There is an extremely pervasive, but not so frequently discussed phenomenon of forgetting. Obviously, people recognize that they can’t remember information, even information they view as important like education. But they also frequently overvalue learning, especially education, believing that our society is only able to run because we spend a large portion of our lives in school. I think that the people who take forgetting seriously enough, are the ones using spaced repetition.
Although people realize that they forget what they learn, people are often hesitant to question the value of education. Students frequently ask “when am I ever going to use this?”, but this is regarded as a myopic viewpoint by adults. They know that education is important for being a productive member of society, good citizen, and enlightened individual.
I don’t want to say that you shouldn’t learn things. My point here is going to be that if it is worth learning once, it is worth retaining the information. Our mind is like a bunch of buckets with holes in them. You fill the buckets up with water, and they start leaking immediately. If you don’t add the water back to the bucket, you’ll eventually find it empty. Similarly, you will forget the majority of what you learn alarmingly quickly.
There are some exceptions of course. Some facts are so salient for whatever reason, that you will never forget them even without repetition. For example, you might never forget the time you went skydiving because it was so intense and interesting. In fact, you may naturally recall the moment reinforcing it without the need for external reinforcement.
Despite the fact that forgetting obviously occurs, it is worth examining more in depth. A German psychologist and pioneer in the study of memory, Hermann Ebbinghaus, did just this when he ran an experiment in which he studied nonsensical syllables and seeing how long it took him to forget them. He found the formula in which, t was time in minutes, b was memorization percentage.
A later—possibly less empirically verified—proposal from Supermemo creator Piotr Wozniak declined exponentially, in which R is retrievability, S is stability and t is time.
These curves are non-linear. The rate of decay of knowledge is quick initially and slower as time progresses. In reality and Wozniak’s example, every reinforcement makes the memories more stable. The shape of the curves should look something like this:
If you find it worthwhile to learn something 100%, then you should use spaced repetition to retain it. It is a rare case where you need to know something 100% and then not know it at all a few weeks later. Unfortunately, that is the situation for a schooling. Many students cram because cramming is actually effective at doing what the student wants, namely getting the grade.
Since decay of knowledge is so rapid, a lot of learning isn’t particularly worthwhile unless it is inherently enjoyable to learn. If you read a book, it is likely that you will retain very little information, perhaps only the core thesis. I think that often this gives someone a strong conviction about a belief, but without the memory knowledge to justify the belief. I call this free-floating convictions.
Ironically, I recall having a discussion with a friend about the book The Case Against Education, in which I felt as though I was unconvincing because I had forgotten the justifying details, but I recalled being thoroughly persuaded by the book. If I look back at the book, I can see numerous citations supporting Caplan’s argument, and yet when I am conversing with my friend, it is impossible to recall everything.
I think that it may be the case that reading non-fiction books is largely overrated if you do not use some sort of mechanism to retain the information—like spaced repetition—or make the summarized information easily accessible. Knowledge is always fleeting. I can’t help people for not taking an interest in learning because I think that in most cases it truly does not matter.
People tend to know information that is critical to their day-to-day life and become largely uninterested in actually knowing information that would help them to understand history, science, philosophy or politics better. If that information is consumed, it is usually for fun. I think that is okay. People do not need to develop domain specific knowledge and opinions in irrelevant topics. In fact, it is often epistemically more responsible to avoid doing so.
We have a culture which values learning for learnings sake, and that is likely good in some ways but harmful in other ways. It is harmful in a lot of ways because people do not apply cost-benefit analysis to their time spent learning information. For school, it is worthwhile as a means to an end, namely to successfully complete classes, but we should not regard the information as being useful far into a person’s life. That knowledge will decay exponentially and be lost anyway.
The good thing about using spaced repetition software (SRS) like Anki is that you become intimately familiar with the forgetting process and get more of a sense of the worth of remembering certain things. I recall spending a great deal of time memorizing a bunch of facts about the nations of the world, but I abandoned it. It was fun for a while, but I really don’t need to know the capital of Serbia or the location of Brunei. If I wanted to know this, I can simply search for the information online.
Gwern wrote an interesting article about spaced repetition in which he quantified the amount of time spent on a single flashcard to determine what is worthwhile to learn. His conclusion, which we could call something like Gwern’s rule of 5 was “don’t use spaced repetition if you need it sooner than 5 days or it’s worth less than 5 minutes”.
The most difficult task, beyond that of just persisting until the benefits become clear, is deciding what’s valuable enough to add in. In a 3 year period, one can expect to spend “30–40 seconds” on any given item. The long run theoretical predictions are a little hairier. Given a single item, the formula for daily time spent on it is Time = 1⁄500 × nthYear−1.5 + 1⁄30000. During our 20th year, we would spend t = 1⁄500 × 20−1.5 + 1⁄3000, or
3.557e-4
minutes a day. This is the average daily time, so to recover the annual time spent, we simply multiply by 365. Suppose we were interested in how much time a flashcard would cost us over 20 years. The average daily time changes every year (the graph looks like an exponential decay, remember), so we have to run the formula for each year and sum them all; in Haskell:sum $ map (\year -> ((1/500 * year**(-(1.5))) + 1/30000) * 365.25) [1..20] # 1.8291
Which evaluates to 1.8 minutes. (This may seem too small, but one doesn’t spend much time in the first year and the time drops off quickly 55.) Anki user muflax’s statistics put his per-card time at 71s, for example. But maybe Piotr Woźniak was being optimistic or we’re bad at writing flashcards, so we’ll double it to 5 minutes. That’s our key rule of thumb that lets us decide what to learn and what to forget: if, over your lifetime, you will spend more than 5 minutes looking something up or will lose more than 5 minutes as a result of not knowing something, then it’s worthwhile to memorize it with spaced repetition. 5 minutes is the line that divides trivia from useful data. 56 (There might seem to be thousands of flashcards that meet the 5 minute rule. That’s fine. Spaced repetition can accommodate dozens of thousands of cards. See the next section.)
To a lesser extent, one might wonder when one is in a hurry, should one learn something with spaced repetition and with massed? How far away should the tests or deadlines be before abandoning spaced repetition? It’s hard to compare since one would need a specific regimens to compare for the crossover point, but for massed repetition, the average time after memorization at which one has a 50% chance of remembering the memorized item seems to be 3-5 days. 57 Since there would be 2 or 3 repetitions in that period, presumably one would do better than 50% in recalling an item. 5 minutes and 5 days seems like a memorable enough rule of thumb: ‘don’t use spaced repetition if you need it sooner than 5 days or it’s worth less than 5 minutes’.
There are likely some other benefits such as the possibility of having an idea in your mind to connect to another idea which is something that is not valued in the time consideration. That is particularly beneficial for a creative thinker or writer. However, very few people have aspirations to be creative non-fiction writers.
There are certain situations in which learning is very valuable and worthwhile. Essential skills that are necessary for work and life. These skills are typically reinforced so frequently that spaced repetition is not necessary. In cases in which that is not true, it is worthwhile to use spaced repetition.
If you’re serious about learning something and you believe it is worthwhile, then it is hard to not justify using technology like Anki. If you intend to cover the content at a later point in time, why not use an algorithm to review the content at the most efficient time?