Metacognition: The Thinking Skill That Decides Whether You Trust the AI or Yourself

Metacognition is the ability to think about your own thinking: to monitor, evaluate, and deliberately regulate your own cognitive processes rather than simply being run by them. It’s a genuinely powerful tool for self-awareness and better decision-making, and it’s become more important, not less, as more of daily decision-making now happens in partnership with AI systems that sound confident whether or not they actually are.

What Metacognition Actually Is

Metacognition, often described as “thinking about thinking,” refers to the awareness and control of our own cognitive processes. The term was introduced by psychologist John Flavell in 1979, and it involves two main components. Knowledge of cognition means understanding what you know and how you learn, your own genuine strengths and weaknesses in processing information and making decisions. Regulation of cognition means the ability to plan, monitor, and evaluate your own cognitive activity: setting goals, choosing strategies, and adjusting course based on real feedback rather than habit.

In simple terms, metacognition means being aware of how you think, recognising your own patterns, and consciously steering them toward a better outcome. It’s a higher-order thinking skill, and one that keeps proving useful in genuinely new contexts as the world changes around it.

Why It Matters for Resilience

Resilience is the capacity to recover quickly from real difficulty, and metacognition plays a genuinely vital role in that process. Facing a real challenge, thoughts can either propel you forward or hold you back, and metacognition is what lets you examine those thoughts critically, catch an unhelpful pattern, and deliberately replace it with something more constructive.

If you catch yourself ruminating on a failure, metacognition is what lets you notice the loop and actually intervene, shifting focus toward what you’ve genuinely learned rather than staying trapped in self-doubt. That shift matters directly for resilience: it’s what keeps a setback from becoming a spiral. Metacognition also directly supports emotional regulation, since being aware of your own emotional response, and what specifically triggered it, is what lets you manage that response deliberately rather than being carried by it during a genuinely stressful moment.

This isn’t just a personal-development claim. A 2024 study by Bajaj, Jain, Singh, and Bajaj, examining employees working in teams, found individual metacognitive ability directly predicts workplace performance, not just personal learning outcomes, with the regulation component (actively monitoring and adjusting your own thinking) mattering more than simply knowing your own cognitive strengths and weaknesses in the abstract.

Metacognition in Leadership

For most people, metacognition is a private skill. For leaders, it’s also a public one. A leader’s thinking rarely stays inside their own head, because the people around them pick up on it directly.

Neuroscience shows the brain doesn’t simply react to events. It predicts what’s about to happen and prepares the body in advance. A leader walking into a meeting carrying a threat-laden forecast from a difficult week isn’t just in a poor mood. Research on emotional contagion shows a group genuinely absorbs that state, usually without anyone consciously noticing it happen. Stress responses synchronise, and the team starts scanning for danger instead of opportunity.

This has direct consequences for psychological safety, one of the strongest known predictors of team performance. Safety isn’t something a leader establishes once. It’s a signal sent continuously, and every interaction either confirms or undermines a team’s genuine expectation that it’s safe to speak up.

Metacognition is how leaders take real responsibility for that signal. Before a significant meeting, decision, or difficult conversation, it helps to pause and ask three questions directly: What am I predicting will happen here? What is that prediction actually based on, and how recent is that information? What would I need to be wrong about for a better outcome to become possible?

Senior leaders have a particular reason to practise this deliberately. Research on self-awareness suggests experience and power tend to reduce it rather than increase it, since leaders receive progressively less honest feedback the higher they rise. Metacognition is the internal check that compensates for the missing external one.

The New Frontier: Metacognition and AI

A genuinely new dimension of metacognition has emerged directly alongside AI adoption. When an AI system offers a recommendation, a forecast, or a confident-sounding answer, the real question isn’t just “is this right?” It’s “how accurately can I judge whether this is right, and do I actually know the difference?” Current research on human-AI collaboration is explicitly studying this as a metacognitive skill: the accuracy of a person’s confidence judgments about their own performance, and increasingly, about an AI system’s performance too.

This connects directly to automation bias, the well-documented tendency to accept an automated system’s output while discounting contradictory information, particularly under time pressure or after repeated positive experience with a tool. Strong metacognition is the actual safeguard against this: it’s what lets someone notice they’re deferring to a confident-sounding output rather than genuinely evaluating it. As AI keeps taking on more of the world’s routine cognitive load, the specific skill of accurately monitoring your own thinking, and knowing when to trust it over a machine’s, and when to trust the machine over yourself, is becoming a genuinely new, practical application of a decades-old psychological concept.

Building Metacognition Deliberately

Reducing the Risk of Cognitive Bias

Cognitive biases are common, genuine pitfalls that quietly undermine resilience, and metacognition is what helps you catch and counteract them. Slow your thinking down deliberately: Daniel Kahneman’s Nobel-winning work on fast and slow thinking found rapid, reflexive judgment reliably produces more bias than a genuinely paced decision does. Deliberately consider the opposite of what you currently believe or plan to do, a well-documented debiasing technique that surfaces flaws in your own reasoning and directly reduces the pull of confirmation bias. And balance real optimism with real realism: metacognition isn’t about becoming pessimistic, it’s about staying resilient without drifting into complacency, holding both the risk and the opportunity in view at once.