Review

<<>>

Review

TL;DR

Review isn't rereading your notes — it's deliberately testing what you've already learned, spaced out over time, so it sticks for the exam. You'll learn why spaced repetition and active recall beat cramming, how to build a review schedule using the forgetting curve, and how to run a review session that actually works.

1. The Mental Model

Your brain forgets predictably — fast at first, then slower. Reviewing isn't about adding new information; it's about interrupting forgetting at the right moments. Each time you successfully pull a memory back out of storage, right before you'd have lost it, that memory gets stronger and more durable. Review is memory maintenance, not memory intake.

2. The Core Material

The Forgetting Curve and Why Timing Matters

In the 1880s, Hermann Ebbinghaus memorized lists of nonsense syllables and tested himself over time. He found that forgetting isn't linear — it's exponential and front-loaded. Roughly:

$$R(t) = e^{-t/s}$$

where $R$ is the fraction you remember, $t$ is time since learning, and $s$ is a "strength" parameter that depends on how well you learned the material. The key insight isn't the exact formula — it's the shape. Without any review, you lose about 50-70% of new material within 24 hours. But each time you review successfully, $s$ grows, meaning the curve flattens and the next review can wait longer.

This gives you a practical rule: review intervals should expand. A common, well-tested schedule looks like:

• First review: 1 day after learning
• Second review: 3 days after that
• Third review: 1 week after that
• Fourth review: 2-3 weeks after that
• Fifth review: 1-2 months after that

This is spaced repetition. The spacing matters more than the total number of reviews — five reviews crammed into one evening teach you almost nothing extra; five reviews spread over six weeks build a memory that survives the exam and beyond.

Active Recall: What a Review Session Should Actually Contain

Here's the mistake almost everyone makes: they "review" by rereading the textbook or highlighting notes again. This feels productive because the material feels familiar — but familiarity is not retrieval. You're recognizing, not recalling, and recognition is a much weaker, much more fragile form of memory.

A real review session replaces rereading with retrieval practice: you close the book and try to produce the answer from memory first, then check. The friction of struggling to recall — even failing and then seeing the answer — strengthens the memory far more than passively re-reading it ever could. This is called the "testing effect," and it's one of the most replicated findings in learning science (Roediger & Karpicke, 2006, found students who took a practice test remembered more a week later than students who reread the material four times).

Concretely, a review session should look like this sequence, not like rereading:

flowchart TD
    A["Pick a topic from your review schedule"] --> B["Close notes. Try to recall it:
    write/say everything you remember"]
    B --> C{"Did you get it mostly right?"}
    C -->|Yes| D["Check notes briefly to fill gaps.
    Push next review further out"]
    C -->|No / partial| E["Check notes fully. Re-explain it
    in your own words"]
    E --> F["Re-schedule this topic sooner
    (1-3 days) for another attempt"]
    D --> G["Mix in a flashcard or practice
    question from an older topic (interleaving)"]
    F --> G

Notice the branch: if recall was easy, you push that topic further into the future (it's becoming durable). If recall failed, you bring it back sooner. This adaptive scheduling is exactly what tools like Anki automate, but you can do it by hand with a simple box system (the Leitner system): index cards move to a "review less often" box when you get them right, and back to "review soon" when you get them wrong.

Interleaving: Don't Review One Topic at a Time

The second common mistake is "blocked" review — spending a whole session on Topic A, then a whole session on Topic B. It feels organized, but it's actually a weaker way to review. Interleaving means mixing different topics or problem types within the same session.

Why it works: when every problem in front of you is about, say, integration by substitution, you don't have to figure out which method applies — you already know, because that's all you've been doing for the last hour. On the real exam, the hard part isn't executing the method; it's recognizing which method a problem calls for. Interleaving forces you to practice that recognition step during review, which is the step blocked practice skips entirely.

Rohrer and Taylor (2007) tested this directly with math problems: students who practiced four problem types in a mixed, interleaved order scored far higher on a later test (about 63%) than students who practiced the same problems blocked by type (about 20%) — even though both groups solved the exact same number of problems during practice. The interleaved group felt like they were doing worse during practice — more mistakes, more effort — but performed dramatically better on the actual test. This is the "desirable difficulty" principle: the review method that feels hardest in the moment is usually the one building the strongest memory.

Building Your Review Schedule

A practical way to run this across a semester:

1. Log every topic the day you learn it, with the date.
2. Set review checkpoints using the expanding-interval rule above (1 day, 3 days, 1 week, 3 weeks…).
3. Each session, pull 3-5 due topics, not one. Interleave them.
4. Grade yourself honestly after each recall attempt — "easy," "hard," "failed" — and adjust the next interval accordingly (easy = push further out, failed = bring back sooner).
5. In exam week, shift to higher-frequency review of your weakest-graded topics, using timed practice questions instead of flashcards, so the retrieval format matches the exam format.

That last point matters: review the way you'll be tested. If the exam is essay-based, your retrieval practice should be writing out explanations from memory, not just recognizing multiple-choice answers.

3. Worked Example

Say you're studying for a biology exam in five weeks and you just learned the Krebs cycle on Monday, September 1.

Session 1 (Sept 1, learning day): You read the material once, then immediately close your notes and try to write out the 8 steps of the Krebs cycle from memory. You get 4 of 8 correct. You check your notes, fix the gaps, and re-explain the cycle out loud in your own words — not the textbook's words.

Session 2 (Sept 2, +1 day): Before touching new material, you pull out a blank sheet and try the Krebs cycle again, cold. This time you get 6 of 8. Because it wasn't perfect, you schedule the next review for only 2 days out, not the full 3.

Session 3 (Sept 4, +2 days): You try again — interleaved this time with a second topic (glycolysis, learned Sept 3). You do one recall attempt at Krebs, then one at glycolysis, then back to Krebs for a follow-up practice question. You get the Krebs cycle fully correct. Because it was easy, you push the next review out to 1 week.

Session 4 (Sept 11, +1 week): Krebs cycle comes up again, now mixed with three other topics from that week. Full recall, no errors. Next review: 3 weeks out.

Session 5 (Oct 2, +3 weeks): You're now in exam week. Instead of a flashcard, you use a practice exam question that applies the Krebs cycle to a scenario (e.g., "what happens to ATP output if this enzyme is inhibited?") — because that's the format the real exam uses. You get it right, confirming the memory survives both time and application pressure.

By exam day (around Oct 6), the Krebs cycle has been retrieved five separate times, each time with increasing spacing, each time self-graded, and the last attempt matched the actual exam format. Compare this to a student who reread the textbook chapter five times the night before — they'll recognize the material, but under exam pressure, with no cues and no textbook to check against, recognition won't produce the recall they need.

4. Key Takeaways

4.1 Most Important Concepts

• Forgetting is exponential, not linear — most of what you lose, you lose fast, which is why the first review (within 24 hours) matters most.
• Spaced repetition expands intervals over time — review sooner when a memory is weak, later once it's strong.
• Active recall beats rereading — closing the book and trying to produce the answer is what builds durable memory; rereading mostly builds false confidence.
• The testing effect is real and large — simply testing yourself on material, even without feedback, improves retention more than restudying it.
• Interleaving builds discrimination skill — mixing topics during review trains you to recognize which method or concept applies, which is the actual hard part of most exams.
• Match your review format to your exam format — if the exam is essay-based or applied, review by writing/explaining, not by recognizing flashcard answers.
• Self-grading drives the schedule — "easy" pushes the next review further out; "failed" brings it back sooner. This adaptivity is what makes spaced repetition efficient instead of arbitrary.

4.2 Common Misconceptions

• "Rereading my notes is reviewing." Rereading builds familiarity (recognition), not recall. It feels productive but produces weak, exam-fragile memory. Active recall — testing yourself without the notes — is what actually works.
• "Cramming the night before covers the same material, so it's just as good." Cramming can boost short-term recognition, but spaced review produces memory that lasts weeks to months, which is what you need for cumulative exams and later coursework.
• "If a review session feels easy, I'm doing it right." Ease usually means you're under-challenging yourself — you're recognizing material you half-know rather than retrieving it cold. The sessions that feel a bit effortful (interleaved, delayed, no notes) are usually the ones building stronger memory.
• "I should review each subject in one long, focused block for efficiency." Blocked review feels organized but weakens your ability to tell topics apart under exam conditions. Interleaving mixed topics in the same session, even though it feels messier, produces better test performance.

4.3 Compare & Contrast

5. Now Try It

Pick one topic you learned in the last week. Set a 20-minute timer and do this, right now:

1. Close your notes completely. For 5 minutes, write down everything you remember about the topic — definitions, steps, examples — from memory only.
2. Open your notes. Compare. Circle everything you missed or got wrong (5 minutes).
3. Rewrite the topic in your own words, as if explaining it to a classmate who's never heard of it, focusing specifically on the parts you circled (5 minutes).
4. Write one practice question about this topic in the same format your exam will use (short answer, calculation, essay prompt). Answer it from memory, no notes (5 minutes).

Success looks like: you can state, honestly, which parts of the topic you now recall without looking — and you've written down a date 2-3 days from now to test yourself on it again.