How to teach yourself anything.
Teaching yourself works when you do three things most people skip: decide why you're learning, find out what you don't know, and test yourself instead of re-reading. Every claim here links to its research.
The method in five steps
- Start with whyName the real situations you're learning for. Why, what, how
- Map the field before you dive inLay out the field from a source outside your head, then mark what you know. Learning from first principles
- Test yourself instead of re-readingRetrieval practice
- Space it outSpaced repetition
- Turn it into workPaul Graham's four steps
The techniques with the best evidence
Retrieval practice
Pulling information out of memory strengthens it more than reading it again. The evidence, and how to do it.
Read the guideSpaced repetition
Spreading practice out beats cramming, and the best gap grows with how long you want to remember. The evidence and how to use it.
Read the guideInterleaving
Mixing different kinds of problems feels harder but improves learning. The evidence and how to use it.
Read the guideSelf-explanation and the Feynman technique
Explaining why each step is true, in your own words, exposes gaps. The evidence behind self-explanation and how the popular Feynman technique uses it.
Read the guideDeliberate practice
What deliberate practice is, what the research does and doesn't show, and how to apply it when teaching yourself.
Read the guideFrameworks for self-directed learning
Ultralearning
Scott Young's nine principles of ultralearning, in one line each, and how to apply them.
Read the guideWhy, what, how
Start every learning project with why you're learning, what the field contains, and how you'll learn it.
Read the guideLearning a field from first principles
Break a field into the problems it exists to solve, map it from an outside source, and find the gaps you couldn't have named.
Read the guideFrom learning to great work
Paul Graham's recipe for great work, and what it implies for how to learn.
Read the guideDrop one myth: learning styles.
Matching lessons to a "visual" or "auditory" style doesn't improve learning. Here's the evidence, and what to measure instead.