Three rules for effective practice

The hidden curriculum of practice: how habits, as well as knowledge, shape learning

Practice is the bridge between knowing and doing. It’s how knowledge turns into skill and skill into mastery. But practice is also where learning most often goes wrong. We assume that practice, by its very nature, leads to improvement. It doesn’t. Repetition alone merely stabilises whatever already exists. Practice does not make perfect, it makes permanent, and therein lies the risk.

What is practice?

The late K. Anders Ericsson’s research on deliberate practice showed that expertise depends less on time spent than on how that time is used.¹ Experts focus on specific weaknesses, practise with clear goals, and receive immediate feedback. The same principle should hold true in school. Effective practice targets the right thing at the right level of challenge and repeats it under conditions that make skilled performance inevitable.

Writing provides a perfect example. Students must juggle ideas, vocabulary, syntax, punctuation, and spelling simultaneously, an unnatural cognitive load. Unless they automate the basics, such as capitalisation or sentence construction, the act of writing overwhelms working memory and these things predictably slip and writng is strewn with countless basic errors. Fluent writers don’t think harder about commas; they no longer need to.

Knowledge problems vs. practice problems

We often confuse knowledge problems with practice problems. Knowledge problems are solved by instruction; practice problems are solved by perfecting the conditions of practice.

If a student doesn’t understand what a metaphor, or cell division is, that’s a knowledge problem. The solution is teaching: explanation, modelling, and examples. But if, for instance, a student knows that French adjectives usually follow the noun, yet continues to write le bleu ciel instead of le ciel bleu, that’s a practice problem. They don’t need reteaching, they need retraining. The difficulty isn’t misunderstanding the rule, it’s failing to apply it automatically. Each time the wrong order slips by unchecked, it becomes further embedded. Only through repeated correction and guided rewriting does the right pattern take hold.

Take capital letters. Every pupil in secondary school can tell you when to use one, but many still write “london” or “i went.” The problem isn’t ignorance but habit. They’ve practised carelessness. Each time unproofed work is accepted the wrong behaviour is reinforced. What’s required isn’t another lesson on proper nouns, but relentless, corrective rehearsal, rewriting until the correct version feels natural.

The same pattern is obvious in sport. I’m currently relearning how to run. Video analysis has revealed a catalogue of faults: my legs cross over as I stride, my foot lands too far in front of my body, and I rely too much on my quads instead of driving with my glutes and hamstrings. The result is that every step becomes a tiny act of braking. None of this feels wrong; in fact, it feels natural because it’s what I’ve practised. My poor running form is effortless. Correcting it, by contrast, demands conscious focus and feels awkward. Yet that discomfort is the point. Improvement lies in replacing an old, automatic pattern with a new one, repeated so often that it becomes second nature. My hope is that through slow, deliberate retraining, the better form will eventually feel as easy as the bad form currently does.

Knowledge problems demand explanation; practice problems demand directed repetition. The teacher’s task is to tell the difference; to diagnose which kind of problem it is, and then design appropriate responses. If we get that diagnosis wrong, we risk either pointlessly reteaching what’s already known or letting bad habits harden unchecked. Get it right, and every act of practice becomes purposeful. Which brings us to three simple rules for getting practice right…

1. Don’t let students practise making mistakes

Every act of repetition strengthens a neural pathway. If what’s repeated is wrong, the brain faithfully encodes the error. Before practice begins, accuracy must be guaranteed.

In writing, that means never allowing students to independently produce extended text before they’ve rehearsed correct forms through modelling and guided construction. Sentence combining, gap-fill scaffolds, or shared writing sessions ensure that the first attempts are accurate, not approximate. Only then can we seek to integrate these routines into independent practice.

The Capital Letter Problem illustrates this perfectly. Students know the rule but fail to apply it consistently because they’ve been allowed to practise failure. The solution is immediate correction and enforced redrafting: no unproofed work is accepted. Over time, the habit of correctness takes hold. Teachers find it almost impossible to write a lower-case proper noun because they’ve over-practised doing it correctly. The same reflex can be built in students, but only through deliberate insistence.

In mathematics, the distinction between knowledge and practice problems appears just as sharply. A student may know that when multiplying two negative numbers the result is positive, yet still write -3 × –4 = –12. The error isn’t conceptual, they can often explain the rule aloud, but habitual. When working quickly, their hand reverts to the more familiar pattern that “a negative with a negative stays negative.” They don’t need another explanation of integer rules; they need retraining through guided rehearsal: slowing down, checking signs, and repeating correct examples until accuracy becomes automatic.

2. Do less for longer

Coverage masquerades as progress. Racing through new content gives the illusion of learning, but durable understanding depends on depth of rehearsal. Mastery arises not from exposure but from sustained, varied repetition. This is where most classrooms go wrong. The rhythm of school life rewards movement - new topics, new units, new assessments - so both teachers and students come to equate pace with progress. But durable, flexible learning is slower, more repetitive, and usually less dramatic.

Cognitive psychologists have long known that mental effort is finite. When students are flooded with too many new ideas, they experience cognitive exhaustion: the working memory becomes saturated, attention fragments, and errors multiply. Kirschner, Sweller and Clark (2006) argue that overloading learners with unstructured novelty leads not to deeper understanding but to cognitive thrashing, the mental equivalent of spinning one’s wheels in mud.² Once students are cognitively spent, no amount of additional exposure helps. In fact, it harms, as their practice becomes increasingly confused

The antidote is to do less for longer. Instead of cramming more material into limited time, we allow rehearsal to consolidate what’s already been introduced. When content is revisited through deliberate variation - different contexts, slightly altered problems, restated formulations - the schema grows richer, more flexible, more resilient. The paradox is that by slowing down, learning speeds up.

This also reframes the relationship between success and struggle. For too long, schools have been caught between the slogans of “productive struggle” and “supportive success,” as if the two were mutually exclusive. But they exist on a continuum. Too much struggle breeds frustration and avoidance; too little leads to complacency. The optimal zone lies where success feels earned, where challenge stretches but doesn’t break. Early accuracy matters not because it’s easy, but because it builds the confidence needed to sustain harder practice later. As Ericsson found, mastery emerges through repeated cycles of small success followed by stretch.³

The same pattern underpins the NHS’s Couch to 5K running programme. It doesn’t ask beginners to run a full 5K on day one; it asks them to run for one minute, walk for ninety seconds, and repeat. Each week, the intervals shift slightly; the bouts of running lengthen, the walking shortens, and the body adapts. Progress comes not from heroic effort but from consistent, structured, incremental strain. The success of the app lies in its calibration: it keeps you just uncomfortable enough to grow, never so exhausted that you give up.

Rather than throwing students into full essays and hoping they’ll develop fluency through exposure, it starts with short bursts of controlled practice - one sentence type, one rhetorical move, one grammatical structure - repeated until automatic. Only then do students extend to paragraphs, and later, full essays. The sequence mirrors effective physical training: gradual overload, deliberate recovery, and the steady conversion of effort into ease.

Doing less for longer is not indulgent. It’s efficient. The student who spends weeks mastering one high-leverage skill will outpace the one who skims ten. As with running, so with writing: progress comes not from variety or volume, but from consistent, purposeful repetition that leaves just enough room for struggle to turn into strength.

In English, this means focusing on a small set of high-leverage skills - sentence variety, cohesion, clarity - and practising them until they become instinctive. Rather than writing full essays weekly, students might spend several lessons perfecting a single analytical sentence pattern: claim, evidence, inference. They combine, rephrase, and manipulate it until fluency emerges.

In science, it might mean returning repeatedly to “energy transfer” across units - cells, forces, ecosystems - so that the schema becomes robust. In maths, it might mean practising one problem type (e.g., expanding brackets) until it is automatic, rather than dabbling across ten.

Doing less for longer allows accuracy to harden into fluency. Only when the mental effort drops can students redirect attention to higher-order thinking.

3. Over-practise: from “I can” to “I can’t not”

Students typically stop practising once they can perform a skill once or twice. Experts stop only when they can’t not perform it. Research on overlearning explains why. Driskell, Willis and Cooper (1992) found that continued practice after initial success significantly improves retention, especially under stress. Rohrer and Taylor (2006) showed that overlearning, spaced over time, makes recall and performance far more durable.⁴

Professional musicians provide perhaps the clearest illustration of overlearning. A concert pianist doesn’t stop practising a piece once they can play it correctly from start to finish. They continue long past that point; repeating difficult passages until every movement is effortless, every transition seamless, every potential hesitation eliminated. The goal is not merely accuracy but automaticity. When performance anxiety, fatigue or distraction strike on stage, there’s no space for conscious calculation; the body must know what to do without thinking. Violinists, for instance, isolate a troublesome bar and loop it hundreds of times at half speed before gradually returning to tempo. Singers rehearse scales until pitch and breath control align instinctively. Orchestral players revisit the same repertoire year after year, not because they’ve forgotten it, but to keep the neural pathways active and the physical movements smooth.

This is overlearning in action: practice continued beyond mastery until correctness becomes the default state. The same principle applies in classrooms. Whether pupils are solving equations, conjugating verbs, or writing analytical sentences, the aim isn’t to do it once but to do it so consistently that success survives pressure and distraction, just as a violinist’s bow finds the string perfectly, every time.

In writing, overlearning means continuing until success is inevitable. A class that can produce one accurate analytical paragraph continues practising until every student can produce five in a row, under timed conditions, without losing precision. A history teacher might have pupils explain a causal chain (“Because X, therefore Y, which led to Z”) until they can’t write it incorrectly. In PE, drills continue past competence until correct technique survives fatigue.

This is appropriate pressure.⁵ Too little, and habits decay; too much, and students become anxious. The teacher’s role is to apply just enough pressure for fluency to emerge, then withdraw it to allow autonomy.

---

The problems of practice are not problems of motivation but of design. Without structure, feedback and repetition, practice merely rehearses mediocrity. With them, it becomes what Ericsson called “a lifelong process of purposeful improvement.”⁶

Don’t apply pressure until students get it right. Apply pressure until they can’t get it wrong.

Good practice doesn’t just make students better, it makes them permanently better. It turns knowledge into habit, accuracy into fluency, and effort into ease. Practice is not something that happens after learning. It is learning, repeated and applied until fluent skill emerges.

1. Ericsson’s work on deliberate practice began with his 1993 paper, “The Role of Deliberate Practice in the Acquisition of Expert Performance” (Psychological Review, 100:3, 363–406), co-authored with Krampe and Tesch-Römer, and was expanded in his 2006 edited volume, “The Cambridge Handbook of Expertise and Expert Performance.” His central claim is that expert ability is not innate but the result of sustained, purposeful practice characterised by four features: (1) tasks designed to improve performance, (2) opportunities for immediate feedback, (3) repetition under controlled conditions, and (4) progressive refinement of goals. Ericsson’s theory directly challenged talent-based explanations of expertise, showing instead that high-level performance depends on structured environments where individuals push beyond their current competence. Later critics, such as Hambrick and Macnamara (2014), have nuanced this view, arguing that deliberate practice explains much, but not all, of performance variance. Still, the evidence remains clear that how we practise determines the ceiling of what we can achieve.

2. Kirschner, P. A., Sweller, J., & Clark, R. E. (2006). Why minimal guidance during instruction does not work: An analysis of the failure of constructivist, discovery, problem-based, experiential, and inquiry-based teaching. On pp. 77–80, they describe how novice learners “lack the schemas” to manage unstructured problem solving, resulting in “heavy working-memory load and consequent interference with learning.” The specific phrase cognitive thrashing is a paraphrase of their description of what happens when working memory is overtaxed by unguided exploration.

3. Ericsson, K. A., Krampe, R. T., & Tesch-Römer, C. (1993), “The Role of Deliberate Practice in the Acquisition of Expert Performance,” Psychological Review, 100(3), 363–406. Ericsson and colleagues describe expertise as the product of iterative cycles of practice: repeated success at a given level followed by deliberate stretch to a slightly harder one. Each loop consolidates what has been mastered before increasing the challenge, ensuring that performance remains at the edge of competence. See also Ericsson, K. A., & Pool, R. (2016),* Peak: Secrets from the New Science of Expertise, which develops this model into the broader principle that mastery arises from “repeated adaptation to slightly greater demands.”*

4. The concept of “overlearning” dates back to early experimental psychology but was systematised in the meta-analysis by Driskell, Willis and Cooper (1992), “Effect of Overlearning on Retention,” Journal of Applied Psychology, 77(5), 615–622. Reviewing forty-five independent studies, they found that continued practice after initial mastery significantly improves long-term retention, particularly under stress or time pressure. Overlearning strengthens the stability of performance, making it less vulnerable to interference or anxiety. Rohrer and Taylor (2006) extended this work to educational settings, showing that overlearning combined with spacing - revisiting material after delay - produces far more durable learning than either alone (“The Effects of Overlearning and Distributed Practice on the Retention of Mathematics Knowledge,” Applied Cognitive Psychology, 20, 1209–1224). Their findings suggest that the apparent redundancy of continuing to practise after success is, in fact, the point: fluency is secured not when we can perform a task once, but when we can no longer perform it incorrectly.

5. The idea of “appropriate pressure” in practice aligns with findings from motor learning and performance psychology, which show that moderate stress enhances focus and consolidation, while excessive pressure impairs performance. Ericsson and Pool (2016) describe this as a defining feature of deliberate practice: tasks must be challenging enough to stretch current ability but not so difficult as to overwhelm it (Peak: Secrets from the New Science of Expertise, London: Bodley Head). Similarly, Yerkes and Dodson’s classic “Law of Arousal” (1908) demonstrated an inverted-U relationship between pressure and performance: too little produces complacency, too much creates anxiety. In both sport and education, effective practice depends on calibrating this tension: enough demand to disrupt comfort, enough safety to sustain effort.

6. Peak, p. 98. Ericsson describes purposeful practice as involving clear goals, sustained attention, feedback, and continual adjustment, all directed toward extending performance beyond current limits. In contrast to routine experience, which stabilises existing habits, deliberate practice reshapes them through intentional refinement over time.