Spaced repetition means reviewing a word again just as you begin to forget it, with gaps that grow after each successful review. It works because of the spacing effect: across hundreds of experiments, reviews spread over time beat the same reviews crammed together. Flashcards add retrieval practice, and recalling a word strengthens memory more than rereading it.
What is the forgetting curve?
In the 1880s Hermann Ebbinghaus memorised lists of nonsense syllables and measured how much time he saved when relearning them after a delay. Plotted against the delay, the savings fall steeply at first and then level off: the forgetting curve. In 2015 Murre and Dros repeated the experiment, again with a single subject, at intervals from 20 minutes to 31 days, and concluded that the curve had indeed been replicated. They also saw a possible upturn at the one-day point, which may or may not be linked to sleep and which they say still needs testing.
The same pattern shows up with foreign-language words picked up from reading. In Waring and Takaki's 2003 study, Japanese learners of English read a graded reader containing 25 target words whose spellings had been changed so that nobody could already know them. Straight after reading they could translate 4.6 of them on average. Three months later the average was 0.9. The authors note that this mirrors memory research: most forgetting happens soon after learning.
How far apart should reviews be?
The spacing effect is one of the best-replicated findings in the psychology of memory. Some of the key studies:
| Study | What it found |
|---|---|
| Bahrick and colleagues (1993) | Four people learned 300 English–foreign-language word pairs and relearned them over nine years. Thirteen review sessions 56 days apart gave retention comparable to 26 sessions 14 days apart. Longer gaps slowed learning slightly but paid off in much better retention. |
| Cepeda and colleagues (2006) | A meta-analysis of 839 assessments of spaced practice, from 317 experiments in 184 articles. The best gap between study sessions grew as the time until the final test grew. |
| Cepeda and colleagues (2008) | Over 1,350 people learned facts, reviewed them after a gap of up to 3.5 months and were tested up to a year later. The best gap was about 20–40% of the test delay for a test one week away, falling to 5–10% for a test a year away. |
| Kornell (2009) | Studying one large stack of flashcards, which spaces each card out, beat studying four small stacks. Spacing won for 90% of participants, yet after the first session 72% believed the opposite. |
Two practical lessons follow. The gap should grow with how long you want to remember something: to keep a word for a year, Cepeda's figures point to gaps of roughly 18 to 37 days, which is 5–10% of 365. And spacing feels less effective than cramming while you are doing it, which is why people keep cramming.
Why does testing yourself beat rereading?
A flashcard makes you retrieve the meaning before you see it. Karpicke and Roediger's 2008 study in Science tested this with foreign-language vocabulary. Once a student could produce a word correctly, they either kept studying it, kept being tested on it, or dropped it. Extra studying had no effect on delayed recall; extra testing had a large positive effect.
The study also found that students' predictions of how much they would remember bore no relation to how much they actually remembered. Put that together with Kornell's result and you have the case for letting a schedule, not your sense of how well you know a word, decide when it comes back.
Why pair flashcards with reading?
Reading brings common words back constantly, and a story recycles its own key words from chapter to chapter, but the rarer ones may never return. In the free chapters of every Spanish series on Contos, about half the different word forms appear only once:
| Level | Different word forms in chapters 1–2 | Forms that appear only once | Share |
|---|---|---|---|
| A1 | 828 | 399 | 48% |
| A2 | 1,311 | 692 | 53% |
| B1 | 1,773 | 1,050 | 59% |
| B2 | 2,308 | 1,403 | 61% |
Waring and Takaki found that learners couldn't give the meaning of any word they had met fewer than eight times when tested three months later. A word that turns up once in your reading needs another way back, and that is the flashcard's job.
In Contos you tap a word to see its meaning and can save it as a card, along with the sentence you met it in. Cards then come back on a fixed ladder of intervals: 1, 3, 7, 16, 35 and 70 days. "Again" brings a card back in five minutes and starts it from the bottom, and "Easy" skips a step. Once a card reaches the 16-day step, the word counts towards the known words that move you up a level. The 16- and 35-day steps fall either side of the gap Cepeda's figures suggest for remembering something a year on.
Read a chapter, save the three words you had to tap most, and review them tomorrow.
Read free in your browserFrequently asked questions
What is the spacing effect?
The finding that information is remembered better when study is spread out over time than when the same amount of study is massed together. Kornell notes it has been shown in hundreds of experiments, and Cepeda's 2006 meta-analysis drew on 839 assessments of it.
What happens if I miss a few days of reviews?
Your due cards wait for you, and a slightly longer gap isn't a disaster. In Bahrick's nine-year study, longer gaps between sessions slowed learning a little but led to substantially better retention. Clear the backlog gradually rather than skipping it.
Why does cramming feel more effective?
Because it makes recall easy in the moment. In Kornell's flashcard experiments, spacing beat massing for 90% of participants, yet after the first session 72% thought massing had worked better. Easy recall today is a poor guide to what you'll remember next month.
Should I use flashcards instead of reading?
No, use them together. Reading gives you words in context and meets common words again and again. Flashcards catch the rarer words that reading alone won't bring back in time. Waring and Takaki suggest graded reading is especially good for strengthening words you already partly know.
Does Contos use an adaptive algorithm?
No. Contos uses a fixed ladder of intervals, 1, 3, 7, 16, 35 and 70 days, moved by your Again, Good and Easy answers. A card reaching the 16-day step counts the word as known for levelling up.