Active Recall vs Re-reading: What Science Actually Says

The landmark study: Karpicke & Roediger 2008

In 2008, Jeffrey Karpicke and Henry Roediger published a study in Science titled "The Critical Importance of Retrieval for Learning" (Karpicke & Roediger 2008). The experiment is one of the cleanest demonstrations in cognitive psychology of how badly re-reading underperforms.

The setup was simple. University students learned a list of 40 Swahili-English word pairs (Swahili was used because the students had no prior exposure to it). The students were split into four groups, each practising the words in a different way:

• Group A: study all 40 pairs, then re-study all 40, repeatedly.
• Group B: study all 40 pairs, then take a recall test on all 40, repeatedly.
• Group C: study all 40, drop the ones recalled correctly, only re-study the missed ones ("study only what you do not know").
• Group D: study all 40, drop the ones recalled correctly, only test on the missed ones.

Crucially, all four groups continued until they had recalled every word once. So they all reached the same "I have learned this" milestone, just by different routes.

One week later, the students returned for a delayed test. The results:

• Groups A and B (re-study everything, or test everything) scored equally well — around 80% retention.
• Group C (re-study only missed items, the strategy most students intuitively use) scored 36% retention.
• Group D (test only missed items) also scored around 80%.

The headline result: students who tested themselves repeatedly retained more than twice as much as students who re-studied. And the strategy most students actually use — "focus on the parts I do not know" — was by far the worst.

Why retrieval practice works

Two main mechanisms explain why retrieval beats re-reading.

First, retrieval is itself a memory modification. When you successfully pull a fact from memory, your brain does not just "check the fact is still there." It updates the memory trace — making it more durable, more accessible, and more connected to whatever else you were thinking about at the moment of retrieval. Re-reading does not trigger this update; you see the words on the page, your brain recognises them as familiar, and the trace is left unchanged.

Second, retrieval reveals what you actually know. When you re-read, your brain tells you "I know this" because the material is familiar. But familiarity and recall are different cognitive operations. Recognising a word in your notes does not mean you can produce it on a blank page. Active recall exposes this gap — sometimes painfully — and the discomfort is exactly the signal you need to direct further study.

Robert Bjork at UCLA describes this as a "desirable difficulty" — a study condition that feels harder, produces worse short-term performance, but builds stronger long-term memory. Retrieval practice is one of the clearest examples. See Roediger & Butler 2011 for a thorough review of the research.

5 study methods compared

Here is how the most common study methods compare, based on multiple meta-analyses of learning research:

Two things to notice. First, the methods every student uses by default — re-reading and highlighting — sit at the bottom of the table. They feel productive but produce the worst outcomes. Second, the gap between the worst methods and the best methods is large: students who switch from re-reading to active recall typically see 40–60 percentage point improvements on delayed tests.

5 ways to do active recall properly

Active recall is not one technique — it is a category of techniques. Five concrete ways to put it into practice:

1. Closed-book self-testing. Read a section of your notes or textbook. Close the book. Write down everything you can remember on a blank page. Then open the book and compare. The gap between what you wrote and what was in the book is your study plan for the next session.

2. The Cornell note-taking system, used properly. Cornell notes are usually taught as a layout (left margin for keywords, right column for content, bottom for summary). What is rarely taught is the actual study method — cover the right column with your hand, look at the keywords on the left, and force yourself to recall the content. The layout is just scaffolding for active recall.

3. The Feynman technique. Pick a concept. Try to explain it in plain language, as if to a 12-year-old, without using jargon. Where you get stuck or fall back to jargon is exactly where your understanding is incomplete. Go back to the source material, fill the gap, and explain again.

4. Past papers under closed-book conditions. The most exam-relevant form of active recall is just attempting past exam questions without notes. The 4-step past papers framework is built around exactly this — treat the first attempt as a retrieval test, not a study session.

5. Flashcards with cloze deletion. Plain flashcards work, but flashcards with sentence-shaped questions that hide a key word (cloze deletion) work better. "The mitochondria are the _____ of the cell" is harder than "What are mitochondria?" because you have to retrieve the specific word, not just describe the concept. Cloze flashcards combined with spaced repetition is the highest-yield daily study habit in the research literature.

Common pitfalls

Active recall is simple, but most students do it wrong in the same few ways.

Confusing recognition for recall. Looking at a multiple-choice answer and thinking "yes, that one" is recognition — a much weaker test of memory than producing the answer from a blank page. If your only "active recall" is multiple-choice flashcards or quiz apps that show you the answer along with three distractors, you are mostly practising recognition. Force yourself to write or say the answer before any options appear.

Looking at the answer too early. The exact moment you check the answer matters. If you check after 30 seconds of struggle, you build the memory trace. If you check after 3 seconds (because the answer felt close), you mostly practise reading the answer. Set a personal rule: at least one full minute of genuine attempt before you check.

Doing it too rarely. Active recall once a week is dramatically better than re-reading three times a week, but the gap between weekly and daily is also large. Spaced retrieval research finds that brief, frequent retrieval sessions beat long, infrequent ones for the same total study time. Five 20-minute sessions across a week is more effective than one 2-hour session.

Forgetting that retrieval and spacing work together. The forgetting curve ensures that material you do not retrieve will fade. Active recall slows the decay only briefly. To make knowledge stick over weeks and months, you need to combine active recall with scheduled review — same material, retrieved again 3 days later, 7 days, 14 days, and so on.

When NOT to use active recall

Active recall is the wrong tool when you have not yet encoded the material. You cannot retrieve information that is not in your memory in the first place. For genuinely new material — a topic you have never been exposed to, in a course you have not started — start with passive encoding: read the textbook chapter, watch the lecture, take notes. Once the material is in, switch to active recall for everything else.

The rough split: spend the first 20% of your time on a topic encoding (reading, watching, taking notes), and the remaining 80% on active recall. Most students invert this ratio and wonder why they cannot remember anything on exam day.

The other case where active recall is the wrong tool: when you are exhausted. Retrieval requires real cognitive effort. After 6 hours of intense study, your retrieval performance drops sharply, and you start "recalling" things incorrectly without noticing. At that point, switch to lighter activities (re-reading is fine here, as a low-effort consolidation activity) or stop and sleep.

Putting it into practice

Active recall is not a hack. It is the default way the brain builds durable memory, and re-reading is the unusual one — a study habit students adopt because it feels productive while actually doing very little. The fix is not complicated: spend less time looking at material and more time trying to reproduce it from memory.

If you want a system built around this principle — question banks designed for closed-book retrieval, with worked solutions for the gaps your retrieval exposes — that is what StudyPilot's library is for. Pick the topic you have been re-reading the longest, switch to closed-book testing, and you will see your scores move within a week.

References

• Karpicke, J. D., & Roediger, H. L. (2008). The critical importance of retrieval for learning. Science, 319(5865), 966–968. https://doi.org/10.1126/science.1152408
• Karpicke, J. D., Butler, A. C., & Roediger, H. L. (2009). Metacognitive strategies in student learning: Do students practise retrieval when they study on their own? Memory, 17(4), 471–479. https://doi.org/10.1080/09658210802647009
• Roediger, H. L., & Butler, A. C. (2011). The critical role of retrieval practice in long-term retention. Trends in Cognitive Sciences, 15(1), 20–27. https://doi.org/10.1016/j.tics.2010.09.003
• Bjork, R. A., & Bjork, E. L. (1992). A new theory of disuse and an old theory of stimulus fluctuation. From Learning Processes to Cognitive Processes, 2, 35–67.