You re-read your notes three times, highlighted the key parts, and felt ready. Then the exam came — and so did the blackout. Millions of students live this every year. It’s not a memory problem. It’s a method problem.
For two decades, cognitive psychology research has stacked up solid evidence: re-reading is one of the least effective study strategies out there. Yet it remains by far the most popular. Why? Because it creates a feeling of familiarity that looks like knowledge — but isn’t.
This article draws on the work of Henry Roediger, Jeffrey Karpicke, John Dunlosky, and Robert Bjork.
Why re-reading “works” (in your head)
When you re-read your notes, your brain recognizes the information. The sentences feel familiar. The diagrams make sense. You nod along thinking: “yeah, I know this.”
Researchers call this the illusion of competence, or more precisely, a fluency bias. The mechanism is simple: the ease with which you process information tricks you into believing you’ve mastered it.
Robert and Elizabeth Bjork at UCLA described this in their work on “desirable difficulties.” Their finding: recognizing is not the same as remembering. Recognizing means saying “yes, I’ve seen this before” with the text open in front of you. Remembering means retrieving the information without support, under pressure, in a different context — exactly what an exam demands.
What the studies say: re-reading rated “low utility”
In 2013, psychologist John Dunlosky (Kent State University) published a landmark meta-analysis: Improving Students’ Learning With Effective Learning Techniques. His team evaluated ten popular learning strategies across four criteria: strength of evidence, diversity of conditions, populations, and materials.
The verdict on re-reading: low utility.
The researchers don’t claim re-reading is totally useless — a second read does offer a marginal benefit over reading once. But that benefit is significantly lower than other strategies. And it drops fast. A third re-reading adds almost nothing over a second.
Highlighting, often paired with re-reading, received the same rating. Dunlosky notes that highlighting can actually hurt learning by creating the impression you’ve “processed” the information, reducing motivation to study it actively afterward.
Roediger and Karpicke’s experiment: reading vs. testing
One of the most striking demonstrations comes from a 2006 study by Henry Roediger III and Jeffrey Karpicke (Washington University in St. Louis).
The protocol was simple. Two groups of students read a text. The first group re-read it multiple times. The second group took a recall test — they were asked to reproduce what they’d read, from memory, without the text.
Result at 5 minutes: the re-reading group remembered slightly more. Makes sense — the information was still fresh.
Result at 2 days: complete reversal. The tested group remembered significantly more. And the gap only widened over time.
This phenomenon has a name: the testing effect. Testing yourself — even without immediate feedback — strengthens memory traces far more durably than simple re-exposure.
Roediger explains it this way: active recall forces the brain to reconstruct the path to the information, strengthening the associated neural connections. Re-reading just follows a path already laid down — smooth but fragile.
The real problem: students don’t know what they don’t know
A particularly cruel aspect of the illusion of competence is that it’s self-reinforcing. The more you re-read, the more familiar the text feels, the more convinced you are that you’ve mastered it. You become, in a sense, blind to your own gaps.
Kornell and Bjork (2007) showed that students who re-read systematically overestimate their mastery. Those who test themselves regularly have a much more accurate assessment of what they know and don’t know.
This is a double benefit of testing: it strengthens memory and calibrates confidence. After a quiz, you know what you don’t know — and that’s valuable information. It lets you target your revision instead of reviewing everything in a loop.
What works: the two strategies rated “high utility”
In the same 2013 meta-analysis, Dunlosky and colleagues identified two strategies rated “high utility” — the only ones to earn that level:
1. Practice testing
Asking yourself questions, taking quizzes, or trying to retrieve information from memory. The format doesn’t matter much: multiple choice, open questions, flashcards, free recall. What matters is that your brain does the work of finding the information rather than re-reading it.
Practice testing is even more effective when spaced over time and combined with feedback. But even testing without correction produces a measurable benefit.
2. Distributed practice
Studying in multiple spaced sessions rather than cramming the night before. Known as the spacing effect, Ebbinghaus observed it back in 1885; modern research has confirmed it robustly.
Concretely: 3 sessions of 20 minutes spread over a week produce better results than one 60-minute session the night before. The brain needs time between exposures to consolidate memory.
The practical problem: who has time to organize all this?
If the evidence is so strong, why do students keep re-reading? One word: friction.
Testing yourself takes more cognitive effort. It’s uncomfortable. When you re-read, everything flows. When you test yourself, you stumble, hesitate, fail — and that’s precisely what makes the method effective. But in the moment, it feels bad.
There’s also a logistics problem. Effective self-testing requires:
- creating relevant questions (time-consuming)
- organizing them (needs a system)
- spacing them over time (needs a schedule)
- identifying weaknesses to target efforts (needs diagnosis)
That’s a lot of work before you even start studying. Which is why digital tools are starting to fill this gap.
Apps like Anki have applied spaced repetition for years, but setup remains heavy: you create every card yourself. More recently, tools like Wizidoo take a different approach: you import your course (PDF, photo, notes), and the app automatically generates adaptive quizzes targeting your weaknesses — with a mastery percentage per chapter so you know where you actually stand. The first course is free on iOS.
The point isn’t that technology replaces effort. It’s that it removes the friction preventing students from using the methods that work.
Re-reading is lying to yourself (without meaning to)
Re-reading isn’t “bad” per se. But it’s dramatically insufficient for anyone who wants to retain knowledge long-term. It creates an illusion of competence, prevents you from identifying gaps, and burns time better spent on active recall.
The work of Roediger, Karpicke, Dunlosky, and Bjork all converges on the same message: test yourself. Ask yourself questions. Try to recall from memory. Identify what you don’t know. And come back to it later — not in a panic the morning of the exam.
It’s not about talent or intelligence. It’s about method. And on that front, the science is settled.
FAQ
Is re-reading completely useless? No, a first re-reading does help compared to reading once. But the benefit drops fast after the second time through. That time would be better invested in active recall (quizzes, retrieval practice).
Why do I feel confident after re-reading? That’s the illusion of competence: familiarity with the text tricks you into thinking you’ve mastered it. Kornell and Bjork’s research shows this confidence is often disconnected from actual memorization.
What’s the most effective study method according to science? Dunlosky’s 2013 meta-analysis rates practice testing and distributed practice as the two most effective strategies. They outperform re-reading, highlighting, summarizing, and most popular techniques.
How do I test myself when I don’t have quizzes? Close your notes and try to recall what you know (free recall), use flashcards, or try apps that generate quizzes from your own courses.