Chapter 16: Make the System Work

Bringing the Learning Principles Together in Everyday GCSE Education

The problem with a good plan

It is Monday morning.

A fifteen-year-old autistic student has five subjects to study this week. There are live online lessons, assignments, revision tasks and several topics that need more practice.

Her parent has prepared a timetable.

At 10 o’clock, she is supposed to begin Mathematics.

The lesson is on algebra. She has studied the topic before, but some of the methods remain uncertain. There is a worksheet, a recorded explanation and a collection of practice questions.

Everything appears ready.

At 9:55, her parent reminds her that Mathematics starts in five minutes.

She says she knows.

At 10:00, she is still looking at her phone.

At 10:05, her parent asks her to open the worksheet.

She says she will.

At 10:10, nothing has happened.

The parent becomes frustrated.

“We agreed we’d start at ten.”

“I know.”

“Then why aren’t you doing it?”

“I don’t know.”

The parent offers to sit beside her.

“I don’t want you watching me.”

The parent suggests doing just the first question.

“I said I’ll do it!”

By 10:20, both are upset. Mathematics has become a confrontation. The student has not attempted a question, and the parent is wondering whether the entire educational plan is failing.

This is where an educational system is tested.

Not when the student is enthusiastic and the work is easy.

Not when a beautifully organised timetable is followed without difficulty.

It is tested when something goes wrong.

The earlier chapters have examined the components of effective learning: accessibility, emotional regulation, motivation, explicit instruction, worked examples, retrieval, feedback, confidence, metacognition, external organisation and increasing learner control.

Those components are individually useful.

But real education does not arrive neatly divided into chapters.

At 10:10 on Monday morning, several problems may be operating simultaneously.

The student might not understand the algebra. She might understand it but be unable to organise the first action. She might fear discovering how much she has forgotten. She might be exhausted, distracted, resentful of being directed, or overwhelmed by a task whose endpoint is unclear.

The parent cannot solve the situation merely by remembering that retrieval practice is effective or that autonomy matters.

They need a way to decide what to do next.

This chapter brings the book’s principles into one working educational system.

Its central proposition is simple:

An effective learning system must help the learner make progress when conditions are imperfect, identify why progress has stopped, and adapt without sacrificing either learning or wellbeing.

The goal is not a timetable that never fails.

It is a system that can recover.

1. The system has six jobs

A successful GCSE education involves more than completing the work set by a school.

The student needs to acquire knowledge, retain it, apply it, demonstrate it in assessments and increasingly manage the processes that make learning possible.

For an autistic teenager, the system must also accommodate individual sensory, emotional and executive-function needs.

We can organise these requirements into six jobs.

Job 1: Make learning accessible

Before asking whether a student is trying hard enough, establish whether the task can reasonably be attempted under the current conditions.

This includes the sensory environment, the student’s available energy, the clarity of instructions, the accessibility of materials and the predictability of the task.

Accessibility is not a reward for cooperation.

It is part of the educational design.

Job 2: Teach what the student does not yet know

Students cannot independently practise knowledge or procedures they have never adequately learned.

Where necessary, the system must provide clear explanations, examples, modelling and guided practice.

The amount of help should depend on the student’s current knowledge.

Job 3: Produce durable learning

Completing an exercise today does not establish that the student will remember the material next month.

The system needs retrieval practice, spaced revisiting, corrective feedback and opportunities to apply knowledge in different situations.

Job 4: Develop accurate confidence

Students need credible evidence of what they can do.

This requires distinguishing success achieved through adult direction from success achieved through the student’s own knowledge, decisions and strategies.

Neither kind of success is worthless, but they tell us different things.

Job 5: Develop learner control

The student should increasingly participate in setting goals, choosing strategies, monitoring progress, requesting assistance and deciding how to use supports.

This does not require abandoning calendars, checklists, assistive technology or human help.

Job 6: Protect sustainability

The system must be workable across weeks and months, not merely during an unusually productive afternoon.

A schedule that produces three excellent days followed by prolonged exhaustion may be less useful than a smaller, consistent workload.

Equally, a comfortable routine that never produces meaningful academic progress is insufficient.

The challenge is to maintain both.

These jobs are interconnected, but they are not interchangeable.

A student can be engaged without learning. They can learn without becoming more independent. They can perform well with assistance but fail to retain the knowledge. They can develop impressive examination skills while experiencing unacceptable distress.

A good system monitors these outcomes separately.

2. Build the week around learning, not attendance

A common mistake in home and online education is to treat the school timetable as the complete educational plan.

The timetable tells us when teaching is scheduled.

It does not necessarily tell us what the student has learned.

Consider a student studying GCSE Biology.

On Tuesday, she attends a lesson about enzymes.

She listens carefully, answers a question and completes a worksheet.

The lesson is marked as done.

Three weeks later, she cannot explain why increasing temperature eventually reduces the rate of an enzyme-controlled reaction.

Was the Tuesday lesson successful?

In one sense, yes. She participated and completed the immediate work.

But the intended knowledge was not demonstrated after a delay.

This does not establish that she learned nothing. It tells us that the available evidence of durable learning is inadequate.

The educational plan should therefore contain three different kinds of activity.

Teaching introduces or clarifies knowledge and procedures.

Practice strengthens the student’s ability to retrieve, explain and apply what has been taught.

Checking establishes what the student can currently do, under what conditions, and with how much assistance.

These activities may overlap. A retrieval attempt can strengthen memory while revealing a knowledge gap. A worked example can teach a method while preparing the student for practice.

Nevertheless, the distinctions help prevent a serious planning error: allocating all available time to new lessons and leaving no room for consolidation.

A practical weekly structure

Imagine that the student has Biology, Mathematics, History, English and Psychology to manage.

Rather than filling every available hour with scheduled tasks, divide the week into four categories:

1. Required commitments: live lessons, deadlines and appointments.
2. New learning: topics requiring explanation, modelling or guided practice.
3. Consolidation: retrieval, correction and application of previously taught material.
4. Flexible capacity: time for unexpected difficulties, missed work, recovery or additional practice.

The fourth category is especially important.

If every hour is allocated, a difficult Monday creates a backlog that spreads through the week.

The timetable becomes a debt-collection system.

A more resilient plan includes capacity for variation.

This does not mean leaving large amounts of time permanently unused. It means recognising that educational tasks have uncertain durations, especially when the student is still developing the knowledge and strategies required to complete them.

The weekly planning conversation

At the beginning of the week, parent and student can review three questions.

What must happen this week?

Identify genuine deadlines and essential commitments.

What knowledge needs strengthening?

Use recent work, retrieval attempts and teacher feedback rather than relying entirely on what the student feels confident about.

What choices can the student make?

These might include the order of independent subjects, the format of a revision session, whether to use a written or digital checklist, or which difficult topic to tackle first.

The adult may initially organise much of this conversation.

The developmental objective is for the student to take increasing responsibility for meaningful decisions.

However, responsibility should follow demonstrated competence. Giving a teenager complete control over a complicated revision schedule before they can estimate task duration or identify priorities may create avoidable failure.

The better question is:

Which planning decision is the student ready to make, and what support would make the next decision possible?

3. Give every study session a clear purpose

A timetable entry such as “History, 45 minutes” tells the student where to be and for how long.

It does not tell them what successful learning would look like.

Even “Revise the causes of the First World War” leaves considerable ambiguity.

Should they read the textbook? Watch a video? Make notes? Answer questions? Memorise dates? Write an essay?

A more useful session begins with a specific learning outcome.

For example:

By the end of this session, I want to explain two reasons why tensions between Austria-Hungary and Serbia increased between 1908 and 1914, using accurate historical evidence.

The student now has a target.

The next question is how to reach it.

If the material is unfamiliar, the session may require teaching.

If it was taught recently, the student might attempt retrieval.

If the student knows the facts but struggles to construct explanations, the session should focus on connecting evidence to causal reasoning.

The same topic can therefore require different educational activities depending on the student’s present competence.

The session card

A practical session card can contain five elements:

This is not intended as a compulsory format for every lesson.

It is external infrastructure that makes the task understandable.

For a student who struggles with initiation, the first action deserves particular attention.

“Study History” is an intention.

“Open the document and answer question one” is an action.

A student may need assistance converting the first into the second without needing assistance with the historical reasoning itself.

That distinction matters.

We should not mistake difficulty organising a task for inability to learn its content.

4. When the student cannot begin

Return to Monday’s Mathematics lesson.

At 10:10, the student has still not started.

The parent has several possible responses.

They could repeat the instruction more firmly.

They could offer a reward.

They could remove the student’s phone.

They could sit beside her and insist that she attempt the first question.

Any of these might produce visible activity under some circumstances.

But none necessarily identifies why she has not begun.

The earlier chapters introduced the importance of diagnosing the barrier before selecting an intervention.

Here that principle becomes an operational routine.

First, establish what is actually happening

Instead of asking, “Why won’t you work?”, the parent might say:

“I can see Maths hasn’t started. Is the problem getting into the task, knowing how to do it, or something else?”

The student might know.

She might not.

If she cannot explain, the parent should not turn the question into an interrogation.

A small, low-pressure diagnostic attempt may be more informative.

“Would it help to look at the first question together, just to see what it needs?”

Suppose the student agrees.

The question asks her to solve a linear equation.

She reads it, looks uncomfortable and says:

“I’ve forgotten all of this.”

Now the parent has useful information.

The obstacle may involve insecure knowledge or retrieval, rather than simple unwillingness.

A worked example or brief reminder might be appropriate.

But suppose she instead says:

“I know how to do it. I just don’t want to do twenty questions.”

The problem has changed.

The issue might be task size, anticipated effort, boredom, uncertainty about the finish condition or conflict over control.

The appropriate response is no longer necessarily more explanation.

Match the response to the barrier

These are hypotheses, not diagnoses.

The parent should observe what happens after an adjustment.

If providing a worked example enables the student to begin, missing procedural knowledge was probably contributing.

If shortening the worksheet improves engagement, task size or perceived burden may have mattered.

If neither helps, further investigation is needed.

An intervention that fails is not proof that the student is being deliberately difficult.

It is information that the proposed explanation may be incomplete.

Avoid the endless negotiation

There is a danger here.

A parent trying to be collaborative may ask so many questions, offer so many choices and discuss so many possibilities that the process itself becomes exhausting.

A student already struggling to begin may now have to analyse their emotions, evaluate six strategies and negotiate a plan.

The diagnostic process should be proportionate.

Sometimes the best response is simply:

“Let’s put the first step on the screen. You can tell me whether you want help with it.”

At other times, the right response is to stop and revisit the task later.

The system should help adults distinguish these situations rather than impose one universal script.

5. Protect the intellectual work

Making a task accessible does not mean removing the learning challenge.

This distinction becomes particularly important when the parent is under pressure to keep the week on schedule.

Imagine a student working on a GCSE English Literature paragraph.

The question asks how a writer presents conflict between generations.

The student has several relevant ideas but cannot organise them into a paragraph.

The parent suggests the opening sentence.

The student writes it.

The parent supplies a quotation.

The student copies it.

The parent explains the quotation’s significance.

The student writes that too.

After fifteen minutes, the paragraph is complete.

It looks excellent.

But what has the student demonstrated?

Perhaps they have copied accurately and followed instructions.

They have not necessarily demonstrated independent selection of evidence, interpretation or paragraph construction.

This is not a reason to avoid modelling.

A student who has never understood how to build an analytical paragraph may need a complete worked example.

The problem arises when adult-supported performance is mistaken for independent competence.

Use a three-stage check

When a task has required substantial help, ask:

Can the student explain what happened?

For example, can they identify why the quotation supports the argument?

Can the student attempt a similar task with less assistance?

They might select their own quotation and explain it using a familiar structure.

Can the student do something comparable after a delay?

A few days later, they might construct a paragraph on a different passage.

These checks provide different information.

Understanding an explanation is not the same as producing one.

Producing an answer immediately after modelling is not the same as retaining the method.

Retaining a method for one familiar task is not the same as recognising when to use it in a new context.

The purpose of support is to make worthwhile thinking possible.

It should not permanently substitute for that thinking.

Support is not one thing

We also need to distinguish instructional assistance from other forms of support.

A paragraph frame may be an instructional scaffold if the objective is eventually to construct paragraphs independently.

A word processor may be an accommodation for a student whose handwriting creates an unnecessary motor burden.

A checklist of assignment deadlines may be useful organisational infrastructure.

A parent supplying the next sentence is direct instructional assistance.

These supports should not all be reduced together.

A student might outgrow the need for a paragraph frame while continuing to benefit from typing and a visual schedule.

The governing principle remains:

Adjust assistance to competence, while retaining supports that make learning accessible and manageable.

6. Turn completed work into useful evidence

A finished worksheet is not the end of the learning process.

It is an opportunity to gather information.

Suppose a student completes ten Biology questions.

Eight are correct.

Two are wrong.

The parent could record 80% and move on.

But that number conceals several possibilities.

The student may have answered eight independently and guessed two incorrectly.

They may have needed substantial prompting on every question.

They may have understood the concepts but confused technical vocabulary.

They may have answered accurately after reading the textbook immediately beforehand, yet be unable to retrieve the material a week later.

A percentage alone cannot distinguish these situations.

Record the kind of success

A lightweight learning record can capture:

• What was attempted?
• What could the student do independently?
• What assistance was required?
• What error or uncertainty remains?
• When should the knowledge be checked again?

For example:

Biology: enzymes

Today: Correctly explained the effect of temperature after reviewing a diagram. Needed a cue to remember denaturation.

Next check: Explain the process from memory later in the week.

This is more informative than “Biology completed.”

It also provides a basis for confidence.

A student who previously needed a full explanation but can now independently describe denaturation has evidence of improvement.

The parent can make that progress visible without exaggeration.

“Last week we had to go through the whole process. Today you explained it without the diagram.”

That is a statement about demonstrated competence.

It is more useful than an unsupported claim that the student is brilliant at Biology.

Avoid turning the record into surveillance

A learning record can become counterproductive if every mistake is documented, every hesitation timed and every session evaluated by the parent.

The student may begin to experience the system as continuous assessment.

Keep records brief.

Make them accessible to the learner.

Use them to make decisions, not to build a case against the student.

As competence develops, the student should increasingly help decide what is recorded and what it means.

7. Design revision around evidence of forgetting

A major challenge in GCSE education is the quantity of material that must remain available over time.

A student might understand the causes of the First World War in September, study civil rights in October and discover in January that the earlier material is difficult to retrieve.

This is ordinary forgetting, not necessarily evidence that the original teaching failed.

The question is whether the learning system anticipates it.

A simple revision loop

After a topic has been taught, the student attempts to retrieve important knowledge.

Errors are corrected.

The material is revisited after a delay.

Later practice introduces different questions, contexts and demands.

Eventually, the student attempts tasks resembling the examination.

The interval between checks should depend on performance and the intended retention period, rather than a rigid rule that every topic must be reviewed after exactly the same number of days.

Consider GCSE Psychology.

The student has learned about Milgram’s obedience research.

Immediately after teaching, she can describe the procedure and findings.

Several days later, she remembers the main findings but confuses the meaning of the agentic state.

That is a useful result.

The revision session should concentrate on the uncertain concept, rather than require an exhaustive repetition of everything she already knows.

A later question might ask her to use agency theory to explain behaviour in a new scenario.

Now the system is checking application, not merely recall.

Keep retrieval appropriately challenging

Retrieval should require the student to reconstruct knowledge.

But difficulty must be calibrated.

A student who cannot recall anything from a large topic may need narrower questions, cues or brief reteaching.

A student who retrieves all the facts easily may need more demanding application.

The aim is neither effortless recognition nor repeated failure.

It is productive reconstruction followed by correction.

For autistic students, predictable formats and explicit expectations may make retrieval more accessible, but there is no justification for assuming that every autistic learner needs the same level of prompting.

8. Make the student an operator of the system

The preceding chapters have described a gradual transfer of educational responsibility.

Here that principle becomes concrete.

Imagine two students with identical weekly revision schedules.

The first student follows a plan created entirely by an adult.

The adult selects the topics, chooses the revision methods, checks the answers, identifies weak areas and decides what happens next.

The student completes the activities.

The second student uses a similar schedule, but increasingly participates in those decisions.

She checks which topics need revisiting, chooses an appropriate method, compares her answers with the specification and requests assistance when necessary.

Both may produce similar amounts of work.

But they are developing different capabilities.

The second student is practising regulation of learning itself.

Transfer decisions, not merely tasks

It is tempting to measure independence by whether the parent can leave the room.

That is an incomplete measure.

A student might work alone for an hour while following instructions designed entirely by someone else.

Another student might ask a parent for help selecting a revision strategy, then independently organise and complete the work.

The second student may demonstrate more meaningful control despite requiring interaction.

A useful progression is to transfer responsibility for particular decisions.

For example, in Mathematics:

Initially: The parent identifies that fractions need practice and chooses the questions.

Next: The student reviews recent errors with the parent and helps select the topic.

Later: The student identifies recurring fraction errors and chooses an appropriate practice set.

Eventually: The student notices the problem, obtains suitable materials, practises, checks progress and seeks help if necessary.

This progression is not necessarily linear.

The student might manage Biology revision independently while still needing substantial support organising English essays.

They might handle planning well during a quiet week but require more assistance during examinations or periods of exhaustion.

Responsibility should respond to actual competence and available capacity.

The ownership test

For any part of the learning system, ask:

“Who currently makes the important decisions?”

Then ask:

“What would the student need to know or practise to take more control of this part?”

The answer might be a new skill.

It might be a checklist.

It might be experience estimating task duration.

It might be learning how to ask a teacher for clarification.

Or it might be discovering that a particular external support should remain in place.

The goal is not to make the student perform every function internally.

It is to increase their effective control over learning.

9. Measure the right things

A functioning educational system needs feedback about its own effectiveness.

But measuring the wrong outcomes can create misleading conclusions.

Suppose a parent introduces a new study routine.

Before the change, the student completed thirty minutes of work each morning.

After the change, she completes sixty minutes.

Has the intervention worked?

Perhaps.

The increase in participation is encouraging.

But we do not yet know whether the student learned more, retained more, became more independent or experienced greater distress.

Those outcomes require separate evidence.

Six dimensions of progress

Wellbeing and sustainability must also be monitored.

They are not merely side effects.

A student who is academically progressing but experiencing persistent distress may need changes to the educational system.

Equally, a student who feels comfortable but is making no progress towards agreed educational goals may need better teaching, clearer expectations or a more appropriate level of challenge.

The solution should not automatically be more pressure or less work.

It should follow an investigation of what is limiting progress.

A short weekly review

A practical weekly review might ask:

What did you learn this week that you could not do before?

Look for specific evidence.

What do you still need to practise?

Use errors, delayed retrieval and feedback.

What made learning harder than it needed to be?

Consider task design, environment, scheduling, instruction and support.

What helped?

Identify strategies that are worth repeating.

What will you take responsibility for next week?

Choose a manageable decision or skill.

This conversation need not become a formal meeting.

For some teenagers, a written review or a few selectable prompts may be easier than an extended discussion.

The important feature is that the information changes future decisions.

10. When the whole week goes wrong

A resilient educational system needs a recovery procedure.

Consider a difficult week.

On Monday, Mathematics produces an argument.

On Tuesday, the student feels exhausted and misses an online lesson.

On Wednesday, an English assignment takes much longer than expected.

By Thursday, the parent has a growing list of unfinished work.

The temptation is to compress everything into Friday.

The student sees the backlog and becomes overwhelmed.

A familiar cycle begins: missed work creates pressure, pressure makes initiation harder, and further missed work increases the pressure.

The problem is no longer a single worksheet.

It is the system’s response to disruption.

Recovery begins with triage

Not every missed activity has equal educational importance.

Some tasks are essential prerequisites for future learning.

Some are required for assessment.

Some are useful practice that can be shortened or rescheduled.

Some are duplicative or can be replaced with a more efficient check of understanding.

The parent and student should identify what genuinely needs recovering.

For example, missing a Biology lesson on cell division may require watching the explanation and completing a short knowledge check.

It may not require reproducing every classroom activity if those activities served the same learning objective.

Similarly, if the student already demonstrates competence in a Mathematics topic, completing every missed practice question may provide little additional value.

But if the student has not learned a foundational method, skipping it to catch up with the timetable may create greater problems later.

Recovery should be based on learning needs, not simply the number of outstanding tasks.

The recovery conversation

A useful conversation might be:

“We’ve got more work listed than we can reasonably do tomorrow. Let’s identify what is essential, what needs teaching and what can wait.”

The student can help make those decisions at a level appropriate to their competence.

The adult should avoid presenting the backlog as a moral debt.

Missed work is a planning problem.

It may also be evidence that the current system is overcommitted.

If the same backlog appears every week, the solution is unlikely to be a permanently more intense Friday.

The baseline plan needs revision.

11. Protect the parent–teenager relationship

In parent-supported education, the adult occupies several roles.

They may be a parent, tutor, organiser, advocate, examiner, emotional support and liaison with the school.

These roles can conflict.

The parent wants the student to learn.

The student wants some control over her day.

Both may be anxious about examinations.

A missed lesson can become emotionally charged because it represents more than a missed lesson.

For the parent, it may raise fears about falling behind.

For the teenager, it may represent another occasion on which someone is disappointed in her.

The educational system must prevent ordinary academic difficulties from becoming repeated relational battles.

Separate the person from the problem

If the student cannot begin an assignment, the problem is not automatically that she is irresponsible.

If the parent becomes frustrated, the solution is not automatically to abandon all expectations.

The task is to identify a workable response while preserving dignity on both sides.

That might mean agreeing that academic disagreements are discussed at a designated time rather than throughout the day.

It might mean separating the parent’s role as emotional support from the teacher’s role in evaluating academic work.

It might mean using written instructions so the parent does not have to repeat reminders.

It might mean arranging for a teacher or tutor to provide feedback on work that has become a recurring source of conflict.

Collaboration is not the absence of boundaries

Autonomy support does not require the teenager to decide whether every educational obligation exists.

A family can maintain clear expectations while giving the student meaningful influence over how those expectations are met.

For example:

“The assignment is due on Thursday. We need a plan for completing it. Would you prefer to work on the first section today or tomorrow morning?”

This is useful only if both options are genuinely available.

False choices undermine trust.

Sometimes the student will disagree with the expectation itself.

That disagreement deserves consideration, particularly if the workload is inappropriate or inaccessible.

But collaboration does not mean that every difficult task disappears.

It means that problems are addressed through explanation, negotiation, appropriate structure and responsive support rather than relying primarily on escalating pressure.

12. A complete example: preparing for a History assessment

Let us bring the system together.

A student has an upcoming GCSE History assessment covering the causes of the First World War.

She has attended lessons on the Moroccan Crises, the Bosnian Crisis, the Balkan Wars, the naval race and the assassination at Sarajevo.

Her notes are extensive.

She says she knows the topic.

Her parent is unsure.

Step 1: Establish the goal

The objective is not simply to reread the notes.

The student needs to recall accurate historical knowledge and use it to construct explanations appropriate to the assessment.

Step 2: Check current knowledge

The student answers five short questions without notes.

She accurately describes the naval race and the assassination.

She confuses the Bosnian Crisis with the Balkan Wars.

She remembers that Serbia and Austria-Hungary were in conflict but cannot explain the significance of nationalism.

This provides a more useful starting point than her general confidence.

Step 3: Select the right teaching

The student does not need another complete lesson on the naval race.

She needs clarification of the events and relationships she has confused.

The parent or teacher provides a concise timeline and explains the relevant distinctions.

The student then reconstructs the sequence.

Step 4: Move to explanation

Once the facts are sufficiently secure, the student attempts:

“Explain why tensions between Austria-Hungary and Serbia increased between 1908 and 1914.”

She knows relevant facts but produces a list rather than a causal explanation.

The adult models how to connect an event to its political significance.

The student then constructs her own explanation.

Step 5: Check independence

The student attempts a similar question without the model.

The parent checks whether she has selected relevant evidence and explained its significance.

If she still requires a paragraph frame, that support can remain temporarily.

Step 6: Revisit after a delay

Several days later, the student answers a different causation question.

This tests whether the knowledge and explanatory method remain available.

Step 7: Transfer control

The student compares her answer with the mark scheme or teacher guidance.

She identifies that she has included accurate facts but needs stronger causal links.

She chooses to practise another explanation question.

The parent remains available if she needs clarification.

Step 8: Review the outcome

The student has done more than complete History revision.

She has retrieved knowledge, corrected a misconception, practised explanation, received feedback and participated in selecting her next task.

But the outcomes should still be recorded accurately.

One successful independent answer is evidence of current performance.

A successful answer after a delay strengthens evidence of retention.

Applying the same reasoning to a different historical problem provides evidence of transfer.

Independently identifying the next learning need provides evidence of developing self-regulation.

These are distinct achievements.

Together, they represent the kind of progress the book is trying to cultivate.

13. The system must change as the learner changes

A learning system is not a permanent set of procedures.

It should evolve.

A teenager who initially needs the parent to prepare every study session may later manage familiar subjects independently.

A student who once required worked examples for basic algebra may become frustrated if the adult continues to explain every step.

A learner who needed extensive prompting to begin revision may eventually use a checklist without reminders.

The system should respond.

But change does not always mean reducing support.

A more demanding GCSE topic may require renewed modelling.

Examination preparation may introduce unfamiliar executive demands.

Illness, stress or sensory difficulties may temporarily reduce available capacity.

A useful system can increase assistance without treating that as regression or failure.

Review support by function

For each support, ask:

What problem does this solve?

Is that problem still present?

Does the support help the student do worthwhile intellectual work, or is it doing that work for them?

Could the student take more control of the support?

Would removing it improve learning, or merely make learning harder to access?

A student may no longer need an adult to choose revision questions but may still need a digital calendar.

They may become capable of independently writing analytical paragraphs while continuing to type rather than handwrite.

They may learn to request a sensory break without needing the adult to identify overload first.

The system is improving when the learner has greater effective capability and control, not merely when fewer supports are visible.

14. What to do tomorrow

The ideas in this chapter do not require an elaborate new educational platform.

A family can begin with one subject and one study session.

Choose a task that regularly creates difficulty.

Define what the student should learn.

Make the first action clear.

Identify what assistance is currently necessary.

Allow the student to attempt meaningful work.

Check what they can do independently.

Correct important errors.

Arrange a later opportunity to retrieve or apply the knowledge.

Invite the student to make one decision about what happens next.

Then review whether the process worked.

If the session fails, examine why.

Was the task too difficult?

Was the instruction unclear?

Was the student unable to begin?

Did the adult provide too much assistance?

Was the workload unrealistic?

Were the sensory or emotional conditions unsuitable?

Did the student have enough control?

Change the part of the system that appears to be causing the problem.

Do not redesign the entire educational programme after every difficult morning.

Small, observable adjustments are easier to evaluate.

A practical experiment might be as simple as replacing a vague instruction with a written first step for one week.

Or allowing the student to select the order of two independent revision tasks.

Or replacing repeated rereading with five retrieval questions followed by correction.

Or having the student identify which kind of help they need before the adult intervenes.

The aim is to learn what works for this particular student.

Not what works for an imaginary average autistic teenager.

15. The deeper objective

Return once more to Monday morning.

The student has not started Mathematics.

In the original version of the morning, the parent interprets the delay as a failure to follow the plan.

The parent presses.

The student resists.

The conflict grows.

Now imagine a different response.

The parent notices that the work has not begun and asks whether the difficulty is starting or understanding.

The student says she has forgotten the method.

They look at one worked example.

She remembers part of the procedure but needs help with the next step.

They work through it together.

She attempts a similar question.

This time, she succeeds.

The parent asks whether she wants another similar question or a slightly harder one.

She chooses the harder question.

She makes an error.

They identify it, and she corrects the answer.

After twenty minutes, the session ends.

Only three questions have been completed.

Measured against the original worksheet, that may look disappointing.

But consider what actually happened.

The student identified a barrier.

She accepted a useful form of assistance.

She reconstructed a mathematical method.

She solved a problem.

She encountered an error and corrected it.

She made a decision about the next challenge.

The parent gained information about what she knows and what needs practice.

There is now a clear basis for the next session.

That is meaningful progress.

Of course, three questions are not automatically sufficient. If the student must develop fluency across a wider range of problems, further practice will be necessary.

Nor does one successful session establish durable learning or independence.

Those outcomes require later evidence.

But the session has moved the student forward.

It has also preserved the possibility that Mathematics can become something she learns to manage, rather than something that repeatedly produces conflict.

This is the distinction at the heart of the book.

We are not trying to build a system that makes a teenager complete every task immediately.

We are trying to build a system through which the teenager acquires knowledge, learns effective strategies, understands difficulties, uses support intelligently and develops increasing control over their education.

A successful educational system does not eliminate difficulty.

It makes difficulty understandable, manageable and increasingly navigable by the learner.

And when the system works well, the most important change is not that the parent becomes better at directing the teenager.

It is that the teenager becomes better at directing her own learning.

Chapter 16: Practical Toolkit

Tool 1: The weekly learning plan

At the start of the week, record:

Keep this visible and revise it when circumstances change.

Tool 2: The five-part session card

Goal: What should I know or be able to do?

Start: What is my first physical action?

Method: Am I learning, practising, retrieving or applying?

Check: How will I know whether I can do it?

Next: What needs to happen afterwards?

Tool 3: The learning evidence record

For important topics, record:

Do not turn this into a record of every mistake. Its purpose is to inform teaching and revision.

Tool 4: The weekly recovery check

When the plan has broken down, ask:

1. What is genuinely essential?
2. What knowledge is currently missing?
3. Which tasks can be shortened, rescheduled or replaced without losing the learning objective?
4. What caused the overload?
5. What must change in next week’s baseline plan?

Tool 5: The learner-control check

Choose one part of the system.

Currently, who decides what happens?

What knowledge or skill would allow the student to make more of that decision?

What support should remain available?

How will we recognise successful transfer of responsibility?

Review one responsibility at a time rather than attempting to transfer everything simultaneously.

Evidence Notes

This chapter integrates evidence from several distinct research traditions. Its overall operating model is a practical synthesis, not an intervention that has itself been experimentally validated.

1. Accessibility and environmental fit

Direct autism research supports considering environmental and contextual barriers when interpreting participation. Kasztelan Chapman, Southall and Grieshaber’s 2025 scoping review identified sensory conditions, relationships, inclusion, identity and wellbeing as important themes in autistic secondary students’ experiences. Li and colleagues’ 2024 study also found an association between difficulties with the physical school environment and reduced school participation.

These findings justify assessing accessibility before attributing disengagement to motivation. They do not establish a single intervention that resolves all forms of school non-participation.

Sources: Kasztelan Chapman et al. (2025), DOI: 10.1007/s13384-025-00838-0; Li et al. (2024), DOI: 10.1177/13623613231225490.

2. Instruction and adaptive assistance

General instructional research supports modelling, guided practice and worked examples, particularly when prior knowledge is limited. Expertise-reversal research indicates that assistance can become less beneficial when learners already possess substantial relevant knowledge.

Applying these principles to autistic GCSE students is reasonable, but the precise form and amount of assistance require individual assessment.

Source: Project Research Priorities 04–06, including the 2025 expertise-reversal meta-analysis discussed there.

3. Retrieval and retention

General educational research strongly supports retrieval practice and distributed learning. The 2026 experiments by Cairney, Haebig and Lucas provide direct evidence of retrieval-practice benefits in autistic adults, including benefits associated with post-retrieval feedback.

This strengthens the case for using retrieval in autistic education but does not establish the effectiveness of a particular GCSE revision timetable.

Source: Cairney, Haebig and Lucas (2026), DOI: 10.1177/13623613261464659.

4. Metacognition and learner regulation

The Education Endowment Foundation recommends explicitly teaching planning, monitoring and evaluation, including modelling the thinking involved.

The 2025 systematic review by Cristescu, Papoudi and Douglas found emerging evidence for self-determination interventions involving autistic young people and people with learning disabilities, while emphasising the interaction between the individual and their environment.

The chapter’s proposed transfer of planning and monitoring decisions is consistent with this evidence. Its exact developmental sequence remains a synthesis.

Sources: EEF, Metacognition and Self-Regulated Learning; Cristescu et al. (2025), DOI: 10.1007/s40489-025-00508-y.

5. Academic self-management

The 2026 meta-analysis by Gahtan and colleagues provides autism-specific evidence that self-management interventions can improve academic engagement in the single-case studies reviewed.

It does not establish equivalent improvements in GCSE attainment, delayed retention or broad independence across subjects.

Source: Gahtan et al. (2026), DOI: 10.1177/10983007251389292.

6. Parent-supported education

The Project’s Research Priority 10 supports combining warm relationships, clear structure, autonomy support and appropriate educational assistance.

However, controlled research on comprehensive parent-supported GCSE programmes for autistic teenagers is limited.

The recovery procedures, session cards and weekly planning system proposed here should therefore be understood as practical tools derived from the evidence, not independently validated interventions.

Evidence Audit

Evidence Debt

Priority 1: Integrated GCSE intervention evidence. The central gap remains the absence of controlled studies testing the complete combination of adaptive instruction, retrieval, metacognition, parent support and learner-controlled planning among autistic GCSE-age students.

Priority 2: Long-term independence. More longitudinal evidence is needed on whether improvements in academic engagement and self-management transfer across subjects and persist without intensive adult prompting.

Priority 3: Home and online education. The proposed scheduling, recovery and responsibility-transfer procedures require direct evaluation in home-educated and online-school populations.

Priority 4: Workload calibration. There is insufficient direct evidence to prescribe ideal daily study durations, buffer proportions or weekly workload for autistic teenagers. These should be adjusted to individual circumstances rather than presented as scientifically established quantities.

Priority 5: Parent–teenager interaction. Research is needed on whether particular collaborative planning and recovery routines reduce conflict while improving durable academic learning.

Editorial Review

This chapter advances the book from individual instructional and self-regulatory strategies to their integration into an everyday educational system.

It deliberately avoids presenting the system as a validated treatment or a universal programme. The examples distinguish completed work from demonstrated learning, and adult assistance from learner-controlled support.

The main editorial risk is overlap with Chapters 12–15. In a final manuscript pass, earlier explanations of help-seeking, metacognition and responsibility transfer should be shortened here wherever they are already fully developed. The distinctive contribution of Chapter 16 should remain operational integration, weekly planning, monitoring and recovery.

The chapter’s tools should also be checked against the precise approved architecture before publication, to ensure terminology and chapter-to-chapter transitions remain consistent.