Flashcards Might not Be the Fundamental Unit of Learning
December 08, 2021
If you don’t have any knowledge of Incremental Writing by Piotr Wozniak and Evergreen Notes by Andy Matuschak, this article might not make too much sense.
I remember participating to a group call where Andy Matuschak had been invited to speak about his work and, more in general, about learning.
One of the most important questions he was looking into was: “What’s the fundamental unit of learning?”
To clarify, it’s like asking what’s the most basic functioning piece you should divide things into. The answer to this question is of utmost importance because it could radically change how we approach learning.
Right now, even if we don’t explicitly say it, we assume the fundament unit is a flashcard. That’s what med students and language learning students use to learn – and I refer to them because they represent the majority of ‘desperate for learning tricks’ seekers.
I have pondered a lot on whether or not flashcards are that effective. I have not written much on the blog because I thought my current answers were not enough. I still continued with my tutoring work, albeit reduced in terms of volume of students I could follow.
And the more I go forward, the more it seems like flashcards are a mockery of what learning should be.
They work for med students because they mimic the USMLE, therefore getting the impression of ‘learning’ instead of ‘training for the exam’. And don’t get me wrong, I think you should do anything to pass an exam and get out of med school once and for all. Few people hate this plight more than me; therefore, I’m not going to judge anyone using flashcards.
With language learning, if you ever tried learning Japanese or Chinese, you should know very well how hard it is to break the 200 words barrier. At a certain point, the churning rate becomes higher than the acquisition rate; unless you somehow stick with it long enough to cram the knowledge in your brain, you will end up forgetting most of what you tried to learn.
So, assuming that what I just said is true, flashcards kind of suck for general learning.
They can ace you a couple of exams, but that seems to be about it.
People praise the use of flashcards – and related software – mostly because of secondhand experience. I distinctly remember a quite popular YouTube video named ‘How I Would Study Subject X with Spaced Repetition Software Y’. The person in question already had a degree and was hypothesizing on how they would study now with such a software. But they never did. I’ll keep the names out in case anyone ever wants to offer me a cushy consulting job.
Do you realize how utterly asinine such a video is?
What direct or implicit experience do you think you have to tell people they should study in a way when you didn’t even try it on your skin?
It seems to me a malignant behavior given how high and how desperate people are to learn something. I don’t believe that such videos are made in good faith. If they are, those people shouldn’t have a degree, they should have a donkey hat to walk around and notify the people watching them of the risks of following their opinions.
I disappeared for almost a year from the public sphere of the SRS community not just because I thought my insights were getting repetitive and stale, but also because I started seeing something wrong with my approach.
I conducted a study together with Alexis, the creator of SMA, Raj, and a Daniel, a Canadian professor. We had people create a massive boatload of flashcards for one exam - Anatomy - trying to see whether or not it would be effective for them. In theory, it should have had a discreet measure of success. But the compliance, after the students formulated the cards, was extremely low.
Some of the most experienced people guided students in the creation – and sharing – of almost 20.000 flashcards which I went over one by one to make sure they would be decent.
Which begs the question:
Why wouldn’t people take advantage of that?
So, I started seeing flashcards not just as something ‘suboptimal’ but also something that people did not enjoy. If you ever had to slog to create thousands of medical flashcards for your own exam, you know what I’m talking about.
Why are we adopting flashcards as the most fundament unit of learning, then?
Well, the Spaced Repetition domain seems very taken with flashcards. Every software has flashcards at its core. Bigger companies just employ them because ‘well, look, those guys look cool and they are using them; I do want a salary bump, let’s propose this to the upper echelons!’
Few people ask the right question.
“Are we sure that flashcards are that good for learning?”
They satisfy the basic insecurities of people who want to remember it all, sure. I mean, they satisfy it on paper. Not really in the real world.
I know some of the greatest users in the SR domain and most of them are starting to think that flashcards are garbage.
So, what’s the fundamental unit of learning?
We drew circles around the question, but we didn’t answer.
One thing is for sure, I don’t like making wild speculations. I like to observe. I was mostly proven wrong throughout my life when something came out of my head based on some incoherent association I came up with. But I also pinned down some great answers here and there.
In my career as a writer, I’ve always told my editor that I have zero ‘creativity’ as people usually perceive it: the exceptional capability to form models out of thin air and have them work in the real world.
I’m terrible at that.
What I’m good at is looking at people who are smarter than me, ripping off their work and trying to see things they didn’t see through associative observation.
So, I’ll rip a few pages out of some experts in this field, repackage it, and call it my invention. That’s what we are going to do.
Without further ado…
I think that we already discovered the fundamental unit of learning!
From what I tested on myself an a few other people, it seems to me that the fundamental unit of learning is already among us. It’s just not used as such by its inventor. I mean, to be fair, it partially is. But it sounds like I’m far smarter if I say it’s not.
A variation of Evegreen Notes by Andy Matuschak is, in my opinion, the fundamental unit of learning. A one note = one concept unit. Basically, what
It came to me when I was trying out that approach to write a business plan and a world-building for a book. I tried to ‘version’ the notes I was writing. I would have the first version, then the second version, I would start commenting – literally typing down the comments in the text – what I thought I got right and wrong from one version to another, observing the evolution of it.
This is probably one of the most remarkable feats I would ascribe to my moronic mind.
I’m not going to go deep into this approach because I’ll write later on about it and because this post might already be too confusing.
To make a long story extremely short, I started noting down with comments, highlights, and so on what was wrong in the first version. Then I would collapse the first version and, the next day or later in the afternoon, I would write down a second version. The forgetting happening in between helped me generalize some concepts, trim others, and, most importantly, see my Thinking Trail.
Oh yeah, baby.
I could see my thoughts laid out in front of me, walk back, look at the errors and try to understand why my brain made such connections. Then, I would note it down and read my thoughts about my previous thoughts/knowledge.
Very meta, I know.
There is so much to say about this and how I came to this conclusion. But I used a variation of this approach for Chinese Flashcards – which I had to abandon for a matter of time – and it worked.
So, I put together some Incremental Writing/Problem Solving and the Evergreen Notes structure – which most important feature is the one note to one concept. Then, I added this version-comparison, which gave birth to a Thinking Trail. The Thinking Trail made it possible for my brain to build all the associations and links I’ve been trying to build for the last year and a half through flashcards.
The approach is so minimal and simple that I could lay it down in three pages. I will not because I’m still testing it in the field and teaching it to some people close to me. I resumed studying to finish my degree and I’m using this approach only. I may create a few flashcards here and there, but I don’t see myself creating more than a hundred flashcards per exam, compared to the several thousand I needed before.
As soon as my observations demonstrate a sensible (great) pay-off for me and other people, I will make a video/course/book about it.
Until then, if you want to learn from me, I resumed my consulting/tutoring for about 25/30 Euro an hour. Price may vary based on my availability and how many people are texting me. You can book two-three people meetings with a custom price. Just make sure your internet connection works properly.
If you want to learn how to do this or you just want tips with your SuperMemo journey, shoot me an email at firstname.lastname@example.org
I’m using Roam Research for this, but I had started using this new approach through SuperMemo topics because I loved the scheduling. Sadly, I need something flexible I can use online and I’d rather do without the scheduling if it means I can move around without having to go through backup setups and so on.
You could use RemNote. You could use Notion.
It doesn’t matter that much, honestly.
The approach could be conducted on paper too, although that would be a bit cumbersome.
It’s designed with compliance, elegance and speed in mind.
If you are confused about what I explained, it’s normal.
The idea is there, but now it’s time to refine it and see how applicable it can be on a larger scale.
A blog about writing and learning.