Chapter 8 Quiz: Spaced Repetition

Before checking answers, write your response to each question from memory. Then score yourself.

1. Who is Hermann Ebbinghaus and what two major discoveries about memory did he make using himself as a subject?

2. Roughly what percentage of newly learned material is forgotten within 24 hours, according to Ebbinghaus's research on nonsense syllables?

3. In the New Theory of Disuse (Bjork), what is the difference between storage strength and retrieval strength? Why does this distinction explain why spaced repetition works?

4. Explain why cramming works for a test tomorrow but fails for long-term retention. Use the storage strength / retrieval strength distinction in your explanation.

5. What is the Leitner box system? Describe how it works with five boxes.

6. In Anki, what do the four response options (Again, Hard, Good, Easy) tell the algorithm? What happens to a card's review interval when you choose "Again" vs. "Easy"?

7. What is the minimum information principle for Anki cards? Give an example of a card that violates it, then redesign it to follow the principle.

8. True or false: you should add new concepts to Anki before you understand them, so the reviews help you learn them. Explain your answer.

9. List three types of material that are well-suited for spaced repetition and three that are not. Explain why each type belongs in its category.

10. A student says: "I'll just naturally space out my studying — I don't need an algorithm." What does the research say about humans' ability to judge their own forgetting, and why is this claim problematic?

11. Why is it better to do 20 minutes of Anki every day than a two-hour session once a week, even though the total time is the same?

12. What is FSRS, and how does it improve on the original SM-2 algorithm?

Answer Guide

1. Ebbinghaus was a 19th-century German psychologist who studied his own memory using nonsense syllables. His two major discoveries: (1) the forgetting curve — memory decays rapidly after learning, with roughly half of new material forgotten within an hour and most gone within 24 hours; and (2) the spacing effect — reviewing material at intervals is dramatically more efficient than massed review, producing better long-term retention with the same or less total study time.

2. Approximately 70% forgotten within 24 hours (for nonsense syllables; real meaningful material is retained somewhat better). Roughly 50% is forgotten within the first hour.

3. Storage strength is how deeply a memory is consolidated in long-term memory — how durable it is. Retrieval strength is how easily you can access the memory right now. High retrieval strength (something is fresh/recently reviewed) does not mean high storage strength. When retrieval strength is low (you're starting to forget), effortful retrieval builds storage strength. Spaced repetition schedules review when retrieval strength is low — making retrieval effortful — which exercises and increases storage strength. Reviews when retrieval strength is high (too soon) are less effective because there's little work to do.

4. Cramming keeps retrieval strength high by continuous recent exposure. But because you're reviewing material that's still fresh, the retrieval is easy — it doesn't require much work, so it doesn't build deep storage strength. Storage strength remains low. After the exam, with no more reviews, retrieval strength drops rapidly. Because storage strength is weak, the memory becomes inaccessible quickly — often within two weeks.

5. The Leitner system uses five boxes. New cards start in Box 1. Correct answers advance a card to the next box. Incorrect answers return a card to Box 1. Review schedules: Box 1 every day; Box 2 every other day; Box 3 weekly; Box 4 biweekly; Box 5 monthly. Result: weak cards are reviewed frequently; strong cards are reviewed rarely.

6. The ratings tell Anki how well you retrieved the card, and the algorithm adjusts the next review interval accordingly. "Again" means you failed to recall — the card goes back to a short interval (hours or the next day). "Easy" means you recalled instantly with no effort — the interval is extended significantly (you won't see that card for a long time). "Good" produces a moderate interval extension based on the card's history. "Hard" produces a small extension or no extension.

7. The minimum information principle: one concept per card. Example violation: "Describe the four chambers of the heart and the type of blood (oxygenated vs. deoxygenated) each contains." This tests four different facts. When you get it wrong, you don't know which fact you missed. Redesigned: four separate cards — one per chamber — each asking about that specific chamber's content and function.

8. False. Anki is a retention tool, not a learning tool. You should understand material before adding it to Anki. Adding cards for content you don't understand produces "orphan cards" — you might memorize the words without understanding their meaning, which is both inefficient and fragile. The correct sequence: understand through reading, lectures, or elaboration; then add to Anki for long-term retention.

9. Well-suited: vocabulary, definitions of technical terms, formulas/equations, historical facts/dates, procedural steps. Not well-suited: deep conceptual understanding (requires elaboration and reasoning, not just recall), complex analysis and reasoning, physical/motor skills (require practice, not card review). The difference: SRS is for material where the task is reliable retrieval of specific information. It cannot build understanding, reasoning ability, or physical competence.

10. Research shows that humans are systematically poor at judging their own forgetting. We tend to feel like we still know things we've actually forgotten, because we confuse familiarity (being able to recognize something if we saw it) with retrievability (being able to produce it from memory). People consistently overestimate how much they'll remember without review. An algorithm based on your actual performance history is far more accurate than your intuition.

11. Daily reviews keep each item on its optimal spacing schedule, reviewing at the right interval for each card. A single long weekly session reviews everything too infrequently for some items (which get forgotten in the meantime) and too frequently for others (wasted reviews on things you already know solidly). The distribution of reviews throughout the week is what enables optimal spacing per item — you can't replicate that in a single session.

12. FSRS (Free Spaced Repetition Scheduler) is a newer algorithm that replaces or supplements SM-2 in Anki. It's based on more recent forgetting curve research and is better calibrated for items at the extremes — very easy items (which SM-2 tends to over-review) and very difficult items (which SM-2 can underserve). It uses a different mathematical model that more accurately captures individual forgetting rates. For most users the practical difference is modest, but it's a meaningful improvement for heavy users.