04 June 2019
Psychology has been studying how our brains learn for decades. Even though there have been breakthroughs, they haven't reached mainstream. That's why most children in school or even university still use inefficient methods. As a result, they feel discouraged by how easy their memories fade away. That's why I want to do an overview in this article of techniques that work. Hopefully more people discover them and put them to good use.
Plenty of people have written about the benefits of using Anki. Anki uses the 3 principles of learning that have the most scientific literature behind them: the testing effect, the interleaving effect, and the spacing effect.
The testing effect says that you will remember better a piece of information if you try to recall it. Even if you fail to remember it and lookup the answer. The effort of trying to remember makes the memory stick better.
The interleaving effect says it's better to mix topics. Don't only do derivatives problems. Mix them with linear algebra or with integration problems. With Anki you will even mix completely different topics, like biology and programming, or whatever interests you.
The spacing effect says that you should avoid re-reading an item of information. Instead, you test yourself at ever increasing intervals. Anki has an algorithm that knows when's the best time to ask you again about something.
This allows you to see the connections between topics. You can visualise the different parts of a system and how they relate. This helps you with semantic encoding, which I explain lower.
Creating a mind map also helps with chunking. When you chunk, you learn all the parts of the whole. Then you use the name of the whole, of the system, to retrieve all the information about the subparts. It's like a library index that helps you retrieve more information (the book).
For example, let's say you're learning biology. You have to memorise all the parts of the eye. Once you have that down, by thinking of the eye, all its components will come to mind.
This comes in connection with using Anki. It's better to make a card with one item of information. For this reason, you should avoid creating a card with a long list of items. To encode a list, you can put each item on its own card. Then, for each element other than the first, you can say "After element X, what does Y do?" or something to that effect.
You've chosen your fact. Now you want to connect it to its surroundings. Every fact refers to a part of a system. Its "surroundings" represent the other parts of the system. Your goal is to highlight these relationships.
By doing this, you'll have more paths that can lead you to the fact, when you want to recall it.
This entails connecting the new piece of information with things you already know. One funny example from personal experience involved encoding the Polish word for "mushroom". It's "grzyb". The "gr" part made me think of Greece. I couldn't associate "zyb" with anything, but I found online that it's the name of a Danish telecommunications company. So I pictured a mushroom dressed in the Greek flag talking on the phone in Danish. I even searched for videos of people speaking in Danish on Youtube.
You can phrase the item of information as a question and the thing you want to remember on the other side. For example, you phrase the definition as a question. An example from one of my cards about linear algebra:
Question: "What's the first axiom of a vector space?"
Answer: "The associativity of addition: (u + v) + w = u + (v + w)"
Another type of card is a cloze card. With this one, you blank the word that you want to remember. For example:
"In probability, ___ is the probability of the data given that the hypothesis has the occurred."
The answer is "the likelihood".
From now on, you let Anki do everything. It will ask you the questions when it "thinks" it's the right time. All you'll have to do is remember the answer. And you will, even if not from the first try.