Spaced repetition

Once you have learned something you don’t want to forget it. In order to secure the knowledge in your mind, you need to review it. But not too often, for that is a waste of energy, and not too seldom, for then you will have forgotten some of the information already.

You will want to space your repetitions properly to ensure that information is retained over the long term. From a look at the forgetting curve below, you will understand the need for frequent repetitions early on as soon as you have learnt something and it also becomes clear that over time you need to review your material less often.

But what if you want to find out about the optimal intervals for doing your revisions? The easy way out is to use a software application that has already made all the calculations for you and simply in due time delivers the piece of information scheduled to be reviewed. You never have to know why and when, just review it whenever it pops up in your software app, one of the most popular free apps being Anki.

If you for any reason want to handle your spaced repetitions manually there are ways to do that. After all, that is what people had to do before the digitalization of culture. And getting the intervals right is important. If you review too seldom you will forget and maybe have to relearn it all. If you review too often, it is a waste of time.

Gwern offers an informative article on spaced repetition frequency, and the following graphs are from this website, with all due thanks.

Ebbinghaus’ original forgetting curve

Wozniak’s Supermemo take on dealing with the forgetting curve using a spaced repetition schedule.

In the Leitner system, which was devised in the 70s for doing spaced repetition efficiently, you have several boxes with cards. The boxes are reviewed at different intervals. Let’s say that the first box is reviewed once a day, the second every other day, the third box every fifth day and the fourth box once every fortnight. If you know the content of the card, you promote it to the next box. If you do not know the card, you degrade it to the box below. In this way, you are making certain that you are always reviewing only the cards that are the most difficult and that need reviewing.

In the Leitner system, correctly-answered cards are advanced to the next, less frequent box, while incorrectly-answered cards return to the first box.

(This illustration is taken from, with due thanks)

Maybe you want to look behind the scene, so to speak, and see how an actual spaced repetition algorithm looks like. To apply it yourself would take a lot of work, but if you are interested, you may read this article:, taken from P.A.Wozniak, Optimization of learning, Master’s Thesis, University of Technology in Poznan, 1990 and adapted for publishing as an independent article on the web. (P.A.Wozniak, May 10, 1998).

And if you are really, really interested, you may want to have a look at the following article: