It’s the night before an exam. You reread the chapter, run a highlighter over the key lines and feel the material growing familiar. By midnight it all makes sense. That feeling is real. It is also a poor guide to what you’ll remember on Friday.
In a 2006 study by researchers at Washington University in St. Louis, students read short prose passages and then either read them again or tried to write down as much as they could recall, with no feedback. On a test five minutes later, the rereaders did better. On tests two days and one week later, the students who had practiced recalling remembered substantially more, even though rereading had left the other group feeling more confident.
A 2008 follow-up in the journal Science found something more unsettling. Students’ predictions of how well they would do were unrelated to how well they did. The study habits that feel productive and the ones that work are often different, and your own sense of progress is not a reliable referee.
Below: what the evidence supports, where learning styles and growth mindset stand, and a plan to try this week.
Quizzing yourself beats reading again
Psychologists call it the testing effect, or retrieval practice: pulling information out of memory strengthens it more than putting it in again. Henry Roediger and Jeffrey Karpicke, who ran the 2006 study, summed it up in one line: “Testing is a powerful means of improving learning, not just assessing it.”
Their Science paper showed how specific the effect is. Students learned foreign-language vocabulary. Once they had recalled a word correctly, some kept restudying it without being tested on it, and others kept being tested on it without restudying. Extra studying did nothing for later recall. Extra testing produced a large gain.
The effect holds up outside the lab. A 2021 meta-analysis in Psychological Bulletin, led by Chunliang Yang of Beijing Normal University, pooled 222 classroom studies covering 48,478 students. Quizzing raised achievement by a medium amount, an effect size of about 0.5. By a standard statistical conversion, that puts the average quizzed student ahead of roughly two-thirds of comparable students who weren’t quizzed. The size of the benefit varied with what the comparison group did instead, whether students got corrective feedback, and how many times they were quizzed.
Federal education researchers reached a similar view earlier. A 2007 practice guide from the Institute of Education Sciences, the research arm of the US Department of Education, gave its top evidence rating, Strong, to using quizzes to re-expose students to key content. The panel, chaired by Harold Pashler of the University of California, San Diego, gave the same rating to just one other recommendation: prompting students to ask and answer deep explanatory questions about the material. The guide dates from 2007 and is written for teachers, but its ratings line up with the newer classroom data.
A quiz in this sense doesn’t need a grade. Flashcards you answer before flipping, a blank page where you write what you remember, or a friend asking you questions all count, as long as you produce the answer before you see it.
Spacing: let some forgetting happen first
Cramming can carry you through tomorrow morning’s test. The problem is what’s left a month later.
A 2006 review in Psychological Bulletin, led by Nicholas Cepeda, combined 839 comparisons drawn from 317 experiments on spreading study sessions apart. The gap between sessions that produced the best long-term retention grew as the time until the final test grew. The right spacing, in other words, depends on how long you need to remember.
A 2008 study put numbers on that. More than 1,350 people learned a set of facts, reviewed them after a gap of up to three and a half months, and took a final test up to a year later. When the test was a week away, the best gap was about 20% to 40% of that week. When the test was a year away, it shrank to about 5% to 10% of the year.
Translated into calendar terms:
| When you need to remember it | Best gap before reviewing (worked out from the 2008 study’s percentages) |
|---|---|
| A test in 7 days | Roughly 1.5 to 3 days after first learning |
| Something you’ll need in a year | Roughly 2.5 to 5 weeks after first learning |
The federal practice guide rates spacing as Moderate evidence, a notch below quizzing to re-expose content, and suggests revisiting key material several weeks to several months after it is first taught. The two methods combine naturally: each spaced review can be a self-test rather than a reread.
Where highlighting, summaries and other habits land
In 2013 five psychologists led by John Dunlosky of Kent State University reviewed ten common study techniques in the journal Psychological Science in the Public Interest. They rated each by how broadly it helps across ages, materials and kinds of tests. The ratings are the authors’ judgment, not an official standard.
Here is how the habits most students use came out, alongside the federal guide where it covered something similar:
| What you might do | Dunlosky team’s rating | Federal guide, 2007 |
|---|---|---|
| Reread notes or a chapter | Low | Not rated |
| Highlight or underline | Low | Not rated |
| Write summaries | Low | Not rated |
| Picture the text as mental images, or use keyword memory tricks | Low | Not rated |
| Ask yourself why a fact is true, or explain each step to yourself | Moderate | Strong, for the related teaching practice of deep explanatory questions |
| Mix different problem types in one session | Moderate | Not rated in this form |
| Test yourself | High | Strong, for quizzes that re-expose key content |
| Spread study across days or weeks | High | Moderate |
“Low” does not mean useless. The Dunlosky team found that rereading and highlighting do not consistently improve performance, and that summarizing helps some students on some tasks under limited conditions. The National Academies’ 2018 report How People Learn II lists helping students retrieve information, summarize and explain it, and space out material among strategies with evidence, and adds that what works depends on what the learner already knows, the kind of material and the goal. For rereading, the Dunlosky team suggested putting practice testing in its place.
“I’m a visual learner” doesn’t hold up
In a 2017 online survey of nearly 3,900 Americans, 93% of the general public agreed that people learn better when taught in their preferred learning style, such as visual, auditory or hands-on. So did 76% of educators and 78% of people with substantial neuroscience training. It was the most widely held of the “neuromyths” the researchers tested.
Preferences are real. Ask people and they will tell you whether they’d rather see a diagram or hear an explanation. The claim that fails is the next step, sometimes called the meshing hypothesis: that teaching each person in their preferred mode makes them learn more.
Testing that properly takes a particular design. Sort students by style, randomly assign each group to different teaching methods, give everyone the same test, and look for a crossover, where visual learners do best with visual teaching and auditory learners do best with something else. A 2008 review by Harold Pashler and three colleagues found virtually no evidence of that pattern, and several well-designed studies that contradicted it. They concluded there was “no adequate evidence base to justify incorporating learning-styles assessments into general educational practice.” They also cautioned that many versions of the idea had simply never been tested.
A 2024 meta-analysis from the University of North Dakota took another look, pooling 21 studies with 1,712 participants. Matched teaching showed a small average benefit, an effect size of 0.31 with a wide margin of uncertainty. Only about a quarter of outcome measures showed the crossover the theory needs, and study quality was low. Weighing those results against the time and money that matching takes, the authors did not consider the gains large or consistent enough for broad adoption.
In this library’s terms, the meshing claim has been tested and not supported: the 2024 meta-analysis that found a small average benefit also found the required crossover in only a minority of measures. Time spent on style quizzes would go further on retrieval and spacing, which the Dunlosky review found help learners across ages and abilities.
The same survey turned up other stubborn myths. A third of US educators agreed that people use only 10% of their brains. Next in line after the learning-styles items was the belief that a common sign of dyslexia is seeing letters backwards, held by 76% of the public and 59% of educators. Dyslexia is a reading disability whose core problem is not visual. The authors describe anecdotal reports from child assessment clinics of families and teachers delaying a referral because a struggling reader wasn’t reversing letters.
Growth mindset helps some students, a little
A growth mindset is the belief that intellectual ability can be developed through effort, better strategies and help. Short programs teaching it have been promoted as a way to raise grades. The research is a good example of a real effect that turned out smaller and more conditional than early enthusiasm suggested.
The largest single test is the National Study of Learning Mindsets, reported in Nature in 2019 by a team led by David Yeager of the University of Texas at Austin. In 65 US public high schools, 12,490 ninth graders were randomly assigned to two online sessions of about 25 minutes each: either the mindset program or a control program about brain functions. Independent research firms gathered and processed the data, and the main analysis was registered in advance.
Among lower-achieving students, the group the study targeted, grades in core courses rose by 0.10 points on the four-point scale, roughly the difference between a 2.5 and a 2.6. The share of those students finishing ninth grade below a 2.0 average fell from 46% to 41%. Gains were larger in schools where classmates tended to choose challenging work and smaller in the highest-achieving schools. An exploratory analysis in 41 of the schools found enrollment in advanced math the next year rose from 33% to 36%. The program told students the brain grows stronger with hard learning, like a muscle. That was a teaching metaphor, not a research finding.
Reviews that pool the wider literature read it differently:
- 2018, Victoria Sisk and colleagues: two meta-analyses found weak overall effects, with some support for benefits among low-income or academically at-risk students.
- 2023, Brooke Macnamara and Alexander burgoyne: across 63 studies and nearly 98,000 students, an average effect of 0.05 that lost significance after adjusting for publication bias, and 0.02 in the six highest-quality studies. Authors with a financial incentive reported larger effects. The pair concluded the apparent benefits are likely due to weak study design, reporting flaws and bias.
- 2023, Jeni Burnette and colleagues: using only randomized studies and a method that models how effects vary between groups, an effect of 0.14 on achievement when programs targeted students expected to benefit and were delivered as designed. The range of likely future effects ran from slightly negative to moderate.
A published commentary by a group that included Yeager and three of his 2019 co-authors argued that Burnette’s method is the better practice, and that applying it to Macnamara and Burgoyne’s own data showed a meaningful effect in at-risk groups.
Graded claim by claim: that a brief mindset lesson transforms achievement for students in general has been measured, and the effect falls below a size families would notice. That it gives struggling students a small lift in supportive schools counts as positive evidence whose interpretation is still contested.
Learning doesn’t stop with age
The National Academies’ 2018 report concluded that people keep learning and growing across the life span, and that how much and how well they learn depends on their choices, motivation, ability to manage their own learning, and circumstances. It also linked complex work, social engagement, physical exercise and adequate sleep with lifelong learning and healthy aging, wording that describes associations rather than proof that any one of them causes better learning.
A randomized trial shows what learning a demanding skill can do. In the Synapse Project in Dallas, published in 2014 in Psychological Science, 221 adults aged 60 to 90 completed a 14-week program. Apart from a comparison group given no activities, each was directed to spend about 15 hours a week on one of five assignments: learning digital photography and photo-editing software, learning to quilt on a computer-driven sewing machine, learning both, joining a social club with outings and games, or doing low-demand activities alone at home, such as listening to classical music and doing word puzzles. The three learning groups averaged about 16.5 hours a week.
The three learning groups together improved more in episodic memory, the memory for specific events and information, than the social and at-home groups. Photography showed the clearest gain, a net effect size of 0.54 against the at-home group, and it was the only comparison that stayed significant after the researchers corrected for running many tests. Quilting alone did not produce a significant gain. The social club produced no cognitive benefit, though the authors said that finding needs more work, and the trial did not report long-term follow-up. What moved memory scores here was the effortful work of learning something new and complicated, not simply staying busy.
When a learning problem needs a professional look
Most study trouble is a technique problem, not a medical one. A few situations are different:
- A child who is struggling to learn to read, whether or not they reverse letters. Waiting for reversals can delay help. Whether to request an evaluation is worth raising with the child’s school and pediatrician.
- An adult with a new difficulty learning or holding on to information, especially one that came on quickly. MedlinePlus, the National Library of Medicine’s consumer health site, says normal aging can mean needing more time to learn or recall, but not dramatic memory loss. It lists medicines, alcohol, low vitamin B12, head injury, depression and stroke among possible causes. Whether a change fits ordinary aging or calls for testing is a judgment for a clinician, who will want to know whether it arrived suddenly or gradually.
- Sudden confusion, or trouble speaking or understanding speech. Both are on the stroke warning-sign list from the Centers for Disease Control and Prevention, which says to call 911 right away.
A one-week study plan built on the evidence
This plan assumes a test seven days out and roughly 45 minutes a day. The structure matters more than the minutes.
- Day 1: Learn it once, carefully, then close the book. Write down everything you can recall. Check it against your notes and mark the gaps.
- Day 2: Leave it alone. Learn the next section the same way, or take the day off if there is only one.
- Day 3: First recall check. Without notes, turn each Day 1 heading into a question and answer it on paper. Review only what you missed. This lands about two days after first learning, inside the window the 2008 spacing study found best for a test a week away.
- Day 4: Ask why. Repeat the recall check for the Day 2 section, and for each key idea explain why it’s true or how it connects to something you already know.
- Day 5: Mix it up. Do practice questions that shuffle topics and problem types instead of grouping them.
- Day 6: Full practice test. Work under test-like conditions, then explain the questions you missed out loud, to a friend or an empty room.
- Day 7: Light check, not a marathon. Run through your list of weak spots with a quick self-quiz. Resist the urge to reread everything.
For material you want to keep for months, such as a language, a professional certification or a new job’s procedures, keep the same pattern but stretch the gaps: a self-test a few weeks later, then another a couple of months after that. Expect the quizzes to feel harder than rereading. That difficulty is not a sign the method is failing.
The single most useful change
Swap some rereading for recall. Close the book and try to produce what you just read, then check. It feels slower and less satisfying, which is exactly why people skip it, and it is the habit with the deepest support in this research, from laboratory experiments to classroom studies involving tens of thousands of students. Space those attempts out over days and weeks, and you have most of what the science of studying offers.
Keep exploring: study drugs and meditation and the brain.
Sources
- Psychological Science. “Roediger HL, Karpicke JD. "Test-enhanced learning: taking memory tests improves long-term retention." Psychol Sci, 2006. PubMed. DOI: 10.1111/j.1467-9280.2006.01693.x.” pubmed.ncbi.nlm.nih.gov
- Science. “Karpicke JD, Roediger HL. "The critical importance of retrieval for learning." Science, 2008. PubMed. DOI: 10.1126/science.1152408.” pubmed.ncbi.nlm.nih.gov
- Psychological Bulletin. “Yang C et al. "Testing (quizzing) boosts classroom learning: A systematic and meta-analytic review." Psychol Bull, 2021. PubMed. DOI: 10.1037/bul0000309.” pubmed.ncbi.nlm.nih.gov
- Institute of Education Sciences, U.S. Department of Education (National Center for Education Research). “Pashler H (Chair) et al. Organizing Instruction and Study to Improve Student Learning (NCER 2007-2004), September 2007.” ies.ed.gov
- Psychological Bulletin. “Cepeda NJ et al. "Distributed practice in verbal recall tasks: A review and quantitative synthesis." Psychol Bull, 2006. PubMed. DOI: 10.1037/0033-2909.132.3.354.” pubmed.ncbi.nlm.nih.gov
- Psychological Science. “Cepeda NJ et al. "Spacing effects in learning: a temporal ridgeline of optimal retention." Psychol Sci, 2008. PubMed. DOI: 10.1111/j.1467-9280.2008.02209.x.” pubmed.ncbi.nlm.nih.gov
- Psychological Science in the Public Interest. “Dunlosky J et al. "Improving Students' Learning With Effective Learning Techniques: Promising Directions From Cognitive and Educational Psychology." Psychol Sci Public Interest, 2013. PubMed. DOI: 10.1177/1529100612453266.” pubmed.ncbi.nlm.nih.gov
- National Academies of Sciences, Engineering, and Medicine. “How People Learn II: Learners, Contexts, and Cultures (2018), Summary.” nationalacademies.org
- Frontiers in Psychology. “Macdonald K et al. "Dispelling the Myth: Training in Education or Neuroscience Decreases but Does Not Eliminate Beliefs in Neuromyths." Front Psychol, 2017. PubMed. DOI: 10.3389/fpsyg.2017.01314.” pubmed.ncbi.nlm.nih.gov
- Psychological Science in the Public Interest. “Pashler H, McDaniel M, Rohrer D, Bjork R. "Learning Styles: Concepts and Evidence." Psychol Sci Public Interest, 2008. PubMed. DOI: 10.1111/j.1539-6053.2009.01038.x.” pubmed.ncbi.nlm.nih.gov
- Frontiers in Psychology. “Clinton-Lisell V, Litzinger C. "Is it really a neuromyth? A meta-analysis of the learning styles matching hypothesis." Front Psychol, 2024. PubMed. DOI: 10.3389/fpsyg.2024.1428732.” pubmed.ncbi.nlm.nih.gov
- Nature. “Yeager DS et al. "A national experiment reveals where a growth mindset improves achievement." Nature, 2019. PubMed. DOI: 10.1038/s41586-019-1466-y.” pubmed.ncbi.nlm.nih.gov
- Psychological Science. “Sisk VF et al. "To What Extent and Under Which Circumstances Are Growth Mind-Sets Important to Academic Achievement? Two Meta-Analyses." Psychol Sci, 2018. PubMed. DOI: 10.1177/0956797617739704.” pubmed.ncbi.nlm.nih.gov
- Psychological Bulletin. “Macnamara BN, Burgoyne AP. "Do growth mindset interventions impact students' academic achievement? A systematic review and meta-analysis with recommendations for best practices." Psychol Bull, 2023. PubMed. DOI: 10.1037/bul0000352.” pubmed.ncbi.nlm.nih.gov
- Psychological Bulletin. “Burnette JL et al. "A systematic review and meta-analysis of growth mindset interventions: For whom, how, and why might such interventions work?" Psychol Bull, 2023. PubMed. DOI: 10.1037/bul0000368.” pubmed.ncbi.nlm.nih.gov
- Psychological Bulletin. “Tipton E et al. "Why Meta-Analyses of Growth Mindset and Other Interventions Should Follow Best Practices for Examining Heterogeneity: Commentary on Macnamara and Burgoyne (2023) and Burnette et al. (2023)." Psychol Bull, 2023. PubMed. DOI: 10.1037/bul0000384.” pubmed.ncbi.nlm.nih.gov
- Psychological Science. “Park DC et al. "The impact of sustained engagement on cognitive function in older adults: the Synapse Project." Psychol Sci, 2014. PubMed. DOI: 10.1177/0956797613499592.” pubmed.ncbi.nlm.nih.gov
- MedlinePlus (U.S. National Library of Medicine). “Memory loss.” medlineplus.gov
- Centers for Disease Control and Prevention. “Signs and Symptoms of Stroke.” cdc.gov