School of Learnerhood | The Learning Loop — Notice, Question, Model, Practise, Feedback, Update

This article follows the School of Learnerhood Master Map. It does not replace the canonical pages for Re-entry and Repair, Responsibility Transfer or the Six Learning Modes. It gives those controls a common learning process.

One-Sentence Definition

The Learnerhood Learning Loop is the recurring process through which a person notices a gap or possibility, forms a question and working model, practises against the world, receives feedback and updates what to believe or do next.

Position: Information Is an Input; Learning Is a Change

A lecture can be excellent and still fail to produce learning in one student. A short conversation can produce a durable change in another.

The difference is not simply the amount of information delivered. Learning requires a change in the learner’s usable model, skill, judgement or behaviour.

This is why the loop begins before instruction and continues after it.

1. Notice — Something Does Not Fit

Learning often begins with a signal: confusion, surprise, error, curiosity, changed demand, repeated friction or a new opportunity.

The signal may be external—a wrong answer, failed task or new system—or internal: “I can follow this example but I cannot explain it,” “this result does not match what I expected,” or “I no longer know how this tool works.”

Without noticing, the learner may keep repeating an obsolete or incomplete model.

2. Question — What Exactly Is the Gap?

A vague question creates vague repair.

“I do not understand maths” is too broad. “I understand how to expand brackets but do not recognise when factorisation is useful” is more actionable.

Question quality improves when the learner can separate concept, recognition, execution, transfer, context and uncertainty.

3. Model — Build the Best Current Explanation

A model is the learner’s current representation of how something works.

It can be a concept, procedure, diagram, causal explanation, checklist, story, formula, rule of thumb or mental simulation.

The model does not need to be perfect before practice begins. It needs to be clear enough to test.

4. Practise — Make the Model Meet Reality

Practice turns an internal model into observable performance.

The learner solves, writes, explains, builds, decides, rehearses or performs. Useful practice should match the capability being developed rather than simply create more activity.

Where possible, practice begins below catastrophic consequence and grows in realism as capability stabilises.

5. Feedback — What Did the World Return?

Feedback can come from a teacher, tutor, mentor, peer, answer key, measurement, customer, experiment, system response or the consequence of the action itself.

Feedback is useful when it helps distinguish what worked from what failed and why.

A score alone may be too compressed. “You lost marks because the method was correct but the algebraic execution failed at this step” gives a more repairable signal.

6. Update — Change the Model or the Action

Learning completes a loop when feedback changes what happens next.

The learner may strengthen the same model, narrow a claim, replace a method, seek a missing prerequisite, change the environment, recruit help, reduce confidence or increase it.

Without update, feedback becomes merely judgement rather than learning information.

Diagnosis: Where the Loop Can Break

• No notice: the learner does not detect error or changing conditions.
• No question: the problem remains global and unlocatable.
• No model: the learner memorises fragments without understanding relationships.
• No practice: understanding remains untested.
• No useful feedback: the learner sees success/failure but not the mechanism.
• No update: the same error repeats despite evidence.

Mechanism: The Loop Changes Shape Across Life

In Infanthood, much of the loop is carried by the environment and caregiver. In Childhood, questions, play and feedback become increasingly visible. In Schoolhood, teachers organise deliberate practice and assessment. Teenhood increasingly requires the learner to interpret social and academic feedback independently. Universityhood asks for self-directed questions and disciplinary evidence. Adulthood makes feedback less neatly packaged. Elderhood may require adaptation and re-entry into changed systems.

The loop persists even while the institution changes.

Practice: A Worked Learning Example

A student repeatedly loses marks in algebra.

• Notice: errors cluster in unfamiliar questions.
• Question: is the problem concept, recognition, execution or transfer?
• Model: the student can perform the method when told which method to use.
• Practise: use mixed problems requiring method selection.
• Feedback: identify whether the student chooses correctly and where execution fails.
• Update: if recognition improves but signs remain inaccurate, the next repair shifts from transfer to execution.

The loop prevents one weak score from becoming one vague diagnosis.

Practice: An Adult Technology Example

• Notice: a familiar banking task is difficult in a redesigned app.
• Question: is financial judgement weak, or is the interface unfamiliar?
• Model: identify the invariant task—recipient, amount, verification, confirmation.
• Practise: rehearse the new interface with a low-stakes example.
• Feedback: note where navigation breaks.
• Update: teach the new interface convention rather than reteaching the whole financial task.

AI and the Learning Loop

AI can assist every part of the loop: surfacing questions, proposing models, generating practice, providing feedback and suggesting revisions.

That assistance creates a control risk. If the AI performs all diagnosis, modelling and checking, the learner may receive a good output while acquiring little transferable capability.

The ownership test is: after using the tool, can the learner still explain what changed, why the answer is credible and what to do when the next case differs?

Boundary: Feedback Is Not Always Truth

Teachers can be wrong. Peers can be wrong. Metrics can optimise the wrong target. AI can fabricate. Institutions can reward compliance rather than capability.

Mature learnerhood therefore evaluates the feedback source as well as the learner’s response to it.

Repair: When Practice Produces Repetition but Not Improvement

More repetition can strengthen a wrong model.

A useful repair is:

stop the repeated task → inspect the current model → locate the error → obtain discriminating feedback → rebuild one component → practise with variation → retest transfer.

Transfer and Measurement: When Has the Loop Produced Learning?

The loop has produced stronger learning when the learner can perform after support reduces, recognise the capability in a changed context, explain the model, detect error earlier and continue updating without waiting for the original instructor.

The purpose of feedback is not to keep the learner permanently inside the loop with one teacher. It is to help the learner become increasingly able to run the loop.

Evidence Context

The loop is eduKateSG’s reader-facing process model rather than a claim that all learning science can be reduced to six stages. It is compatible with broader evidence supporting active learning, feedback, metacognition, retrieval, practice and iterative updating while preserving the important differences among domains and learners.

Next Route

The next Learnerhood foundation asks who should increasingly operate the loop: Self-Direction and Metacognition — Knowing What You Know, What You Do Not, and What to Do Next.

---

School of Human Life | Stage 4 Continuity | School of Learnerhood. Checked 24 August 2026.