The night before an exam, students across the country engage in a familiar ritual: highlighters in hand, they reread their notes and textbooks, marking passages that seem important. Many will stay up late, cramming material in a final push. These strategies feel productive. They're also largely ineffective.

A landmark 2013 review published in Psychological Science in the Public Interest by John Dunlosky at Kent State University and colleagues evaluated ten common learning techniques used by students. Their conclusion was striking: highlighting, rereading, and cramming—among the most popular study strategies—provide minimal benefits for long-term retention. Meanwhile, the techniques that actually work remain underused, in part because few students ever learn about them.

Understanding what the research shows can transform how students prepare for tests, reducing wasted effort while improving both performance and retention.

• • •

Why Popular Study Methods Fall Short

Surveys of university students reveal just how widespread ineffective strategies are. Research by Kornell and Bjork at UCLA found that 84 percent of students at an elite university studied by rereading their notes or textbooks. A study by Hartwig and Dunlosky found that 83.6 percent of students reported massing their study sessions—cramming—while 66 percent relied primarily on rereading.

Why don't these approaches work? The problem lies in how memory functions. Rereading creates what psychologists call the "illusion of competence": as text becomes more familiar, students mistake fluency for genuine understanding. The words feel easier to process, which the brain interprets as mastery. But recognition and recall are different cognitive processes. Being able to recognize information when you see it doesn't mean you can retrieve it independently during an exam.

Highlighting faces similar limitations. Students often struggle to identify which material is genuinely important, and even when highlighting does improve memory for marked passages, it can actually impair recall of surrounding context. As Dunlosky and colleagues concluded, most studies show no benefit of highlighting over simply reading.

• • •

The Testing Effect: Learning Through Retrieval

The single most robust finding in learning science may be the "testing effect"—the discovery that actively retrieving information from memory strengthens that memory far more than passive review. This phenomenon has been documented in hundreds of studies across diverse populations and subject areas.

Research by Henry Roediger and Jeffrey Karpicke at Washington University in St. Louis demonstrated that students who studied a passage once and then took a practice test remembered significantly more than students who studied the passage four times. The effect was particularly pronounced when measured after a delay: testing produced superior long-term retention even though it felt less effective to students in the moment.

A 2021 meta-analysis published in Psychological Bulletin, examining classroom quizzing across multiple studies, confirmed that testing boosts learning with a medium to large effect size. The benefits extend beyond simple memorization to transfer and application of knowledge.

Why does retrieval work so well? Current theories suggest that the effort of pulling information from memory changes how that information is stored, creating stronger and more accessible memory traces. Retrieval also helps students identify gaps in their knowledge—if you can't recall something during a practice test, you know what to study further.

The practical implication is straightforward: instead of rereading notes, students should close their books and try to recall the material, whether through flashcards, practice problems, or simply writing down everything they can remember about a topic.

• • •

Spacing: Why Cramming Backfires

Even students who use retrieval practice often undermine its benefits by cramming. A substantial body of research demonstrates that spacing study sessions over time produces dramatically better retention than massing practice into a single session.

A 2025 study published in Academic Medicine involving over 26,000 physicians found that spaced repetition was significantly superior to massed study for both learning and knowledge transfer. Participants in spaced repetition conditions outperformed controls by roughly 15 percentage points on assessments administered months after initial learning. Double-spaced repetitions—returning to material twice over time—proved even more effective than single repetitions.

Research in undergraduate STEM courses has found similar patterns. A 2024 meta-analysis in the International Journal of STEM Education examined spaced retrieval practice across nine introductory courses and found consistent benefits, though the magnitude varied by subject and implementation.

The spacing effect appears to work because distributed practice requires the brain to repeatedly reconstruct memory traces rather than simply maintaining information in working memory during a cramming session. Each retrieval event at a delay strengthens the underlying memory in ways that massed practice cannot replicate.

For students, this means starting to study well before an exam and returning to material at intervals. A study session Sunday, another Wednesday, and a final review Friday will typically outperform the same total time spent studying Thursday night alone.

• • •

Sleep: The Overlooked Performance Factor

Sleep deprivation has become normalized among students, particularly during exam periods. Yet research consistently links sleep to academic performance in ways that cramming cannot compensate for.

A 2019 study published in npj Science of Learning tracked MIT students using wearable activity monitors throughout an academic term. The findings were notable: sleep quality, duration, and consistency correlated with grades, with sleep measures accounting for nearly 25 percent of the variance in academic performance. Crucially, sleep on the single night before a test showed no relationship to performance—what mattered was sleep patterns in the weeks leading up to exams.

This finding was replicated in a 2023 study published in PNAS tracking first-year college students at three universities. Every additional hour of average nightly sleep early in the semester was associated with a 0.07 increase in end-of-term GPA. The effect persisted even after controlling for previous academic performance.

Research from the University of Madrid published in 2023 in Sleep and Biological Rhythms found that sleep quality during examination periods was positively associated with academic performance, while partial sleep deprivation showed negative associations with grades. Students classified as "good sleepers" performed better than those with poor sleep quality.

The mechanisms connecting sleep to test performance involve memory consolidation—the process by which the brain stabilizes and integrates newly learned information during sleep. Cutting sleep to gain extra study time may actually be counterproductive, trading effective consolidation for diminishing returns from late-night cramming.

• • •

Managing Test Anxiety

Test anxiety affects a substantial portion of students and can meaningfully impair performance. A 30-year meta-analytic review published in the Journal of School Psychology by von der Embse and colleagues found negative associations between test anxiety and academic achievement, with cognitive worry appearing particularly detrimental.

Effective interventions do exist. A 2019 meta-analysis published in the Journal of Anxiety Disorders by Huntley and colleagues at the University of Liverpool examined randomized controlled trials for test-anxious university students. Combined cognitive-behavioral approaches showed the strongest effects, though the researchers noted that most interventions produced only medium effects for reducing anxiety and weak effects for improving performance—suggesting significant room for improvement.

More recent research has explored mindfulness-based interventions. A 2024 meta-analysis in Frontiers in Psychology found that mindfulness-based interventions demonstrated moderate to large effects on reducing test anxiety across 18 studies involving over 1,200 participants.

Research from the University of Liverpool published in 2023 in Frontiers in Education found that metacognitive beliefs—particularly beliefs about the uncontrollability and danger of worry—predicted test anxiety and subsequent exam performance. This suggests that helping students recognize and challenge catastrophic thinking about exams may be particularly beneficial.

Some degree of anxiety is normal and may even enhance performance through increased alertness. The goal isn't to eliminate nervousness entirely but to prevent anxiety from overwhelming cognitive resources needed for test-taking.

• • •

What Students and Educators Should Do Differently

The gap between research evidence and common practice represents a significant missed opportunity. Students continue using ineffective strategies not because they're lazy, but because they were never taught alternatives. A survey of educational psychology textbooks found that effective learning strategies are largely overlooked, leaving future teachers without the knowledge to instruct their students.

For students, the evidence points toward several concrete changes: replace rereading with self-testing, spread study sessions across days or weeks rather than cramming, prioritize sleep during exam periods, and address anxiety through cognitive strategies rather than avoidance.

For educators, low-stakes quizzes at the beginning of class can leverage the testing effect while providing students experience with retrieval practice. Cumulative exams encourage distributed study. Explicit instruction about which strategies work—and which don't—may help students develop more effective study habits.

The research on effective learning is remarkably consistent across ages, subjects, and contexts. The challenge lies in translating that evidence into practice. Students who make even modest changes to their approach—testing themselves instead of rereading, spacing their study sessions, protecting their sleep—position themselves to learn more efficiently and perform better when it matters most.

Sources