How to start thinking about a problem

It is eleven at night, something is unresolved, and you have been turning it over for an hour. That hour felt like work. Very little of it was. Worry, replaying a conversation and rehearsing a reply are all activities the mind performs on a problem without moving it, and they are comfortable in a way real thinking is not, because they never produce an output you could be wrong about.

Thinking, in the sense this note means it, has an input, a few steps and something at the end you could show someone. The test is blunt: at the end of the hour, is there a sentence on a page that was not there before? If not, the hour was rehearsal.

Use the stem: the problem is ___ and I will know I have solved it when ___. Both halves matter, and the second half is the one that does the work. A problem with no finish condition cannot be finished, which is exactly why it keeps coming back.

Expect the first attempt to be wrong, and expect that to be useful. "I am behind on my exam" becomes, on the second attempt, "I do not know which of the six topics I am weakest on" — which is not a mood, it is a question answerable in about ten minutes. The vague version had no next step. The specific version has an obvious one.

There is a ladder here and it is worth knowing you are on it: a topic becomes a worry, a worry becomes a question, and a question becomes a decidable question. "I want to be more productive" is a topic. "Which two hours of my day produce nothing, and why?" is decidable.

Take everything you believe about the situation and put each piece into one of three bins: what is known, what is assumed, what is felt. All three are legitimate. Only one of them is a to-do list.

The assumption bin is where the work is, because an assumption is a fact you have not checked yet and one of them is usually load-bearing. Consider a job offer in another city: "the base salary is eighteen per cent higher" is known, "I would be letting my team down" is felt, and "the commute would be worse" is assumed — and it is the assumed one that, if wrong, flips the decision. Checking it costs twenty minutes on a map.

The question to end this step with is the one that names that item: which single assumption, if wrong, changes what I would do?

A page of ugly sentences beats an hour of circling, and the reason is mechanical rather than motivational: you cannot hold six competing considerations in mind and compare them at the same time, but you can look at six lines. The one-page think-through is question, what I know, what I assume, options, what I will do, and what would change my mind.

Nobody else has to read it. It is a workspace, not a document.

Two options is usually one option and an excuse. Forcing a third changes what the first two look like, and including "do nothing" as a real candidate — costed, not dismissed — is what stops a decision being made by momentum.

There is a published trick worth using here. Assume it is a year from now and the plan failed, then list the reasons why. That framing is Gary Klein's pre-mortem, and it draws on a 1989 finding by Mitchell, Russo and Pennington that imagining an outcome as already certain increased the number of reasons people could generate for it — reported there as roughly thirty per cent, in a laboratory task, in 1989. Quoted with its context because that is the only honest way to quote it, and useful anyway: it is much easier to explain a failure than to predict one.

A position you cannot state a falsifier for is not a conclusion; it is an attachment. Writing the falsifier down before you act is what makes a later review about the process rather than about the luck.

Then stop, on purpose. Over-thinking a reversible choice is itself a mistake. Ask what undoing this would cost: choosing a note-taking app is reversible and deserves two minutes, and signing a year's lease is not and deserves the whole page. Spending equal thought on both is how people end up with a carefully chosen app and a badly chosen lease.

Tomorrow, before you look at this page again, try to write the five steps from memory. You will get three or four, and finding the missing one is worth more than reading all five again.

That is not a stylistic preference. Practice testing and distributed practice were the two techniques rated highest for general utility in Dunlosky and colleagues' 2013 review of ten study techniques, and Roediger and Karpicke showed that being tested beats re-studying on a delayed test even though re-studying wins after five minutes. The fluency of re-reading is the trap: it feels like knowing, and it is the feeling that fades.

One honest limit. Those findings come from people learning factual material, and whether they reach a thinking procedure like this one has not been shown; this note does not assume it has.

This is a procedure, not a measurement. Running it produces no score, no level, no type and nothing that describes you — Brain Logic never says you are a strategist, only that you can ask what a strategist asks. Counts of what you did are the only numbers that exist in it.

And there is no evidence that completing modules like this one changes performance on any timed task, on any test, or at work. If that evidence ever exists it will be a study of this product, and until then the claim is not made.

A grade belongs to one claim, in one population, for one preparation — never to a compound, a technique or an article. The weakest link is the part of the chain between the evidence and the thing you would actually do that is doing the least work.