Metacognition: What It Is, and How to Use It to Study Smarter

What is metacognition?

Metacognition means thinking about your own thinking. It is the part of your mind that watches the rest: noticing that you have read the same paragraph three times without taking it in, realising you are not sure of an answer, or deciding that a problem needs a different approach.

The term was popularised by the developmental psychologist John Flavell, whose 1979 paper "Metacognition and cognitive monitoring" described it as a new area of research. Since then psychologists have usually split it into two parts:

• Monitoring: judging your own mental state. "Do I understand this?" "Will I remember this next week?" "How sure am I of this answer?"
• Control: acting on those judgements. Rereading a section, switching strategies, spending more time on a weak topic, or deciding you are ready to stop.

Control can only be as good as monitoring. If you wrongly think you know something, you stop studying it.

Everyday examples of metacognition

• Realising halfway through a conversation that you have misunderstood the question, and asking again.
• Checking a map because you are not confident about a turn, rather than because you know you are lost.
• Knowing that you remember names badly, so you repeat a new colleague's name back straight away.
• Feeling a word is "on the tip of your tongue": a judgement that the memory is there, even though you cannot reach it.
• Noticing that you understood a lecture while listening, but cannot explain it to a friend afterwards.

That last one is the most important for studying, because the feeling of understanding and actual understanding often come apart.

Why learners misjudge what they know

Familiarity feels like knowing

A survey of 177 college students by Karpicke, Butler & Roediger (2009) found that most repeatedly reread their notes or textbook, and relatively few tested themselves. The authors argued that students experience illusions of competence: after reading something several times it feels familiar, and that familiarity is mistaken for being able to recall it.

A 2024 experiment shows the illusion directly. In Seban et al. (2024), 186 psychology students either reread a text or read multiple texts. In the rereading group, their judgements of how well they had learned rose after each reading, and they were overconfident, especially when tested a day later.

Explanations feel deeper than they are

Rozenblit & Keil (2002) showed that people believe they understand how things work in far more detail than they do, and that this "illusion of explanatory depth" is stronger for explanations than for facts, procedures or stories. Many certification questions test exactly this kind of how-and-why knowledge.

Effective methods feel worse

Learners also misjudge strategies. In Kornell (2009), spacing flashcard practice beat massing for 90% of participants, yet after the first session 72% believed massing was more effective. In Kornell & Bjork (2008), people rated grouped (blocked) examples as better for learning even after their own test results had shown mixed examples worked better. More on this in desirable difficulties.

A nuance: learners are not hopeless judges

It is easy to overstate this. Weissgerber & Rummer (2022) ran three preregistered experiments with 306 participants. When learners were asked to imagine a test one to two weeks away, their judgements often favoured testing over rereading. In other words, people may be less wrong about the value of retrieval practice than earlier studies suggested, especially when they think about the long term rather than tomorrow.

The practical lesson is not "your judgement is useless", but "judge with the right question and the right evidence". Asking "does this feel familiar?" invites the illusion. Asking "can I answer a question on this, without notes, next week?" and then actually testing it gives you a much better reading.

Why accurate self-judgement matters for results

Better monitoring is linked to better learning. In Dunlosky & Rawson (2012), college students practised recalling key-term definitions and judged after each attempt how much of their answer was correct; they kept going until they had judged each definition correct three times. Across two studies, students whose judgements were more accurate retained more on a test two days later, and this could not be explained by feedback, effort or the number of practice trials. Because practice on each definition ended once a student judged it correct, an overconfident student dropped definitions before they were really learned.

Learning about how memory works can also change behaviour. In Brown-Kramer (2021), students who completed an assignment on evidence-based learning strategies used fewer low-utility and more high-utility strategies, and their course performance improved by about a third of a letter grade compared with a control semester. Metacognitive knowledge includes knowing which beliefs are myths: for example, Pashler et al. (2008) found virtually no evidence that matching teaching to a learner's preferred "learning style" improves learning.

Warning signs your judgement is misleading you

You cannot feel an illusion of competence from the inside; that is what makes it an illusion. But some situations make one likely:

• "I recognise all of this." Recognition is easier than recall. Multiple-choice exams give you options, but scenario questions still require you to recall the principle that rules out the distractors.
• "I just read it." Judging straight after reading measures what is in short-term memory, not what will be there on exam day.
• "That session felt productive." Smooth sessions are often easy sessions. Hard retrieval that felt frustrating may have taught you more.
• "I got it right, so I know it." On a four-option question, a pure guess is right a quarter of the time. Without a confidence rating you cannot tell a guess from knowledge.
• "I keep scoring well on the same practice test." After a few attempts you may be remembering the answers, not the material. Use fresh questions or leave a gap of weeks.

Metacognitive strategies for studying

1. Test before you judge. Do not decide you know a topic until you have answered questions on it without notes.
2. Delay your judgement. Judge your learning a day or more later, not right after reading, when familiarity is highest.
3. Rate your confidence on every practice question (sure / unsure / guess) before checking. This is the core of calibration, below.
4. Explain it out loud. If you cannot explain a mechanism simply, the illusion of explanatory depth has caught you; see the Feynman technique.
5. Grade strictly. Partly right is wrong for study purposes. Dunlosky & Rawson's results suggest lenient self-grading leads you to stop too soon.
6. Plan, monitor, review. Before a session, decide what you will be able to do at the end. Afterwards, test whether you can.

How to use this for a certification exam

The most useful metacognitive skill before an exam is calibration: making your confidence match your accuracy. Here is a protocol using our free practice tests.

1. Predict your score. Before you start, write down what percentage you expect on a free practice test, for example the CISSP practice test or the AWS practice test.
2. Rate each answer. As you answer, mark each one sure, unsure or guess. Do not look at explanations yet.
3. Score and sort. Put every question into one of the four boxes in the table below.
4. Compare. How far was your predicted score from your actual score? How many "sure" answers were wrong? Wrong-but-sure answers are your most dangerous gaps, because you would never have studied them.
5. Act on each box, following the table.
6. Repeat after a week or two. Calibration is improving if your predicted score gets closer to your real one and the sure-but-wrong box shrinks.

Do not book the exam on a feeling of readiness. Book it on consistent, calibrated practice results. For how retrieval and spacing fit in, see active recall, the testing effect and the study science overview.