Metacognition: Thinking About Your Thinking

Most of your thinking runs without supervision. You reach conclusions, form impressions, and make choices before you have any sense of how you got there. That is efficient, and often it works. But it means the moments when your thinking is going wrong feel exactly like the moments when it is going right — from the inside, being confidently mistaken and being correct are indistinguishable.

Metacognition is the skill that breaks this symmetry. When you can watch your own thinking as it happens, you gain a second layer of control: you can catch a hasty inference, notice that a problem is harder than it looked, and tell the difference between knowing something and merely feeling that you do. Every other critical-thinking habit depends on this one, because you cannot correct an error you never notice.

Monitoring and regulation

The developmental psychologist John Flavell introduced the term metacognition in the 1970s, describing it as knowledge and cognition about cognitive phenomena — thinking about thinking. It has two working parts. Monitoring is awareness: sensing that you are confused, that an answer is on the tip of your tongue, or that you have stopped understanding a page you are still reading. Regulation is the response: slowing down, rereading, planning a different approach, or deciding you have done enough. Good thinkers are not the ones who never feel confused; they are the ones whose confusion reliably triggers a change of tactics.

Calibration and overconfidence

Calibration is the match between how confident you feel and how often you turn out to be right. A well-calibrated person who says they are ninety percent sure is right about nine times in ten. Most people are overconfident: their certainty outruns their accuracy, and the gap is largest precisely where it is most dangerous — in areas they know little about but assume they understand. Treating confidence as data about the world, rather than a feeling to be checked, is one of the most common failures of self-aware thinking.

Fast and slow thinking

The psychologist Daniel Kahneman popularised a useful distinction in his 2011 book Thinking, Fast and Slow: a fast, automatic, intuitive mode he calls System 1, and a slow, effortful, deliberate mode he calls System 2. Fast thinking is superb for the familiar and the routine and terrible when a situation only looks familiar. Metacognition is largely the art of knowing when to override the fast answer and pay the cost of the slow one.

A riddle: a bat and a ball cost £1.10 together, and the bat costs £1 more than the ball. How much is the ball? Almost everyone's fast answer is ten pence. It arrives instantly and feels obviously right. Slow it down: if the ball were ten pence and the bat a pound more, the bat would be £1.10 and the total £1.20, not £1.10. The correct answer is five pence. Nothing here is hard — the difficulty is noticing that the confident first answer deserves a check at all. That noticing is metacognition doing its job.

Metacognition is not the same as endless self-doubt. Imagine a surgeon who, mid-operation, begins second-guessing every well-drilled movement, or a driver who consciously deliberates over each gear change. Here the fast, automatic system is the expert, and forcing slow deliberation degrades performance. The skill is not to distrust intuition everywhere but to know where your intuition has earned its authority — through genuine, feedback-rich practice — and where it has not. Regulation applied in the wrong place is its own kind of error.

In 1999 the psychologists Justin Kruger and David Dunning published a study in the Journal of Personality and Social Psychology titled "Unskilled and Unaware of It." Across tests of grammar, logic, and humour, they found that the lowest-scoring participants dramatically overestimated their own rank — those in the bottom quarter often believed they were above average. Their explanation is the interesting part: the very knowledge needed to perform well is the same knowledge needed to judge your performance. If you do not know the rules of grammar, you cannot see your grammatical mistakes, so incompetence hides itself.

Tellingly, the study also found that when poor performers were trained in the skill, their self-assessments improved — they became able to recognise how badly they had done. The effect is often exaggerated in popular retellings into a simple claim that the ignorant think they are geniuses, which overstates it; the real finding is a modest, measurable gap in self-insight that shrinks as competence grows. It remains a clean demonstration of why metacognition cannot be taken for granted.

• "Feeling certain means I am right." Confidence is a feeling produced by your mind, not a measurement of the truth; the two often come apart.
• "Metacognition means doubting everything." It means checking the right things at the right time, not distrusting well-earned intuition everywhere.
• "The Dunning-Kruger effect says stupid people think they are brilliant." The actual finding is a smaller gap in self-assessment that narrows as skill improves.
• "Fast thinking is bad thinking." Fast thinking is excellent in familiar, feedback-rich domains; it fails mainly when a novel problem merely looks familiar.

Before your next decision or answer of any weight, pause and write down a confidence figure — a rough percentage that you are right. Then find out the outcome and compare. Do this a dozen times and you will have real data on your own calibration, and a sharper feel for when your certainty can be trusted.

Think Like a Maester: Watch the thinker as well as the thought — the feeling of being right is the last thing you should trust to tell you that you are.

1. What are the two working parts of metacognition, and how do they differ?
2. What does it mean to be well-calibrated, and why is overconfidence dangerous?
3. In Kahneman's terms, when is fast thinking most likely to mislead you?
4. What did Kruger and Dunning actually find, and how is the effect commonly overstated?
5. Give one situation where slowing down and deliberating would make your thinking worse, not better.

Metacognition is thinking about your own thinking: monitoring it as it happens and regulating it when it needs a change of course. It matters because being confidently wrong feels exactly like being right, so without a second layer of awareness your errors stay invisible. The core skills are noticing confusion and treating it as a signal, calibrating confidence against real accuracy rather than mistaking a feeling for a fact, and knowing when to let a fast intuitive judgement stand and when to pay for slow deliberation. Kruger and Dunning showed how the least skilled are often least able to see it, precisely because the knowledge that would fix the work is the same knowledge that would reveal the flaw. Build the habit of watching the thinker, not just the thought.

• Cognitive Biases
• Mental Models: A Toolkit for Thinking