What Is Metacognition?
Metacognition—literally "thinking about thinking"—represents one of the most significant frontiers in cognitive science research. The term, coined by developmental psychologist John Flavell in the late 1970s, refers to the awareness and understanding of one's own thought processes. Unlike standard cognition, which involves processing information, metacognition involves monitoring and controlling that processing.
Research demonstrates that metacognitive abilities predict academic performance more reliably than traditional intelligence measures. A landmark study by Schraw and Dennison (1994) developed the Metacognitive Awareness Inventory, which revealed that skilled learners consistently outperform their peers not through superior memory or intelligence, but through superior awareness of their own cognitive processes. Students in the top quartile of metacognitive awareness score an average of 0.75 standard deviations higher on achievement tests than those in the bottom quartile—a substantial effect that traditional IQ measures fail to capture.
The distinction matters because metacognition is trainable. Unlike fluid intelligence, which remains relatively stable across adulthood, metacognitive skills can be deliberately cultivated through specific practices and interventions. This discovery has profound implications for education, professional development, and self-directed learning throughout the lifespan.
Flavell's Metacognition Model
John Flavell's (1979) foundational model established the architecture of metacognition as comprising two primary components: metacognitive knowledge and metacognitive experiences. This framework remains the foundation for contemporary research and intervention design.
Metacognitive knowledge encompasses what individuals know about cognition generally—beliefs about how the mind operates, assumptions about learning processes, and understanding of which strategies work under which conditions. Metacognitive experiences, by contrast, are the conscious awareness that emerges during learning: the feeling of knowing, the sense that something doesn't make sense, the recognition that you're losing the thread of an argument.
Flavell's model emphasized that these components interact dynamically. When metacognitive experiences reveal gaps or problems, they activate metacognitive knowledge, which in turn generates strategies to address the difficulty. This cyclical process—experience triggering knowledge triggering action—forms the core of self-regulated learning.
Knowledge Components: Person, Task, and Strategy
Schraw and Dennison (1994) expanded Flavell's framework by identifying three critical knowledge categories that metacognitively skilled learners possess:
Declarative knowledge involves knowing about oneself as a learner—what your strengths and weaknesses are, what kinds of tasks you find challenging, what your attention span permits before fatigue sets in. Research by Veenman and colleagues (2006) found that learners with accurate declarative knowledge selected appropriate difficulty levels 67% more often than those with inaccurate self-perceptions.
Procedural knowledge involves knowing how to execute learning strategies—the steps involved in summarization, the procedures for self-testing, the mechanics of spaced repetition. Crucially, procedural knowledge alone is insufficient; the same strategy applied without metacognitive monitoring often fails.
Conditional knowledge—the most underdeveloped in novice learners—involves knowing when and why to apply particular strategies. Conditional knowledge enables learners to match their approach to the demands of the task. Studies of expert learners consistently show they possess rich conditional knowledge networks that guide strategy selection in context-sensitive ways.
Regulation and Control: Monitoring Your Learning
Metacognitive regulation refers to the deliberate actions learners take to monitor, evaluate, and adjust their cognitive processes. Nelson and Narens (1990) developed a influential framework describing the feedback loop between cognitive processing at the "object level" and metacognitive monitoring at the "meta level."
The monitoring component involves ongoing assessment of learning progress. This includes:
- Judgment of learning (JOL): estimating the likelihood of remembering something later. Dunlosky and Metcalfe (2009) demonstrated that learners' JOL accuracy—measured as calibration between confidence and actual performance—varies dramatically across individuals and contexts, with expert learners showing calibration errors less than half the magnitude of novice learners.
- Feelings of knowing (FOK): the sense that certain information is accessible even when it cannot currently be retrieved. This sensation guides strategic decisions about how long to persist with retrieval attempts versus seeking external information.
- Confidence judgments: post-retrieval assessments of answer correctness that inform subsequent study decisions.
The control component involves adjusting cognitive strategies based on monitoring information. When monitoring reveals confusion, skilled learners respond by re-reading, seeking additional examples, switching strategies, or allocating additional study time to the problematic material.
Comprehension Monitoring Strategies
One of the most practically important metacognitive skills involves monitoring reading comprehension. Research by Pressley and colleagues (1992) identified several evidence-based comprehension monitoring strategies:
Self-questioning involves generating questions about text content and verifying answers during subsequent reading. In controlled studies, students trained in self-questioning showed 34% better retention of expository text compared to control groups reading identical material without the intervention.
Paraphrasing and elaboration require learners to restate ideas in their own words, revealing gaps in understanding. A classic study by Pressley and colleagues found that when students encountered the statement "The impact was softened by the cushions," those who spontaneously paraphrased as "The cushions softened the impact" showed superior encoding and later recall.
Bridge-building involves actively connecting new information to prior knowledge. Markman (1979) demonstrated that readers who spontaneously detected inconsistencies between new information and their existing knowledge showed dramatically better comprehension than those who failed to notice contradictions—even when intelligence and reading ability were controlled.
Coherence monitoring involves tracking whether material is forming a coherent mental representation. When local coherence breaks down—sentences that don't connect to prior context, unexplained terminology—skilled readers recognize the problem and take corrective action, whether through re-reading, inference, or external help-seeking.
Self-Regulation in Learning: The Research
Zimmerman's (1989) self-regulated learning model describes how expert learners systematically approach acquisition of new skills and knowledge. His research program, spanning decades and thousands of participants, identified consistent patterns distinguishing self-regulated learners from their less effective counterparts.
Self-regulated learners exhibit characteristic forethought processes before engagement: they analyze task demands, activate relevant prior knowledge, set specific proximal goals, select appropriate strategies, and maintain high self-efficacy beliefs about their capability to succeed. Their performance phase involves deliberate self-observation during execution—monitoring strategy effectiveness, tracking progress toward goals, noting emotional states that might interfere with performance. Their self-reflection phase involves evaluating outcomes against goals, attributing success or failure to controllable factors, and adjusting approaches for future attempts.
The research demonstrates that these patterns predict academic outcomes powerfully. A meta-analysis by Dignath and Büttner (2008) examining 39 intervention studies found that metacognitive instruction improved student performance by an average effect size of 0.74 standard deviations across all subject domains and grade levels. Particularly striking was the finding that effects were sustained at follow-up assessments conducted weeks or months after intervention ended.
Notably, self-regulated learning capacity increases with age and experience, but the trajectory depends heavily on instruction and practice. Children as young as five can be taught basic metacognitive monitoring skills, with developmentally appropriate interventions showing effect sizes of 0.5 or higher in elementary populations.
Practical Protocols for Metacognitive Training
Translating research into practice requires structured protocols that can be implemented consistently. The following approaches have strong empirical support:
The PLAN Protocol (Karpicke & Blunt, 2011): Before reading new material, spend five minutes actively recalling what you already know about the topic—this forces assessment of prior knowledge readiness. During reading, pause every 15-20 minutes to summarize what was learned in your own words without looking at the text. After reading, create a brief retrieval practice quiz for yourself and test recall before reviewing.
Comprehension Monitoring Checklist (Schraw, 1998): During reading or learning, periodically ask: "Can I state the main idea in my own words? Can I give an example not in the material? Do I have adequate background knowledge for this? Does this connect logically to what came before? What is still confusing?"
The 2-2-2 Review Method: After initial learning, review within 2 hours, then within 2 days, then within 2 weeks. At each review, explicitly assess: "What can I now do that I couldn't do before? What remains unclear? What strategy should I use to address remaining gaps?"
Strategy Selection Training: Before beginning a study session, explicitly write down: the specific learning goal, the strategy you intend to use and why you believe it's appropriate, how you will monitor success, and what you will do if monitoring reveals the strategy isn't working.
Measuring Your Metacognitive Growth
Assessing metacognitive development requires measuring both knowledge and regulatory skill. The Metacognitive Awareness Inventory (MAI; Schraw & Dennison, 1994) provides validated subscales for knowledge of cognition, regulation of cognition, and monitoring accuracy. While designed for research contexts, its items can inform self-assessment.
Practical indicators of metacognitive growth include: increased accuracy in predicting what you will and won't remember, earlier detection of comprehension failures, more strategic allocation of study time based on perceived difficulty, and greater flexibility in strategy selection when initial approaches fail.
The goal of metacognitive development isn't perfection—it's calibrated awareness that enables continuous improvement. Skilled metacognitive learners don't eliminate confusion or mistakes; they recognize and address them efficiently when they occur. This capacity for self-diagnosis and self-correction distinguishes learners who continue growing throughout their lives from those who plateau early.
Metacognition represents a fundamental shift in how we conceptualize learning: from passive absorption of information to active orchestration of cognitive processes. The research is unambiguous that this shift is possible, trainable, and consequential for outcomes across every domain of human endeavor.