What Is Metacognition? Meaning, Examples, and Three Ways to Train It

Metacognition is the capacity to treat your own thinking and understanding as an object, monitor it, and adjust your approach when needed. Put simply, it is the ability to go and look at how much you currently understand. This article covers the definition from psychology, everyday examples, and three procedures you can use today.

What metacognition is: a framework from 1979

The term metacognition was introduced by Flavell (1979) as 'cognition about cognitive phenomena.' That paper organizes it into four components: metacognitive knowledge, metacognitive experiences, goals or tasks, and actions or strategies.

For practical use, a two-layer split is easier to work with:

• Metacognitive knowledge: what you know about your own strengths and about which method suits this task
• Metacognitive control: checking whether you currently understand, and changing method when you do not

The second is the hard one. Knowing that review matters is a separate question from whether you can accurately estimate your own understanding in the moment. For where this sits relative to cognition itself, see what cognitive abilities are.

What it looks like in practice

It sounds abstract, but it happens constantly.

• Reading a textbook, noticing that your eyes passed over a paragraph without it going in, and rereading
• Thirty seconds into a problem, judging that this approach will not finish in time and switching
• Watching a listener's expression change and realizing you skipped a premise
• Before an exam, distinguishing 'I could solve this' from 'I have actually solved this'

What they share is attention on the state of the work rather than the work itself. The rule in matrix reasoning to move on if no approach forms within 30 seconds is metacognitive control exactly.

The core of metacognition is the checking step

Metacognition can be drawn as a loop: plan, do, monitor, adjust. Monitoring is the step most often skipped, because completing the work itself produces a sense of accomplishment and carries you into the next thing without verifying the result.

When checking drops out, self-rated understanding drifts away from actual performance. The extreme form of that divergence is the Dunning-Kruger effect: judging your performance in a domain accurately requires competence in that same domain, which makes the structure recursive.

Three procedures for training it

Metacognition is a procedure, not a personality trait. Each of these moves the checking step outside your head so it actually runs.

1. Write the prediction first: before solving, write one line on how long it will take and whether you will get it right. Compare with the measurement afterward. The direction of the gap is your personal bias.

2. Confirm by explaining: say the material aloud as if teaching it, with no references. Wherever you stall is exactly what you only felt you understood.

3. Classify your errors: sort mistakes into did not know, misunderstood, ran out of time, and misread. What to do about them only diverges after that sorting.

None of this acquires new ability; it corrects how existing ability is used. The distinction runs through can IQ be raised as well. Given how rarely training effects spread beyond the trained task, improving strategy is one of the few realistic moves available.

Measure the gap in your own estimate

Whether your estimate is accurate is not knowable until you measure it. Before taking BrainRank's free IQ test, write down the IQ you expect and which of the four domains you think is your strongest. The results from 20 questions in about 10 minutes report spatial reasoning, pattern recognition, logical reasoning, and classification separately, so you can read the gap against your prediction directly. Results are estimates for self-understanding, not an official psychological assessment.