Active recall beats rereading and highlighting when the test comes a day or more later. In the experiments that set them side by side, students who closed the book and wrote down what they remembered kept more after two days or a week than students who read the same pages again, even when the rereaders went through the text about four times as often. Highlighting, measured against plain reading, helped memory a little and did not reliably help understanding.
This guide is the head-to-head: what each study compared, what it measured, the figures as the papers report them, and where the evidence is thin. Why self-testing and spacing work in general is in our guide to how to study effectively; how to do it well is in retrieval practice: how to use it.
EdCitation publishes this guide. Five papers were read in the original on 2 October 2026 and four only in their abstracts, which we name. We searched in EdCitation's Find sources, built six of the references from their DOIs in Cite a source, and wrote the week's plan near the end ourselves; no study tested it.
Is active recall better than rereading?
Active recall is better than rereading for anything you need to remember beyond the same day. Rereading wins on a test a few minutes later; recall wins after a delay. Active recall here means retrieval practice, the testing effect in researchers' terms: answering from memory, with the text out of sight, before you look.
Rereading wins at five minutes, recall wins after two days
Roediger and Karpicke (2006) had 120 undergraduates read a short passage and then either read it again or write down all they could recall, with no feedback. The rereaders did better five minutes later. Two days later recall led, with an effect size of d = 0.95, and a week later d = 0.83. The recall group's score after a week matched the rereaders' after two days: in the authors' reading, one test held off five extra days of forgetting.
A second experiment made the gap starker. Students who spent four five-minute periods reading got through the passage about 14 times in all; students who read once and then took three recall tests read it about three times. After a week the readers recalled 40% of the ideas and the testers 61%.
Studying again after you know it added nothing
Karpicke and Roediger (2008) taught students 40 Swahili-English word pairs and changed what happened to a pair once it had been recalled correctly: it stayed in study, in testing, in both, or was dropped. A week later, further study after that first success made no measurable difference; further testing made a large one. Combining conditions, the authors report d = 4.03, and the ranges did not overlap: 10 to 60% recalled by students whose pairs left the tests, 63 to 95% by those who kept being tested.
Which study technique has the best evidence?
Practice testing has the best evidence of any technique a student can use alone. Dunlosky et al. (2013) rated it, with spaced practice, of high utility, and rated rereading and highlighting low. The table sets each technique beside what it was compared with.
| Technique | What the evidence rated it | Source |
|---|---|---|
| Practice testing (active recall) | High utility; in 222 classroom studies, quizzing raised achievement by g = 0.499 against a mix of control conditions | Dunlosky et al., 2013; Yang et al., 2021 |
| Read, recite, review | Beat rereading on free recall at once and a week later; with engineering texts, also on multiple-choice and problem questions | McDaniel et al., 2009 |
| Answering questions after one reading (text open) | Beat rereading after 15 minutes and after two days | Weinstein et al., 2010 |
| Note-taking while reading | No better than rereading on free recall; matched read-recite-review on problem solving, but slower | McDaniel et al., 2009 |
| Rereading straight away | Low utility; little or no gain over one reading of a textbook chapter | Dunlosky et al., 2013; Callender and McDaniel, 2009 |
| Rereading after a gap | Beat rereading straight away on delayed tests | Dunlosky et al., 2013 |
| Highlighting your own text | Low utility; against plain reading, an effect of 0.36 on memory and 0.20 on comprehension | Dunlosky et al., 2013; Ponce et al., 2022 |
| Text highlighted for you | An effect of 0.44 on both memory and comprehension | Ponce et al., 2022 |
| Finding and citing these studies | Each reference looked up in the publisher's record | EdCitation's Find sources and Cite a source |
Read each figure against its own comparison: highlighting's against reading without highlighting, Yang's against a mix of controls. No study we read pitted highlighting against recall, so the numbers cannot be subtracted.
Is rereading effective?
Rereading is effective only in a limited way: the second reading does most of what rereading can do, it works better after a gap than straight away, and in direct comparisons it has consistently lost to self-testing (Dunlosky et al., 2013).
One reading against two
Callender and McDaniel (2009) ran four experiments in which students read textbook chapters or a Scientific American article once or twice, knowing a test was coming, then answered multiple-choice and short-answer questions or wrote summaries. In their abstract, rereading did not significantly raise scores apart from a few exceptions, and stronger and weaker readers showed the same pattern. Their conclusion: with real course material, rereading at once may do little or nothing for the tests students sit. We read the abstract; the full paper is paywalled.
Dunlosky and colleagues add that most of the gain comes with the second reading, in studies where it came straight away, and that rereading after a gap beat rereading at once on delayed tests. No experiment had tested rereading in a real course; the classroom-side evidence was surveys, one of which found a small positive link between self-reported rereading and grades.
Read, recite, review: recall between the readings
McDaniel et al. (2009) changed one thing about rereading: between the two readings, students set the text aside and said aloud everything they remembered. In the first experiment, 72 Washington University undergraduates studied four passages of 221 to 283 words. Free recall straight after was 63% of the ideas with read-recite-review, 51% with note-taking and 46% with plain rereading; a week later, on passages not tested before, 33%, 21% and 16%. Multiple-choice and short-answer inference scores did not differ.
The second experiment used two harder texts, on pumps and on brakes, of 864 and 915 words. Read-recite-review again led on free recall, and now beat rereading on multiple-choice and problem-solving questions as well. Note-taking matched it on problem solving but took longer: 25.4 minutes against 21.5. One caveat: in the first experiment reciting took more time than rereading, 13.5 minutes against 9.2, so the comparison was not minute for minute.
Questions against a second read
Weinstein et al. (2010) gave 29 volunteers, most of them undergraduates, three 575-word passages. After one reading, they reread a passage, answered eight set questions on it with the text still in front of them, or wrote and answered eight questions of their own. Fifteen minutes later they got 57% of test questions right after rereading against 72% after either question task (d = 0.96 and 0.75). With a two-day delay and 24 new volunteers, rereading again came last.
The questions were answered with the passage open, a milder form of recall than a closed-book test, and writing your own questions took at least twice as long as answering set ones.
Does highlighting work?
Highlighting works a little for remembering what you marked, does not reliably help understanding, and nearly all of its evidence compares it with reading alone, not with recall. Dunlosky et al. (2013) rated it low, because most studies found no benefit beyond simply reading. The classic experiments are set out in how to remember what you read.
What a 2022 meta-analysis found
Ponce et al. (2022) pooled 36 published articles from 1938 to 2019, giving 85 effect sizes, each comparing students who read a text with or without highlighting. Highlighting done by the students themselves gave an average effect of 0.36 on memory tests and 0.20 on comprehension tests, which the authors read as improving memory but not comprehension. Text already highlighted by an instructor did better, 0.44 on both. Students' own highlighting helped college students (0.39) but not school pupils (0.24). We read the abstract; the paper is paywalled.
Where the studies disagree
Yue et al. (2015) found an overall benefit of highlighting on a test a week later, larger when the two readings came back to back than when they were spaced, with no cost to the parts left unmarked. Students who doubted highlighting gained more from it than its fans did. We read the abstract only. That sits against Dunlosky's verdict, though the review itself suggests that students who mark too much get little from it.
Our reading: highlighting is not harmful in itself, but it is a weaker use of the minutes than recall, and nobody we read has put the two head to head.
Why does rereading feel like it works?
Rereading feels like it works because students are as confident after it as after recall, or more so, while their later scores are lower. In Roediger and Karpicke's second experiment, students who had only read rated how well they would remember the passage in a week at 4.8 on a seven-point scale, against 4.0 for those who had been tested.
In Karpicke and Roediger's word-pair study, every group predicted it would recall about half the list, 20.3 to 22.0 of 40 pairs, though real scores ran from about a third to about 80%. Weinstein's volunteers predicted 63.0 out of 100 after rereading and 63.4 after answering questions, yet scored 15 points lower after rereading. A feeling of knowing is not a measure of it; a closed-book test is.
Where is the evidence thin?
The evidence is thin wherever real courses come in. What these studies cannot tell you:
- Long, hard material over a term. Most passages ran to a few hundred words or were word lists, and the final tests came within a week.
- Whether it holds beyond one campus. Three of the four experimental papers we read in the original were run with volunteers at Washington University in St Louis.
- Highlighting against recall. No direct comparison in anything we read.
- Four full papers. Callender and McDaniel, Ponce, Yue and Yang were read as abstracts; Yang's does not give the figure against rereading alone.
What does a week of active recall instead of rereading look like?
A week built on these studies swaps the second reading for a recall and keeps rereading for checking. This plan is ours, drawn from the studies above; no study tested it as a whole. The studying is your own work, and so are the answers: no tool can recall for you.
- Monday: read, recite, review. Read a section once, close it, write or say all you remember, then read it again to check what you missed.
- Mark after, not during. If you highlight, do it on the checking read, and mark only what your recall missed.
- Tuesday: turn the gaps into questions and answer them with the notes shut, for about ten minutes.
- Wednesday: new material, same routine, then five minutes recalling Monday's section on a blank page.
- Thursday: answer every question again, including the ones you got right, as Karpicke and Roediger's results suggest.
- Friday or Saturday: one closed-book test on the week, then reread only what you got wrong: rereading after a gap, never straight after the first read.
- For an essay, look the research up as you go. Search EdCitation's Find sources for the claim a paragraph needs, and build the reference in Cite a source as you note the point.
How can EdCitation help you check the studies?
EdCitation is the best tool for checking the research behind a study tip, because it looks every reference up in the publisher's record and never writes one, which no chatbot can promise: all six references we built for this guide came back from Crossref's records. It writes no part of anyone's work: it finds the papers, builds the references and checks them, and the recall, the notes and the essay are the student's own.
Find the studies
Find sources searches about 300 million published works by topic or by the claim a sentence needs to support. On 2 October 2026, "retrieval practice versus rereading highlighting" returned 1,311 works, with results from OpenAlex. The first was Hinze and Rapp's 2014 paper "Retrieval (Sometimes) Enhances Learning: Performance Pressure Reduces the Benefits of Retrieval Practice", in Applied Cognitive Psychology, cited 63 times; the second a 2019 review of retrieval practice in classroom settings in Frontiers in Education, cited 77 times; the third McDermott's 2020 "Practicing Retrieval Facilitates Learning" in the Annual Review of Psychology, cited 126 times.
Cite them
From the DOI 10.1111/j.1467-9280.2009.02325.x, Cite a source returned this from the Crossref record, with no retraction notice:
McDaniel, M. A., Howard, D. C., & Einstein, G. O. (2009). The read-recite-review study strategy: Effective and portable. Psychological Science, 20(4), 516–522. https://doi.org/10.1111/j.1467-9280.2009.02325.x
That entry is right as it stands, title in APA 7's sentence case. One needed a hand: the record for Weinstein and colleagues' paper lists the third author as "Henry L. Roediger" without the "III" on the paper's byline, so the entry came back "Roediger, H. L." and we added ", III" ourselves.
Check the list, and what it costs
Verify references takes a pasted list or a whole paper, checks each entry against the publishers' records, flags retracted papers, and never shows "could not check" as "not found". All three tools are free with no account. A free account keeps a reference list and a reading list; Pro is $8 a month and Max $24, as Pricing sets out.
Quick questions
Is active recall better than rereading?
Active recall is better than rereading on tests a day or more later. In Roediger and Karpicke's experiment, recall beat rereading after two days (d = 0.95) and a week (d = 0.83), though rereading won at five minutes.
Does highlighting help you remember?
Highlighting helps memory a little and comprehension not reliably: Ponce and colleagues' 2022 meta-analysis found effects of 0.36 and 0.20 against reading without highlighting. Text highlighted by an instructor did better.
Is reading a chapter twice a waste of time?
Reading a chapter twice is not a waste, but straight away it added little in Callender and McDaniel's experiments. Recalling between the two readings, or leaving a gap of a few days before the second, does more.
Is note-taking better than rereading?
Note-taking beat rereading on problem-solving questions in McDaniel and colleagues' second experiment, matching read-recite-review there, but it took longer and gave no better free recall than rereading.
Where can I find the studies comparing study techniques?
Search EdCitation's Find sources by the claim, for example "retrieval practice versus rereading", then build each reference from its DOI in Cite a source.
References
- Callender, A. A., & McDaniel, M. A. (2009). The limited benefits of rereading educational texts. Contemporary Educational Psychology, 34(1), 30–41. https://doi.org/10.1016/j.cedpsych.2008.07.001
- 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. https://doi.org/10.1177/1529100612453266
- Karpicke, J. D., & Roediger, H. L., III. (2008). The critical importance of retrieval for learning. Science, 319(5865), 966–968. https://doi.org/10.1126/science.1152408
- McDaniel, M. A., Howard, D. C., & Einstein, G. O. (2009). The read-recite-review study strategy: Effective and portable. Psychological Science, 20(4), 516–522. https://doi.org/10.1111/j.1467-9280.2009.02325.x
- Ponce, H. R., Mayer, R. E., & Méndez, E. E. (2022). Effects of learner-generated highlighting and instructor-provided highlighting on learning from text: A meta-analysis. Educational Psychology Review, 34(2), 989–1024. https://doi.org/10.1007/s10648-021-09654-1
- Roediger, H. L., III, & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249–255. https://doi.org/10.1111/j.1467-9280.2006.01693.x
- Weinstein, Y., McDermott, K. B., & Roediger, H. L., III. (2010). A comparison of study strategies for passages: Rereading, answering questions, and generating questions. Journal of Experimental Psychology: Applied, 16(3), 308–316. https://doi.org/10.1037/a0020992
- Yang, C., Luo, L., Vadillo, M. A., Yu, R., & Shanks, D. R. (2021). Testing (quizzing) boosts classroom learning: A systematic and meta-analytic review. Psychological Bulletin, 147(4), 399–435. https://doi.org/10.1037/bul0000309
- Yue, C. L., Storm, B. C., Kornell, N., & Bjork, E. L. (2015). Highlighting and its relation to distributed study and students' metacognitive beliefs. Educational Psychology Review, 27(1), 69–78. https://doi.org/10.1007/s10648-014-9277-z