"How do you guys actually study, beyond just rereading notes?" comes up constantly in study forums, and it points at something real: rereading is the most common study method and also one of the least effective ones on record. Decades of cognitive psychology research point to a different answer, and it has been hiding in plain sight since at least 2006.
Quick Answer: testing yourself beats rereading, by a wide and consistent margin
Retrieving information from memory (active recall, also called practice testing) produces better long-term retention than rereading notes, and the gap holds up across ages, subjects, and time delays. A landmark 2013 review in Psychological Science in the Public Interest rated practice testing as "high utility" and rereading as "low utility," the same conclusion a 2006 Psychological Science study reached directly by comparing the two head to head.
Why rereading feels productive even though it barely works
The core problem is what psychologists call the fluency illusion. When you reread a paragraph you already studied, the words feel familiar, and that familiarity gets misread as understanding. Your brain is recognizing the text, not retrieving the underlying knowledge, and recognition is a much weaker memory signal than recall. You can reread a chapter on for-loops five times and feel confident, then freeze the moment someone asks you to write one from scratch, because rereading never tested whether you could produce the information without the page in front of you.
Henry Roediger and Jeffrey Karpicke's 2006 study in Psychological Science demonstrated this directly. Students studied prose passages, then either restudied the material repeatedly or took repeated recall tests on it with no feedback. On a test given five minutes later, the repeated-study group actually did slightly better, which is exactly why rereading feels like it works in the moment. But on tests given two days or a week later, the group that had been tested on the material outperformed the rereading group by a wide margin, even though the tested group had reported feeling less confident about what they knew. Testing was the harder, less comfortable choice in the moment, and the better one a week later.
That finding held up at scale. Dunlosky, Rawson, Marsh, Nathan, and Willingham's 2013 review in Psychological Science in the Public Interest evaluated ten common study techniques against the available evidence and sorted them into utility tiers. Practice testing and distributed (spaced) practice were the only two techniques rated high utility, meaning they reliably improved learning across different ages, subjects, and types of test. Rereading and highlighting, the two most commonly reported study habits among students, were rated low utility: not useless, but not something to build a study plan around.
The fluency illusion is worse, not better, for video-based learning. Rereading a page at least forces your eyes to move at reading speed and lets you notice if you skipped something. Rewatching a YouTube tutorial can happen at 1.5x or 2x speed while you half-listen, which produces an even stronger sense of familiarity for an even weaker memory trace. This is why so many self-taught developers report the same failure mode: they can follow along with a tutorial video and nod at every step, then open a blank editor for the same task and freeze, because following along was never a memory test in the first place.
The ten study techniques, ranked by what actually works
The 2013 review's tiers are the most cited ranking in this area of research. Utility here means how consistently a technique improved test performance across the studies reviewed, not how much effort it takes.
| Technique | Utility rating (Dunlosky et al., 2013) | Effort to start | Best for |
|---|---|---|---|
| Practice testing (active recall) | High | Low, needs only questions or flashcards from material you already have | Long-term retention, exam prep, technical interviews |
| Distributed (spaced) practice | High | Moderate, needs a schedule or spacing app | Retaining material across weeks or months instead of forgetting it after a test |
| Self-explanation | Moderate | Low | Understanding why something works, not just that it works |
| Elaborative interrogation | Moderate | Low | Connecting a new fact to what you already know |
| Interleaved practice | Moderate | Moderate, requires mixing topics instead of blocking them | Telling similar problem types apart (a common weak spot in math and programming) |
| Summarization | Low | Low | A first pass to identify what matters, not retention on its own |
| Rereading | Low | None | Initial familiarity only; replace with practice testing as soon as possible |
| Highlighting or underlining | Low | None | Marking a text for a later pass; does not build recall by itself |
| LearnPath (quiz-gated YouTube path) | Built on the high-utility techniques above | Free tier; Pro $12.99/mo | People who want practice testing and spacing automated instead of hand-built |
LearnPath is not magic here, it is a delivery mechanism for the two techniques the research actually rates highest: every video in a path is followed by a quiz on that video's transcript before the next one unlocks, which is practice testing applied to video content instead of textbooks, and the path resequences based on quiz performance, which functions like spacing by pushing weak topics back around instead of letting them drop. The honest comparison is that you can build the same discipline yourself with flashcards and a calendar reminder; LearnPath's role is removing the setup cost, not inventing a new technique.
For a deeper look at the spacing half of this table, see our piece on spaced repetition, and if your rereading habit is really a rewatching habit, how to take notes on YouTube videos covers the timestamp method that turns passive rewatching into material you can actually self-test on later.
How to actually switch from rereading to active recall
-
Turn your existing notes into questions before you review them again. Take whatever you already have (notes, a textbook chapter, a video transcript) and rewrite each key point as a question. "For-loops iterate over a sequence" becomes "What does a for-loop iterate over, and write one that prints each item." This conversion is the whole trick; once the material is in question form, rereading is no longer possible, only testing.
-
Close the source before you answer. The entire benefit of active recall depends on retrieving the answer from memory, not from the page. If you can see the notes while answering, you are back to a recognition task, which is what makes rereading feel easy and work poorly.
-
Check your answer immediately, and be specific about what you missed. Do not just mark right or wrong. Note whether you got the concept but missed a detail, or missed the concept entirely, because those two failures need different fixes: a detail gap needs one more retrieval attempt, a concept gap needs to go back to the source material first.
-
Space your retries instead of repeating them back to back. Testing yourself repeatedly in one sitting on the same material has diminishing returns quickly, because you start recalling the answer you just gave, not retrieving it fresh. Retest the same material the next day, then after three days, then after a week, stretching the interval each time you get it right.
-
Interleave topics instead of finishing one before starting the next. If you are studying three related concepts (three sorting algorithms, three grammar rules, three chart types), mix your practice questions across all three instead of drilling one to mastery before moving on. This costs more in the moment because it feels harder and slower, which is exactly the signal that it is working: the difficulty comes from your brain relearning to tell the concepts apart, which is the actual skill you need on a real test or in real code.
-
Track failure rate, not hours studied. Hours spent rereading is not a useful metric because the activity has a low ceiling regardless of time invested. Track how many of your self-test questions you get right on the first try over successive rounds; a rising first-try success rate is the signal that retrieval, not just exposure, is happening.
A worked example makes this concrete. Say you just watched a 20-minute video on Python dictionaries. The rereading version of studying is rewatching the video, or scrolling back through your notes on it. The active-recall version is closing the video and writing down, from memory: what a dictionary is, how to add and remove a key, and one situation where you would use one instead of a list. Then you check your answers against the video and specifically note which of the three you got wrong. That five-minute exercise, repeated the next day and again three days later, produces more retained knowledge than watching the same video twice.
Frequently Asked Questions
Is rereading notes actually a waste of time?
Not a total waste, but low-value. A 2013 review in Psychological Science in the Public Interest rated rereading as low utility because it does not consistently boost performance on tests, even though most students rely on it heavily. Use it only for a first pass, then switch to self-testing.
What is active recall and how do I actually do it?
Active recall means retrieving information from memory without looking at your notes, then checking what you got wrong. In practice: close the notes, write or say everything you remember about a topic, then compare against the source and fix the gaps. Flashcards and practice questions are both forms of it.
How is active recall different from spaced repetition?
Active recall is the technique (testing yourself instead of rereading). Spaced repetition is the schedule (reviewing at increasing intervals instead of all at once). They are usually used together: you space out active-recall sessions over days and weeks instead of cramming them into one sitting.
Does highlighting or underlining help at all?
Barely. The same 2013 review rated highlighting as low utility across age groups and subjects, because it does not require processing why something matters, only that it looks important. Passive marking rarely transfers into better recall on a later test.
Is the Feynman technique as effective as practice testing?
The Feynman technique is a form of self-explanation, which the research rates as moderate utility, one tier below practice testing's high rating. It is strong for catching gaps in understanding a concept, but it does not substitute for repeatedly retrieving facts and procedures from memory over time.
How long before switching to active recall actually improves retention?
Effects on a single test show up within days: Roediger and Karpicke's 2006 study measured the retention gap at just one week. The real payoff compounds further out as spaced retrieval accumulates over multiple review cycles, so stick with it past the first week before judging whether it is working.
How does LearnPath build active recall into a YouTube-based learning path?
LearnPath quizzes you on each video's transcript before unlocking the next one, which forces retrieval instead of passive watching. Performance on those quizzes also branches the path toward harder or easier material, so the retrieval practice adapts to what you have and have not actually retained.