The Emergence of Metacognition: When Children Begin to Know What They Know

Lead

"I understand!" — followed immediately by doing something completely different. "I can do it," said with total conviction, then abandoned seconds later. "I know that one," and then silence when you ask for an explanation.

If you spend time with four- to six-year-olds, these moments accumulate. It is tempting to interpret them as small deceptions, or as overconfidence. But there is a more useful frame: the ability to accurately monitor one's own understanding — what researchers call metacognition — is still being built at this age. The gap between "I know it" and actually knowing it is not a character flaw in a five-year-old. It is a developmental fact.

What Metacognition Means

The term "metacognition" was conceptualized by the psychologist John Flavell in the 1970s [1]. Flavell defined it as the ability to think about one's own thinking — specifically, to understand what you know and don't know, to monitor your cognitive states, and to regulate your thinking processes accordingly.

Flavell distinguished two components: metacognitive knowledge (one's beliefs about one's own cognitive capacities and how the mind works in general) and metacognitive experiences (the moments of awareness that arise during an actual cognitive task) [1]. The feeling "this problem is hard" is a metacognitive experience. The belief "I'm good at numbers" is metacognitive knowledge.

Schraw and Moshman later offered a complementary framework [2], distinguishing knowledge of cognition from regulation of cognition. Within knowledge of cognition, they identified three subtypes: declarative knowledge (knowing that something is the case), procedural knowledge (knowing how to do something), and conditional knowledge (knowing when and why to apply a strategy). Which of these a young child possesses, and in what order they are acquired, remains an active area of research.

What Five- and Six-Year-Olds Can Actually Do

Research on false belief tasks provides an important reference point for understanding metacognition in early childhood. The false belief task, developed by Wimmer and Perner in 1983, tests whether a child can understand that another person holds a belief that the child knows to be false [3]. The capacity to represent another mind's mistaken belief is a central component of what researchers call Theory of Mind, and typically emerges around ages four to five.

But representing someone else's mistaken belief and monitoring your own understanding are distinct capacities. The latter — what is sometimes called self-directed metacognitive monitoring — remains unstable in five- and six-year-olds. Overconfidence is common: children at this age frequently believe they know something they do not.

Kuhn addressed this developmental trajectory in a 2000 review [4]. She traced the progression from an early phase in which children are certain of their knowledge and struggle to recognize its limits, toward a later capacity for "knowing what you don't know." The five- to six-year-old is, in Kuhn's account, standing at the threshold of this transition — not yet reliably on the other side of it.

The Over-Claiming Problem

One of the consistent findings in the developmental literature is that young children — including those in the four-to-six age range — tend to overestimate both what they know and what they will remember. In a well-known set of experiments on what researchers call "judgments of learning", children were shown a set of items and asked to predict how many they would be able to recall later. Young children's predictions are reliably more optimistic than their performance [4]. This is not random error; it is a systematic bias in the direction of confidence. The child is not guessing randomly about their own competence — they are consistently overestimating it.

The practical implication is that the child who says "I know how to do it" and then proceeds to struggle is not reporting dishonestly. Their metacognitive monitoring is simply not yet calibrated. Understanding this changes the adult's response. Frustration ("you said you knew") addresses the symptom as if it were willful. Curiosity ("let's figure out where it got hard") addresses the underlying developmental reality.

What Observation Studies Show

Whitebread and colleagues took a different approach: instead of clinical tasks, they observed children ages three to five in naturalistic preschool settings, looking for metacognitive and self-regulatory behavior as it actually appears in daily life [5]. Using an observational coding tool they developed called CHILD 3-5, they identified 582 episodes of metacognitive or self-regulatory behavior across their sample.

The observed behaviors included: spontaneously changing strategy when something isn't working, showing evidence of planning before starting a task, and producing self-evaluative comments — "I did it," "that didn't work" [5]. These behaviors are easy to overlook in the flow of daily childcare. Directed attention is needed to register them.

An important implication of this work is that early metacognition is largely procedural rather than declarative. A child who cannot articulate "I don't understand this" in words may nonetheless, when stuck, quietly try a different approach. The metacognition is present; it simply operates at the level of action rather than speech. Expecting a five-year-old to verbalize their understanding accurately, then being surprised when they cannot, mistakes the form that this ability takes at this age.

What Adults Can Do

Metacognitive development is influenced by the environment — by the quality of adult interaction in daily life. This is worth saying with some care, because "teach metacognition" can be misread as a simple prescription. The developmental timing varies considerably between children, and there is no robust evidence that forced or formal training programs produce reliable gains at these ages.

That said, certain kinds of interaction create conditions in which metacognitive awareness can develop.

Make it possible to say "I don't know." Asking "Did you understand?" invites a yes or no. Asking "What part was confusing?" invites the child to locate and name the limit of their understanding — which is the cognitive act metacognition requires.

Treat failure as material for problem-solving. When something goes wrong, the question "What do you think didn't work?" asks the child to observe their own process, not simply to receive a correction. The shared act of noticing is more useful, at this age, than the provision of the right answer.

Distinguish knowing from doing. "Do you understand?" collapses the two. A phrase like "It's possible to know something but still find it hard to explain, right?" creates conceptual space between partial and complete understanding. That space is precisely where metacognitive growth happens.

From a parenting-record perspective: keeping notes on both the moments when a child says they know something and the moments when they demonstrate it creates a personal log of that child's metacognitive accuracy over time. This is not an assessment tool. It is a way of sharpening your own observation — of learning to distinguish declarative confidence from procedural readiness in the particular child in front of you.

Summary

When a child says "I understand!" and immediately does something different, they are not lying. They are operating at a developmental stage in which the ability to accurately monitor their own comprehension is genuinely incomplete.

This capacity — metacognition, conceptualized by Flavell [1] and documented in naturalistic settings by Whitebread and colleagues [5] — is present in rudimentary form by ages five to six, but it is unstable and highly context-dependent. That is not a weakness to be corrected. It is a phase to be understood.

Watching a child gradually develop the ability to know what they don't know is one of the quieter, richer experiences that the early years offer.

References

1. Flavell JH. Metacognition and cognitive monitoring: a new area of cognitive-developmental inquiry. Am Psychol. 1979;34(10):906–911. doi:10.1037/0003-066X.34.10.906.
2. Schraw G, Moshman D. Metacognitive theories. Educ Psychol Rev. 1995;7(4):351–371. doi:10.1007/BF02212307.
3. Wimmer H, Perner J. Beliefs about beliefs: representation and constraining function of wrong beliefs in young children's understanding of deception. Cognition. 1983;13(1):103–128. doi:10.1016/0010-0277(83)90004-5.
4. Kuhn D. Metacognitive development. Curr Dir Psychol Sci. 2000;9(5):178–181. doi:10.1111/1467-8721.00088.
5. Whitebread D, Coltman P, Pasternak DP, Sangster C, Grau V, Bingham S, Almeqdad Q, Demetriou D. The development of two observational tools for assessing metacognition and self-regulated learning in young children. Metacogn Learn. 2009;4(1):63–85. doi:10.1007/s11409-008-9033-1.