Skip to main content
Vocabulary research

How many times do you
need to see a word?

Roughly 8–12 encounters for a word you meet while reading — individual studies report anywhere from 6 to more than 20. The count falls when each encounter is a recall attempt rather than a glance, and falls further when the encounters are spread across days instead of stacked in one sitting. The hard part was never the number; it’s getting that many spaced repetitions into an ordinary week.

See the evidence Run the numbers

Sources cited in full at the foot of the page

A vocabulary flashcard shown on an iPhone the moment a blocked app is opened, prompting a single recall attempt

The short answer: 8–12 encounters

There is no magic number, but the research clusters tightly enough to plan around. Four findings do most of the work:

  • About 10 encounters is the usual threshold. Nation and Wang (1999) found words were substantially more likely to be learned after around ten meetings in context, though even then acquisition wasn’t guaranteed. Pigada and Schmitt (2006) reported the same shape: a marked jump in learning at roughly ten or more exposures.
  • The published range is 6 to 20+. Different studies land on 6, 8, 10, 12 or more than 20, depending on the word, the learner and how learning was measured. Treat 8–12 as the planning figure, not a guarantee.
  • Repetition is a medium-sized effect, not a magic one. A meta-analysis of 45 effect sizes from 26 studies (N = 1,918) put the correlation between number of encounters and incidental vocabulary learning at r = .34 (Uchihara, Webb & Yanagisawa, 2019). Frequency matters — it just isn’t the only thing that does.
  • Spacing was one of the moderators. The same meta-analysis found the size of the repetition effect varied with, among other things, whether the encounters were spaced. When you meet the word again changes how much that meeting is worth.

So the honest one-line answer: plan for about ten good encounters per word, and care as much about their spacing and quality as their count.

Not all encounters are worth the same

“Seeing” a word and “being asked for” a word are different events in memory. Roediger and Karpicke (2006) had learners either re-read material or take recall tests on it the same number of times. Re-reading won when the final test came five minutes later — and lost badly when it came two days or a week later. Retrieval, not exposure, is what survives the delay.

That’s why a flashcard you have to answer buys more progress per encounter than a word list you skim, and why the 8–12 figure — measured on incidental reading — is an upper bound rather than a target when you’re actively recalling.

Types of word encounter compared by what happens in memory and roughly how many are needed
Kind of encounter What it does to memory Encounters needed
Meeting it while reading Meaning inferred from context; recognition builds slowly ~8–12, sometimes 20+
Re-reading a word list Familiarity without recall — feels easy, fades fast Many, and low value each
Answering a flashcard Active retrieval — the effortful attempt is the learning Fewer, and each one counts
Answering on a spaced schedule Retrieval timed near the edge of forgetting Fewest — retrieval plus spacing compound

Prefer the classic deck-based version of the bottom row? See how a passive scheduler compares as an Anki alternative.

When the repetitions happen matters as much as how many

Hermann Ebbinghaus mapped the forgetting curve in 1885: new material drops away sharply in the first days unless it’s revisited. Every repetition schedule since has been an attempt to place encounters where that curve is steepest.

The modern synthesis is unambiguous. Cepeda, Pashler, Vul, Wixted and Rohrer (2006) reviewed 839 assessments of distributed practice across 317 experiments in 184 articles and found spaced study reliably outperforms massed study — and that the optimal gap between encounters grows as the delay before the test grows. Ten encounters in one evening and ten encounters over three weeks are not the same ten encounters.

In practice that means an expanding schedule. A Leitner system — the box method spaced-repetition apps are built on — produces one automatically: get a word right and it moves to a slower box; get it wrong and it drops back to a fast one.

An example expanding-interval schedule for reaching about ten encounters of a single word
Leitner box Gap before the next attempt Where you are with the word
Box 1 Same day Brand new, or just got it wrong
Box 2 ~1 day Recognised once; still fragile
Box 3 ~3 days Recalled without the context clue
Box 4 ~1 week Surviving a real delay
Box 5 ~2–4 weeks Effectively known; occasional check-ins

Illustrative intervals in the Leitner tradition; exact gaps vary by implementation. LearnScreen uses a Leitner schedule of this shape.

Where ten encounters per word actually come from

Here’s the practical problem. Ten spaced retrievals per word, across a 20-word rotation, is 200 recall attempts — and a study session you have to remember to start is the least reliable way to produce them. The most reliable source of repeated moments in a modern day is the phone itself.

Reviews.org’s 2026 cell-phone usage survey (fielded Q4 2025, ~1,000 US adults) found people check their phone 186 times a day — about 11.6 times an hour. That is an enormous, already-existing budget of repetitions, currently spent on feeds.

Worked example converting daily phone pickups into vocabulary encounters
Step Figure Where it comes from
Phone checks per day 186 Reviews.org 2026 survey, US adults
Share landing on a shielded app 15% Assumption — the apps you chose to block
Recall attempts per day ~28 One flashcard per shielded open
Words in the active rotation 20 A typical working set
Days to ~10 encounters each ~7 28 ÷ 20 ≈ 1.4 per word per day

Arithmetic, not a promise. Your pickup count and blocked-app share will differ, and a spaced scheduler deliberately doesn’t spread encounters evenly — it front-loads the words you keep missing and backs off the ones you know.

Getting the encounters
without scheduling them

The repetitions have to happen. They don’t have to be planned by you. LearnScreen attaches each one to a habit you already repeat dozens of times a day.

1

Choose the trigger apps

Pick the apps that pull you in — Instagram, TikTok, X, games. They’re shielded at the OS level through Apple’s Screen Time API, so there’s no VPN or DNS trick to route around it.

2

Each open becomes a retrieval

A vocabulary card appears before the app does. You answer it — a real recall attempt, the kind of encounter that survives a delay — and the app unlocks for a set window.

3

The schedule picks the word

A Leitner spaced-repetition schedule chooses which word you see. Words you missed come back fast; words you know back off. The spacing is handled without you tracking anything.

  • 1,080+ curated words across 18 topics in 11 languages — English, German, Spanish, French, Italian, Japanese, Korean, Polish, Brazilian Portuguese, Russian and Mandarin Chinese.
  • Your own words too — add any word with translation, transcription and an example sentence, and it enters the same rotation.
  • Works offline. Cards and shields run with no network once installed; iCloud backup is opportunistic and uses your own private database.
  • No account. No signup, no email, no password — nothing between installing and the first encounter.

What one encounter
looks like

A few seconds, one word, one recall attempt — repeated on a schedule you never have to keep.

A single vocabulary flashcard replacing a blocked app, asking the learner to recall the translation Selecting which apps and categories should be shielded so each open triggers an encounter The eleven target languages whose vocabulary can enter the spaced-repetition rotation Progress screen tracking how many words have accumulated enough encounters to count as learned

Related reading

Repetition FAQ

How many times do you need to see a word before you remember it?
For a word met incidentally while reading, estimates cluster around 8–12 encounters, with individual studies reporting anywhere from 6 to more than 20. Nation and Wang (1999) found words were more likely to be learned after about ten encounters; Pigada and Schmitt (2006) saw a marked jump at roughly ten or more. There’s no single magic number — it depends on the word, the context, and what kind of encounter it is.
Does seeing a word count the same as being tested on it?
No. Roediger and Karpicke (2006) found that after two days or a week, learners who had been tested retained substantially more than learners who had restudied the same material the same number of times — even though re-reading looked better on an immediate test. A recall attempt is a stronger encounter, so fewer of them are needed.
How far apart should the repetitions be?
Spread out, with widening gaps. Cepeda, Pashler, Vul, Wixted and Rohrer (2006) synthesised 839 assessments from 317 experiments and found spaced study reliably beats massed study, with the best gap growing as the delay before the test grows. A workable schedule is same day → next day → three days → a week → two weeks → a month — which is what a Leitner system automates.
How many encounters can a normal day realistically give you?
More than most study plans do. Reviews.org’s 2026 survey of ~1,000 US adults found people check their phone 186 times a day, about 11.6 times an hour. If 15% of those pickups triggered one flashcard, that’s roughly 28 recall attempts a day with no study time set aside — enough to carry a 20-word rotation past ten encounters each in about a week.
Do I have to study to get the repetitions?
The repetitions have to happen; they don’t have to be scheduled by you. LearnScreen shields the apps you choose using Apple’s Screen Time API and shows a vocabulary card when you open one, so encounters accumulate inside phone use you were already doing. A Leitner schedule picks the word, so spacing is handled too.

Sources

  1. Uchihara, T., Webb, S., & Yanagisawa, A. (2019). The Effects of Repetition on Incidental Vocabulary Learning: A Meta-Analysis of Correlational Studies. Language Learning, 69(3), 559–599. Record
  2. 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. Record
  3. Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249–255. Record
  4. Ebbinghaus, H. (1885). Memory: A Contribution to Experimental Psychology. Full text
  5. Nation, P., & Wang, K. (1999). Graded readers and vocabulary. Reading in a Foreign Language, 12(2), 355–380.
  6. Pigada, M., & Schmitt, N. (2006). Vocabulary acquisition from extensive reading: A case study. Reading in a Foreign Language, 18(1), 1–28.
  7. Reviews.org (2026). Cell Phone Usage Stats: Americans Check Their Phones 186 Times a Day. Survey of ~1,000 US adults, fielded Q4 2025. Report

Last reviewed 27 July 2026.

Start collecting
encounter number one

No account. Works offline. The next time you reach for a shielded app, that’s one repetition down.

Download for iOS