The simple definition
Metacognition, coined by developmental psychologist John Flavell in 1979, refers to cognition about cognition — or more precisely, the ability to monitor and regulate your own mental processes.
It has two core components: metacognitive knowledge (what you know about how you think, what you're good at, where you tend to fail) and metacognitive regulation (your ability to actively control your thinking — planning, monitoring, and evaluating your performance in real time).
Why it matters more than IQ
Fluid intelligence — raw mental processing power — is largely determined by genetics and peaks in your late twenties. Metacognition is different: it is trainable at any age, and the effects are large.
A landmark meta-analysis by de Boer, Donker, and van der Werf (2018) reviewed 67 studies and found metacognitive training produced an effect size of g = 0.63 on academic performance — larger than homework, larger than class size reduction, and comparable to one-on-one tutoring. Hidayat et al.'s controlled study found an even larger ES ≈ 1.11 when training was structured and sustained.
But the benefits extend far beyond academic performance. Normann and Morina (2021) conducted a meta-analysis comparing metacognitive therapy (MCT) to Cognitive Behavioural Therapy (CBT) — the gold standard for anxiety and depression. Metacognitive therapy outperformed CBT with an effect size of g = 0.69. The mechanism: MCT targets how you relate to your thoughts rather than the content of those thoughts.
Why it matters more than EQ
Emotional intelligence became a cultural phenomenon in the 1990s — but the scientific case for it has always been weaker than the hype. Metacognition, by contrast, is quietly outperforming it in every category that matters: effect size, trainability, and durability.
A 2004 meta-analysis by van Rooy and Viswesvaran covering 69 studies found EQ predicted job performance at ρ = 0.23 — a modest result. More critically, studies by Locke (2005) and Schulte et al. (2004) found that after controlling for general cognitive ability and the Big Five personality traits, EQ's incremental predictive power dropped to near zero — accounting for roughly 1–2% of additional variance at most. The 2010 meta-analysis by Joseph and Newman confirmed this pattern: EQ's apparent predictive power largely reflects its overlap with existing personality and intelligence measures, not a distinct construct.
Metacognition does not have this problem. De Boer et al. (2018) found that metacognitive training predicts performance improvements above and beyond baseline cognitive ability — the gains are not just a repackaging of IQ. This is because metacognition operates at a different level: it is not a trait you possess but a process you execute, and it governs how you deploy whatever raw intelligence and emotional awareness you have.
The trainability gap is equally stark. Mattingly and Kraiger's 2019 meta-analysis of EQ training programmes found an average effect size of d ≈ 0.35 — and noted that effects typically decay rapidly after training ends, with few programmes producing durable change. Metacognitive training, by contrast, consistently shows larger and more persistent effects: the de Boer et al. effect of g = 0.63 comes from studies measuring outcomes weeks and months after training concluded.
The deeper reason metacognition outperforms EQ is structural. Emotional regulation is itself a metacognitive process. Adrian Wells' Metacognitive Model (2009), supported by decades of clinical data, shows that emotional distress — anxiety, rumination, depression — is driven not by emotions themselves but by beliefs about thinking: the conviction that worrying is protective, that ruminating is necessary to solve a problem, that intrusive thoughts must be controlled. These are metacognitive beliefs, and they are what Metacognitive Therapy targets directly. This is why MCT outperformed CBT at g = 0.69 in Normann and Morina's meta-analysis — it treats the operating system, not just the output.
Put simply: EQ tells you what you feel. Metacognition determines what you do with it. You can have high emotional intelligence and still ruminate compulsively, overestimate your certainty, fail to update your beliefs when evidence changes, and make systematically poor decisions under pressure. Metacognitive skill is the layer above EQ that decides whether any of those emotional signals get used effectively.
The two layers of metacognition
Metacognitive knowledge is what you know about your own cognition:
- Person knowledge — how your cognition compares to others, what conditions you think best in, which subject areas you're actually strong in versus which you merely feel confident about
- Task knowledge — knowing what makes a problem easy or hard for you, how different question types tax your thinking differently
- Strategy knowledge — knowing which mental approaches work for which kinds of problems
Metacognitive regulation is what you do with that knowledge in real time:
- Planning — selecting strategies before tackling a problem
- Monitoring — tracking your comprehension and progress as you think
- Evaluating — assessing how well your approach worked and what to change
The metacognition deficit hiding in plain sight
Most people have a severely distorted model of their own cognition. Research on the “” (Pronin, 2008) shows that people consistently believe their own reasoning is less biased than that of others — even when shown direct evidence to the contrary.
Studies of show that when people say they are “99% certain” on general knowledge questions, they are wrong 40% of the time (Fischhoff et al., 1977). Most people have never received calibrated feedback on their — they have no accurate model of where their knowledge ends and guesswork begins.
This is the gap MindFrame closes. Every session produces precise, quantified feedback on the distance between your self-assessment and your actual performance — the measurement that makes metacognitive improvement possible.
How metacognition is trained
The research is clear on what works:
- Immediate, precise feedback on the accuracy of your confidence estimates (not just whether you got the answer right)
- Spaced repetition — distributed practice across many sessions rather than massed learning in a single sitting
- Error analysis — structured review of errors to identify patterns in your thinking
- Strategy instruction — explicit training in specific metacognitive strategies, not passive exposure
- Transfer prompts — connecting training insights to real-world decisions
MindFrame is built around all five. Your Brier Score measures probability calibration. Your Calibration Error tracks confidence accuracy. Your mode breakdown reveals which types of thinking you overestimate. Your Composite Score tracks improvement across sessions. And every session ends with a reflection prompt to consolidate what you learned.
Metacognition across the lifespan
Metacognitive ability develops throughout childhood and adolescence — children younger than 8 have very limited metacognitive monitoring — and continues to develop with deliberate practice into adulthood.
Critically, it does not decline with age the way fluid intelligence does. Older adults often have richer metacognitive knowledge than younger adults — they have a more accurate model of their strengths and limitations, even if processing speed has declined. This is partly why experience remains valuable even as raw cognitive speed falls.
What MindFrame trains — specifically
MindFrame targets five components of metacognition that research identifies as most impactful and most trainable:
- Calibration — the match between confidence and accuracy
- Reasoning quality — the logical structure and validity of your arguments
- Bias recognition — identifying cognitive biases in your own and others' thinking
- Working memory utilisation — effective use of limited attentional resources
- Belief updating — adjusting your views in proportion to evidence
Each of MindFrame's 14 game modes targets a specific combination of these components, with challenges at three difficulty levels, AI-evaluated reasoning, and precise performance metrics — so you always know exactly where you stand and what to work on next.