Metacognition: why teaching students to think about their learning matters more than ever

That is the promise of metacognition - a word that may sound complex but refers to something very practical: helping students take control of their own thinking. It is about building learners who can plan their approach, monitor their progress, and adjust when they encounter difficulties.

In short, metacognition is learning how to learn by thinking about thinking. In today’s classrooms, especially as artificial intelligence (AI) becomes part of how students access and interact with knowledge, this skill has never been more important.

What is metacognition?

Metacognition means being aware of and managing your own thinking. A metacognitive student asks:

• What do I already know about this topic?
• How will I tackle this task?
• Is my approach working, or do I need to change it?

These habits may seem like common sense, but they often do not develop on their own. The good news is that they can be taught.

Research from several hundred randomised controlled trials shows that students explicitly taught metacognitive strategies make significantly more progress, especially those from disadvantaged backgrounds. They become more independent, confident, and better equipped to handle challenges. In our Lifelong Learners programme, supporting Palestinian refugees in Jordan, we have seen just how powerful this can be. This programme helps refugees build a wide range of life skills (from financial literacy to job applications and CV writing) embedded in an English language approach. Students were taught to think in this way, not only in individual lessons but throughout the programme, building reflections into a learning journal.

What does the evidence say?

Based on the randomised controlled trial evidence, the Education Endowment Foundation (EEF) (a UK-based organisation that reviews international education research) ranks metacognition and self-regulated learning as one of the most effective and affordable.

These ideas are grounded in how the brain learns and functions. As Richard and colleages noted in the chapter on metacognition in their book, Neuroscience for Teachers:

Left to its own devices, the brain would happily loiter at the bottom of the hierarchy of its abilities, relying on mental shortcuts wherever possible.

Getting thinking to higher levels of cognition requires effort and literally more brain energy.

What does it look like in practice?

The EEF recommends a seven-step approach that teachers can use in any subject:

1. Activate prior knowledge: help students recall what they already know.
2. Teach strategies explicitly: show how to plan, monitor, and evaluate.
3. Model your own thinking: let them hear your thought process as you solve a problem.
4. Support memorisation: reinforce strategies until they become familiar.
5. Use guided practice: work with students as they try out strategies.
6. Shift to independent practice: let them take the lead.
7. Encourage reflection: ask what worked, what did not, and why.

This is not just general good teaching. It is about being deliberate: making visible the mental processes that strong learners use, then helping all students build those habits for themselves. Critically, target subject-specific learning goals, teaching children to think at each stage of the process.

In practice, this can look as follows:

1. Planning - before the task
   Students ask:
   • What do I already know about this topic?
   • What is the goal, and which strategy fits best?
   
   Example: In maths, choosing between methods to solve an equation.
   
   

2. Monitoring - during the task
   Students ask:
   • Is this working?
   • Do I understand, or should I change approach?
   
   Example: In history, recognising that an argument is drifting off-topic.
   
   

3. Evaluation - after the task
   Students ask:
   • What worked, what did not, and why?
   • How can I improve next time?
   
   Example: In art, evaluating use of techniques against original intent.
   
   

This builds independence by making thinking visible, structured, and purposeful.

What about AI?

Artificial intelligence tools (such as ChatGPT and others) are now part of how many students learn. These tools can write essays, answer questions, and explain complex topics.

Used well, AI can be incredibly helpful. But it can also encourage passivity. It can provide quick answers without real understanding. It can present incorrect information with confidence. It can bypass learning instead of supporting it.

That is where metacognition becomes essential. A metacognitive learner does not simply accept what AI says. They ask:

• Does this match what I have learned?
• Do I trust the sources behind this answer?
• What do I still need to understand better?

Teachers can model this in real time:

“Let us check if this AI response is accurate based on what we know from class.”
“Watch how I rephrase the question to get a better answer.”
“What could this response be missing?”

This kind of thinking can transform AI from a shortcut into a partner in learning.

Why does metacognition matter now?

In a recent speech to Southeast Asian education leaders at the 2025 CRPN conference, Tony shared striking results from the PISA 2022 global survey. Students were asked whether they:

• Asked for help when they did not understand something;
• Linked new problems to things they had learned before.

In both cases, fewer than half responded positively.

That tells us something important: metacognition is not happening enough. As AI tools become more widespread, the risk of students outsourcing their thinking will only increase.

Equipping students and teachers with the power of metacognitive strategies

Metacognition is not a luxury. It is one of the most powerful tools we have to help students succeed - not just in school, but throughout life. And it is especially important now, in a world where AI offers fast answers without deep understanding.

This includes equipping teachers to explicitly teach metacognitive strategies, skills that underpin student self-regulation, problem-solving, and long-term success. Prioritising this in professional development ensures the workforce is prepared to embed metacognition systematically across the curriculum.

When students learn how to learn, they do not just do better on tests. They become more confident, more reflective, and more prepared for the challenges of the future.

And that is something no AI can replace.