“What is the reference whole?”
Maya hears that question so often in tuition that she can almost predict when it is coming.
At first, the tutor asks it.
Then the tutor pauses and looks at her.
Then Maya begins saying it aloud before the tutor does.
Weeks later, nobody says anything.
Maya reads a percentage question, pauses for half a second, circles the starting quantity and writes the calculation correctly.
The words are no longer spoken.
But the question is still there.
It has moved.
From tutor.
To student.
From external prompt.
To self-question.
From visible scaffold.
To internal routine.
That movement is one of the most important transitions in good tuition.
A tutor prompt has finished its job when the learner begins asking the useful question without the tutor.
That is Prompt Migration.
Quick Read: Prompt Migration in One Sentence
Teach a small, task-specific metacognitive prompt explicitly, use it when the learner needs it, gradually shift responsibility for asking it from tutor to student, fade the external cue, and verify that the learner can self-trigger the question in fresh, delayed and authentic work.
This article owns a narrow learning mechanism.
The Escalation Packet asks what the learner should bring when help is needed.
The Help Gradient asks how much help should enter.
The Help Ledger asks which cognitive jobs still came from outside.
Prompt Migration asks:
How does a useful external cue become the learner’s own self-question?
1. External Prompts Are Borrowed Executive Functions
When a tutor asks:
“What is the question asking?”
“What do you know already?”
“Which strategy fits?”
“Does this answer make sense?”
the tutor is doing more than speaking.
The tutor is temporarily supplying a planning, monitoring or evaluating function.
The question directs attention.
It interrupts autopilot.
It activates prior knowledge.
It asks the learner to inspect strategy.
It can trigger checking or recovery.
In the early stage, that executive function is partly external.
The destination is not permanent external prompting.
The destination is internal control.
2. Prompt Migration Is a Form of Internal Scaffolding
Current evidence guidance on metacognition and self-regulated learning uses a helpful phrase:
internal scaffolding.
At first, the learner may need explicit teaching, modelling, guided practice and verbal or written prompts.
Over time, those supports can become habitual ways of planning, monitoring and evaluating learning.
The physical scaffold disappears.
The cognitive function remains.
Prompt Migration is the small-scale mechanism by which this can happen.
3. Prompt Migration Is Not Just “Ask Yourself Questions”
Generic advice is easy to give.
“Think about your thinking.”
“Ask yourself questions.”
“Be metacognitive.”
These statements are too broad to control a live problem.
Useful self-questioning is normally task-linked.
“What is the reference quantity?”
“What word in the question controls the answer form?”
“What is the outermost operation?”
“What observation must my Science explanation account for?”
These prompts are concrete enough to change behaviour and general enough to travel across related tasks.
That is the design target.
4. The Three Main Prompt Families: Plan, Monitor, Evaluate
Metacognitive research commonly organises self-regulation around planning, monitoring and evaluation.
The same structure is useful for tuition.
Planning prompts help before or at the beginning of a task.
- What is the task?
- What do I already know?
- What strategy fits?
- What resources will I need?
- What should I watch out for?
Monitoring prompts operate during performance.
- Does this step still make sense?
- Am I answering the actual question?
- What evidence supports this?
- Has my method created unnecessary complexity?
- Am I repeating a failed move?
Evaluation prompts operate after or at checkpoints.
- Does the answer fit the conditions?
- What worked?
- Where did the first weak point appear?
- What would I change next time?
- Can I reproduce this without the support I used?
A mature learner needs prompts in all three families.
5. Recovery Prompts Deserve Their Own Family
School work is not always a clean sequence from plan to success.
Students get stuck.
They discover contradictions.
They forget.
They panic.
They need a restart routine.
Useful recovery prompts include:
- Where is the last line I still trust?
- What changed immediately before the error?
- Can I test a simpler case?
- What are the two most plausible rules I may be mixing?
- What is the smallest help I need?
- Can I skip, mark and return?
These questions convert “I’m stuck” into action.
Prompt Migration is especially valuable here because examination environments require recovery without a tutor.
6. A Good Prompt Is Short Enough to Fire Under Pressure
A prompt can be educationally beautiful and practically useless if it is too long.
Imagine asking yourself during an examination:
“What prior learning strategies have I historically applied to semantically equivalent but structurally different mathematical representations, and what does that imply about my present selection architecture?”
No.
Better:
“What structure do I see?”
The best prompts are often compressions of much larger teaching ideas.
A short question should reactivate the larger model.
7. A Good Prompt Does Not Contain the Answer
“Use elimination.”
That is not a metacognitive prompt if method selection is the learning job.
It performs the decision.
Better:
“Which method keeps the numbers simplest?”
“Use the original value.”
Again, perhaps too directive.
Better:
“What is the comparison measured from?”
A migratable prompt triggers a cognitive operation rather than supplying its result.
8. A Good Prompt Has a Trigger Condition
Knowing a prompt is not enough.
The learner must know when to use it.
Prompt:
“What is the reference quantity?”
Trigger:
a comparison, percentage change, rate or relative measure appears.
Prompt:
“What evidence?”
Trigger:
the learner makes a claim about a text or phenomenon.
Prompt:
“What is the outermost operation?”
Trigger:
a nested or composite mathematical expression appears.
Prompt Migration succeeds only when both question and trigger migrate.
9. The Prompt Migration Ladder
A practical migration sequence is:
Stage 1 — Tutor asks.
The question is fully external.
Stage 2 — Tutor asks, learner explains why.
The learner begins understanding the function of the prompt.
Stage 3 — Tutor pauses at the trigger.
The learner is expected to recall the prompt.
Stage 4 — Learner says the prompt aloud.
External tutor language becomes student self-talk.
Stage 5 — Learner uses a short written cue.
A checklist, margin symbol or one-word reminder supports self-scaffolding.
Stage 6 — External cue disappears.
The learner self-triggers in fresh tasks.
Stage 7 — Prompt compresses.
Full verbal question becomes a fast internal check.
Stage 8 — Prompt generalises.
The learner adapts the metacognitive function across related problems and subjects.
This is an explanatory teaching framework, not a validated scale.
10. Stage 1: The Tutor Must Model the Question Properly
Migration cannot begin from a bad prompt.
The tutor should explain not only the question, but why the question matters.
“When a percentage question appears, we need to know what quantity the comparison is measured against. That is why I ask: what is the reference whole?”
The student now learns the prompt as strategy, not ritual.
This matters because metacognitive strategies are more likely to become flexible when learners understand how, when and why they help.
11. Stage 2: The Learner Must Explain the Prompt’s Job
Ask:
“Why do I keep asking you that?”
If the learner says:
“Because you always do,”
the prompt has not migrated conceptually.
If Maya says:
“Because percentage mistakes often come from dividing by the wrong reference quantity,”
the strategy is becoming explicit.
Understanding the prompt’s job supports transfer beyond one worksheet.
12. Stage 3: Replace the Question With a Pause
This is a simple but powerful move.
The familiar trigger appears.
The tutor does not ask the familiar question.
The tutor waits.
If the learner self-prompts, the migration has begun.
If not, the tutor can provide the cue.
The pause creates a retrieval opportunity.
It also produces diagnostic information:
does the student recognise that this is the moment the prompt belongs?
13. Stage 4: Make Self-Talk Audible
Students often need a visible bridge between tutor talk and silent internal thought.
Ask them to say the prompt.
“What evidence?”
“What is the structure?”
“Does this answer fit the question?”
This is not childish.
Thinking aloud makes invisible regulation observable.
It lets the tutor hear whether the student’s self-question is useful, correctly timed and correctly interpreted.
Over time, external speech can become quieter self-talk.
14. Stage 5: Use Written Prompts as Temporary Bridges
A small written scaffold can help when spoken prompting is fading.
At the top of a Mathematics page:
Read → Structure → Choose → Execute → Check
On an English comprehension bookmark:
Question → Evidence → Inference → Scope
On a Science answer card:
Observation → Mechanism → Outcome
These can be useful intermediate scaffolds.
But they should not become permanent wallpaper.
If the learner never performs without the card, the scaffold has not migrated.
15. Stage 6: Remove the Prompt Deliberately
Support rarely disappears by accident.
If the tutor keeps asking the useful question forever, the student may continue waiting for it forever.
So remove it.
Fresh question.
No verbal cue.
No prompt card.
No chapter heading revealing the method.
Watch.
If the learner self-triggers, the migration is real.
If not, restore the smallest useful scaffold and continue training.
16. Stage 7: Prompts Compress Into Fast Checks
Expert self-regulation is often fast.
A learner no longer needs to verbalise:
“What is the nature of this task, what strategies are available, and what resources do I need?”
They may glance at the problem and think:
“Structure?”
Then choose well.
Compression is not loss of metacognition.
It can be evidence that the routine has become efficient.
The important question is whether the compressed check still fires at the right moment.
17. Stage 8: The Learner Starts Designing Their Own Prompts
This is a higher level of self-regulated learning.
The student notices a recurring failure and invents a useful question.
Hana keeps writing beyond the scope of comprehension questions.
She creates:
“Answer only what was asked?”
Ethan repeatedly expands expressions too early.
He creates:
“Preserve structure first?”
Jia Jun writes Science answers that jump from condition to outcome.
He creates:
“What is the causal middle?”
Now metacognition is not merely taught.
It is generative.
18. The Prompt Trigger Is More Important Than the Prompt Wording
Students sometimes memorise a beautiful checklist and still fail to use it.
Why?
The missing capability is not prompt knowledge.
It is prompt activation.
The learner knows:
“check units.”
But nothing triggers the check at the end of the problem.
Prompt Migration must therefore train cue–action links.
When I see a final numerical answer → check units.
When I make a textual claim → find evidence.
When two methods seem plausible → ask what feature discriminates them.
When my working becomes much longer than expected → inspect structure.
Migration is prompt plus trigger.
19. Prompts Should Be Subject-Specific Before They Become General
“Monitor your understanding.”
Too abstract for many learners.
In Mathematics:
“Does this operation preserve equality?”
In English:
“Which word in the text proves that inference?”
In Science:
“What mechanism connects condition to outcome?”
In A-Math:
“What is the outermost structure?”
These prompts live inside the knowledge domain.
Only later may the learner abstract a broader principle such as:
“Before acting, identify the structure that determines the method.”
20. Do Not Teach Metacognition as a Detached Thinking Lesson
Prompt Migration works best when embedded inside real subject work.
Students do not need another worksheet called:
“Thinking About Thinking.”
They need to ask useful questions while solving Mathematics, reading English, explaining Science and writing essays.
The point is not reflection for its own sake.
The point is better action inside authentic learning.
21. Prompt Migration Can Fail Through Over-Prompting
A tutor creates twenty excellent prompts.
The student now has another problem:
which prompt should I use?
Prompt libraries can become cognitive clutter.
Prefer a small number of high-value questions tied to recurring decision points.
One good prompt that fires reliably is more useful than twelve clever prompts the learner cannot retrieve.
Prompt systems should compress over time.
22. Prompt Migration Can Fail Through Permanent Tutor Timing
The tutor always asks the prompt at exactly the right moment.
The learner becomes excellent at answering it.
But the learner never learns when to ask it.
This creates hidden dependence.
The content of the question migrated.
The timing did not.
Therefore prompt fading must remove both wording and external timing.
23. Prompt Migration Can Fail Through Answer-Shaped Questions
“Shouldn’t you use the original value?”
Technically a question.
Functionally an answer.
“Wouldn’t Product Rule be better?”
Again, answer-shaped.
Migratable prompts should preserve a meaningful decision for the learner.
“What is the comparison measured from?”
“What is the outermost operation?”
These questions guide attention without resolving the choice prematurely.
24. Prompt Migration Can Fail Through Checklist Ritual
A learner mechanically ticks:
read question.
check units.
show working.
But the checks do not actually change behaviour.
This is metacognitive theatre.
A prompt should cause inspection.
“Check units” should mean:
identify expected dimensions, compare with final answer, repair if inconsistent.
The prompt is useful only if the underlying cognitive operation is real.
25. Prompt Migration Can Fail Through Prompt Dependence
A student refuses to start unless the prompt card is present.
The scaffold has become a gatekeeper.
This is not a reason to remove it abruptly.
Use graded fading.
Full question.
Then one keyword.
Then a symbol.
Then no visible prompt.
Each stage asks the learner to carry more of the activation internally.
26. Use the Help Ledger to Decide Which Prompt Must Migrate
The Help Ledger says:
interpretation — independent.
method selection — tutor prompted.
execution — independent.
checking — parent prompted.
Now we know exactly what needs Prompt Migration.
Not the whole subject.
Two triggers.
Method selection.
Checking.
This precision prevents generic independence training.
27. Use the Help Gradient to Choose the Migration Starting Point
If a learner currently needs a full explanation, the target is not yet self-prompting.
The knowledge must first exist.
If a learner succeeds after one orienting question, prompt migration may be ready.
The Help Gradient therefore tells us how close a support is to internalisation.
Large support may need teaching.
Small recurring cue may need migration.
28. Use the Exit Ramp to Retire the External Prompt
Prompt Migration is one mechanism inside The Exit Ramp.
The tutor once asks the question.
The learner later asks it.
Eventually the external question disappears.
That is a transfer of ownership.
A good exit plan should identify recurring tutor prompts and ask:
Which of these should become student self-prompts before support reduces?
29. Use the Transfer Gate to Prove Migration
A learner self-prompts in the exact worksheet where the prompt was trained.
Good.
Now change the context.
Different wording.
Mixed topic.
Delayed return.
Time pressure.
School paper.
The Transfer Gate asks whether the prompt has become portable regulation rather than worksheet-specific ritual.
30. Use the Flight Recorder to Track Prompt Dependence Over Time
January:
tutor asks “what evidence?” every time.
February:
Hana asks it aloud.
March:
Hana underlines evidence without speaking.
April:
school comprehension shows evidence selection independently.
This is a meaningful trajectory.
The Flight Recorder keeps the history visible.
31. Prompt Migration Is Different From Memorising a Mnemonic
A mnemonic stores content or sequence.
A metacognitive prompt controls cognitive activity.
The two can overlap.
But they are not identical.
“PEEL” may remind a student of paragraph components.
“What does this evidence prove?” asks the learner to inspect reasoning.
The latter can regulate quality even after the visible acronym disappears.
Prompt Migration is about internal control, not merely memory aids.
32. Prompt Migration Is Different From Generic Scaffolding Fade
Scaffolding can include worked examples, templates, visual organisers, partial solutions and tutor modelling.
Prompt Migration owns one specific subset:
questions and cues that regulate the learner’s own planning, monitoring, checking and recovery.
A writing frame may fade because the learner no longer needs the structure visible.
A prompt migrates when the learner begins performing the regulation internally.
33. Prompt Migration Is Different From Automaticity
Automaticity often means executing a familiar process with little conscious effort.
Prompt Migration concerns regulatory triggers.
The learner may calculate automatically but still need a conscious prompt to choose when calculation is appropriate.
Or a self-prompt may itself become highly compressed and automatic.
The important distinction is:
automatic execution is not the same as automatic self-regulation.
34. Primary 1: Start With Tiny Prompt Pairs
Young learners do not need abstract metacognitive vocabulary.
Use tiny questions.
Before reading:
“What sound first?”
After writing:
“Does it end?”
Before adding:
“How many altogether?”
The tutor models.
The child repeats.
Later the child starts doing the check without the question being spoken.
Prompt Migration can begin very early.
35. Primary 2: Build “Before I Ask” Self-Questions
Ethan raises his hand quickly.
The tutor teaches three self-questions:
- What is the question asking?
- What part do I know?
- Where exactly am I stuck?
These prompts support better Escalation Packets.
They also slow premature handover.
The child learns that asking for help begins with inspecting one’s own state.
36. Primary 3 and 4: Science Needs Causal Self-Questions
Science answers often fail in the middle.
Condition.
Then outcome.
No mechanism.
Teach:
“What is happening in the middle?”
Or:
“What must the particles be doing?”
At first Jia Jun hears the tutor ask.
Later he asks himself before submitting.
The prompt becomes a quality-control routine.
37. Primary 5: Build a Small Prompt Toolkit Before PSLE Volume
Primary 5 is a good time to establish a compact self-questioning toolkit.
English:
“What exactly was asked?”
Mathematics:
“What is the relationship?”
Science:
“What is the causal middle?”
General recovery:
“Where is the last step I trust?”
Keep the toolkit small.
Train it repeatedly in subject contexts.
By Primary 6, these should increasingly fire without adult prompting.
38. Primary 6: Prompt Migration Must Survive Full Papers
Topical worksheets help teach prompts.
Full papers test whether they self-trigger.
No chapter heading.
No tutor standing beside the desk.
No parent saying “check.”
The learner must recognise the cue in the task itself.
That is why PSLE preparation should not only practise answers.
It should practise self-triggered regulation under mixed conditions.
39. Secondary 1: Prompt Migration Protects the Independence Transition
Secondary 1 increases organisational and academic independence at the same time.
Prompts therefore need to include executive routines:
- What is due first?
- What needs my highest attention?
- What can I finish independently?
- What should I package for help?
- What evidence shows I am done?
Parents may initially ask these questions.
The destination is the student asking them.
Prompt Migration applies to organisation as well as subject cognition.
40. Secondary 2: Strategy Selection Needs Self-Questions
As topics accumulate, the challenge shifts from executing known methods to choosing between them.
Useful prompts:
- What representation exposes the relationship?
- Which method reduces unnecessary work?
- What feature rules out the alternative?
- What result should I expect roughly?
These questions help build adaptive expertise rather than a larger pile of procedures.
41. Secondary 3: Additional Mathematics Needs Structural Self-Questioning
A-Math is dense with structural decisions.
Teach prompts such as:
- What is the outermost operation?
- Is there nesting inside it?
- Which identity is suggested by this form?
- Am I preserving useful structure or expanding blindly?
- What restrictions exist before I simplify?
The aim is not to make the student narrate all five questions forever.
The aim is to build a fast structural scan.
Prompt Migration turns explicit questions into mathematical habits.
42. Secondary 4: Prompts Must Survive Examination Speed
A prompt that works only when the learner has five quiet minutes is not examination-ready.
Under speed, prompts need compression.
“What is the command word?”
“Structure?”
“Evidence?”
“Units?”
“Scope?”
Practise these during timed work until they no longer create disruptive overhead.
Self-regulation must become fast enough to coexist with performance.
43. JC: The Learner Should Maintain Their Own Prompt Library
By JC, a mature learner can keep a small personal library built from actual failure patterns.
Not fifty generic study tips.
Five questions with high leverage.
For example:
- What assumption am I making?
- What would falsify this interpretation?
- What is the cleanest representation?
- Where is the first line I no longer trust?
- What would I do if the obvious method were unavailable?
These prompts should evolve as the learner changes.
A prompt that solved last year’s weak link may eventually be retired.
44. English: Prompt Migration for Comprehension
Hana answers beyond the scope of the question.
The tutor repeatedly asks:
“What exactly was asked?”
Eventually Hana writes a small “Q?” in the margin before answering.
Later the mark disappears.
The question remains internal.
Another useful comprehension prompt:
“What in the text earns this inference?”
This keeps evidence and interpretation connected.
45. English: Prompt Migration for Writing
Writing prompts should regulate reasoning, not force formulaic prose.
Instead of a permanent rigid paragraph frame, migrate questions such as:
- What am I trying to prove?
- What evidence actually supports it?
- Have I explained the connection?
- What would a reasonable objection be?
- Does this paragraph move the argument forward?
These can guide mature writing across many structures.
The surface form can change.
The regulatory questions remain useful.
46. English: Prompt Migration for Vocabulary
A learner chooses a near-synonym.
Useful self-questions:
- What shade of meaning do I need?
- What emotion or intention does this word imply?
- Would a native speaker use it naturally in this sentence?
- What nearby word am I confusing it with?
These prompts help vocabulary move from dictionary recall to precise lexical choice.
47. Mathematics: Prompt Migration for Number and Percentage
Maya’s reference-whole prompt is a classic example.
Other useful questions include:
- What does this number represent?
- Compared with what?
- Should the answer be larger or smaller?
- What units should survive?
- Can I estimate before calculating?
These are metacognitive prompts because they regulate interpretation and checking, not just arithmetic.
48. Mathematics: Prompt Migration for Algebra
Algebra benefits from structural self-questioning.
- What is the goal of this transformation?
- What must remain equivalent?
- Can I simplify before expanding?
- What happens to every term under this operation?
- Can I substitute back to check?
The tutor can model these questions while thinking aloud.
Then ask the student to verbalise.
Then fade.
49. Science: Prompt Migration for Explanation
Science explanations often benefit from a three-question loop:
- What changed?
- What mechanism explains the change?
- How does that mechanism produce the observed outcome?
Another:
- What is observation?
- What is inference?
- What evidence supports the inference?
The prompts help students regulate scientific reasoning rather than memorise answer phrases.
50. Science: Prompt Migration for Experimental Reasoning
Before evaluating an experiment:
- What variable changed?
- What was measured?
- What else could explain the result?
- Was the comparison fair?
- What claim does this evidence actually justify?
These questions can migrate from teacher talk into scientific habit.
That is a more durable outcome than memorising “independent variable” as a definition without using the concept to regulate reasoning.
51. Additional Mathematics: Prompt Migration for Differentiation
Ethan sees a complicated expression.
The tutor’s old sequence:
- What is the outermost structure?
- Is there nesting?
- Which rule governs the outside?
- Which rule appears inside?
- Can I preserve structure while differentiating?
Over time, this becomes a fast structural scan.
The learner no longer chants the questions.
But the decisions improve because the prompts have migrated.
52. Additional Mathematics: Prompt Migration for Trigonometry
Useful prompts:
- What form do I want?
- Which identity moves me toward that form?
- Am I transforming one side or both?
- Have I introduced a restriction?
- Can I verify with a simple value where appropriate?
These are better long-term regulators than “use this identity” because they train choice rather than obedience.
53. Prompts Can Migrate From Parent to Student Too
At home, a parent repeatedly asks:
“What homework is due tomorrow?”
“Did you pack your calculator?”
“Have you checked the units?”
These can become invisible family dependencies.
Prompt Migration offers a route:
parent asks.
student repeats.
student writes reminder.
parent waits.
student self-triggers.
parent retires prompt.
This protects family relationships because responsibility moves visibly instead of becoming an endless argument about nagging.
54. The Parent’s Prompt Migration Test
- What question do I keep asking?
- What function does it perform?
- Should my child eventually ask it?
- Have I explained why the question matters?
- Can I replace the question with a pause?
- Can my child use a temporary written reminder?
- Can I remove that reminder later?
- Does the routine survive without me?
The parent’s goal is not silence immediately.
It is successful transfer of the regulatory function.
55. The Tutor’s Prompt Migration Test
- What recurring question am I asking?
- Which cognitive operation does it trigger?
- Is the learner capable of that operation once prompted?
- Does the learner understand why the prompt works?
- What task feature should trigger it?
- Can the learner say the prompt aloud?
- Can I replace my prompt with a pause?
- Can a smaller written cue bridge the transition?
- Can the external cue disappear?
- Does self-questioning survive delay and mixed practice?
- Can the prompt compress without losing function?
- Can the learner adapt it to a new context?
This turns a good tutor question into a planned transfer mechanism rather than a permanent feature of the lesson.
56. The Student’s Prompt Migration Test
An older learner can ask:
- What question does my tutor keep asking me?
- What mistake is that question preventing?
- What kind of task should make me ask it myself?
- Can I write a shorter version?
- Can I notice the trigger without somebody else pointing it out?
- Can I use the prompt in another subject or context?
- Can I stop using the written reminder yet?
- Which prompts are now automatic enough to retire?
This is self-regulated learning becoming deliberate rather than accidental.
57. Prompt Migration Should Be Tracked in the Help Ledger
The Help Ledger might show:
February: evidence prompt supplied by tutor.
March: evidence prompt supplied by written card.
April: learner asks “what proves it?” aloud.
May: no visible prompt; evidence selected correctly.
This is a clean transfer story.
The external cognitive job has migrated inward.
58. Prompt Migration Can Lower Reactivation Cost
A forgotten skill does not always need full reteaching.
Sometimes one old self-prompt can reactivate the system.
Ethan forgets how to begin a graph interpretation.
He remembers:
“Axes first.”
That tiny internal cue reopens the larger routine.
This can reduce Reactivation Cost.
Self-prompts become part of the learner’s recovery architecture.
59. Prompt Migration Can Reduce Interference
Two rules compete.
The student repeatedly chooses the wrong one.
A discriminating prompt can become a permanent internal boundary check.
Percentage change versus percentage error:
“Reference to what?”
Product Rule versus Chain Rule:
“Outer structure?”
Observation versus inference:
“Seen or concluded?”
The Interference Test identifies the competition.
Prompt Migration turns the discriminating question into a self-managed defence.
60. Prompt Migration Can Support Supersession
An old heuristic is being replaced by a more accurate model.
A temporary prompt can help inhibit the obsolete trigger.
Old habit:
“Expand first.”
New self-prompt:
“What structure should I preserve?”
The prompt helps the learner interrupt the old automatic response while the newer model becomes dominant.
This is one way The Supersession Test can become practical action.
61. Prompt Migration and AI-Assisted Learning
AI tools can provide prompts instantly.
That can be useful.
But a learner can also become prompt-dependent on the tool.
A better use pattern is:
- ask the tool for questions rather than answers;
- identify which prompt actually helped;
- rewrite that prompt in the learner’s own words;
- close the tool;
- use the prompt on a fresh task;
- later test whether the prompt self-triggers without the tool.
This preserves the possibility that digital scaffolding becomes internal scaffolding rather than permanent external regulation.
62. Digital Prompts Should Become Less Necessary Too
Research on digital metacognitive prompts has found benefits for self-regulated learning, but it also raises an important design question:
Does the learner regulate because the system keeps prompting, or because regulation has become self-initiated?
A digital environment can help teach planning, monitoring and evaluation.
Eventually, the learner should encounter tasks without the digital prompt.
Otherwise the interface may be regulating the learner more than the learner is regulating themselves.
63. The Prompt Fading Test
Take one recurring prompt.
Ask:
- Can the learner explain its purpose?
- Can the learner identify the trigger?
- Can the learner say the question before the tutor?
- Can the learner use a shorter version?
- Can the tutor replace the question with silence?
- Can the learner use it without a written reminder?
- Can the learner self-trigger in mixed practice?
- Can the learner self-trigger after delay?
- Can the learner use the same regulatory idea in a new context?
If yes across these steps, the prompt is migrating successfully.
64. The Prompt Retirement Test
Not every self-prompt needs to remain consciously active forever.
A prompt can retire when:
- the trigger is recognised reliably;
- the cognitive operation happens without deliberate rehearsal;
- performance is stable across contexts;
- removing the prompt does not reduce quality;
- the prompt no longer deserves working-memory space.
Retirement does not mean the question is forgotten.
It means the function has become sufficiently embedded.
65. The Prompt Reactivation Test
Months later, an old error returns.
Before reteaching the entire topic, reactivate the old self-question.
“Reference whole?”
“Evidence?”
“Outer structure?”
If the capability returns immediately, the prompt remains a useful recovery key.
This is cheaper than full reteaching and connects Prompt Migration to the wider maintenance architecture.
66. The Five-Question Universal Core
Although subject-specific prompts are usually best, many learners benefit from a small universal core.
- What is the task?
- What do I know?
- What strategy fits?
- Is it working?
- What will I change?
These map broadly onto planning, monitoring and evaluation.
They are not enough for every subject problem.
But they provide a stable skeleton for self-regulated learning.
67. The Three-Question Recovery Core
When stuck:
- Where is the last step I trust?
- What exactly changed after that?
- What is the smallest next move or smallest help I need?
This triad connects Prompt Migration directly to The Escalation Packet and The Help Gradient.
The learner attempts recovery first.
Then escalates precisely if needed.
68. The Prompt Migration Record
A lightweight tutor note can say:
Prompt: What evidence?
Trigger: learner makes inference from text.
Current state: tutor prompted.
Next state: learner says prompt aloud.
Exit test: mixed comprehension, no cue.
Or:
Prompt: Outer structure?
Trigger: composite A-Math expression.
Current state: student written cue “outer?”
Next state: remove written cue.
Exit test: mixed timed differentiation.
That is enough to make prompt fading intentional.
69. Prompt Migration Is a Better Independence Metric Than Silence
A quiet student is not necessarily independent.
They may be stuck silently.
A talkative student is not necessarily dependent.
They may be verbalising an increasingly self-owned strategy.
A better question is:
Who is triggering the useful regulation?
If the learner increasingly triggers it themselves, independence is growing.
70. Prompt Migration Protects Against Tutor Dependency
A student may appear dependent on the tutor when the actual dependency is tiny.
The learner can read.
Reason.
Calculate.
Explain.
But waits for one recurring question:
“Check the condition.”
Do not respond with more tuition volume.
Migrate the prompt.
A large-looking dependency may collapse after one small regulatory function becomes internal.
71. Prompt Migration Protects Against Parent Nagging Loops
“Have you done your homework?”
“Have you checked your bag?”
“Have you revised?”
Repeated external prompting can become a stable family system.
The child waits.
The parent reminds.
Both become frustrated.
Prompt Migration reframes the problem.
Which reminder should become a self-reminder?
What trigger can the learner use?
What temporary scaffold bridges the transfer?
When can the parent retire?
This turns nagging into a transfer plan.
72. Prompt Migration Protects Against Empty Reflection
“What did you learn today?”
“It was okay.”
Reflection can become low-information routine.
Better evaluation prompts:
- Where did your first strategy stop working?
- What cue helped you recover?
- What will you notice sooner next time?
- What support can we remove now?
These questions connect reflection to future action.
73. Prompt Migration Protects Against Strategy Forgetting
Students often learn a strategy during one topic and fail to use it later.
The problem may not be forgetting the strategy.
It may be forgetting to trigger it.
Prompt Migration explicitly binds strategy to cue.
When I see conflicting evidence → compare source strength.
When the algebra explodes → inspect structure.
When I make a claim → demand evidence.
The self-question becomes the retrieval route into the strategy.
74. The Long Return: Experts Carry Internal Prompts
Experts often appear to “just know what to do.”
But expertise frequently contains compressed questions.
A doctor:
what dangerous alternative must I exclude?
An engineer:
what assumption controls this model?
A journalist:
what is the source for this claim?
A programmer:
what changed immediately before the failure?
A mathematician:
what structure is invariant?
These may no longer be spoken as full sentences.
But regulation remains.
School can begin building this architecture deliberately.
75. Back to Maya
The tutor gives Maya a new percentage question.
No reminder.
Maya reads.
Pauses.
Circles the original quantity.
Solves.
The tutor says nothing.
A week later, percentage error appears.
Maya pauses again.
This time she circles the accepted value.
She has not memorised one denominator.
She has internalised a question:
Reference to what?
That question now travels.
It distinguishes related rules.
It protects against old errors.
It does not require the tutor’s voice.
The prompt did not disappear.
It migrated.
Prompt Migration in One Page
1. Treat recurring tutor questions as temporary external regulation.
The destination is student self-regulation.
2. Use task-specific metacognitive prompts.
Concrete questions migrate more easily than vague “think harder” advice.
3. Teach the prompt’s purpose.
The learner should know what mistake or decision the question regulates.
4. Teach the trigger.
Knowing the question is not enough; the learner must recognise when to ask it.
5. Use planning, monitoring, evaluation and recovery prompts.
Self-regulated learning requires more than end-of-task reflection.
6. Make self-talk audible during transition.
Let the learner say the prompt before it becomes silent internal scaffolding.
7. Use written prompts temporarily.
Cards, symbols and checklists can bridge migration but should not become permanent dependencies.
8. Replace tutor questions with pauses.
Silence creates a retrieval opportunity for self-prompting.
9. Fade external prompts deliberately.
Do not wait for independence to happen accidentally.
10. Test cold transfer.
Use mixed, delayed and authentic tasks without the original cue.
11. Allow prompts to compress.
Long explicit questions can become fast internal checks.
12. Prune prompt clutter.
A few reliable high-value questions beat a large generic checklist.
13. Let students design prompts from recurring errors.
Self-regulation becomes generative when learners build their own control questions.
14. Migrate parent prompts too.
Homework, planning and checking routines should increasingly self-trigger.
15. Retire prompts when the function is stable.
The question can disappear consciously when the underlying regulation reliably remains.
Where This Fits in eduKatePunggol
- How Tuition Works at eduKatePunggol — the broad tuition relationship.
- The Weak Link — identify the first limiting mechanism.
- The Learning Simulator — create safe conditions where prompts can be trained and faded.
- The Sparring Partner — ask questions that create intelligent resistance rather than automatic rescue.
- The Flight Recorder — track movement from tutor-prompted to self-prompted performance.
- The Learning Dispatcher — route remaining external regulation to the correct environment.
- The Exit Ramp — transfer support ownership toward the learner.
- The Priority Queue — prioritise recurring external prompts with high consequence and transfer value.
- The Intervention Threshold — preserve enough independent attempt before prompting.
- The Translation Layer — identify precisely which regulatory function is missing.
- The Transfer Gate — verify self-questioning in changed and authentic contexts.
- The Diagnostic Probe — distinguish missing knowledge from missing self-triggering.
- The Evidence Threshold — decide when repeated prompt dependence warrants a plan change.
- The Maintenance Window — check whether self-prompts remain available after delay.
- The Reactivation Cost — use old self-prompts as low-cost recovery keys when learning cools.
- The Interference Test — turn discriminating cues into self-managed boundary checks.
- The Supersession Test — use self-prompts to interrupt obsolete heuristics while better models take control.
- The Escalation Packet — teach self-questions that locate and package uncertainty before help-seeking.
- The Help Gradient — identify small recurring cues ready for migration.
- The Help Ledger — identify which important cognitive jobs still depend on outside prompts.
This completes Batch 05:
- The Escalation Packet — what should the learner bring when asking for help?
- The Help Gradient — how much help should enter?
- The Help Ledger — which cognitive jobs still came from outside?
- The Prompt Migration — how does the recurring external cue become the learner’s own self-question?
The student-side support loop is now complete:
Attempt → locate the edge → escalate precisely → receive bounded help → record what help supplied → convert useful recurring cues into self-prompts → fade the external trigger → verify independent transfer.
Research Foundations and Evidence Boundaries
Prompt Migration is an explanatory tuition framework. The exact eight-stage ladder in this article is not a validated psychometric scale. The stronger evidence supports the underlying ideas: explicit teaching of metacognitive strategies, self-regulated learning, question prompts, self-talk, guided practice, carefully faded scaffolding and increasing learner independence.
- Education Endowment Foundation — Metacognition and Self-Regulation: Developing Independent Learners (24 August 2026). EEF describes planning, monitoring and evaluating learning, and notes that these processes can become habitual “internal scaffolding.” It also highlights self-talk in which learners pose questions to themselves.
- Education Endowment Foundation — How to Use Visual, Verbal, and Written Scaffolding to Support Pupils’ Use of Metacognitive Strategies (14 January 2026). EEF describes a least-support-first approach and the use of verbal, visual and written scaffolds to build increasing independence and ownership.
- Education Endowment Foundation — Promoting Metacognitive Talk in the Classroom (2026). The resource describes how external metacognitive talk can become internalised and provides prompts structured around task, strategy and self.
- Education Endowment Foundation — Updated Guide to Metacognition and Self-Regulation (2025). The updated guidance emphasises explicit teaching, modelling, metacognitive talk, question prompts and “least help first” scaffolding so responsibility can move toward the pupil.
- Brown University Sheridan Center — Metacognition. Brown organises metacognitive prompts around planning, monitoring and evaluation and provides concrete self-questions learners can use to regulate goals, strategies and progress.
- Guo — Using Metacognitive Prompts to Enhance Self-Regulated Learning and Learning Outcomes: A Meta-Analysis of Experimental Studies in Computer-Based Learning Environments, Journal of Computer Assisted Learning (2022). The meta-analysis found positive effects of metacognitive prompts on self-regulated learning activities and learning outcomes, while highlighting the importance of prompt specificity, adaptability and feedback.
- Australian Education Research Organisation — Supporting Self-Regulated Learning. The practice guide emphasises explicit teaching and modelling of strategies that help students plan, monitor, evaluate, identify gaps and seek guidance as they gradually take greater control of learning.
- Royal Society of Chemistry Education — Why Metacognition Is Vital for Successful Independent Learning. The article gives subject-specific examples of self-questioning and explicitly notes that repeated teacher prompts can become automatic student self-prompts over time.
These sources do not prove that every teacher question will automatically internalise, or that all learners benefit from the same prompt sequence.
They support a more careful conclusion:
Self-regulated learning can be strengthened when useful planning, monitoring and evaluation strategies are explicitly taught, modelled and scaffolded in authentic subject tasks, with responsibility gradually shifted toward the learner.
Prompt Migration is one practical way to make that shift visible.
The tutor asks.
The learner answers.
The learner begins asking.
The tutor fades.
The question remains.
Then, eventually, even the full question can disappear—because the thinking it was designed to trigger has become part of how the student works.
