Metacognition is thinking about your own thinking: knowing how you learn, noticing whether you really understand something, and changing course when you do not. Everyone does it — it is what happens when you realise you have read a page without taking it in. The part that matters is accuracy, and that is where the surprise lies: metacognition is only weakly related to IQ, and it predicts academic performance even after intelligence is taken into account.
What metacognition means
The word was introduced in 1979 by the developmental psychologist John Flavell, who wanted a name for what children know about their own memory and understanding, and how that knowledge changes as they grow. The idea has since spread across psychology and education, and researchers now usually divide it in two:
| Part | What it covers | Example |
|---|---|---|
| Metacognitive knowledge | What you know about your own mind, about tasks and about strategies | “I forget names unless I write them down”; “a multiple-choice exam needs different preparation from an essay” |
| Metacognitive regulation | Planning an approach, monitoring how it is going, and evaluating the result | Stopping to check a calculation that came out suspiciously round; switching method when the first one stalls |
An influential 1990 model by Thomas Nelson and Louis Narens describes regulation as a loop between two levels. Information flows up from the thinking itself to a “meta-level” that watches it — monitoring, such as the feeling that you know an answer — and instructions flow back down — control, such as deciding to study a topic again or to stop. Everything useful about metacognition depends on the upward signal being accurate. If the feeling of knowing is wrong, every decision built on it is wrong too.
Examples of metacognition
- Before a task: asking what a good result looks like, how long it should take, and which approach suits it.
- During a task: noticing that you have lost the thread of an argument; sensing that an answer came too easily and checking it.
- After a task: asking which mistakes were careless and which came from not knowing, and what you would do differently.
- About yourself: knowing that you understand diagrams better than long paragraphs, or that you overestimate how fast you work.
Is metacognition linked to IQ?
This is one of the questions Google's own results show beside the word, and it is the one an intelligence site is best placed to answer. The short answer: weakly, and less than most people assume.
The largest synthesis is a 2018 meta-analysis that pooled 149 samples from 118 studies. It found that metacognition correlated only weakly with intelligence, and weakly with academic performance — and that it still predicted academic performance once intelligence was controlled for. The strength of the link also depended on how metacognition was measured, which is a reminder that questionnaires about your study habits and laboratory tests of your confidence are not measuring quite the same thing.
The laboratory evidence points the same way. In a 2010 study in Science, researchers separated people's ability to tell their own correct decisions from their incorrect ones from how well they actually performed the task. That introspective ability varied substantially between people performing at the same level, and it was associated with the structure of the anterior prefrontal cortex, a region distinct from the one doing the task itself.
Put together: being good at a kind of problem and knowing when you have got it right are related but separable abilities. A person with a high IQ can be a poor judge of their own understanding, and a person of average ability can be an excellent one. That is also why metacognition matters more for learning than for any single test score — it steers what you study next. The best-known argument about judging your own performance, and why the famous curve is weaker than its reputation, is on our page about the Dunning–Kruger effect.
How researchers measure it
Questionnaires ask people how they plan and check their work, but the sharper measure compares confidence with accuracy, answer by answer. Someone with good metacognition is confident when right and unsure when wrong. A 2014 review of these methods separates three things that are easy to blur:
- Metacognitive bias: how confident someone is overall, whether or not they are right.
- Metacognitive sensitivity: how well their confidence distinguishes their correct answers from their incorrect ones.
- Metacognitive efficiency: that sensitivity relative to how well they performed the task.
The distinction matters because a confident person is not necessarily a metacognitively skilled one. Confidence that goes up and down with actual accuracy is the skill; confidence alone is closer to a personality trait.
Why your sense of knowing can mislead you
The most useful findings about metacognition are about where the upward signal goes wrong — and they are consistent enough to act on.
- Rereading feels better than it works. In a 2006 study, students who restudied a passage were more confident about remembering it than students who tested themselves on it. After two days and after a week, the students who had tested themselves remembered substantially more.
- Fluency is taken for learning. In a 2008 study, people learning to recognise painters' styles did better when the paintings were mixed together than when each painter's work came in a block. Most still rated the blocked schedule as more effective — even after their own test results had shown the opposite.
- Judging too soon inflates confidence. A judgement of how well you have learned something is more accurate when it is made a little while after studying than immediately afterwards — the delayed-judgement effect, reported in 1992 — because by then you are checking whether you can retrieve it, not whether it is still in front of you.
Metacognitive strategies that are backed by evidence
A 2013 review rated ten common study techniques on how widely their benefits hold across learners, materials and tests:
| Rating | Techniques |
|---|---|
| High utility | Practice testing; distributed (spaced) practice |
| Moderate utility | Elaborative interrogation (asking “why is this true?”); self-explanation; interleaved practice |
| Low utility | Summarisation; highlighting; the keyword mnemonic; imagery for text learning; rereading |
The two most popular techniques — rereading and highlighting — sit in the bottom row, and the review recommends replacing rereading with practice testing. That is a metacognitive lesson as much as a study tip: the techniques that feel productive are often the ones that give the least accurate signal about what you know. Turned into habits:
- Plan: before starting, write down what you need to be able to do at the end, not what you will read.
- Monitor by recall: close the book and try to explain the idea or answer a question. What you cannot produce is what you do not yet know.
- Wait before judging: decide what to revise the next day, not straight after studying.
- Evaluate: after a test, sort errors into “did not know” and “knew but slipped”; they need different fixes.
Planning, monitoring and switching approach also draw on executive function, the set of control processes that let you hold a goal and inhibit distractions; and the habit of questioning your own conclusions is the core of critical thinking.
How metacognition develops, and whether it can be taught
A 2021 review drawing on neuroscience and education summarises the developmental picture: knowledge about one's own thinking appears around age 5, deliberate control around age 8, and both keep developing into adolescence. Early on the skills are tied to particular activities — a child may monitor their reading well and their arithmetic poorly — and they become more general with age.
The same review reports that metacognitive instruction has positive effects on learning and achievement from primary school to university, with the largest effects when students are taught a combination of strategies, with an emphasis on planning, together with knowledge about when and why each one works. Unlike the claims made for most ways to raise IQ, this is an area where teaching has measurable effects on how well people learn.
Trust and scope notes
This page explains published research in general terms. Our IQ test measures reasoning; it does not measure metacognition, and your result does not say how accurately you judge your own thinking. The research above suggests that the two are only loosely connected.
IQ Revealed is not affiliated with, endorsed by or connected to the researchers, universities, journals or publishers named on this page. Their names are used only to identify the work being described.