3.7 — How to Study Anything: The Full Method

Everything in Part 3, assembled into a single sequence you can run on a certification, a language, a technical subject, an instrument, or this book.

It is written as seven phases. They are not rigid — you will loop back constantly — but the order matters, and the most common cause of a stalled effort is doing phase four before phase one.

Phase 1 — Decide what "knowing it" means

Before anything else, write down what you will be able to do when you are finished. Not "understand statistics" but "read a paper's results section and say whether the conclusion is supported."

Why this is first. The target determines the method. Being able to recognise something requires flashcards. Being able to produce it requires production practice. Being able to decide something requires practice deciding under the same ambiguity you will actually face. Most wasted study is aimed at a target the person never specified, and so defaults to reading, which serves almost no target well.

Three questions to answer in writing:

What will I actually do with this? Pass a test, hold a conversation, build something, teach it.

How will I know I have it? Name the concrete demonstration.

How long do I need to keep it? A month, a career, forever. This sets your spacing schedule (3.2), and it is a genuine decision — some material is legitimately disposable.

Phase 2 — Get the map before the territory

Skim the entire subject badly, fast, before studying any of it properly. A contents page, an overview article, a two-hour introductory video at speed, the chapter summaries. Two to four hours for a large subject.

You will understand very little and that is expected. What you are building is the framework new material will attach to (3.1), and without it every fact arrives with nothing to connect to.

Then draw the map from memory. One page: the main areas and how they relate. It will be wrong. Keep it, and correct it as you go — the corrections are some of the most valuable learning you will do, because each one is a place where your model was wrong and you found out.

Phase 3 — Find the load-bearing 20 per cent

Every subject has a small core that everything else depends on, and a long tail that can be looked up.

How to find it. Look at what the introductory chapters spend most time on. Look at what appears in every chapter regardless of topic. Ask anyone competent what they use daily. In an exam-driven subject, look at the distribution of past questions.

Study the core to the point of fluency and the tail to the point of recognition. The commonest failure in self-directed learning is uniform effort across material of wildly unequal importance, usually because the material was studied in the order the book presented it.

Phase 4 — Build understanding before memory

Do not start memorising things you do not understand. Retrieval practice on a meaningless string produces a durable memory of a meaningless string (3.3).

For each core concept:

Get an explanation, then produce your own. Read or watch, then close it and write it in your own words as though for a specific person who does not know the subject.

Ask why until it bottoms out (3.4). Three levels is usually enough to hit something you already accept.

Find two examples from different areas, one from your own life or work.

Draw it. The relationship, the flow, the structure — from memory, on paper.

And find the boundary. What would break this? Where does it not apply? Understanding a concept means knowing its limits, and this question is what separates knowing a rule from being able to use it.

The test for this phase is the explanation test: can you explain it, out loud, to someone unfamiliar, without notes, and answer one follow-up question? If you stall, you have found exactly what you do not understand — which is the point of the exercise, not a failure of it.

Phase 5 — Practise retrieval, spaced and mixed

Now the core is understood, make it permanent.

Daily, or every session: finish by closing everything and writing what you remember. Three minutes.

Facts onto cards — your own, one item each, added slowly (3.2). Only what you genuinely need instantly available.

Problems mixed, never blocked (3.4). Shuffle types. If the book is organised in blocks, work from a mixed set you build yourself.

Reviews scheduled at day one, day three, day ten, in the calendar, dated.

And test before you feel ready. The failed attempt is doing work.

Phase 6 — Use it on something real

The gap between studied and usable is closed only by use, and this phase is where most self-study quietly stops.

Find the smallest real application and do it badly. Speak the language to one person for five minutes. Build the tiny broken version. Do the analysis on your own data. Explain the topic to a colleague who will ask questions.

Two things happen. You discover what you actually cannot do, which is never what you expected. And the material acquires the retrieval cues of a real context, which is the difference between knowledge that appears when you need it and knowledge that appears the following morning in the shower.

Phase 7 — Maintain, or let it go on purpose

Anything unused decays. Decide deliberately which parts you are keeping.

Keep — a short spaced review, decreasing over time, plus occasional real use. Once a subject is well learned, maintaining it costs very little.

Let go — and write down where to find it again. Knowing that a thing exists and where it is documented is a legitimate and efficient state of knowledge for most of the tail.

Fitting it to a week

A realistic template, and the specific numbers matter less than the shape.

When	What	How long
Session start	Recall last session from memory	5 min
Main block	New material, understanding first	45–60 min
Second block	Mixed practice problems	30 min
Session end	Blank page, write what you learned	5 min
Every session	Scheduled reviews due today	10 min
Weekly	Redraw the map, correct it	15 min
Weekly	Use it on something real	30–60 min

Two deep blocks (3.6), a written objective for each, and one honest weekly count of undivided hours.

The five ways this goes wrong

Reading as the main activity. The commonest failure by a wide margin. Reading feels productive, is measurable in pages, and produces recognition rather than knowledge.

Collecting resources instead of using one. Four courses, three books, a folder of saved articles. Pick one primary source, finish it, and use the others to fill gaps. The collecting is avoidance wearing a productive costume.

Optimising the system. The perfect card deck, the perfect note structure, the perfect schedule. All of it is easier than the difficult material, which is why people do it.

Waiting until ready to practise. There is no readiness threshold; the attempt is part of the learning.

No feedback loop. Studying for months without ever finding out whether it worked. Build the check in phase 1 and use it.

Why this book is built the way it is

Since this method is the one the book itself follows, it is worth stating the choices plainly.

Every chapter opens on a concrete case rather than an outline, because a fact with something to attach to is remembered and a fact arriving into empty space is not (3.1).

Volume and Part introductions give the map before the detail, which is phase 2.

Terms are defined at first use, always, because material you cannot decode cannot be encoded — that is the undesirable difficulty of 3.1, and it is pure waste.

Mechanisms are drawn, because two channels beat one when they are integrated (3.4).

Every chapter ends in something to do, because phase 6 is where knowledge becomes usable and it is the phase readers skip.

And chapters cross-reference constantly — the same idea meeting you in a different Part, months later, is spacing and interleaving happening without a schedule.

What the book deliberately does not do is quiz you. Volumes from V onward carry no self-test boxes, no flashcards and no recall panels, and that is a decision rather than an omission. A summary of a page you have just read is the fluency illusion in printed form (3.1): it feels like consolidation and it is recognition. The retrieval has to be yours, from a closed book, and no box on the page can do it for you.

Which is the one instruction to take from this Part into all nine volumes.

What to do with this

Pick the thing you are learning right now and write phase 1 down today. What will you be able to do, how will you know, and for how long do you need it. Ten minutes, and it will change what you do tomorrow.

Then do the one thing that separates people who finish from people who do not: close the book and produce something, every single session. A paragraph, a diagram, an explanation to someone, a worked problem with the source shut. Everything else in this Part is refinement. That one habit is the method.

Next: Part 4 begins the second half of this volume — not what your mind can do, but what it feels, and how to run it. 4.1 starts with the question that turns out to be much harder than it looks: what an emotion actually is.