedded in the engineering — you just answer questions and the system handles the rest.

What makes this different from Passmedicine/AMBOSS: The spaced repetition is not a bolt-on feature — it is the core architecture. Every interaction feeds the adaptive engine, which continuously refines your personalised review schedule. The system gets smarter the more you use it.

Cost: Free. All of this — adaptive difficulty, spaced repetition, curriculum mapping, clinical vignettes — is free for every clinician and student.

A practical study plan using spaced repetition

Here is a 12-week board exam study plan that leverages spaced repetition principles with iatroX:

Weeks 1–4 (Foundation phase):

Complete a first pass through all topics in your exam blueprint using iatroX Quiz (mixed-topic mode).
Answer 30–50 questions per day.
The system identifies your baseline performance across all topics.
Focus on understanding, not memorisation. Use Ask iatroX to explore the explanations behind questions you get wrong.

Weeks 5–8 (Targeted phase):

Switch to adaptive mode. The system now prioritises your weakest areas.
Answer 40–60 questions per day.
Spaced repetition is now actively scheduling reviews of material from weeks 1–4.
Use Brainstorm for topics where you struggle with clinical reasoning, not just factual recall.

Weeks 9–12 (Consolidation phase):

Increase to 60–80 questions per day.
The system is now surfacing previously learned material at optimal review intervals — reinforcing retention of material from weeks 1–4 while consolidating newer learning.
Simulate exam conditions: timed blocks, mixed topics, no aids.
Use performance analytics to identify any remaining weak spots for targeted last-minute review.

This plan leverages three evidence-based principles simultaneously: spaced repetition (optimal timing), active recall (question-based practice), and interleaving (mixed-topic presentation). The result is retention that persists through exam day and into clinical practice — not just until the exam is over.

Beyond exams: spaced repetition for practising clinicians

Spaced repetition is not just for students. Practising clinicians face the same forgetting curve — you learned the management of rare conditions during training, but if you have not seen a case in two years, that knowledge has decayed.

iatroX Quiz allows practising clinicians to:

Maintain clinical knowledge through regular, low-effort quiz sessions (5–10 questions per day, 5 minutes)
Prepare for revalidation by evidencing CPD through question-based learning
Stay current as quiz content is updated to reflect new NICE guidelines and evidence

The daily habit of answering 5 clinically contextualised questions, with spaced repetition ensuring you revisit material you are at risk of forgetting, is one of the most efficient CPD activities available — and it is free.

The comparison table

Feature	Anki	AMBOSS	Passmedicine	iatroX Quiz
Spaced repetition	Yes (SM-2 algorithm)	Partial (within Study Plan)	Partial (Knowledge Tutor)	Yes (core architecture)
Adaptive difficulty	No	Partial	No	Yes (AI-driven)
Curriculum-mapped	No (user-created)	Yes (USMLE, international)	Yes (UK postgrad)	Yes (UKMLA, AKT, MRCP, USMLE, MCCQE, AMC, ABIM, ABFM, PSA, PANE)
Question format	Flashcards	Clinical SBAs	SBAs	Clinical SBAs / EMQs
Clinical context	Minimal	Strong	Moderate	Strong (guideline-grounded)
UK-specific	Depends on deck	Partial	Strong (UK postgrad)	Strong (UK + US + CA + AU)
Cost	Free (desktop/Android)	$29–$39/month	Subscription	Free
Mobile apps	iOS ($25 one-time) / Android (free)	iOS / Android	Web-heavy	iOS / Android / Web