You’ve been in the library for hours with your History or Biology syllabus printed out. Your desk looks like a battlefield: four different colors of highlighters, sticky notes, a half-finished coffee, and the absolute certainty that if you keep this up, you’re getting a perfect score. You highlight. You reread what you highlighted. You write a beautiful summary, in handwriting that’s actually a pleasure to look at. You close the book, feeling like you just finished a mental marathon.
And then you go into the exam and your mind goes blank.
It’s not that you’re not smart, or that the material is impossible. Your brain has been lying to you shamelessly, and you fell for it. What you just described isn’t studying. It’s an illusion. And cognitive science has been shouting this for years, but the message doesn’t seem to be reaching classrooms.
Today we’re going to talk about why the way you’ve been taught to study is basically a waste of time, and what actually works. What works is more uncomfortable, harder, and it’s going to make you feel like you don’t know anything. But it’s the only way for knowledge to actually stick.
The illusion of competence
To understand why we get this wrong, we first need to understand what the hell happens in your head when you read a text. When you read something for the second or third time, your brain processes the information very fluently. You recognize the words, you understand the sentences, you connect ideas. That ease of processing translates into a subjective feeling of mastery. “I understand it, therefore I know it.”
That’s the illusion of competence. You’re confusing familiarity with knowledge. It’s like when you hear your favorite song and feel like you could sing it from memory, but when someone takes the music away and says “sing it,” you realize you only remembered the chorus and the rest is an embarrassing mumble.
The problem is that highlighting and rereading feed this illusion. They don’t demand any real cognitive effort. Your brain is a lazy master, and if you give it the option to do the bare minimum, it will. And when the exam comes, the cue is gone. There’s no teacher giving you hints, no text in front of your eyes for you to recognize. You have to generate the information from scratch. And that’s where your study method collapses.
The technique nobody uses: retrieval practice
This is where retrieval practice comes in. The concept is ridiculously simple, but it goes against everything we’ve been taught. It consists of trying to bring information to mind without looking at your notes. It’s quizzing yourself, taking practice tests, trying to explain a topic out loud without looking, or simply closing the book and writing down everything you remember.
When you do this, your brain suffers. And that suffering is exactly what you need.
By trying to retrieve a memory, you’re strengthening the neural connections associated with that information. You’re forcing your brain to rebuild the path from scratch. Every time you get something wrong and then correct the error, or every time it costs you real effort to recall a concept and you finally pull it out, you’re burning that knowledge in.
The mechanism behind this is the testing effect. It’s not that exams are good for evaluating you; it’s that the act of trying to remember (the test itself) modifies your memory and makes it more likely you’ll recall that information in the future. The test isn’t a measure of learning; the test is the learning.
The study that changed everything (and the ones that came after)
If there’s a foundational study in this field, it’s the one by Henry Roediger and Jeffrey Karpicke, published in 2006 [1]. These researchers at Washington University (United States) designed an elegant, devastating experiment. They split students into two groups. Both groups studied a scientific text. One group studied it by reading the text four times in a row. The other group read it once and then did three retrieval tests (trying to recall what they’d read without looking).
Then they measured them. First, at five minutes. Here, the group that reread four times scored slightly higher. They felt more confident. But the real test came a week later. When they took the final exam, the group that had done retrieval practice beat the rereading group by 50%. What’s more, when students were asked which method they thought had helped them more, the overwhelming majority said rereading. Their intuition was completely wrong.
But this didn’t just happen in the United States. A study conducted in Scotland by researchers at the University of Edinburgh [2] took retrieval practice into real schools, with children between 8 and 12 years old studying new geographical facts. The group that practiced retrieval remembered significantly more information four days later, and the effect replicated weeks later in a different, larger sample. What works in a sterile laboratory works in the chaos of a real school.
And if you think this is just a couple of isolated studies, brace yourself for the final blow. In 2013, John Dunlosky and a team of researchers from several universities in the United States and the United Kingdom published a massive review [3]. They didn’t run a single experiment; they analyzed decades of research, hundreds of studies, to evaluate the real usefulness of ten common study techniques.
The result was a massacre for traditional methods. Highlighting, summarizing, rereading, and keyword mnemonics were all rated “low” utility. Retrieval practice and distributed practice (spacing out your study over time) were the only two techniques rated “high” utility. No nuance. No “it depends.” The evidence converges brutally: if you’re not testing yourself, you’re not learning.
The testing effect isn’t just useful for remembering loose facts or dates. A study by Karpicke and Blunt (2011) [4] showed that retrieval practice is superior to building concept maps (a technique usually considered quite advanced) even for learning complex concepts and drawing new inferences. When you try to remember something, you’re not just accessing a static file; you’re reorganizing the information, making it more flexible and applicable to new situations. Your brain isn’t a hard drive, it’s a muscle that adapts to effort.
Why we resist what works
If retrieval practice is so effective, why do we keep highlighting like maniacs? Because effective learning is uncomfortable. Cognitive psychologists call this desirable difficulty [5].
When you study the easy way (rereading, highlighting), your performance during the study session looks excellent. You feel good. But that performance is superficial and fades fast. When you study with desirable difficulty (testing yourself, trying to recall without looking), your performance during the session looks worse. You get things wrong, you get frustrated, it’s hard. But that difficulty is exactly what produces deep, lasting learning.
We’ve gotten used to measuring our learning by how good we feel while studying, instead of measuring it by what we’re able to do the next day without help. It’s a massive calibration error.
Desirable difficulty doesn’t mean you have to suffer pointlessly. A study by Soderstrom and Bjork (2015) [6] clarified that not just any difficulty is good. If the material is so hard that you don’t have the prior knowledge to tackle the problem, or if the difficulty is irrelevant to the learning goal (like reading a text in a horribly blurry font, which has sometimes been proposed as a “desirable difficulty” but actually just increases cognitive load without improving long-term memory), then learning suffers. The difficulty has to be relevant: it has to force you to process the meaning of the material, not just decipher the format.
So far we’ve talked about retrieval practice at the individual level. But what happens when we do it in a group? There’s a hypothesis that collaborative retrieval, where a group of students quiz each other without looking at their notes, could be even more effective than individual retrieval, because of the social pressure and peer explanation involved. However, there’s also a risk that errors spread if there’s no teacher or reference material to correct false recollections. While there are preliminary studies suggesting collaborative retrieval improves motivation and reduces anxiety, the evidence on whether it beats individual practice in pure long-term retention is still an open, active debate in educational psychology [7], with prior studies reporting mixed results depending on context. In my opinion, doing it in a group reinforces motivation and consistency, two very important factors in studying.
And under stress?
Acute stress physically destroys your ability to access your memories, but retrieval practice makes you practically immune to this block. A team of researchers put a group of people through an experiment as fascinating as it is slightly evil. They taught a set of concepts using the two methods we already know: traditional rereading and retrieval practice. The next day, they split the participants. Half took the exam on the concepts in a calm, relaxed environment. The other half were subjected to what’s known in clinical psychology as the Trier Social Stress Test. If you’re not familiar with this masterpiece of methodological torture, you’ll love it: it involves forcing the person to give an impromptu speech in front of a panel of judges in white coats instructed to maintain an expression of complete contempt and boredom, and then making them mentally subtract by thirteens starting from a huge prime number, shouting at them to start over every time they make a mistake. It’s a nightmare scientifically designed to send cortisol levels through the roof.
What they discovered when they tested people on the concepts learned the day before was incredible. The stressed students who had studied by rereading suffered massive amnesia; their brains, flooded with cortisol, cut off access to the hippocampus and couldn’t retrieve the information. They drew a total blank. But the stressed students who had studied by forcing themselves to recall through testing showed almost no drop in performance. They remembered exactly as much as their calm, unstressed peers sitting in the control room. Retrieval practice had built memory pathways so thick, solid, and automatic that not even the extreme panic of a brain in survival mode could shut them down.
How to start using this
I’m not telling you to stop reading; you need a first exposure to the material. But as soon as you’ve read a section, close the book. Ask yourself: “What did I just read?” Try explaining it out loud, as if you were teaching it to a high schooler who has no idea about the topic. If you get stuck, open the book, check what you’re missing, close it again, and repeat — saying it out loud reinforces the content in memory.
Use flashcards, but don’t look at the answer before you respond. Take practice tests. Find past exams and do them without checking the solution until the end. And for the love of God, stop highlighting. If you need to mark something, do it after you’ve tried to recall the concept, and mark only the keywords that serve as a cue for retrieval — not entire paragraphs painted yellow that look like a rainbow of desperation.
Conclusion
Cognitive science leaves no room for doubt on this point. The evidence is solid, replicated across multiple contexts, and converges in the same direction: retrieval practice is the most effective study technique that exists, and it is, by far, the least used by students.
Highlighting, summarizing, and rereading are activities that make you feel like you’re working, but that have minimal real impact on your ability to remember and apply information. If you want to actually learn, you have to accept that effective learning feels like effort, like a fight against your own mind. You have to stop passively consuming information and start demanding it from your brain.
Next time you sit down to study, remember this: if it doesn’t cost you effort, you’re not learning. Close the book, quiz yourself, and stop fooling yourself.
References
[1] Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249-255.
Label: Reliable. Foundational study, widely replicated and supported by multiple later meta-analyses.
[2] Ritchie, S. J., Della Sala, S., & McIntosh, R. D. (2013). Retrieval practice, with or without mind mapping, boosts fact learning in primary school children. PLoS ONE, 8(11), e78976.
Label: Reliable. Field study with primary school children in Scotland, with the effect replicated within the same article in a second, larger sample.
[3] Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students’ learning with effective learning techniques: Promising directions from cognitive and educational psychology. Psychological Science in the Public Interest, 14(1), 4-58.
Label: Reliable. Exhaustive review of the global literature, considered the gold standard in evaluating study techniques.
[4] Karpicke, J. D., & Blunt, J. R. (2011). Retrieval practice produces more learning than elaborative studying with concept mapping. Science, 331(6018), 772-775.
Label: Mixed reliability.
[5] Bjork, R. A., & Bjork, E. L. (2011). Making things hard on yourself, but in a good way: Creating desirable difficulties to enhance learning. In M. A. Gernsbacher et al. (Eds.), Psychology and the real world: Essays illustrating fundamental contributions to society (pp. 56-64). Worth Publishers.
Label: Theoretical framework. Theoretical book that establishes and defines the concept of “desirable difficulty,” the basis for much of the empirical research that followed.
[6] Soderstrom, N. C., & Bjork, R. A. (2015). Learning versus performance: An integrative review. Educational Psychologist, 50(2), 127-141.
Label: Reliable. Integrative review that clarifies and nuances the concept of desirable difficulty, debunking mistaken applications of it.
[7] Imundo, M. N., Clark, C. M., Bjork, E. L., & Paquette-Smith, M. (2025). The effects of collaborative practice testing on memory for course content in introductory psychology. Teaching of Psychology.
Label: Recent. Recent classroom study, without independent replications yet; the authors themselves note that prior studies on collaborative testing have produced mixed results.