Why spaced repetition works
You forget most new words within days. Spaced repetition is the one study method built around that fact — here is what it does, why it works, and how Mullawo uses it.
You learn twenty words on Monday. On Friday you remember four. That is not a failing memory — it is how memory works, and it was measured more than a century ago.
The forgetting curve
In the 1880s Hermann Ebbinghaus taught himself lists of nonsense syllables and tested how many he could recall after an hour, a day, a week. The result is the forgetting curve: recall drops steeply in the first day, then flattens. Most of what you will forget, you forget almost at once.
The same experiments found the way out. Every time you recall something just before you would have forgotten it, the curve resets — and the next drop is slower. Review once and a word lasts a day. Review again at the right moment and it lasts a week. Then a month. This is the spacing effect, and it is one of the most replicated findings in the psychology of learning.
What "spaced" actually means
Spaced repetition is not "review often". It is "review at the moment the word is about to slip". Too early is wasted effort: the word was still there. Too late and you are learning it from scratch again.
The trick is that the right moment is different for every word and every person. Water in Spanish you will keep after two reviews. An irregular Georgian verb might need eight. A system that reviews everything on the same schedule spends your time on the easy words and loses the hard ones.
So a spaced-repetition app keeps a schedule per word. You see a card, you say whether you knew it, and that answer moves the word's next review:
- Knew it — the interval grows. A day becomes three, three become a week, a week becomes a month.
- Almost — the interval grows a little, or holds.
- Didn't know — the word comes back tomorrow, and its intervals start short again.
Mullawo uses the SM-2 algorithm for this — the same family of scheduling that Anki and SuperMemo use. Each word carries its own ease, a number that rises when you keep answering right and falls when you stumble, so the words you find hard stay in front of you more often without you having to manage anything.
Why it beats rereading
Rereading a word list feels productive. It is mostly not. Recognising a word on a page is easy and tells you little; what fixes a word in memory is retrieving it — seeing the prompt and producing the answer yourself. Every card in Mullawo is a retrieval: you see the word in a language you know, you try to produce it in the one you are learning, and only then you flip.
Two things make this stronger than it sounds:
- Effort is the point. The harder a retrieval, the more it strengthens the memory — as long as you succeed. A word that is about to slip and gets pulled back is worth more than a word you never had to think about.
- Honesty is free. Nobody is grading you. Tapping Didn't know is not a failure; it is the information the schedule needs.
Five minutes, every day
Spacing works in days and weeks, so it needs days and weeks. Twenty minutes on Sunday does less than four minutes on each of five days, because only the daily version puts reviews where the curve says they belong.
That is why Mullawo asks for a small daily session rather than a long one. Twenty cards, a streak counter, a reminder at the hour you choose. The point of the streak is not the number — it is that the reviews happen on the days the schedule set them.
Several languages at once
There is one more thing spacing lets you do. Because every word has its own schedule, adding a language does not double the work — it adds cards that live on their own timelines. In Mullawo one card carries the word in every language you are learning: see water, flip, and get Wasser, agua and eau together. The languages you already know sit on the front as the prompt.
Learners tell us the languages reinforce each other: a word you know in Spanish is a hook for the same word in Portuguese. The schedule does not care — it just moves each one to the day it is needed.
Where to start
Pick one language you know and one you want. Do the first twenty cards. Come back tomorrow, when the app will show you exactly the ones that are about to slip — and none of the ones that are not.
That is the whole method. The rest is showing up.