Why a word list doesn't stick

What happens to a list you have read

A list of thirty words reads in two minutes and leaves a feeling of work done. A day later a few random lines remain — usually the first and the last. That is not a sign of poor memory: memorization works this way for everyone, and it has been measured in the lab dozens of times.

That feeling of work done is deceptive twice over. Rereading gives a sense of familiarity — the text is recognized, seems familiar, and the brain takes the recognition for knowledge. Testing shows the opposite: people who reread the material are more confident about what they know and answer worse than those who tried to recall.

The first thing that works: trying to recall

The difference between “looking at the card” and “first trying to recall, then looking” is the largest of anything ever tested in this field. And it has been tested a lot, mind you: almost two hundred experiments.

An important detail: a mistake does not spoil the result. A failed recall attempt works almost as well as a successful one — provided the correct answer is shown right away. That is exactly why an exercise owes you the answer instead of leaving you with “wrong”.

Second: come back, but not right away

The second technique works on its own and adds up with the first. The same time spread across days yields more than the same time in one sitting. It has been tested in the lab and in ordinary classrooms — the result is the same.

How far apart — that question has a measured answer. The optimal gap depends on how long you need to remember: about 20% of that term on a horizon of weeks and 5–10% on a horizon of a year. If you need the word in a month, repeat it in about a week; if in a year — in about a month.

Testing yourself a day or two later is more useful than on the same day. A lesson that ends by asking what it itself gave five minutes ago tests short-term memory — and teaches nothing.

What else helps, but less

• Compose the answer yourself rather than pick from ready options.
• Say it aloud rather than to yourself. The gain is small but free.
• Mix topics within one session instead of grinding a single one. But not everywhere: no difference was found on connected texts, and for individual words the checks are fewer than commonly believed.

Note the order of magnitude. The difference between recalling and rereading is bigger than all the other techniques combined. If you change one thing, change this.

The caveats without which the numbers lie

Just do not expect miracles. All these techniques give a noticeable, stable, but not magical gain — something like “twelve words out of twenty instead of nine”. None of them cancels the fact that the language will take hundreds of hours.

One more honest caveat: different researchers got quite different figures. That the technique works is beyond doubt. Exactly how much — far less is known.

What this is based on

Every number above comes from a study, linked next to it. The mark “checked little” means one study stands behind the conclusion — or that researchers still argue about it.

• Recalling beats re-reading. Covering the translation and trying to remember it does more than reading it ten more times.checked many times
  Adesope, Trevisan & Sundararajan, 2017, Review of Educational Research 87(3) · 188 experiments, 272 comparisons. g = 0.61; against plain re-reading 0.51, against passive review 0.93.
• It is better to test yourself the next day than at the end of the same session. That is why a word comes back to you tomorrow.checked many times
  Adesope, Trevisan & Sundararajan, 2017, Review of Educational Research 87(3) · g = 0.82 when tested a day or two later, against 0.56 on the same day.
• Studying a little every day gets more into your head than spending the same total time in one sitting.checked many times
  Donovan & Radosevich, 1999, Journal of Applied Psychology 84(5) · 63 studies compared, difference d = 0.46 — a sizeable figure for a learning effect.
• The longer you want to keep a word, the rarer its repetitions should be. To hold it for a week, review every day; to hold it for a year, about once a month.checked many times
  Cepeda et al., 2008, Psychological Science 19(11) · Over 1,350 participants, tested after delays of up to a year. The best gap is about 20% of the target interval on a horizon of weeks, and 5–10% on a horizon of a year.
• An answer you worked out yourself sticks better than one handed to you.checked many times
  Bertsch et al., 2007, Memory & Cognition 35(2) · 86 studies, 445 comparisons, d = 0.40.

Next

• How many words you need to knowHow many words to learn to understand a conversation, and how many to read without a dictionary. And why “I know almost all the words” does not yet mean “I understood everything”.
• Do streaks and rewards helpWhat is proven about gamification in learning, why rewards for participation undermine motivation, and what is known about streaks — an honest walkthrough with numbers and links.
• Is it too late to start as an adultWhat is known about the critical period: the data on 669 thousand people, the criticism of it, and what it means for an adult starting a language from scratch.
• How to break down an unfamiliar scriptA practical order for taking on an unfamiliar script: how many characters at a time, why to come back, why you must not start with words, and what to do about look-alike characters.