Most of us were taught what to learn throughout school – facts, formulas, dates, definitions. But very few of us were ever explicitly taught how to learn. This gap matters more than it might seem. In a world where knowledge evolves rapidly and careers shift unpredictably, the ability to acquire new skills and knowledge efficiently – what researchers and educators call “learning to learn” – is arguably the most valuable competency a person can develop. It is not just a study trick; it is a foundational life skill that sits at the heart of lifelong education.
Table of Contents
Learning to learn: what it means and why it matters
At its core, “learning to learn” refers to the observation that prior experience with learning tasks can improve how efficiently a person learns in the future. When you practise learning in a particular domain – or even across different domains – you gradually build mental strategies and frameworks that make future learning faster and more effective. Psychologists have also referred to this process as deutero-learning, learning set formation, and transfer of learning, all pointing to the same underlying phenomenon: learning itself is a skill that improves with practice.
The significance of this concept is not limited to the classroom. UNESCO’s landmark 1996 report, Learning: The Treasure Within, chaired by Jacques Delors, identified four foundational pillars of education for the 21st century: learning to know, learning to do, learning to live together, and learning to be. Embedded within the very first pillar – learning to know – is the explicit acknowledgement that education must go beyond transmitting content. It must also mean learning to learn, so as to benefit from the opportunities education provides throughout life. In other words, UNESCO placed the capacity to learn as a prerequisite for everything else education is meant to achieve.
Early experiments: how the idea was discovered
The scientific study of “learning to learn” has roots going back further than most people realise. One of the earliest formal conceptualisations came from the anthropologist and systems thinker Gregory Bateson in 1942. Bateson coined the terms proto-learning and deutero-learning to describe a pattern he had observed both in humans and animals: when a subject is exposed to a series of similar learning tasks, they do not just learn each task – they progressively get better at learning such tasks. As Bateson noted, the experimental subject “becomes more and more skilled in the solving of problems.” He later renamed these as Learning I and Learning II, with the latter referring specifically to learning how to learn more efficiently over time.
However, it was the American psychologist Harry F. Harlow whose empirical work gave the concept its most influential scientific form. In his landmark 1949 paper, “The Formation of Learning Sets,” published in the Psychological Review, Harlow described a series of experiments conducted with rhesus monkeys at the University of Wisconsin-Madison. He presented the monkeys with a series of object-discrimination problems – each time, they had to pick the correct object from a pair to receive a food reward. Critically, each problem involved entirely new objects.
At first, the monkeys learned each new problem gradually, making errors before finding the solution. But as they worked through more and more problems, something striking happened. By around the two-hundredth problem, the monkeys could solve a brand-new discrimination task almost perfectly after just a single trial – effectively figuring out the rule with a single attempt. Harlow described this as the formation of a learning set: the monkeys had not just learned how to solve individual problems; they had learned how to learn that type of problem. They had internalised a general strategy – what Harlow described as a “win-stay, lose-shift” rule – and could now apply it instantly to novel situations.
What is a learning set?
The term learning set, as Harlow used it, refers to the strategies, dispositions, and mental frameworks that a learner develops through repeated exposure to similar types of problems. It is not content knowledge – it is procedural and strategic knowledge about how problems of a certain type are solved. Once a learning set is formed, the learner no longer needs to go through the full trial-and-error process every time. They bring pre-formed strategies to new problems, dramatically reducing the time and errors involved in acquiring new skills.
This idea resonates strongly with what modern cognitive science calls metacognition – the capacity to think about one’s own thinking and learning. Metacognition is awareness and control of thinking for learning, and students with strong metacognitive skills are positioned to learn more and perform better than peers who are still developing their metacognition. Just as Harlow’s monkeys built up strategies over repeated problems, learners who develop metacognitive awareness build a toolkit of strategies they can deploy strategically and flexibly across different learning contexts.
The structure of metacognitive knowledge
Researchers have broken down metacognitive knowledge into three distinct types, each of which contributes to a learner’s capacity to learn effectively:
- Declarative knowledge – knowing about yourself as a learner: your strengths, weaknesses, and what learning strategies exist.
- Procedural knowledge – knowing how to use specific learning strategies.
- Conditional knowledge – knowing when and why to apply a particular strategy, depending on the task at hand.
Together, these three forms of knowledge allow a learner to plan their approach before tackling a task, monitor their progress during it, and evaluate their performance afterwards. Experts possess more knowledge that is better organised and integrated than novices, but they also have highly developed metacognitive skills – they are more aware of themselves as learners and regularly reflect to understand why their chosen strategy is working or not. This mirrors exactly what Harlow observed in his monkeys: the more experienced problem-solvers did not just have more practice – they had developed a reflective, strategic approach to new problems.
How learning sets develop through experience
A key insight from Harlow’s work and from decades of subsequent research is that learning sets are not innate – they are built up through deliberate experience. The more diverse the range of problems a learner encounters, the richer and more flexible their learning set becomes. This is why breadth of experience matters as much as depth. A student who has only ever studied in one way or in one subject is likely to struggle when confronted with an unfamiliar type of learning task, while a student who has been exposed to varied problem-solving demands is better equipped to adapt.
Educators play a direct role in nurturing this development. Research on pre-service teacher education has identified authentic and relevant assessment tasks, prior communication of instructions and evaluation criteria, and frequent performance-focused feedback as key components for promoting lifelong learning and metacognitive development. In other words, classrooms that give students structured opportunities to reflect on how they are learning – not just what they are learning – actively develop the learning sets that will serve them well beyond the classroom.
Practical strategies to develop the ability to learn
Research consistently identifies several practical approaches that help learners build a robust learning set over time:
- Self-testing and retrieval practice: Rather than passively re-reading material, actively recalling information strengthens memory and reveals gaps in understanding, sharpening self-assessment skills.
- Reflective journaling: Writing about what worked and what did not after a study session or exam builds the habit of analysing one’s own learning process – a core metacognitive skill.
- Interleaving practice: Mixing up different types of problems or topics during study, rather than blocking them, has been shown to improve long-term retention and the ability to transfer learning to new contexts.
- Seeking feedback and acting on it: Students with well-developed metacognition can identify concepts they do not understand, select appropriate strategies, implement those strategies, evaluate their effectiveness, and adjust their plans based on outcomes. Feedback – from teachers, peers, or self-assessment – is the fuel for this cycle.
- Thinking about the “why”: When working through a problem, pausing to ask “why am I doing this step?” and “how does this connect to what I already know?” promotes deeper processing and accelerates the formation of transferable learning strategies.
The lifelong value of learning to learn
One of the most important implications of Harlow’s learning set research – and of the broader body of work on metacognition – is that the benefits compound over time. A learner who has deliberately practised learning across a variety of contexts in their early years arrives at new challenges with a far richer strategic toolkit than one who has simply accumulated content knowledge. This is precisely the spirit behind UNESCO’s vision of lifelong learning, which envisages a society where everyone can learn according to their individual needs and interests, anywhere and anytime, in an unrestricted, flexible, and constructive way.
That vision is not idealistic fantasy – it rests on a very concrete psychological reality. Just as Harlow’s monkeys became near-instant problem-solvers after sufficient practice, human learners who are consistently challenged to reflect on and refine their learning strategies gradually become faster, more adaptable, and more confident acquirers of new knowledge and skills. The capacity to learn is not fixed at birth or sealed in childhood. It is, as the evidence firmly shows, a skill in its own right – one that can be taught, practised, and steadily improved throughout a lifetime.
Teachers, therefore, bear a dual responsibility: to transmit knowledge, yes, but also to model and cultivate the process of learning itself. A classroom that treats learning as a transparent, reflective activity – rather than a hidden magic trick performed by naturally “smart” students – does something more powerful than deliver content. It equips learners with the most portable skill they will ever carry: the ability to keep learning, whatever life brings next.
What do you think? If schools were to explicitly teach “how to learn” alongside subject content, which strategies do you believe would make the greatest difference for students – and why? And in your own experience as a learner or educator, at what point did you first become aware of your own learning process?
References
- https://unevoc.unesco.org/home/UNESCO+Publications/lang=en/akt=detail/qs=5426
- https://www.unesco.org/en/articles/reworking-four-pillars-education-sustain-commons
- https://link.springer.com/article/10.1007/BF02382299
- https://www.researchgate.net/figure/A-Harlow-1949-trained-monkeys-to-choose-one-of-two-objects-within-a-pair-by_fig2_364319133
- https://pmc.ncbi.nlm.nih.gov/articles/PMC8734377/
- https://tll.mit.edu/teaching-resources/how-people-learn/metacognition/
- https://www.frontiersin.org/journals/education/articles/10.3389/feduc.2022.879238/full
- https://campus.kennesaw.edu/faculty-staff/cetl/teaching-resources/teaching-essentials/metacognition-key-self-directed-learning.php
- https://uil.unesco.org/fileadmin/keydocuments/LifelongLearning/en/UNESCOTechNotesLLL.pdf
Leave a Reply