Cramming¶

Strategy · All strategies
Evidence · 1 claim (1 against) · 2 studies (1 causal, 1 quant-synthesis), q3–q4 · 1 of 2 report an effect size

Description¶

Cramming concentrates all study of a topic into one or a few sessions immediately preceding an exam, rather than distributing it over time (spacing) or interleaving topics. It typically involves rereading notes and highlighting — high-fluency, low-retrieval activities — under time pressure and sleep deprivation. It can produce adequate performance on an imminent test while leaving knowledge fragile and rapidly decaying.

Design Implications¶

Cramming exploits the spacing effect: massed practice feels effective because material remains in working memory and retrieval is easy, but distributed practice produces substantially better long-term retention for the same total study time [~S]. The subjective fluency of a cram session is a poor proxy for learning — learners mistake ease of processing for durable encoding, a metacognitive illusion that makes cramming self-reinforcing. Because cramming relies heavily on rereading rather than retrieval, it skips the retrieval practice that drives durable memory, and sleep loss immediately after study further impairs consolidation.

Context¶

Requirements¶

• A deadline or assessment that makes massed study rational in the learner's eyes
• Access to notes, slides, or summaries to reread
• Sufficient unbroken time block immediately before the assessment

Constraints¶

• Retention collapses within days to weeks; massed study produces far worse delayed-test performance than spaced study of equal duration [~S]
• Rereading-based cramming builds recognition familiarity, not recall ability — learners fail when the exam requires generation or transfer rather than recognition
• Sleep deprivation after a late-night session disrupts memory consolidation, offsetting even the short-term gains
• Promotes surface strategies (memorizing isolated facts) over schema building; under time pressure, working memory is overloaded and deep encoding suffers Cognitive overload degrades learning. [-S]
• Models poor self-regulation: learners who habitually cram show weaker planning and monitoring skills [~W]

Implementation Variability¶

• Instructor-structured spacing — schedule multiple low-stakes quizzes across the term so learners cannot defer all study to the end; cumulative quizzing forces periodic retrieval
• "Minimum viable spacing" — even two sessions separated by a day outperform one session of double length; useful advice for learners who will not adopt full spaced schedules
• Cramming as review — a final consolidated session on top of prior spaced study is far less harmful than cramming as the only study; the distinction matters for advising
• Interleaved cramming — if time is fixed and short, mixing problem types within the session still beats blocking them

Target Learners¶

• Most harmful for learners building foundational knowledge intended for later courses, where prior-knowledge decay compounds
• Least harmful for one-off assessments with no cumulative follow-on — though even here retrieval-based study outperforms rereading
• Common among novices who lack the metacognitive awareness to distinguish fluency from learning; explicit instruction on desirable difficulties reduces reliance on cramming [~M]

Target Learning Goals¶

• Poorly suited to durable retention, transfer, and cumulative skill building
• Marginally serviceable for short-term recognition of isolated facts
• Counterproductive for goals requiring schema construction, since massed single-session study does not support the repeated, varied encoding that builds flexible knowledge

Instructions¶

1. Diagnose the pattern: ask learners when they studied and how (rereading vs. self-testing) — most crammers are also rereaders.
2. Teach the fluency-vs-learning distinction with a demonstration: two groups study the same list, one rereads, one self-tests; test both after five minutes and after a week.
3. Replace the single session with a spaced schedule using Spaced Practice principles — short sessions at expanding intervals.
4. Convert rereading time into retrieval: closed-book recall, flashcards, or practice problems, per the testing effect.
5. Structure the course so cramming is not viable: cumulative low-stakes quizzes, distributed assignments, and final exams weighted toward material from the whole term.

Related Strategies¶

• Spaced Practice — the direct alternative; same total time, distributed across sessions
• Retrieval Practice — replaces passive rereading with self-testing that strengthens memory
• Interleaving — mixes problem types within and across sessions, countering blocked massed study
• Cumulative Quizzing — course structure that makes deferring study impossible
• Favor learning strategies that rely on long-term memory and integrate new material with existing cognitive structures

Examples¶

• Roediger & Karpicke (2006) — students who repeatedly studied a passage predicted better delayed recall, but students who practiced retrieval remembered far more one week later; the rereaders' confidence was a cramming-style illusion.
• Dunlosky et al. (2013) — rated rereading and highlighting (the core cramming behaviors) as low utility, while distributed practice and practice testing received the highest utility rating across hundreds of studies.
• Cumulative quiz design in introductory STEM courses — weekly quizzes covering all prior material, a common restructuring used to eliminate end-of-term massing.

Key Sources¶

• Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., & Rohrer, D. (2006). Distributed practice in verbal recall tasks: A review and quantitative synthesis. Psychological Bulletin, 132(3), 354–380. doi:10.1037/0033-2909.132.3.354
• Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249–255. doi:10.1111/j.1467-9280.2006.01693.x
• Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students' learning with effective learning techniques. Psychological Science in the Public Interest, 14(1), 4–58. doi:10.1177/1529100612453266
• Kornell, N., & Bjork, R. A. (2008). Learning concepts and categories: Is spacing the "enemy of induction"? Psychological Science, 19(6), 585–592. doi:10.1111/j.1467-9280.2008.02127.x