12 Best Study Methods, Ranked by Research Evidence (2026)

How we ranked these

Every method below carries an evidence badge, and the badge is about the research, not about how popular the method is. A technique can be everywhere on YouTube and still be untested.

Where a pooled effect size exists, we quote it. Where one does not, we say so rather than implying a number. Four claims that circulate widely — the exact per-condition scores in Karpicke and Blunt’s concept-mapping study, the “optimal gap is 10% of your retention interval” rule, the 15–30% attention drop attributed to Mackworth, and any benefit from “brain foods” — are not in this article, because we could not verify them against a primary source.

Effect sizes appear as d or g. Roughly: 0.2 is a nudge, 0.5 moves a middle-of-the-class student to about the top third, and 0.8 is a large jump.

The 12 study methods at a glance

If you read nothing else, read this table. It is the whole article in one screen: what each method is good for, how strong the evidence is, and what it costs you in time.

Group 1Retrieval: get it out of your head

The first three methods share one mechanism. You pull the material out of memory instead of putting it back in. That act of pulling is what does the work, and it is why all three feel harder than reading.

Active recall means closing the book and producing the answer from memory. Roediger and Karpicke tested it against straightforward rereading in 2006, and the result is the cleanest demonstration in the whole field.

Students read a prose passage. One group reread it four times. One read it three times and took one recall test. One read it once and took three tests, with no feedback at any point. Five minutes later the rereaders were ahead. A week later they had lost more than half of it.

Rowland pooled 159 effect sizes from 61 studies in 2014 and put the overall benefit of testing over restudying at g = 0.50. Dunlosky and colleagues reviewed ten common learning techniques and rated only two as high utility. This was one of them.

Wing, Marsh and Cabeza watched it happen inside a scanner. Students studied 192 word pairs; half were shown again, half had to be retrieved. A day later they recalled 63% of the retrieved pairs against 51% of the re-shown ones. Retrieval specifically drove activity in the anterior hippocampus, lateral temporal cortex and medial prefrontal cortex, with the hippocampus coupling more tightly to prefrontal regions during successful recall.

And it holds outside the lab. Agarwal, Nunes and Blunt reviewed 50 classroom experiments covering 5,374 students, in real schools with real material.

Those three bars are the instructions. Check your answer every time (0.73 against 0.39). Leave at least a day between studying and testing (0.69 against 0.41). Use a prompt you have to answer rather than options you can recognise (0.61 against 0.29).

Most lists rank past papers separately from active recall, near the bottom, without noticing they are the same thing. They are. A past paper is a retrieval test someone else wrote, which is why Charts C and D apply here unchanged.

What a past paper adds on top of retrieval is the format, the clock and the nerves. You learn how the questions are worded, how long you can afford to spend, and what sitting the thing feels like — which is the part Agarwal’s classroom work links to lower test anxiety.

• Do it timed, first attempt, closed book.
• Mark it against the real scheme, not your memory of the scheme.
• Log every wrong answer by type of mistake, not by question number.
• Redo the wrong ones a week later, not the same day.

SQ3R — survey, question, read, recite, review — appears on every study-skills list as a proven method. It is not. The survey and question steps are hard to standardise and have not been rigorously tested; the Learning Scientists’ review of the literature says the full method has very little empirical support.

The part that has been tested is the middle: read a section, close it, recite what you remember, then review. McDaniel, Howard and Einstein put that against note-taking and against rereading in 2009.

Group 2Scheduling: when you study

These two methods do not change what you do. They change when you do it, and they cost no extra hours.

Cepeda, Pashler, Vul, Wixted and Rohrer reviewed 839 assessments from 317 experiments across 184 articles. Distributed practice beat massed practice consistently, and it is the second of the two techniques Dunlosky rated high utility.

The nuance worth taking from Cepeda is about the gaps themselves. The further away your exam, the longer the gaps between reviews should be. Five reviews crammed into one evening is the worst version of this method. The same five reviews across five weeks is the best one.

This has the strongest data in the article. Rohrer, Dedrick, Hartwig and Cheung ran a randomised controlled trial with 787 seventh-graders across 54 classes, 15 teachers and five schools. Half the practice was blocked — all of one problem type, then all of the next. Half was interleaved, with the types mixed.

One month after the last practice session, the interleaved group scored 60.7% against 37.6%, an effect size of d = 0.83. Every one of the 15 teachers saw a positive effect.

Group 3Understanding: making sense of it

Retrieval and spacing get material to stay. These three are about whether it made sense in the first place.

The branded four-step version — pick a concept, explain it to a child, find the gaps, go back and simplify — has not been tested as a package. The mechanism underneath it has been tested a great deal. Bisra, Liu, Nesbit, Salimi and Winne pooled 69 effect sizes from 64 studies of self-explanation and found g = 0.55. That puts it above active recall on Chart B.

Most write-ups of this method never tell you when to stop. Here is the criterion: you are finished when you can explain it with no notes and no jargon in under two minutes, and a friend who does not take your subject can repeat it back to you.

Nesbit and Adesope pooled 67 effect sizes from 55 studies covering 5,818 participants. Concept maps were associated with better retention, but the effect ranged from small to large depending on how the maps were used and what they were compared against. That spread is the finding.

Karpicke and Blunt put concept mapping head to head with retrieval practice in Science in 2011. Retrieval practice won — on inference questions as well as factual ones, and even when the final test was building a concept map.

You have probably heard that handwriting beats typing. That comes from Mueller and Oppenheimer’s 2014 study, “The Pen Is Mightier Than the Keyboard”, one of the most cited findings in study advice.

Urry and colleagues ran a direct replication in 2021 with 142 students, 74 on laptops and 68 writing longhand. The laptop users again wrote more words and more verbatim text. They also scored no worse on the quiz. Morehead and colleagues reached similar conclusions in a separate replication.

So the format is not the point. What matters is whether your notes are structured so you can quiz yourself from them a week later. Cornell’s cue column exists precisely to make that easy — cover the page, work down the cues, answer each one. That is the reason it works, and it is method 1 wearing a layout.

Group 4Foundations: the conditions

These four are not study techniques. They are the state you are in while you use one.

Okano, Kaczmarzyk, Dave, Gabrieli and Grossman put Fitbits on 88 MIT chemistry students for a semester and compared their sleep against their grades. Duration, quality and consistency together explained about 25% of the variance in performance.

The surprising part is the timing. Sleep on the single night before a test showed no relation to the score. Sleep across the preceding week and month did.

Erickson and colleagues ran a randomised trial with 120 older adults in 2011. A year of aerobic exercise increased anterior hippocampal volume by 2%, reversing one to two years of age-related decline. The control group’s hippocampus shrank over the same period, and spatial memory improved in the exercise group.

Be careful with what that means for you. These were older adults with age-related hippocampal shrinkage to reverse. Reading it as an exam-week tip for a healthy twenty-year-old is an extrapolation, and nobody has run that trial.

There is no randomised trial establishing that 25-minute blocks with five-minute breaks beat any other schedule for studying. The nearest supporting work is Norman Mackworth’s 1948 vigilance research, which found sustained attention degrading after roughly twenty to thirty minutes. That is suggestive of a break interval, not this one.

Zeigarnik’s work on interrupted tasks points the same loose direction: unfinished tasks are easier to resume. Pushing the other way, Sophie Leroy’s work on attention residue suggests people struggle to disengage when interrupted, so a break may not restore as much as assumed.

Keep it anyway, for a different job. Pomodoro does not help you remember things; it helps you start. Treat it as a procrastination tool and it earns its place.

Put your phone in another room rather than face-down on the desk. Work in the same place often enough that sitting down is a cue. Pick background noise you can ignore, or none.

Two things to drop. “Learning styles” — matching material to your supposed visual, auditory or kinaesthetic type — has no supporting evidence, and matching the method to the material works better anyway. And be sceptical of anything marketed as a brain food; we went through those trials one by one, and the evidence for specific foods improving cognition in healthy young adults is thin. Eat regularly, drink water, and spend the attention elsewhere.

Stop doing these four things

Four habits take up most students’ revision time and return the least. Each one has a direct replacement.

They survive for the same reason. All four make the material feel familiar while you are working, and familiarity is what students use to judge whether revision is going well. Rereading a page you have read three times is smooth, so it registers as learning. Producing the same content from a blank page is halting and uncomfortable, so it registers as failure — right up until the exam, where the blank page is the thing you are actually being asked to do.

The fix is not to work harder at these four. It is to spend the same time on the methods in the table below, which will feel worse for about a fortnight.

Your week, assembled

Twelve methods is a list. One schedule is a plan. Thirty minutes a day, six days, covering retrieval, spacing, interleaving and explanation without any of them needing extra hours.

Two things make it work. Monday’s material comes back on Thursday and again on Sunday, which is the start of the day 1, 3, 7, 16, 35 schedule rather than a separate task. And Tuesday deliberately mixes topics you have already covered, so the week gets harder as it goes rather than easier.

Questions students actually ask

Sources

Every link below goes to the paper, its publisher record, or the author’s own copy — never to a blog summarising it.

1. 01Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning. Psychological Science, 17(3), 249–255. PubMed
2. 02Rowland, C. A. (2014). The effect of testing versus restudy on retention. Psychological Bulletin, 140(6), 1432–1463. PubMed
3. 03Dunlosky, J., et al. (2013). Improving students’ learning with effective learning techniques. Psychological Science in the Public Interest, 14(1), 4–58. APS
4. 04Wing, E. A., Marsh, E. J., & Cabeza, R. (2013). Neural correlates of retrieval-based memory enhancement. Neuropsychologia, 51(12), 2360–2370. PubMed
5. 05Agarwal, P. K., Nunes, L. D., & Blunt, J. R. (2021). Retrieval practice consistently benefits student learning. Educational Psychology Review, 33(4), 1409–1453. Springer
6. 06Karpicke, J. D., Butler, A. C., & Roediger, H. L. (2009). Metacognitive strategies in student learning. Memory, 17(4), 471–479.
7. 07McDaniel, M. A., Howard, D. C., & Einstein, G. O. (2009). The read-recite-review study strategy. Psychological Science, 20(4), 516–522. Full text (PDF)
8. 08Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., & Rohrer, D. (2006). Distributed practice in verbal recall tasks. Psychological Bulletin, 132(3), 354–380. PubMed · Author page
9. 09Rohrer, D., Dedrick, R. F., Hartwig, M. K., & Cheung, C.-N. (2019). A randomized controlled trial of interleaved mathematics practice. Journal of Educational Psychology. Full text (PDF)
10. 10Rohrer, D., Dedrick, R. F., & Stershic, S. (2015). Interleaved practice improves mathematics learning. Journal of Educational Psychology, 107(3), 900–908. ERIC
11. 11Bisra, K., Liu, Q., Nesbit, J. C., Salimi, F., & Winne, P. H. (2018). Inducing self-explanation: a meta-analysis. Educational Psychology Review, 30(3), 703–725. Springer · ERIC
12. 12Nesbit, J. C., & Adesope, O. O. (2006). Learning with concept and knowledge maps: a meta-analysis. Review of Educational Research, 76(3), 413–448. Author page
13. 13Karpicke, J. D., & Blunt, J. R. (2011). Retrieval practice produces more learning than elaborative studying with concept mapping. Science, 331(6018), 772–775. Science
14. 14Urry, H. L., et al. (2021). Don’t ditch the laptop just yet: a direct replication. Psychological Science, 32(3), 326–339. Sage
15. 15Morehead, K., Dunlosky, J., & Rawson, K. A. (2019). How much mightier is the pen than the keyboard for note-taking? Educational Psychology Review, 31, 753–780. Springer
16. 16Okano, K., Kaczmarzyk, J. R., Dave, N., Gabrieli, J. D. E., & Grossman, J. C. (2019). Sleep quality, duration, and consistency are associated with better academic performance. npj Science of Learning, 4, 16. Nature
17. 17Erickson, K. I., et al. (2011). Exercise training increases size of hippocampus and improves memory. PNAS, 108(7), 3017–3022. PNAS

Reviewed every six months. Last checked 13 September 2026.