Spaced repetition is more than memorization – it’s also generalization.

by Justin Skycak on

And if you want to get the most out of your review, you need to engage in spaced, interleaved retrieval practice.

Spaced review is not only about memorization – it’s also about generalization.

Just like when you reread a book or rewatch a movie that you haven’t seen in a while – you see things that you didn’t notice before. You come in with a different mental state, and you come out with some new perspectives.

Now, this might sound like speculation… but guess what?

This “spacing improves generalization” hypothesis is empirically supported.

For instance, a review by Smith & Scarf (2017) recounted multiple studies demonstrating that “spacing not only benefits the learning and retention of specific items but improves the generalization of learning.”

And while spaced repetition is often mischaracterized as “just memorizing flashcards,” it seems that when comparing spaced to massed practice, the greatest benefit occurs for complex generalizations!

image

(another quote from Smith & Scarf, 2017)

So however much you think that spaced practice is improving your directly practiced skills as compared to massed practice… it’s improving your complex generalizations even more.

This makes spaced review a superpowered learning technique in hierarchical knowledge structures.

It not only

  1. keeps your existing knowledge fresh, but also
  2. helps you generalize that knowledge --

and both of these things help you seamlessly transition into progressively more advanced skills.

∗     ∗     ∗

And you know what else helps you promote generalization? Interleaving, also known as mixed/varied practice.

The idea behind interleaving is that, after you’ve learned a skill to the point that you can execute it accurately and consistently, you’ll maximize your learning if you start “mixing it up” with other skills.

This will make the skill feel more challenging, but in a good way – you’re challenging yourself to pull the information out of your brain completely from scratch, without already having it primed at the front of your mind (which would make the task artificially easy).

Interleaving has been shown to enhance two specific types of learning that are necessary components of generalization (see Rohrer, 2012 for a review).

  • Discrimination Learning: matching problems with the appropriate solution techniques. (For instance, the equations $x^2+3x+2=0$ and $x+3x+2=0$ look similar but require wildly different solution techniques.)
  • Category Induction Learning: recognizing general features that distinguish problems requiring different solution techniques.

And the great part about interleaving and spaced practice is that you don’t have to choose one or the other.

You can do both simultaneously – spaced, interleaved review. They fit together naturally.

How? Just do minimal effective doses of spaced practice, and mix up a large number of those minimal effective doses within any given practice session.

∗     ∗     ∗

Now, there’s one thing to watch out for that many people get wrong, that prevents them from capitalizing on the full benefit of spaced, interleaved review.

During your review, you need to try your hardest to retrieve information from memory. You need to engage in retrieval practice.

To transfer information into long-term memory, you need to practice retrieving it without assistance.

It’s the very act of retrieving information from memory that transfers the information to long-term memory.

Each time you successfully recall a fuzzy memory, it stays intact longer before getting fuzzy again.

But the crucial insight is that when you load the information into working memory for use, the specific way you need to load it is by pulling it from long-term memory.

If you load information into working memory by looking at reference material instead of pulling from long-term memory, then you’re not strengthening your retention.

It’s like you’re going to the gym to lift weights, but you’re just going through the motions and letting your spotter lift the weight for you. No strength is being developed.

The only time when the spotter should help you lift the weight is when you can’t lift it despite trying your hardest.

And even then, the spotter should only give you just enough assistance to get you over the edge of lifting the weight. The spotter should be doing as little as possible while ensuring that you manage to eek out a successful rep.

In the same way, the only time you should look at reference material during review is if you can’t recall something after trying your hardest.

And while it’s okay to check reference material as a last resort, you should only peek once, and then try to solve the problem without looking again.

∗     ∗     ∗

I’ve forgotten why I started writing this post, so I guess this is a good place to stop.

Looking back through what I’ve written, here’s the TLDR:

If you want to get the most out of your review, you need to engage in spaced, interleaved retrieval practice.