Chapter 8 Key Takeaways: Spaced Repetition
The Core Insight
The same amount of study time, distributed across multiple sessions over days and weeks, produces roughly three times the long-term retention of massed study (cramming). The timing of review matters as much as the content of review.
What the Research Shows
- [Evidence: Strong] The spacing effect is one of the most replicated findings in psychology, confirmed across 130+ years of research, all age groups, and almost every category of material.
- [Evidence: Strong] Distributed practice produces 30-50% better long-term retention than massed practice using the same total study time.
- [Evidence: Strong] Medical students who use Anki consistently outperform non-users across subjects and assessments.
- Cramming produces high short-term performance followed by steep forgetting — "a loan you can't repay."
The Key Concepts
Forgetting curve (Ebbinghaus): Memory decays rapidly after learning — roughly 50% lost within an hour, 70% within 24 hours for novel material — then flattens.
Spacing effect: Reviewing material at intervals is dramatically more efficient than massed review for long-term retention.
Storage strength vs. retrieval strength (Bjork): Storage strength = how deeply a memory is consolidated. Retrieval strength = how easily you can access it now. High retrieval strength (fresh material) ≠ high storage strength. Spaced review works because it forces retrieval when retrieval strength is low — and effortful retrieval builds storage strength.
Optimal review timing: Review each piece of information right at the edge of forgetting — when retrieval is effortful but still possible.
Minimum information principle: One concept per card. Multi-fact cards are harder to review accurately and slower to process.
The Tools
Leitner box: Five-compartment physical flashcard system. New cards start in Box 1. Correct answers advance cards to next box; incorrect answers return them to Box 1. Review frequencies decrease from Box 1 (daily) to Box 5 (monthly).
Anki: Free digital spaced repetition software. Uses the SM-2 or newer FSRS algorithm to calculate optimal review timing per card based on your performance history.
Card types: Basic (question/answer) and cloze deletion (fill-in-the-blank). Use both. Design all cards as recall prompts, not recognition prompts.
The Non-Negotiable Habits
- Review consistently — ideally daily. Skipped reviews create debt. The algorithm assumes consistency.
- Start new material same-day. The optimal first review is within hours of first exposure.
- Understand before encoding. Anki retains what you've understood — it doesn't teach understanding.
- Make your own cards. Or at minimum, understand each card before adding it to your deck. Card-making is itself an encoding activity.
What Belongs in SRS (and What Doesn't)
| Good fit | Poor fit |
|---|---|
| Vocabulary, definitions | Complex reasoning |
| Formulas, equations | Physical skills |
| Historical facts/dates | Creative tasks |
| Named structures (anatomy) | Concepts you don't yet understand |
| Procedural steps | Material you won't need long-term |
| Foreign language vocabulary |
Common Mistakes to Avoid
- Adding cards before understanding the material
- Making cards too broad (multiple facts per card)
- Using recognition-format cards instead of recall-format cards
- Skipping reviews and letting backlogs accumulate
- Using someone else's deck without reviewing/editing cards
- Believing you'll "naturally" space reviews without a system — you won't
Domain Applications at a Glance
| Domain | Spaced repetition looks like |
|---|---|
| Language learning | Vocabulary decks (production + recognition direction) |
| Medical education | Anatomy, pharmacology, pathology Anki decks |
| Academic study | Definitions, formulas, key facts for any subject |
| Music | Scheduled revisiting of previously learned pieces |
| Professional development | Key terms and frameworks in new domains |
The One-Sentence Version
Reviewing material at increasing intervals — right at the edge of forgetting — builds dramatically more durable long-term memory than reviewing it all at once, using the same total study time.