Definition
Metacognition is the capacity to think about one's own thinking. A student using metacognition does not merely work through a problem — she observes herself working through the problem, notices where her understanding breaks down, and deliberately adjusts her approach. The concept encompasses two interrelated capacities: knowing what you know (and what you do not), and regulating how you think in response to that knowledge.
Psychologist John Flavell coined the term in 1976, defining metacognition as "one's knowledge concerning one's own cognitive processes or anything related to them." Flavell's formulation distinguished between metacognitive knowledge (stable beliefs about how memory, attention, and learning work) and metacognitive experiences, the moment-to-moment feeling of comprehension or confusion that signals whether a strategy is working. Both dimensions matter in the classroom. A student who knows she learns better by self-quizzing than re-reading (metacognitive knowledge) but keeps re-reading anyway (poor metacognitive regulation) is not fully metacognitive.
The practical implication for educators is direct: metacognition is not a personality trait some students have and others lack. It is a set of teachable skills. Research consistently shows that students who receive explicit instruction in metacognitive strategies outperform peers who do not, regardless of prior ability level.
Historical Context
Metacognition as a formal construct emerged from developmental psychology in the 1970s, though its intellectual roots run deeper. Flavell's early work on metamemory — children's understanding of how their own memory functions, laid the groundwork for the broader construct. His 1979 paper in American Psychologist, "Metacognition and Cognitive Monitoring," established the vocabulary that researchers still use today.
Ann Brown at the University of Illinois advanced the concept significantly through the late 1970s and 1980s. Brown distinguished between "knowing about" cognition and "regulating" cognition, a division that remains foundational. Her 1978 chapter "Knowing When, Where, and How to Remember" demonstrated that poor readers differed from skilled readers not in raw ability but in their awareness of comprehension failure, they did not notice when they stopped understanding.
Cognitive psychologists Robert Sternberg and Howard Gardner contributed parallel frameworks in the 1980s that emphasized executive control processes, the mental supervisor that directs which strategies get applied and when. Barry Zimmerman's work in the 1990s integrated metacognition into the broader model of self-regulated learning, showing that high-achieving students cycle through forethought, performance monitoring, and self-reflection as a habitual practice rather than an occasional behavior.
The Education Endowment Foundation's systematic review, led by Alex Quigley, Dunstan Muijs, and Eleanor Stringer in 2018, synthesized this decades-long research tradition into practical teacher guidance, bringing metacognition from academic journals into mainstream professional development.
Key Principles
Metacognitive Knowledge Has Three Dimensions
Flavell identified three types of metacognitive knowledge that together determine how well a learner understands her own cognitive situation. Person knowledge is what someone believes about herself as a learner: "I remember information better when I draw diagrams." Task knowledge is understanding how different tasks demand different cognitive approaches — a multiple-choice test requires different preparation than an essay. Strategy knowledge is knowing which cognitive tools exist and when to apply them, summarizing, self-quizzing, elaborative interrogation, spaced practice.
Students with rich metacognitive knowledge across all three dimensions make better decisions about how to study, how long to spend on a difficult problem before seeking help, and which parts of a text deserve close attention. Students who lack this knowledge often work hard without learning effectively, they mistake familiarity with mastery, re-read instead of self-test, and misjudge their preparedness before assessments.
Metacognitive Regulation Is an Active Process
Regulation is metacognition in motion. It involves three phases: planning (setting goals, selecting strategies, allocating time before a task), monitoring (tracking comprehension and progress during a task), and evaluating (reviewing what worked and what did not after a task).
Students rarely develop regulatory skills spontaneously. Most default to passive re-reading or re-exposure because it feels familiar and comfortable, even though the evidence for its effectiveness is weak. Teaching regulation explicitly, modeling the planning-monitoring-evaluating cycle before asking students to do it independently, produces durable improvements in how students approach learning across subjects.
Metacognition Is Domain-Specific
A student who monitors her comprehension effectively in history class may not transfer that skill automatically to algebra. Metacognitive awareness is partially general, some students are broadly more reflective, but substantial transfer requires deliberate practice in each new domain.
This has a concrete implication for curriculum design: metacognitive instruction embedded within subject-specific content outperforms generic study skills courses taught in isolation. Teaching a student to notice when a mathematical argument stops making sense produces better results when the instruction happens during mathematics, not during a separate "learning to learn" period.
Explicit Instruction Accelerates Development
Students who are told what to do (solve this problem) develop metacognition far more slowly than students who are also told why a strategy works and when to use it. The EEF recommends that teachers explicitly label metacognitive strategies, model their own thinking aloud, and gradually release responsibility to students through structured scaffolding.
Teacher think-alouds are a particularly powerful tool here. When a teacher narrates her own uncertainty, "I'm reading this paragraph and I realize I'm not sure what 'osmosis' means in this context, so I'm going to re-read the previous sentence and then check the glossary", she makes the invisible process of monitoring visible and imitable.
Metacognitive Development Is Age-Sensitive but Never Too Late
Young children have limited metacognitive awareness; they consistently overestimate how much they know and underestimate how long tasks will take. Metacognitive capacity develops substantially through adolescence and continues to grow into adulthood. However, even primary-age students benefit from developmentally appropriate metacognitive instruction, particularly strategies for recognizing confusion and asking for help.
Adults, including pre-service teachers, frequently hold inaccurate beliefs about effective study strategies. A 2013 review by John Dunlosky and colleagues found that two of the most popular student study habits, re-reading and highlighting, have "low utility" relative to strategies like practice testing and distributed practice, yet students overwhelmingly prefer the less effective methods.
Classroom Application
Structured Reflection Routines
Regular, low-stakes reflection builds metacognitive habit. Exit tickets that ask "What was the muddiest point in today's lesson?" or "Rate your confidence in explaining this concept to a classmate" do more than surface comprehension data for the teacher — they train students to self-assess in the moment.
A middle school science teacher might close each lab with a three-question protocol: What did I understand well today? Where did I get stuck? What would I do differently next time? Over a semester, this routine builds students' vocabulary for describing their own cognitive states and their comfort with honest self-evaluation.
Pre-Task Planning and Post-Task Review
Before a significant assignment, ask students to write down: what strategies they plan to use, what they predict will be difficult, and how long they estimate it will take. After the assignment, return these predictions and ask students to compare their plan with what actually happened.
A high school history teacher assigning a document-based question might have students complete a planning sheet before writing: "I will use evidence from which documents? What counterargument do I expect to address? Where do I think I'll struggle most?" After grading, students review their predictions alongside the teacher's feedback. The comparison, where they overestimated or underestimated their performance, is itself the learning.
Modeling Uncertainty in Subject Content
Across all grade levels, teachers can make metacognition concrete by narrating genuine intellectual uncertainty during instruction. When working through a math problem on the board, a teacher might say, "I tried this approach and got stuck, so I'm switching to working backwards." This models that monitoring one's strategy and adjusting course is normal expert behavior, not a sign of failure.
For younger learners, framing metacognitive prompts with accessible language, "Does this answer make sense? How do you know?", builds the same underlying habit in age-appropriate terms.
Research Evidence
The Education Endowment Foundation's 2018 meta-analysis, drawing on over 50 studies, found metacognition and self-regulation interventions produce an average of seven additional months of learning progress compared to students who do not receive such instruction. The effect is consistent across age groups and subject areas, and the cost to implement is low relative to other high-impact strategies.
John Hattie's 2009 synthesis of over 800 meta-analyses in Visible Learning placed metacognitive strategies among the most powerful influences on student achievement, with an effect size of approximately 0.69 — well above the 0.40 threshold Hattie used to designate meaningful impact. Hattie's data suggested that what students think about their own learning predicts achievement as reliably as what teachers do instructionally.
Deanna Kuhn and David Dean's 2004 work in Psychological Science demonstrated that explicit metacognitive training, specifically, having students predict their performance and then compare predictions to actual results, improved both science reasoning and students' calibration of their own knowledge. Students who went through the intervention were more accurate about what they knew and, crucially, more willing to revise incorrect beliefs.
A 2012 review by Yvette Harris and Jamillah Graham found metacognitive instruction particularly beneficial for low-achieving and disadvantaged students. Students who begin with weaker academic backgrounds show the largest gains from metacognitive training, partly because high-achieving students often develop informal metacognitive strategies independently. The implication is that explicit instruction is most important for students who need it most, it reduces, rather than widens, achievement gaps.
Mixed results appear in studies of metacognitive training delivered outside the context of subject-area learning. Generic study skills programs that teach metacognition in isolation from content show smaller effect sizes than embedded instruction. The skill needs to be practiced in the domain where it will be used.
Common Misconceptions
Metacognition is just reflecting at the end of a lesson. End-of-lesson reflection is one metacognitive tool, but metacognition operates continuously during learning, not only afterward. The monitoring component — catching confusion as it happens, is often more valuable than post-task review. Teaching students to pause mid-problem and check whether their strategy is working produces different outcomes than asking them to reflect only at the end.
Metacognitive students know more content. Metacognition is about how students think, not how much they know. A student can have excellent metacognitive awareness about a domain she knows little about, she knows she knows little, she knows what kinds of questions to ask to learn more, and she knows which strategies will help her understand. Content knowledge and metacognitive skill develop in parallel but are not the same thing. A student can be highly knowledgeable but have poor metacognitive accuracy (overconfident), or highly metacognitively aware but still building content knowledge.
Metacognition develops naturally with age and does not need to be taught. While some metacognitive capacity develops through normal cognitive maturation, research is clear that explicit instruction accelerates development substantially, particularly for regulatory skills. Adolescents and adults do not automatically become accurate judges of their own learning. Dunlosky's 2013 review found that university students consistently choose less effective study strategies, not from laziness but from genuine metacognitive inaccuracy. Without instruction, many students never discover that their instinctive approach to studying is inefficient.
Connection to Active Learning
Metacognition is both enabled by and necessary for active learning. Active learning methodologies require students to do something with knowledge — argue, apply, explain, connect, and that doing surfaces the gaps in understanding that metacognitive monitoring then processes. The relationship is circular: more active engagement produces more metacognitive data, and stronger metacognitive awareness makes active engagement more productive.
Chalk Talk is a silent discussion protocol where students respond to a central question in writing, building on each other's posted ideas. The silence and anonymity give students unusual space to notice what they actually think, versus what they would say aloud to manage social impression. As students read others' contributions, they naturally compare those ideas to their own, a form of spontaneous self-assessment. Teachers can amplify this metacognitive dimension by asking students, after the chalk talk, to identify which posted idea most challenged their prior thinking and why.
Save the Last Word structures discussion so that one student speaks last after her peers have responded to a text passage she selected. The preparation phase, choosing a passage and drafting a final response without knowing what others will say, requires students to anticipate disagreement and clarify their own reasoning in advance. This planning-under-uncertainty is metacognitive work: students must ask themselves what they actually think and why, and commit to it before getting social confirmation.
Hexagonal Thinking asks students to arrange concepts on hexagons and explain the connections between adjacent tiles. The physical act of moving hexagons and articulating connections makes thinking visible, to the student, to peers, and to the teacher. Students frequently discover mid-arrangement that a connection they assumed was obvious is harder to articulate than expected. That moment of difficulty is metacognitive data: it tells the student where their understanding is surface-level.
These methods connect naturally to Bloom's Taxonomy. Metacognition lives primarily at the higher levels of Bloom's revised framework, the ability to evaluate and create requires the learner to assess the quality of her own thinking, not just produce it. Critical thinking and metacognition are deeply intertwined: critical thinkers question their own reasoning and recognize cognitive biases, which are metacognitive acts. Self-regulated learning incorporates metacognition as its core control mechanism, learners who regulate their own behavior do so by drawing on metacognitive awareness of what is and is not working.
Sources
- Flavell, J. H. (1979). Metacognition and cognitive monitoring: A new area of cognitive-developmental inquiry. American Psychologist, 34(10), 906–911.
- Brown, A. L. (1978). Knowing when, where, and how to remember: A problem of metacognition. In R. Glaser (Ed.), Advances in Instructional Psychology (Vol. 1, pp. 77–165). Erlbaum.
- Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students' learning with effective learning techniques: Promising directions from cognitive and educational psychology. Psychological Science in the Public Interest, 14(1), 4–58.
- Quigley, A., Muijs, D., & Stringer, E. (2018). Metacognition and Self-regulated Learning: Guidance Report. Education Endowment Foundation.