On Learning And Memorizing


Joshua Mitchell / July 21, 2020

5 min read

Memorization gets a bad rap. People frequently ask, in the age of Google, why bother remembering something?

I propose that all learning is memorization. Here’s a thought experiment:

Pick a subject you don't know. Pretend you spent the next 5 minutes learning it.

Awesome - let's compare your brain before to your brain after. What's the physical difference?

There has to be one. Otherwise, how would you explain a measurable increase in knowledge?

The answer, although there is certainly more to it, is newly formed connections and myelination (i.e. strengthening of existing connections).

If I told you that fuego meant fire in Spanish (and you didn’t already know that) it would cause a certain set of connections to be made in your brain. Once those connections get made, you would know what fuego means.

Similarly, if I, instead of telling you what fuego means, pressed a magic button that caused those same connections to form in your brain, the end result would be the same: you would know what fuego means.

How does this translate to learning more complex ideas? Consider a measurement for how “deep” you learned something.

Memorizing “fuego = fire in Spanish” would be shallow.

Doing a PhD in Physics would be considered deep.

What’s the brain “before and after” for the PhD scenario?

The answer, if you factor out aging and other processes unrelated to learning, is the same: connections.

Anything you know (or could know) is a result of neural connections in your mind. Passing a quiz, defending a dissertation, or any other manifestation of learning is possible exclusively because of connections that you didn’t have by default.

Since these connections represent new information available to you, they are, for all intents and purposes, memory. In this way, forming those connections is equivalent to memorizing.

To be clear, I define memory functionally, rather than anatomically. I’m not familiar with which types of brain tissue are responsible for which types of memory. For simplification purposes, my definition of memory is any part of the brain that encodes information that can be retrieved at a later date. If it’s used for storing information, it’s memory.

It’s possible this doesn’t sit right with you. Maybe you have the following objections:

  • There’s a big difference between rote memorization and deeply understanding a concept. If everything is memorization, how do you explain that?

Rote memorization is a technique. It is the act of memorizing something by repeating it over and over. It has nothing to do with the amount nor complexity of what you are memorizing. You can rotely memorize something very complicated.

On the other hand, doing a PhD in Physics is non-rote memorization. Many questions are asked, many experiments are done, and many perspectives are examined. All of this results in a set of memories being formed - memories in the traditional sense (fun times at the lab) and in the informational sense (connections that enable him or her to solve Physics problems).

  • This sounds like a semantic argument. What society calls deeply learning something you seem to be relabeling as memorizing. What is there to gain here?

I am arguing both semantics and practicality. I feel that understanding something deeply is mythicized and romanticized (e.g. Deep Learning becoming a buzzword in the business ecosystem), which makes it resistant to measurement and rigor. A lack of a concrete standard means people more frequently underestimate (imposter syndrome) and overestimate (the Dunning-Kreuger effect) their expertise.

A second order effect of this lack of precision is that people rely more deeply on social signals to judge expertise, increasing the amount of political games people (even the true experts) must play.

Thinking of knowledge in terms of memorization instead of abstract individual expertise shifts the focus from the person to the knowledge itself. The narrative becomes, “I have this information in my brain,” instead of, “I am (not) an expert.” A craftsman mindset becomes more prevalent.

  • What are the implications for my learning process? How can I take advantage of this idea?

Anything we know is because we memorized it. What, how, and why we think is because of the information our brain has encoded. We are able to stand on the shoulders of giants because we memorized their best ideas. This allows us to reach even higher level concepts and become giants ourselves. Ideally, we make our ideas accessible to those who come after, and the cycle continues.

A quick recap:

  • If you think your idea is original but you haven’t learned the foundations of its domain of origin, it probably isn’t.
  • Original ideas will, almost by definition, come from unexpected combinations of existing connections.
  • The stronger and wider our baseline connections, the more surface area for us to have original ideas.
  • Rote memorization is okay - some ideas are worth drilling - but more intricate forms of learning are more effective (and more fun).

Thank you to the Compound Writing members who reviewed this post: Joel Christiansen, Tyler Wince, Stew Fortier, Dan Hunt, and Casey Rosengren.

Discuss on Twitter