Learning design
Games as retrieval practice: what the format buys you and what it does not
A vocabulary game is retrieval practice with better ergonomics. Retrieval practice — being asked to produce an answer rather than review one — is among the best-supported findings in learning research, and a game is simply a pleasant wrapper that gets you to do more of it. That is a genuine advantage. It is also a narrow one, and this post is equally about the four things the format cannot do.
What a vocabulary game actually is
Strip the animation off a MemDecks game and what remains is a sequence of retrieval attempts with immediate feedback. WordBase Quiz asks you to recognise. Word Type asks you to produce a spelling. Word Mahjong asks you to match. Alphabet Quest asks you to decode a script. Each is a prompt, a required answer, and a correction.
That is exactly the shape the learning literature says works, and it is the same shape as a flashcard. The game adds no new learning mechanism. What it changes is how many repetitions you are willing to do before stopping, and that is not nothing — a technique's real-world value is its effect size multiplied by how much of it you actually do.
Four things the game format genuinely adds
- Volume. People do more repetitions in a form they enjoy. The best-designed review schedule you abandon in week two beats nothing, but it loses to a worse method you keep using.
- Varied retrieval routes. Recognising a word, spelling it, and matching it under time pressure are different cognitive tasks. Practising a word several ways builds more routes back to it than practising it one way many times.
- Time pressure, where appropriate. Fluency is retrieval that is fast, not merely correct. A timed game trains the speed that an untimed flashcard never touches.
- A social frame. Three MemDecks games are playable against another person. Practising because someone is waiting for your turn is a different and often stronger motivation than practising because a queue said so.
Four things games cannot do
Stated plainly, because the games-fix-everything claim is common and false.
- They cannot make recognition into production. A multiple-choice game trains picking the right option out of four. That is a real skill and it is not the skill of saying the word when nobody offers you options. If a game gives you the answer to choose from, it is not testing what you think it is.
- They cannot teach grammar. No MemDecks game teaches a rule. Vocabulary games teach vocabulary.
- They cannot fix timing. A game does not decide when you should meet a word again. MemDecks does not schedule at all — see how a card comes back.
- They can hide effort as fun. The most dangerous failure mode of the format: a game that is enjoyable and cognitively cheap feels like progress and produces very little. Difficulty is the active ingredient, and a game that removes it removes the benefit.
The design line: keep the difficulty, move the friction
The principle a vocabulary game has to hold is that the difficulty of retrieval is not the friction to remove. What can be removed is everything around it: making cards by hand, finding a picture, deciding what to study today, waiting for something to load, keeping score.
MemDecks tries to hold that line in specific ways. Games are built from your own collection rather than a canned word list, so their difficulty tracks your vocabulary. Hints cost score rather than being free. And card progress moves per game answer rather than per session, so the number reflects retrievals rather than attendance.
How to actually use them
| If you want to train… | Play | Because |
|---|---|---|
| Recognition, fast | WordBase Quiz | Four options, quick rounds, immediate correction. |
| Production and spelling | Word Type | You build the word letter by letter, no options offered. |
| Reading a new script | Alphabet Quest | Decoding inside sentences you already understand. |
| Speed under pressure | Canon Potato or Words Tetris | The clock is the point. |
| New words rather than old ones | Phrase Builder | It is the only mode that teaches unfamiliar vocabulary. |
The practical rule is one sentence long: if a game has stopped being hard, it has stopped being useful, and you should switch to one that asks for more.