Self Awareness: How To Know Yourself When Your Reasons Are Made Up

Self awareness begins with knowing yourself well enough to distrust the reasons you give for your own behaviour. Ask yourself why you chose the job, ended the friendship or argued in that meeting, and an answer arrives within a second, detailed and confident. A great deal of the time it is not the reason. It is not a lie either. Your brain generates an explanation for your behaviour automatically, after the decision is already made, and it generates one whether or not a real reason is available to report. The way out is not to introspect harder. It is to build checks your explanation machinery cannot reach.

The experiment that started all of this

In 1977 the psychologists Richard Nisbett and Timothy Wilson published a paper in Psychological Review with an unusually blunt title: "Telling More Than We Can Know: Verbal Reports on Mental Processes." Their argument was that people have very poor access to the processes that actually drive their judgements, while remaining entirely confident that their access is excellent.

Their best known demonstration was almost absurdly simple. They laid four pairs of stockings in a row on a table in a shop and asked passers by which pair was the best quality. The pairs were identical. People chose the rightmost pair far more often than any other, which is a well documented position effect.

Then came the part that matters. Asked why they had chosen that pair, nobody mentioned position. They talked about the knit, the sheerness, the elasticity, the finish. When the position effect was put to them directly, they denied it, politely and often with some amusement.

Notice what did not happen. Nobody said they did not know. Nobody said the pairs looked identical and they had simply picked one. The absence of a real reason did not produce the feeling of not knowing. It produced a reason.

Decision making runs ahead of the explanation

The obvious objection to the stockings study is that choosing between identical objects is a strange situation. Perhaps people only invent reasons when there is genuinely nothing to report, and ordinary decision making is more honest than that.

The answer arrived almost thirty years later from a group at Lund University in Sweden. Petter Johansson and his colleagues showed participants two photographs of faces and asked which was more attractive. The participant chose. The experimenter then slid the chosen photograph across the table and asked why. On some trials, using a simple piece of sleight of hand, the experimenter handed over the photograph the participant had just rejected.

Most people did not notice. More importantly, they explained the choice anyway, giving reasons for preferring the face they had actively turned down moments earlier, sometimes citing details that belonged specifically to the swapped photograph. The explanation was not retrieved from memory. It was generated on demand to fit whatever was in front of them.

The Lund group has since run this manipulation on taste preferences, moral positions, political survey answers and even people's own recorded speech. The pattern holds across all of them.

What the numbers show is not that people are stupid. It is that the mismatch between what you chose and what you are holding is not the kind of thing your mind checks. It checks whether an explanation can be produced, and an explanation can almost always be produced.

Why your brain does this on purpose

This is not a defect that slipped in by accident.

Michael Gazzaniga spent decades working with patients whose brain hemispheres had been surgically separated to control severe epilepsy. Because the hemispheres no longer communicated, he could show an instruction to one of them alone. Show the right hemisphere the word "walk" and the patient stands and walks. Ask why, and the answer comes from the left hemisphere, which never saw the word. It does not say it has no idea. It says something like, "I'm going to get a drink."

Gazzaniga called this system the interpreter. Its job is to produce a coherent running account of your own behaviour.

That account is enormously useful. You act for a hundred reasons you have no access to, and you live among people who need to predict you and be predicted by you in return. A narrative makes that possible. It simply is not an accurate narrative, because accuracy was never the specification. Coherence was. A system optimised for coherence will always produce a story, and it will produce one most eagerly when there is no story to retrieve.

What self awareness is actually good at

Here is the line that makes this usable rather than merely unsettling.

Introspection is reasonably reliable about the contents of your mind, and unreliable about the causes of your behaviour.

You genuinely do know that you feel anxious, that you prefer the blue one, that this room feels too warm. Those are readouts, and they are broadly trustworthy. What you do not have access to is why. Why you are anxious. Why the blue one. Why this room and not the identical one yesterday.

Contents: fairly trustworthy. Causes: not trustworthy at all.

Almost all the bad advice people give themselves comes from treating a confident answer about causes as though it were a readout about contents. "I'm not motivated" is a content report and it is probably accurate. "I'm not motivated because I'm lazy" is a causal claim, generated by the interpreter, and it has roughly the evidential standing of a guess about a stranger.

Four checks that work

Since the explanation arrives regardless, the only reliable move is to build checks the explanation machinery cannot get at. These four are small enough to actually do.

1. Predict before you decide. Write down what you expect to choose and why, before you choose. A prediction on paper cannot be quietly rewritten afterwards, which is exactly what makes it useful. The gap between the prediction and the outcome is real information; the story you tell afterwards is not.

2. Change one variable. If you believe you rejected the proposal on the numbers, ask whether you would have rejected the same numbers presented by somebody else, or on a day you had eaten lunch. If the answer wobbles, the numbers were not doing the work you thought.

3. Ask what would change your mind, and answer it specifically. Name the evidence that would flip you. If nothing would, then the reason you stated is not the reason you hold the position. This one is uncomfortable and it is the fastest of the four.

4. Wait a day. Explanations get less convenient with time. The version you reach for twenty four hours later, when the social pressure of the moment has gone, is usually closer to the truth than the one that arrived instantly.

There is a fifth habit worth adding, which is simply to notice the speed of your own answers. An instant answer is not evidence that you know something. Instant is what generation produces, because generating is faster than searching. An answer that begins "I'm not actually sure, let me think" is the one more likely to be doing real work.

A worked example

You come out of a meeting having argued hard against a proposal. Asked why, you say the numbers do not hold up and the timeline is unrealistic. Both statements are true about the proposal.

You had also skipped lunch. You had been contradicted twice that morning. And the person presenting once took credit for something of yours in front of your manager.

Nothing in your account is false. The numbers really were weak. The question that self awareness actually turns on is narrower and harder: were the weak numbers why you argued, or were they merely why you were able to?

Run check two. Would you have argued as hard against the same slide deck from a different colleague? If the honest answer is no, you have learned something that no amount of reflection on the numbers would have given you.

When your stated reason is the real one

This is not a counsel of total scepticism, and treating it as one is its own mistake.

Your stated reason is usually reliable in two situations. The first is when the reason was written down or spoken aloud before the decision, because then it is a record rather than a reconstruction. The second is when the decision followed an explicit procedure you can point at, such as a checklist, a budget threshold or a rule you set in advance. Procedures leave evidence. Impressions do not.

That is the practical case for writing things down before you act, and it is a better argument for it than productivity ever managed.

What kind of explainer are you?

Most people lean towards one characteristic failure mode: instant certainty, endless second-guessing, or a comfortable middle that never quite gets tested. The quiz below takes about two minutes and sorts you into one of four patterns, each with the checks most likely to help you specifically.

Frequently asked questions

Does this mean I have no free will?

No, and none of the research claims that. These studies are about self knowledge, not about whether you are choosing. Their finding is that your access to your own reasoning is poor, which is a claim about the reporting system rather than the deciding system. You can choose perfectly well and still be wrong about why you chose. Those are separate questions, and conflating them is the most common misreading of this entire body of work.

If introspection is unreliable, why bother with it at all?

Because it is unreliable about one specific thing. Introspection reports the contents of your mind reasonably well: moods, preferences, sensations, what you are currently thinking about. It is causal explanation, the "why", that it handles badly. Knowing which half you are doing at any moment is most of the skill, and it is learnable in an afternoon.

How is this different from lying to myself?

Lying requires knowing the truth and saying something else. Confabulation involves no such knowledge. The explanation your brain produces feels exactly like a memory, arrives with the same confidence as one, and carries no internal marker distinguishing it from a genuine report. There is no seam to feel for, which is precisely why the effect survives being explained to you.

Can I train myself out of it?

Not really, and the honest answer matters here. Knowing about the effect does not switch it off, which has been demonstrated repeatedly. What changes with practice is not the generation of explanations but your willingness to treat them as provisional. That is why the four checks are external, on paper or in a procedure, rather than a mental habit.

Is human psychology research on this actually settled?

The core finding is unusually robust for human psychology. The Nisbett and Wilson paper is nearly fifty years old and still stands; the choice blindness work has replicated across faces, tastes, moral attitudes, political answers and speech. What remains debated is the scope, meaning how far the effect extends into high-stakes and well-rehearsed decisions, where the evidence is thinner.

What should I do first?

Pick one decision this week that you would normally make on instinct. Before you make it, write one sentence naming what you expect to choose and why. Look at it again on Friday. That single habit does more for self awareness than an hour of reflection, because it produces the one thing reflection cannot: a record made before the explanation machinery had anything to explain.

Connect

Questions or thoughts? Reach out anytime.

Email

subscribe to our newsletter

hello@minddecoded.com

© 2026. All rights reserved.

Terms of Service Privacy Policy Disclaimer