Capable, Confident, Independent Helping an autistic teenager become a capable, confident and increasingly independent learner.

Chapter 9: Make Knowledge Stick How to turn today’s understanding into knowledge that survives next week, next term and the examination

A teenager finishes a Biology lesson on respiration.

They have followed the explanation, answered the teacher’s questions and completed the worksheet. When asked whether they understand, they say yes. Their answers are correct.

The lesson appears successful.

Three weeks later, they encounter a GCSE question:

Explain why muscle cells need an increased rate of respiration during exercise.

They remember that respiration has something to do with energy. They recognise the words glucose and oxygen. But they cannot assemble an explanation.

The parent is puzzled. They covered this. The teenager understood it. They even got the questions right.

What happened?

Possibly nothing unusual. The student may have understood the explanation at the time without developing knowledge that could later be retrieved and used independently.

Understanding something while it is being explained is not the same as remembering it later. Remembering a definition is not the same as using it to explain a biological process. Completing a worksheet with notes available is not the same as answering an unfamiliar examination question.

These are different achievements.

And if we fail to distinguish them, we can spend months teaching material that seems to have been learned but gradually disappears.

This chapter is about preventing that disappearance.

It is about helping an autistic teenager build knowledge that is not merely familiar but available: knowledge they can recall, explain, connect, apply and use without an adult reconstructing it for them.

The solution is not simply more revision.

It is better-designed encounters with knowledge, spread across time, with opportunities to remember, make mistakes, receive correction and try again.

The central principle is straightforward:

Learning must be designed for what the student can do later, not merely for what they can do at the end of today’s lesson.

  1. The illusion of having learned Consider two ways of revising GCSE History.

In the first, a teenager reads three pages about the causes of the First World War. They highlight important sentences, reread the material and review a summary.

By the end, everything feels familiar.

In the second, the teenager reads the same material, closes the book and attempts to answer three questions:

The first session may feel smoother. The second may feel more difficult.

But the second contains something the first largely lacks: an attempt to reconstruct knowledge from memory.

That attempt matters.

Research on retrieval practice, often called the testing effect, consistently finds that attempting to retrieve previously learned material can improve later retention compared with simply studying it again.

A systematic review of retrieval practice in educational settings found substantial benefits across many classroom studies. Research on distributed practice likewise shows that learning spread across time generally produces better delayed retention than concentrating the same practice into one session.

These are findings from general learning science, not discoveries unique to autism.

There is now also emerging direct evidence that autistic learners benefit from retrieval practice. In 2026, Cairney, Haebig and Lucas reported experiments involving autistic and non-autistic adults. Retrieval practice benefited both groups, and corrective feedback after retrieval attempts was useful, particularly following errors.

That strengthens the case for using retrieval practice with autistic learners.

It does not prove that the same methods, schedules or effects will apply identically to autistic teenagers preparing for GCSEs. The experiments involved adults learning relatively constrained memory material, not adolescents learning an entire secondary-school curriculum.

Nevertheless, the practical implication is substantial.

We should not assume that rereading, watching explanations or completing supported exercises is sufficient evidence of durable learning.

We need to give memory opportunities to work. Performance is not learning This distinction runs throughout the book, but it becomes particularly important here.

A student may perform well because the teacher has just demonstrated the method. They may recognise the correct answer because it is visible among several options. They may complete a question because the necessary information is open in front of them.

These are legitimate activities. They can contribute to learning.

But they do not establish that the knowledge will remain accessible.

It helps to distinguish six outcomes: Outcome What it tells us Engagement The student participated Immediate performance The student could complete the task now Learning Knowledge or skill has changed Retention The change persists after a delay Transfer The knowledge works in a different situation Independent regulation The student can increasingly organise and check their own learning

A student might complete twenty algebra questions correctly immediately after watching a demonstration. That is encouraging performance evidence.

If they can solve similar questions a week later without the demonstration, we have stronger evidence of retention.

If they can recognise when an unfamiliar word problem requires that algebraic method, we have evidence of transfer.

If they can notice that they are forgetting the method, arrange further practice and seek a specific explanation when necessary, they are also developing control over their learning.

Our educational objective includes all of these, but they cannot be measured by the same worksheet. 2. Retrieval is an attempt, not an examination The word test can cause problems.

For some teenagers, especially those who have experienced repeated academic failure, a test means judgement. It means discovering what they cannot do, perhaps in front of someone whose opinion matters.

The learning process we want is different.

Retrieval practice means attempting to bring previously encountered knowledge back into mind.

It can take many forms.

A teenager might name the parts of an animal cell from an unlabelled diagram, explain a psychological experiment without consulting notes, recall the steps in solving an equation or describe the Four Noble Truths.

The important feature is that the answer is not simply being copied or recognised.

Memory is being asked to reconstruct it. Start with a question the learner can answer Imagine a student has just learned the distinction between respiration and breathing.

Instead of asking them to explain everything they know about respiration, begin with:

Is respiration the same as breathing?

Then:

What happens during respiration?

Then:

Why do living cells need respiration?

These questions are not equally demanding.

The first asks for a distinction. The second requires recall of a process. The third requires an explanation of purpose.

For a student with insecure knowledge, the first question may be an appropriate starting point. For a student who already understands the distinction, it may be too easy to contribute much.

The aim is not to make retrieval effortless. Nor is it to make it so difficult that the student repeatedly fails without learning anything.

The aim is to create an achievable attempt that produces useful information. Use cues without confusing them with independent recall Autistic learners are not a uniform memory group. However, research has identified group-level differences in some forms of unsupported, self-organised recall. This makes it especially important not to treat an open-ended memory demand as the only legitimate form of retrieval.

Suppose the teenager cannot answer:

Tell me everything you remember about Milgram’s obedience experiment.

That failure is ambiguous.

They may not know the material. They may know individual facts but struggle to organise them. They may be uncertain where to begin. The open-ended demand itself may be excessive.

Try a more structured sequence:

  1. What was Milgram investigating?
  2. Who were the participants?
  3. What were they instructed to do?
  4. What was the main finding?
  5. What does the experiment suggest about obedience to authority? The structure supplies retrieval cues. It does not supply the answers.

If the student can answer these questions, we have learned something important: knowledge is accessible with organisation that was absent from the original task.

That does not mean we should permanently abandon unsupported recall. It means we should teach the organisation needed to make it possible.

Later, the student might use the five questions as a self-generated checklist. Eventually, they may be able to explain the experiment without consulting it.

The goal is increasingly effective retrieval, not proving that the learner can manage an unnecessarily unstructured demand. 3. What to do when the student cannot remember A retrieval attempt will sometimes fail.

That is not automatically a problem. It can reveal exactly where further teaching is needed.

The problem begins when failure is allowed to become the entire experience.

Consider this exchange:

“What’s the function of mitochondria?”

“I don’t know.”

“We did this yesterday.”

“I can’t remember.”

“Think harder.”

The student has received no new information. They may now be less willing to attempt the next question.

Compare a different approach:

“What’s the function of mitochondria?”

“I don’t know.”

“They’re involved in releasing energy. Can you remember the name of the process?”

“Respiration?”

“Yes. Mitochondria are the site of most aerobic respiration in the cell. Let’s try that again. What’s their main function?”

“They’re where most aerobic respiration happens.”

“Good. We’ll check that again tomorrow.”

The second exchange has several useful features.

The adult supplies a cue, provides the correct information and creates another opportunity for retrieval. The student is not left repeatedly guessing.

But the final answer is still highly supported and immediate. It should not be mistaken for durable independent knowledge.

The question needs to return after a delay. Match the response to the difficulty When a student cannot retrieve an answer, there are several possibilities.

They may never have understood the material. In that case, they need teaching, not repeated testing.

They may understand but have forgotten a particular fact. A brief correction and another retrieval attempt may be sufficient.

They may remember facts but not the relationship between them. They need an explanation that makes the connection clear.

They may know the answer but struggle to organise a response. A cue or outline may reveal the knowledge and provide a structure to practise.

They may be exhausted, distressed or unable to engage productively. Continuing to ask questions may produce little useful information.

The adult’s job is to identify which of these problems is most likely.

A useful rule is:

If retrieval fails, make the next attempt more learnable, not merely more insistent.

That might mean providing the first word, narrowing the question, showing a diagram, explaining a missing connection or returning briefly to a worked example.

Once the student can produce the answer, check it again after an appropriate delay.

Correcting an answer today is useful. Remembering the correction tomorrow is stronger evidence of learning. 4. Space learning across time Retrieval practice becomes particularly powerful when it is combined with spacing.

Suppose a teenager studies the structure of an animal cell on Monday.

They practise naming the organelles until they can identify every one. They finish the session with a perfect score.

There are two possible approaches.

One is to consider the topic complete and move on.

The other is to return to it on Tuesday, then later in the week, then the following week and again after a longer interval.

The second approach is more likely to reveal whether the knowledge is becoming durable. Why repetition on the same day is insufficient When students repeatedly answer the same question in a short period, their performance often improves rapidly.

But that improvement may partly reflect the fact that the answer is still highly accessible from the previous attempt.

The student is not necessarily rebuilding the memory after a meaningful delay.

Spacing changes this.

By allowing time between encounters, we create opportunities to retrieve knowledge when it is less immediately accessible.

The learner may have to work harder to remember.

That effort can be useful, provided the task remains achievable and feedback is available.

This does not mean that greater difficulty always produces greater learning. A student who cannot remember anything, receives no useful correction and repeatedly experiences failure is not benefiting simply because the activity is difficult.

The purpose of spacing is durable retrieval, not manufactured frustration. A practical spacing schedule There is no single scientifically established schedule that is optimal for every autistic teenager, every GCSE subject or every type of knowledge.

A useful starting arrangement might be: Encounter When Activity Initial learning Monday Teach, explain and practise First return Tuesday Three short retrieval questions Second return Friday Recall and correct Third return Following week Apply without notes Fourth return Two to four weeks later Mixed revision Later returns As needed Exam-style application

This is a practical scheduling example, not a validated autism-specific interval prescription.

The intervals should respond to what happens.

If the teenager recalls the information accurately and easily after several delays, the next return can be further away.

If they repeatedly forget a critical idea, revisit it sooner and investigate whether the original teaching established adequate understanding.

If they can remember definitions but cannot explain relationships, simply increasing the frequency of definition questions will not solve the underlying problem.

They need different questions. The danger of starting again every time A common problem in home education is the repeated restart.

The teenager studies a topic, forgets part of it and the parent responds by reteaching the entire lesson.

Sometimes reteaching is necessary. But often the student retains substantial knowledge and needs only a small amount of correction.

Imagine a Psychology revision session on Asch’s conformity experiments.

The student remembers the procedure, the line-judgement task and the broad result, but cannot remember the difference between normative and informational social influence.

Rather than beginning the whole topic again, isolate the gap.

Explain the distinction, ask for an example of each and revisit it later.

The knowledge already retained should become the foundation for the correction.

This is more efficient and helps the teenager recognise that forgetting one component does not mean losing everything. 5. Build a retrieval system that does not depend on the parent remembering everything The previous chapters established the importance of external structure. Here we apply that principle to memory.

A teenager studying eight GCSE subjects cannot reasonably be expected to keep every topic, every revision interval and every forgotten fact in mind.

Nor should the parent have to remember them all.

A retrieval system can hold that organisational burden.

The system needs to answer four questions:

  1. What have we studied?
  2. What needs checking?
  3. When should we check it again?
  4. What did the student demonstrate when we checked? It does not need to be elaborate.

A notebook, spreadsheet, flashcard application or simple collection of topic cards can work. The Retrieval System For each topic, record a small number of meaningful retrieval prompts.

For example: GCSE Biology: Respiration Example topic card

RETRIEVAL QUESTIONS

  1. What is the purpose of respiration?

  2. Write the word equation for aerobic respiration.

  3. Explain why respiration increases during exercise.

  4. Compare aerobic and anaerobic respiration.

    LATEST CHECK Q1 secure Q2 secure Q3 partial Q4 needs teaching

    Next action: Practise the exercise explanation and teach the comparison before retesting it.

Notice that this record does not reduce everything to a single score.

It distinguishes what is known from what is partly known and what has not yet been adequately taught.

It also records the next instructional action.

That matters because a revision system should not simply generate more tests. It should direct attention to what will improve learning. Avoid turning the tracker into another source of pressure A system intended to support learning can become counterproductive if it grows too large.

Hundreds of flashcards, multiple dashboards, elaborate colour codes and daily performance targets can create an administrative burden greater than the learning they support.

For some autistic teenagers, the visible accumulation of overdue tasks may become particularly discouraging.

The system should therefore remain small enough to operate.

A reasonable starting point is three to five questions for each important topic, with additional questions introduced as needed.

The student does not need to retrieve every fact every day.

Nor does every answer require a score.

The important question is whether the system produces useful, repeated opportunities to retrieve and apply knowledge. Who operates the system? Initially, the parent or teacher may select the questions, arrange the schedule and record the results.

That is appropriate if the teenager cannot yet manage those demands independently.

But there is a difference between providing a system and permanently controlling it.

As competence develops, the student can begin to make choices:

“Which topic needs another check?”

“Would you rather answer these aloud or type them?”

“Do you think this one is secure enough to leave for two weeks?”

The adult can still provide guidance.

Eventually, the teenager may manage much of the process themselves, using the same calendar, reminders and checklists.

Those external supports do not need to disappear.

What changes is the student’s control over them. 6. Teach students to distinguish familiarity from knowledge One of the most difficult parts of revision is deciding what has actually been learned.

A teenager may read a page and think:

I know this.

They may recognise the material, understand the sentences and remember having seen the diagram.

But recognition is not the same as independent recall.

This is a problem for learners generally, not just autistic learners.

A useful way to address it is to compare predictions with actual performance.

Before a retrieval attempt, ask:

“Do you think you could explain the three main causes of the First World War without your notes?”

The teenager might answer:

“Yes, probably.”

Then invite them to try.

Afterwards, compare the prediction with what happened.

Perhaps they recalled two causes but could not explain the third.

The important response is not:

“You thought you knew it, but you didn’t.”

It is:

“You remembered two of the three. The third needs more work. Now we know what to practise.”

This develops a more accurate understanding of learning. Do not turn every retrieval session into a reflection exercise Metacognition matters, but excessive questioning about thinking can become another demand.

A teenager who has just struggled to remember a Biology definition may not benefit from being asked to explain their entire cognitive process.

Sometimes the most useful response is simply to provide the correction and try again.

Reflection becomes valuable when it leads to a decision.

For example:

“I recognised the answer when I saw it, but I couldn’t produce it myself. I need to practise recalling it without the options.”

That is useful strategy knowledge.

The learner has identified a difference between recognition and recall and selected an appropriate response.

We will develop this more fully in the chapter on teaching the learner to learn.

For now, the principle is:

Use retrieval to discover what the learner knows, and use the result to decide what happens next. 7. Move from remembering facts to using knowledge Retrieval is necessary for much GCSE learning, but factual recall is not the final destination.

A student who remembers the definition of conformity has not necessarily learned to apply it to a new social situation.

A student who can state the quadratic formula may not recognise when to use it.

A student who remembers historical events may not be able to construct an argument about their relative importance.

We need to move beyond remembering isolated answers. Stage one: Recall the essential knowledge In History:

What happened during the Bosnian Crisis of 1908?

The student recalls the central events. Stage two: Explain relationships Why did the Bosnian Crisis increase tensions between Austria-Hungary and Serbia?

Now the student must connect facts. Stage three: Compare and discriminate How did the Bosnian Crisis differ from the Balkan Wars in its effects on European tensions?

Now the student must distinguish related events. Stage four: Apply and evaluate How far was nationalism responsible for the outbreak of war in 1914?

The student must select relevant knowledge, construct an argument and weigh competing explanations.

These are different cognitive demands.

Success at one stage does not guarantee success at the next.

However, secure factual knowledge makes the later stages more manageable because the student does not have to reconstruct every underlying fact while simultaneously organising an argument. Why varied practice matters Imagine a Maths student who has learned to solve equations of the form:

(3x + 5 = 20)

They practise ten similar equations and become accurate.

That is useful.

But suppose the next question is:

(5 - 2x = 17)

The structure is related, but the student must handle a negative coefficient.

Or suppose the problem is presented in words.

The student must now identify that an equation is needed.

To develop flexible competence, practice needs to include variation.

Initially, the variation should be controlled. Too many changes at once may overwhelm a novice.

Later, problems can become more diverse.

The eventual goal is not simply to remember a procedure when told which procedure to use.

It is to recognise the problem, select an appropriate strategy and carry it out.

This is where retention begins to support transfer. 8. Mix old and new learning A GCSE course continually introduces new material.

Without a deliberate revision system, the student’s attention can become dominated by whatever was taught most recently.

Older knowledge gradually receives less practice.

One solution is to include a small amount of previous material in current sessions.

A Biology retrieval session might contain:

This also creates opportunities to connect topics.

For example, understanding respiration helps explain how cells release energy, while knowledge of homeostasis helps explain how the body responds to changing demands. Interleaving is not randomising everything Interleaving means arranging practice so that related topics or problem types are mixed rather than always studied in separate blocks.

It can help learners distinguish between problem types and select appropriate strategies.

But it is not automatically superior in every situation.

A student who has never learned to solve a particular kind of equation may need a block of closely related guided examples before being asked to discriminate between several equation types.

Likewise, an autistic teenager who finds switching between tasks difficult may experience substantial additional demands if revision jumps rapidly between unrelated subjects.

A practical compromise is to mix related knowledge within a predictable structure.

For example, a Maths session might begin with five minutes of familiar equations, move to a new method and finish with two questions requiring the student to choose between the old and new methods.

The intellectual challenge is in recognising which method fits.

There is no educational benefit in adding unnecessary confusion about what activity comes next.

Mix the knowledge when doing so serves learning. Keep the surrounding routine as predictable as the student needs. 9. When retrieval practice becomes a source of anxiety Retrieval practice can expose gaps in knowledge.

That is one of its strengths.

But for a teenager who associates not knowing with humiliation, criticism or loss of control, that exposure can feel threatening.

A parent may intend a five-minute memory check as a helpful activity.

The teenager may experience it as another opportunity to fail.

We should not assume that every refusal of retrieval practice is caused by anxiety. The task may be too difficult, badly timed, repetitive, unclear or simply unpleasant.

But neither should we assume that the solution is to increase pressure.

The design of the retrieval activity matters. Make the purpose explicit A useful explanation might be:

“We’re checking which bits your memory can bring back today. Anything missing tells us what to practise next. This isn’t a grade.”

The student should be able to see that an incorrect answer leads to information and assistance, not punishment. Preserve meaningful choice Some teenagers prefer written questions. Others find oral questioning easier.

Some prefer to answer privately before showing their responses.

Some find flashcards manageable but dislike being questioned unexpectedly by an adult.

Where these preferences do not undermine the learning objective, they can be accommodated.

For example, a student can retrieve Biology definitions by typing them, speaking them or writing them on a small whiteboard.

If the target is biological knowledge, handwriting is not necessarily the skill being assessed.

If the target is producing a handwritten examination answer, handwriting may eventually need to be practised as a separate component. Keep retrieval predictable A short routine can reduce uncertainty.

For example:

“At the beginning of Biology, we’ll do three questions from previous topics. You can answer on paper or aloud. We’ll check them together, then move on.”

The student knows the task, its length and what happens afterwards.

This is different from unexpectedly interrupting them throughout the day with questions.

Surprise questioning is not required for retrieval practice to work. Respond to distress without abandoning learning If a student becomes distressed during retrieval, the immediate objective may need to change.

Perhaps the questions are too difficult. Perhaps they have had an exhausting day. Perhaps the activity has become associated with conflict.

Reduce the demand, change the format or return to teaching.

Then consider how to rebuild a manageable retrieval routine.

The longer-term goal is not to protect the student from ever discovering that they have forgotten something.

It is to help them experience forgetting as a normal, manageable part of learning.

That requires repeated encounters in which a gap is identified, corrected and eventually overcome.

The experience we want is:

I couldn’t remember that yesterday. I practised it. Today I can.

Not:

Every time someone tests me, I discover another reason I’m bad at this. 10. A complete example: learning GCSE Psychology Consider a teenager studying Milgram’s obedience experiment.

The parent wants the teenager to remember the procedure, understand the findings, explain agency theory and apply the research to an unfamiliar examination question.

The student has previously watched a lesson and completed a worksheet.

Here is how the learning might develop over several weeks. Day 1: Establish understanding Begin by checking whether the student understands the experiment.

Ask them to explain what Milgram wanted to investigate.

If necessary, provide a short explanation of the procedure, using a diagram or written sequence.

Then check the essential facts.

Who were the participants? What were they instructed to do? What did the apparent electric shocks represent? What was the principal finding?

At this stage, the student may need prompts and notes.

The priority is accurate understanding.

Finish with two or three questions answered without looking at the material.

Correct any important errors. Day 2: Retrieve and reconstruct Begin with a brief retrieval activity.

Ask the student to describe the experiment using five headings:

Aim, participants, procedure, results, conclusion.

If they cannot recall the result, provide a cue or correction.

Then ask them to reconstruct the explanation.

Do not require them to rewrite the entire lesson.

The purpose is to recover and strengthen the essential knowledge. Day 5: Explain rather than list Return to the topic.

This time ask:

Why did Milgram believe his findings helped explain obedience to authority?

The student must connect the procedure and findings to a psychological explanation.

If they can recall the facts but cannot explain the relationship, teach that relationship explicitly.

Then ask for another explanation in their own words. Week 2: Compare and apply Introduce a question about how obedience might change when the authority figure is less physically present.

The student must apply their knowledge to a variation.

They may need to retrieve findings about proximity and authority.

If the relevant knowledge is missing, correct it.

If the facts are known but the student cannot structure an examination response, model the response structure.

Those are different problems. Week 4: Independent examination practice Present an unfamiliar question requiring the student to explain a factor affecting obedience.

Allow the student to attempt it without notes, using any agreed access arrangements.

Mark the answer against appropriate criteria.

If the student remembers the knowledge but omits an explanation, focus on examination technique.

If the student cannot remember the knowledge, schedule further retrieval.

If they can answer with a checklist but not without it, decide whether the checklist is still a useful learning scaffold or whether independent organisation should now be practised.

The decision depends on the current objective. What has developed? Over these sessions, the student has had opportunities to:

Rather than simply knowing that the Psychology worksheet was completed, they can see which knowledge has survived, which explanations remain weak and which skills are becoming more independent.

That is the purpose of the system. 11. A realistic weekly routine For a teenager studying several GCSE subjects, the greatest danger is making retrieval practice into an enormous additional programme.

If the student already finds the school day demanding, adding thirty minutes of compulsory revision to every subject may be unsustainable.

The solution is to integrate retrieval into existing learning.

A possible routine is: Point in the week Activity Start of each subject session 3–5 minutes of previous-topic retrieval After new teaching 2–3 questions checking the central ideas Midweek Short return to one topic that needs correction End of week 10–15 minutes of mixed retrieval across selected subjects Following weeks Return to topics according to demonstrated retention

These durations are practical starting points, not evidence-based minimum doses.

For a student who is currently struggling to engage, even two successful retrieval questions may be a worthwhile beginning.

For a student who is comfortable with the routine and preparing for examinations, longer independent sessions may be appropriate.

The amount should reflect the learner’s capacity, the curriculum demands and the evidence of what is being retained. Do not revise everything equally Some knowledge is already secure.

Some is partially learned.

Some has not yet been properly understood.

Treating all three categories identically wastes time.

A simple decision rule is: The Retrieval Decision Rule Can the student retrieve and use the knowledge after a delay?

Judge the answer against the actual learning objective, not just whether it sounds familiar.

Yes, securely

Increase the interval and later test application.

Partially

Correct the gap and retrieve again soon.

Not understood

Teach or model before further testing.

Cannot engage

Investigate access, capacity and timing.

This is deliberately simple.

A parent should be able to use it without becoming a specialist in memory research.

The central idea is to let the student’s actual performance determine the next action. 12. When the system is not working Imagine that after three weeks of retrieval practice, the teenager still cannot remember important Biology terminology.

The parent might conclude that the student needs more repetition.

Perhaps they do.

But repetition is only one possibility.

The original explanation may not have established meaningful understanding. The questions may be too open-ended. The student may be confusing similar terms. The intervals may be too long for the current level of learning. The format may introduce unnecessary executive or language demands.

Or the student may remember definitions but struggle to retrieve them under particular conditions.

The correct response is to investigate. Look for the pattern If the student remembers answers immediately but not the next day, shorten the interval and strengthen initial learning.

If they recognise answers but cannot recall them, provide retrieval cues and practise producing answers rather than only selecting them.

If they remember isolated facts but cannot explain a process, teach the relationships.

If they can explain orally but cannot produce a written answer, investigate writing and examination demands separately.

If they can retrieve knowledge at home but struggle in timed assessments, consider the demands of time pressure, unfamiliar questions, executive organisation and access conditions.

If retrieval practice consistently leads to distress, examine the difficulty, format, predictability, feedback and relationship around the activity.

These patterns are more informative than a single low score. Avoid interpreting every difficulty as an autism characteristic Autistic students differ considerably in their memory profiles, interests, language abilities, executive functioning and responses to instruction.

A particular teenager may have excellent factual memory but difficulty applying knowledge.

Another may understand complex ideas but struggle to recall precise terminology.

Another may remember material accurately yet find open-ended questioning difficult.

Diagnosis alone cannot tell us which is true.

The purpose of assessment is to discover the individual learner’s actual needs. 13. Helping the learner take ownership A successful retrieval system should eventually become something the teenager can use, rather than something that happens to them.

That change does not have to occur all at once.

Initially, the adult may decide which topics to retrieve, prepare the questions, arrange the schedule and identify the gaps.

Later, the teenager might select a topic from a prepared list.

Then they might choose which questions need another attempt.

Eventually, they may begin noticing gaps without adult prompting.

“I keep forgetting the difference between normative and informational social influence. I’ll put that back into tomorrow’s revision.”

That statement represents more than factual knowledge.

The student has recognised a learning problem, identified a relevant strategy and initiated a response.

This is an early form of independent learning regulation.

It may still depend on a calendar, flashcard application or reminder.

That is not a failure of independence.

The learner is increasingly operating the system themselves. A weekly ownership conversation Once the routine is established, a short weekly conversation can help transfer control.

Ask:

“What did you remember better this week?”

“Which topic still needs work?”

“What would help you practise it?”

Initially, the student may not know.

Offer two meaningful options.

“We could use the diagram again, or we could try three short questions. Which would be more useful?”

Over time, the student may become better at selecting an appropriate method.

The aim is not to demand sophisticated reflection after every exercise.

It is to help the teenager build practical knowledge about what improves their learning.

We will return to this in the chapters on metacognition, external organisation and handing over responsibility.

For now, the important distinction is between a student who completes revision because the parent directs every step and one who increasingly understands what revision is for and how to use it.

Both may be learning.

But only the second is developing greater control over the process. 14. What progress should look like At the beginning of this chapter, our Biology student could answer questions immediately after the lesson but struggled several weeks later.

Suppose we now introduce a retrieval system.

During the first week, they can recall the definition of respiration but cannot explain why it matters during exercise.

After correction and further practice, they can explain the relationship with prompts.

A week later, they can answer the question without prompts.

A month later, they can apply the same knowledge to a question about muscle cells and oxygen supply.

Eventually, they notice when they have forgotten a related concept and arrange to revisit it.

These are different kinds of progress.

The immediate answer demonstrates performance.

The later answer provides evidence of retention.

The unfamiliar question provides evidence of application and transfer.

The student’s decision to revisit a weak area demonstrates developing regulation.

Not every topic will follow this sequence smoothly. Some knowledge will require repeated reteaching. Some questions will remain difficult. Some skills will develop more quickly than others.

That is normal.

The important thing is that the educational system can detect the difference between a temporary success and a durable capability.

A completed worksheet tells us that work happened.

A successful delayed retrieval tells us something about what remains.

An unfamiliar application tells us something about what the learner can do with it.

And when the teenager can increasingly organise those opportunities themselves, the learning process is beginning to belong to them.

Putting it into practice: The seven-day experiment The reader does not need to redesign the entire GCSE programme immediately.

Begin with one subject and one topic.

Choose something the teenager has already studied, such as cell structure, algebraic equations, a History event or a Psychology experiment.

Prepare five short retrieval questions covering the most important knowledge.

On the first day, ask the questions without notes. Allow the student to answer in an accessible format. Correct errors and briefly reteach anything that was not understood.

On the following day, ask the questions again, changing the wording where appropriate.

A few days later, repeat the retrieval and include one question requiring explanation or application.

At the end of the week, return to the topic once more.

Record what happened.

Did the student remember more? Which ideas remained difficult? Did they require fewer cues? Could they explain something they previously could only name? Was the routine manageable?

If the experiment was useful, retain it.

If it was not, change one part of the design and try again.

The objective is not to demonstrate that a particular technique always works.

It is to establish a sustainable way of finding out what the learner actually remembers and what they need next. The chapter in one page Make Knowledge Stick THE PRACTICAL LEARNING CYCLE Teach and establish understanding

Attempt retrieval

Identify gaps and provide correction

Retrieve again after a delay

Vary and apply the knowledge

Increase learner control of the process

Keep these distinctions visible:

Final thought One of the most discouraging experiences in education is discovering that something which seemed understood has been forgotten.

For a teenager who already doubts their academic ability, this can feel like evidence that learning is pointless.

But forgetting does not mean that the original effort was wasted, nor that the student is incapable of learning.

It means that the knowledge may need further retrieval, correction, organisation or application.

A good learning system expects this.

It does not treat every forgotten answer as a failure. It makes forgetting visible early enough to do something about it.

Over time, the teenager can begin to recognise a different pattern.

They learn something. They forget part of it. They retrieve what they can. They correct what is missing. They return to it. More remains available.

Eventually, knowledge that once required substantial help becomes something they can use.

And that is the real achievement.

We are not trying to make the student remember everything immediately. We are teaching them how to build knowledge that lasts.

Evidence notes The following sources underpin the chapter’s principal claims. Claim Evidence status Principal source Retrieval practice improves delayed retention compared with restudy in many settings Strong general evidence Agarwal et al. (2021), classroom systematic review Distributed practice generally improves retention Strong general evidence Cepeda et al. (2006), meta-analysis; later classroom syntheses Retrieval practice and corrective feedback can benefit autistic learners Emerging direct autism evidence, adult participants Cairney, Haebig & Lucas (2026) Autistic learners may experience difficulties with some forms of unsupported, self-organised recall Autism-specific evidence, heterogeneous findings Autism memory research synthesised in Research Priority 03 Executive-function differences can create additional demands around organising revision Strong group-level evidence, individual variability Executive-function meta-analysis (2024) 2024 - Executive function in children with neurodevelopmental conditions- a systematic review and meta-analysis.pdf Structured, personalised supports can improve accessibility Autism-specific guidance and research Autism CRC (2026) National_Guidance_for_best_practice_in_inclusive_education_for_autistic_students.pdf A complete parent-supported GCSE retrieval system improves attainment and independence Plausible synthesis, not directly established This book’s integration of learning science and autism education research

The EEF’s Metacognition and Self-Regulated Learning guidance also supports explicit strategy teaching, modelling and developing students’ ability to plan, monitor and evaluate learning. EEF - Metacognition and Self-Regulated Learning - Guidance Report.pdf

The chapter’s retrieval tracker, decision rule, weekly routine and seven-day experiment are practical applications of the evidence. They are not individually validated intervention packages. Evidence audit Strongly supported: Retrieval practice, distributed learning, informative correction and the distinction between immediate performance and delayed retention are well grounded in general cognitive and educational research.

Supported with autism-specific qualifications: Emerging controlled research demonstrates retrieval-practice benefits among autistic adults. Research on executive function, memory and educational accessibility provides reasons to individualise retrieval demands, cues, organisation and environmental conditions.

Reasonable extrapolation: Combining these findings into a GCSE revision system for autistic adolescents, with adaptive intervals and progressive student ownership, is a defensible practical synthesis. Its complete effectiveness has not been directly demonstrated.

Important boundary conditions: Retrieval should follow adequate teaching, correction must be informative, anxiety and access needs matter, and improving factual retention does not automatically establish conceptual understanding, transfer or examination attainment. Evidence debt Before publication, the following points require final source verification or further investigation:

  1. Verify the complete bibliographic details and outcome measures of Cairney, Haebig and Lucas (2026), including the precise role of post-retrieval feedback.
  2. Check the latest classroom meta-analytic estimates for distributed practice, including how effects vary by retention interval, subject and learner age.
  3. Review direct research on autistic adolescents’ cued recall, free recall and self-organised retrieval to avoid overgeneralising from heterogeneous memory findings.
  4. Establish whether newer autism-specific research has evaluated spaced retrieval, mixed practice or curriculum-level retention in secondary education.
  5. Verify the extent to which retrieval-practice interventions improve transfer and examination outcomes rather than factual retention alone.
  6. Examine evidence on retrieval-related distress and tolerability in autistic adolescents, particularly where assessment anxiety or previous educational adversity is present. Editorial assessment: This chapter advances the book from effective initial teaching to durable learning without repeating the earlier instructional chapters. It introduces one central practical instrument, the Retrieval System, while leaving detailed metacognitive strategy selection and learner-controlled planning to the later chapters.

← All chapters