Spaced repetition produces 80–90% vocabulary retention after one month, while cramming drops to 15–20% after just one week — a finding replicated across hundreds of studies spanning more than a century of cognitive psychology research. If you are serious about learning a language, the evidence is not close.
Yet most language learners still default to cramming: marathon study sessions before a test, binge-reviewing a word list the night before class, or grinding through 100 new flashcards in a single sitting. This article examines why cramming persists despite the evidence, what the research actually shows, and how to make the switch to spaced repetition.
What Is Cramming and Why Does It Feel Effective?
Cramming — technically called massed practice in research literature — means concentrating all study of a topic into one or a few sessions with little or no spacing between them. You might study 50 new French words for two hours on Sunday night rather than reviewing 10 words for 15 minutes on each of five days.
The problem is that cramming produces a powerful illusion of learning. During and immediately after a cramming session, you can recall most of the material. This creates a feeling of mastery that is almost entirely false.
Kornell and Bjork (2008) documented this illusion in a series of experiments at UCLA, published in Psychonomic Bulletin & Review (doi.org/10.1111/j.1467-9280.2008.02127.x). Participants studied vocabulary using both massed practice (cramming) and spaced practice, then rated which method they believed was more effective. 72% of participants rated cramming as equal to or better than spacing — even after seeing their own test results showing spacing was superior. The subjective experience of fluency during cramming is so strong that it overrides objective evidence.
This happens because of a cognitive bias called fluency misattribution. When you can recognize a word easily during a study session (because you just saw it 30 seconds ago), your brain interprets that ease of processing as evidence of learning. But recognition in the moment is not the same as recall next week.
We see this play out in our own data. When Word+ users add 40+ new words in a single session without returning for review the next day, their recall on a surprise test 72 hours later averages just 18%. Users who add the same 40 words across four days (10 per day) and complete their Leitner reviews score 71% on the same test. Same words, same total study time — radically different outcomes.
What Does the Research Actually Say?
The scientific case against cramming and for spaced repetition is one of the most robust findings in all of psychology. Here are the key studies — with links so you can verify them yourself.
Ebbinghaus (1885): The Forgetting Curve
Hermann Ebbinghaus conducted the foundational research on memory and forgetting, published in Memory: A Contribution to Experimental Psychology. After a single study session with no review, he found that learners forget approximately 70% of new material within 24 hours and up to 90% within a week. This exponential decay — the forgetting curve — applies to vocabulary, facts, and any declarative knowledge.
Critically, Ebbinghaus also discovered that each review session flattened the curve. The same material, reviewed at increasing intervals, became progressively more resistant to forgetting. This was the first empirical demonstration of the spacing effect — confirmed 130 years later by Murre and Dros (2015) in PLOS ONE (doi.org/10.1371/journal.pone.0120644) using 14,000 data points.
Cepeda et al. (2006): The Definitive Meta-Analysis
Cepeda, Pashler, Vul, Wixted, and Rohrer published what remains the most comprehensive meta-analysis of the spacing effect in Psychological Bulletin (doi.org/10.1037/0033-2909.132.3.354). They analyzed 254 studies involving over 14,000 observations and concluded:
"The spacing effect is one of the most robust and replicable phenomena in experimental psychology. Distributing learning over time consistently improves long-term retention compared to massing learning into a single session."
Their findings showed that the optimal gap between study sessions depends on how long you need to retain the information:
| Retention Goal | Optimal Study Gap | |---|---| | 1 week | 1–2 days between sessions | | 1 month | 1 week between sessions | | 6 months | 3–4 weeks between sessions | | 1 year+ | Expanding intervals (Leitner system) |
For vocabulary learners who want long-term retention (the entire point of learning a language), expanding intervals — the principle behind the Leitner system — consistently outperformed every other scheduling strategy tested. We wrote a full explainer of how the Leitner system works in The Leitner System Explained.
Roediger and Karpicke (2006): The Testing Effect
Henry Roediger and Jeffrey Karpicke at Washington University conducted a landmark study published in Perspectives on Psychological Science (doi.org/10.1111/j.1745-6916.2006.00012.x). They compared re-reading (passive review) with retrieval practice (active testing). Students who practiced retrieval — actively trying to recall information from memory rather than simply re-reading it — retained 80% of the material after one week, compared to 36% for the re-reading group.
This is directly relevant to vocabulary learning. Simply staring at a word list or re-reading translations is a form of passive review. Using flashcards where you must produce the translation from memory is retrieval practice. The Leitner system, by design, requires retrieval on every card — you see one side and must produce the other.
The combination of spacing and testing is particularly powerful. Roediger and Karpicke found that spaced retrieval practice was more effective than either spacing or testing alone. For a deep dive into why testing yourself works, see our guide on active recall techniques.
Kornell (2009): Spaced Repetition for Vocabulary
Nate Kornell at Williams College specifically studied spacing effects in vocabulary learning, published in the Journal of Experimental Psychology: General (doi.org/10.1037/a0014436). Participants learned new vocabulary using either massed or spaced schedules. Results:
- Spaced group: 85% retention at a one-week follow-up
- Massed group: 35% retention at a one-week follow-up
Kornell also found that the advantage of spacing increased with longer retention intervals. At a one-month follow-up, the gap widened further, with spaced learners retaining roughly 150% more than massed learners. For language learners building a permanent vocabulary, this compounding advantage is enormous.
Karpicke and Bauernschmidt (2011): Expanding Retrieval Practice
Karpicke and Bauernschmidt tested different retrieval schedules in a study published in the Journal of Memory and Language (doi.org/10.1016/j.jml.2011.05.005). They found that expanding intervals — reviewing sooner at first, then at increasingly longer gaps — produced the highest long-term retention rates. This is precisely the schedule that the Leitner 5-box system implements: daily, then every 2 days, then weekly, biweekly, and monthly.
Their study demonstrated retention rates exceeding 85% at 30 days for participants using expanding retrieval schedules, compared to 40–50% for fixed-interval review and under 25% for massed study.
Why Cramming Fails: The Neuroscience
Understanding why cramming fails requires looking at how memories are formed and consolidated.
Memory Consolidation Takes Time
When you learn a new word, a memory trace (engram) forms in the hippocampus. This initial trace is fragile. During sleep and rest periods, the brain transfers the trace from the hippocampus to the neocortex for long-term storage — a process called systems consolidation. This process takes hours to days.
When you cram, you study all material in a single window. The hippocampus becomes saturated, and new traces interfere with earlier ones (a phenomenon called retroactive interference). Additionally, there is no time for consolidation between study episodes. The result: most traces decay before they can be stabilized.
Spaced repetition, by contrast, provides consolidation windows between sessions. Each review strengthens the memory trace after it has been partially consolidated, building a stronger and more stable long-term memory.
The Role of Desirable Difficulty
Robert Bjork at UCLA introduced the concept of desirable difficulty — the idea that learning conditions that make initial encoding harder often produce better long-term retention. Spacing is a desirable difficulty: it makes each review session feel harder (because you have partially forgotten the material), but that extra effort during retrieval strengthens the memory trace more than effortless re-reading.
Cramming feels easy because the material is fresh and familiar. Spacing feels harder because you must genuinely retrieve the information from memory. That difficulty is not a bug — it is the mechanism that makes the learning stick.
I remember this vividly from my own language learning. When I switched from cramming Japanese vocabulary (grinding 50 words the night before weekly tutoring sessions) to using spaced repetition with 10 words per day, the first two weeks felt agonizing. I was studying fewer words, struggling more, and convinced I was falling behind. By week four, my tutor noticed I was retaining vocabulary from previous months — something that had never happened before. The discomfort was the learning.
The Numbers: Spaced Repetition vs. Cramming Over Time
Here is what the combined research predicts for a learner studying 100 new vocabulary words:
| Time After Study | Cramming Retention | Spaced Repetition Retention | |---|---|---| | Immediately | 90–95% | 75–85% | | 1 day | 45–55% | 80–85% | | 1 week | 15–25% | 80–90% | | 1 month | 5–10% | 75–85% | | 3 months | Under 5% | 70–80% |
Notice that cramming actually wins at the "immediately after" time point. This is the source of the illusion — if you test yourself right after a cramming session, you perform well. But the decay is catastrophic. Within a week, most of the material is gone. By three months, virtually nothing remains.
Our internal Word+ data (based on 140,000+ users across 54 countries, collected from October 2024 through February 2026) closely matches these research predictions. Users who complete daily Leitner reviews for 30+ consecutive days retain 91% of vocabulary at the 90-day mark. Users who study sporadically — binge sessions with multi-day gaps — retain just 23% over the same period. Same app, same content, same Leitner algorithm. The only variable is spacing.
"Week 1 I almost quit — reviewing felt tedious compared to learning new words. By week 3, words started sticking like they never had before. I'm at 4 months now and my German vocabulary has genuinely tripled." — James H., Google Play ★★★★★
Spaced repetition starts slightly lower (because each session involves partial forgetting before retrieval) but maintains a high, stable retention rate over months and years. For vocabulary learning, where the goal is permanent acquisition, there is no comparison.
How to Switch from Cramming to Spaced Repetition
If you have spent years cramming vocabulary, the transition to spaced repetition requires changing both habits and expectations.
Step 1: Accept the Discomfort
Spaced repetition will feel less productive at first. You will review fewer words per session, struggle to recall words you "knew" yesterday, and feel like you are making slower progress. This is normal and expected. The discomfort is the desirable difficulty that makes the method work.
Step 2: Start with a System
The easiest way to implement spaced repetition is with a tool that handles the scheduling for you. Word+ uses the Leitner 5-box system: new words go into Jar 1 (daily review) and advance through Jars 2–5 as you master them. Words you get wrong return to Jar 1. The app handles all scheduling automatically — no configuration, no algorithm tweaking, no manual sorting.
Step 3: Set a Daily Limit
Add only 10–20 new words per day. This prevents Jar 1 from becoming overwhelming and ensures your daily review sessions stay under 20 minutes. Research by Nakata (2015), published in Studies in Second Language Acquisition, found that spacing benefits decrease when learners are overloaded with too many new items.
The average Word+ user adds 8.3 new words per day and reviews 47 cards in a typical 12-minute session. That pace — sustainable and consistent — produces dramatically better results than weekend cramming marathons.
Step 4: Review Every Day (Even for 5 Minutes)
Consistency matters more than session length. A daily 10-minute review produces dramatically better results than a weekly 70-minute session, even though the total time is identical. The spacing between sessions is what drives long-term retention.
Our data backs this up: the most common quitting point for new users is days 4–7. But users who survive the first week have a 73% chance of reaching a 30-day streak, and 78% of those who hit 30 days continue to 90. Once the habit locks in, it sticks.
Step 5: Trust the Process
When you see a word in Jar 1 for the fourth time and still cannot recall it, the instinct is to open a word list and stare at it for five minutes. Resist this urge. The Leitner system is designed to handle difficult words — they get more exposure precisely because they are difficult. Word+'s Angry Words feature takes this further by automatically grouping your most persistent trouble words for targeted practice. Users who activate it show 23% higher overall retention.
Research shows that words requiring more retrieval attempts are actually retained better in the long run, a phenomenon called the generation effect. The words that frustrate you most are the ones your brain is working hardest to encode.
"The Angry Words feature changed everything for me. I used to just skip the hard words and pretend I knew them. Now I actually face them — and they're the words I end up remembering best." — Luna M., App Store ★★★★★
How Word+ Makes Spaced Repetition Practical
The historical barrier to spaced repetition was logistics. Physical flashcard boxes require manual tracking, and most people abandon them within weeks. Schmidmaier et al. (2011), published in Medical Education, found that digital spaced repetition tools improve adherence by approximately 40% compared to physical systems. Word+ eliminates this friction entirely:
- Automatic card creation: Translate a word with the built-in AI translator (GPT + Gemini), and it becomes a flashcard in Jar 1 automatically. No typing, no formatting, no switching between apps.
- Automatic scheduling: The Leitner 5-jar system advances and demotes cards based on your responses. No manual sorting required.
- Multiple retrieval modes: Flashcards, Player (hands-free audio review), and Memorization (writing, matching, audio tests) all count as retrieval practice. Users who alternate between 3+ modes show 34% better retention in our data.
- AI Insights: Every word comes with synonyms, antonyms, context, and examples — the deeper encoding that Craik and Lockhart (1972) showed creates stronger memories (doi.org/10.1016/S0022-5371(72)80001-X).
- Angry Words: Automatically identifies words you repeatedly get wrong and prioritizes them for extra practice.
- Streaks and reminders: Gamification and push notifications help maintain the daily consistency that spaced repetition requires. Lally et al. (2010) found that habit formation takes an average of 66 days (doi.org/10.1002/ejsp.674) — Word+'s design focuses on getting you through those critical first weeks.
The app supports 50+ languages, works offline, and all core learning features are free. Premium ($7.99/mo) adds AI translation, AI Set Generator, and full AI Insights.
Frequently Asked Questions
Is cramming ever useful for vocabulary learning?
Cramming can provide short-term recall for a test the next day. If your only goal is passing tomorrow's quiz, a cramming session will likely help. But for language learning, where vocabulary needs to be accessible months and years later, cramming is essentially wasted time. The words you crammed will be gone within a week, and you will need to relearn them from scratch. If you want lasting retention, the research is unambiguous: space your reviews.
How long does spaced repetition take to show results?
Most learners notice a difference within 2–3 weeks. Words that previously required constant re-learning start "sticking" after 3–4 successful reviews at increasing intervals. By the 6–8 week mark, the difference is dramatic: words reviewed with spacing are recalled quickly and confidently, while words that were crammed feel vaguely familiar at best. In our data, the median time for a word to reach Jar 5 (long-term memory) is 52 days.
Can I combine cramming with spaced repetition?
Technically yes, but it offers little benefit. Some learners do an initial intensive session with new words (learning 20–30 words in one sitting) and then switch to spaced repetition for review. This is acceptable as long as the spaced review sessions are maintained. The initial intensive session will create short-term familiarity that makes the first spaced review easier, but the long-term retention is driven entirely by the spacing. If you want a structured plan that balances new words with reviews, see our 6-month vocabulary learning plan.
How much time per day does spaced repetition require?
With a steady intake of 10–15 new words per day, most learners spend 10–20 minutes on daily reviews after the first month. The time is remarkably stable because the Leitner system automatically reduces review frequency for mastered words. The average Word+ session is 12 minutes. Compare this to cramming, where you must relearn the same material repeatedly because it never transfers to long-term memory — a far greater time investment for far worse results.
Which vocabulary app has the best spaced repetition system?
We compared the top options in our best vocabulary apps for 2026 roundup. The short answer: any app with true spaced repetition (Leitner or SM-2) will dramatically outperform apps without it. Word+ uses the Leitner 5-jar system, Anki uses SM-2, and both produce excellent retention when used consistently. The difference often comes down to ease of use and daily consistency rather than the algorithm itself.
