Should MRCPCH revision match the exam format?

MRCPCH revision should match the examination selectively, not constantly. Use the current computer-based single-best-answer format to rehearse navigation, option comparison, pacing and sustained attention. Use other formats - free recall, explanation, flashcards and untimed questions - to build and diagnose knowledge. Match the mental process that the assessment requires, then vary the cues so success does not depend on one screen, wording pattern or familiar option set.

This article covers non-clinical MRCPCH examination preparation only. It contains no clinical teaching, diagnosis, treatment, prescribing, dosing, safeguarding or patient-management guidance. The learning research is not MRCPCH-specific and does not establish that any practice format improves MRCPCH scores or pass rates.

The direct answer: match the demand, vary the route

The current MRCPCH theory examinations are computer-based and use single-best-answer questions. That makes some format-specific rehearsal sensible: candidates need to read a lead-in, discriminate among five options, record an answer on screen and manage the available time.

It does not follow that every revision activity should look like the examination. Learning and assessment have different jobs. A study task may deliberately remove options, require an explanation or slow the process down so that a weak memory, hidden guess or faulty distinction becomes visible.

Match practice to the examination when the target is performance under examination conditions. Vary practice when the target is durable, flexible knowledge. A complete plan needs both.

Use MATCH: Map the target, Acquire accurately, Test across formats, Calibrate in the official format and Hold separate evidence.

Key points

• Transfer-appropriate processing predicts that learning is more likely to support later performance when practice engages processes relevant to the final task. Relevant similarity is functional - not simply using the same colour screen, chair or timer.
• Meta-analyses show that retrieval practice can transfer across formats and to application or inference tasks, although transfer varies with initial learning, response relationships, feedback and the transfer demand.
• Exact-format rehearsal is most informative after a knowledge base exists. Use it to expose navigation, pacing, discrimination and endurance problems; do not let repeated familiar items or interface fluency masquerade as readiness.

What "matching the exam" actually means

Surface similarity is not the main target

Two tasks can look alike while requiring different thinking. A familiar question-bank item and a first-seen examination item may share a five-option layout, yet the familiar item can be answered from recognition of wording. Conversely, a closed-book explanation and a single-best-answer question may look different but both can require retrieval of the same distinction.

The useful question is therefore not "Does this page resemble TestReach?" It is "Which process needed on the day is being practised?" Possible targets include:

• retrieving relevant knowledge from a stem;
• identifying what the lead-in asks;
• comparing several partly plausible options;
• committing to the best available answer;
• moving on under time pressure;
• navigating a computer interface; and
• sustaining accurate work across a representative block.

Treat each as a separate practice target. A realistic-looking session is not automatically a realistic cognitive task.

Learning practice and performance rehearsal are different

During learning, support can be useful: untimed work, immediate feedback, explanation, diagrams, source checking and repeated retrieval. During performance rehearsal, support is deliberately reduced so that the candidate can observe what happens under target conditions.

Confusing these phases creates two errors. The first is making early learning unnecessarily difficult by timing every first encounter. The second is making late rehearsal unrealistically supportive by pausing, checking notes or reviewing feedback after every item.

The timed-versus-untimed guide separates those jobs in more detail.

What the evidence says

Transfer-appropriate processing is about process overlap

Morris, Bransford and Franks introduced the transfer-appropriate processing framework through experiments in which the value of an encoding activity depended on the later test. Semantic processing supported a conventional recognition test, whereas rhyme-focused encoding was better for a rhyme-focused test.

The enduring principle is not that shallow practice is generally desirable. It is that a learning activity cannot be judged without specifying the later performance it is meant to support.

For MRCPCH preparation, exact option selection can help practise option discrimination. It is less suitable for proving that the answer can be generated without option cues. The later assessment matters, but so does the diagnostic question being asked now.

Practice testing usually beats simple restudy

Adesope, Trevisan and Sundararajan's meta-analysis found an overall advantage for practice testing over restudy and other non-testing conditions. Effects varied with practice-test features, learner and study characteristics, outcome and methodology.

Rowland's meta-analysis and later applied reviews likewise support retrieval practice as a learning strategy, while emphasising moderators such as feedback, retention interval, test form and successful initial retrieval. This evidence supports using questions as more than mock examinations: a practice question can itself strengthen learning when it requires retrieval and is followed by useful feedback.

Retrieval can transfer across formats

Pan and Rickard reviewed 192 transfer effects from 122 experiments. Practice testing produced positive transfer on average relative to non-testing re-exposure. Transfer was not confined to identical final tests: benefits appeared across test formats and for application and inference questions, although results were weaker or inconsistent in some transfer categories.

This matters because "practise exactly as tested" is too simple. Matching can help, but varied retrieval does not automatically lose its value. Free recall may reveal missing knowledge; explanation may reveal a broken relationship; an SBA may reveal poor discrimination. These tasks can complement one another.

Initial learning and feedback affect transfer

Pan and Rickard identified initial test performance, response congruency and elaborated retrieval practice as important moderators. Transfer is less likely when the original attempt is mostly unsuccessful and never corrected.

Feedback is therefore part of the design. A wrong answer followed by a glance at the correct letter is not equivalent to reconstructing the relevant distinction, checking it against an authoritative source and retrieving it again later.

The evidence does not justify a universal percentage at which candidates should switch formats. Judge the transition by error patterns and the quality of independent retrieval, not by a borrowed threshold.

Verify the current MRCPCH target before rehearsing it

RCPCH's current structure page states that FOP, TAS and AKP theory examinations are computer-based and use single-best-answer questions. FOP and TAS each contain 100 questions in two hours. AKP contains 120 questions split across two papers, with 60 questions and two and a half hours per paper. RCPCH last updated the structure page on 27 August 2026.

RCPCH also provides online specimen papers so candidates can experience the computer-based presentation. The College's "how you sit your exam" page says theory examinations have been delivered on the TestReach platform since 2025 and explains current test-centre and remote-invigilation arrangements.

These are time-sensitive facts. Check the current RCPCH pages and your admission document before planning final rehearsal. Do not infer your sitting's interface, timing, venue or adjustments from an old screenshot, recollection or third-party mock.

The online-versus-test-centre guide covers venue-specific preparation. This article addresses the learning decision: how much of revision should reproduce the assessment process.

Use the MATCH framework

M - Map the target process

Choose one target before choosing a format.

Target	Useful practice form	What it can reveal
Retrieve without cues	Free recall or short-answer prompt	Missing or inaccessible knowledge
Explain a relationship	Closed-book explanation	Broken reasoning or unsupported links
Discriminate near alternatives	First-seen SBA	Option comparison and lead-in errors
Navigate the interface	Current official specimen paper	Scrolling, selection and review workflow
Manage time	Timed representative block	Pacing, moving-on and review decisions
Sustain performance	Longer planned rehearsal	Attention and endurance patterns

One session can have more than one target, but recording them separately makes the result interpretable.

A - Acquire accurately before simulating pressure

When material is new or unstable, begin with a format that supports accurate understanding and feedback. This may include a worked explanation, untimed questions or a brief source check. Then close the source and retrieve.

Do not use "exam realism" to justify repeated guessing. Timing a task that has not yet been learned measures speed plus uncertainty; it does not repair the uncertainty.

The open-versus-closed-book guide shows how to move from supported input to an independent attempt.

T - Test across formats

Use a small ladder rather than one permanent format:

1. state the principle without options;
2. explain the key distinction;
3. answer a changed cue or comparison;
4. attempt a first-seen SBA; and
5. return after a delay.

Not every topic needs every rung. Choose the smallest variation that tests whether knowledge survives beyond the original wording. If performance collapses only when options disappear, recognition may be carrying the answer. If free recall is sound but SBAs fail, discrimination or lead-in interpretation may need work.

C - Calibrate with official-format rehearsal

Use RCPCH specimen papers for current official orientation. Preserve them for planned checkpoints rather than repeatedly consuming the same items until their wording is familiar. The sample-paper guide explains diagnostic, repair and rehearsal uses.

For an exact-format checkpoint:

• use a computer and current official or independently authored material;
• remove notes and feedback until the planned block ends;
• use a predeclared time boundary;
• record a provisional answer before moving on;
• separate unanswered, guessed and low-confidence items; and
• review the process afterwards, not only the score.

Rehearsal does not need to copy irrelevant discomfort. Use your approved reasonable adjustments, safe equipment and the conditions documented for your sitting.

H - Hold separate evidence

Do not combine unlike observations into one readiness number. Track at least:

• first-seen SBA performance;
• familiar or repeated-item performance;
• untimed retrieval accuracy;
• timed pacing and completion;
• error cause; and
• delayed performance.

The question-bank score guide explains why a percentage needs context. Interface familiarity, content retrieval and decision quality can improve at different rates.

A worked non-clinical example

Suppose a candidate is learning the current structure of the theory examinations. Reading the RCPCH page establishes the authoritative facts. A free-recall prompt then asks for the components and their current structure without displaying options. A comparison prompt asks which features differ between FOP/TAS and AKP. Finally, the candidate opens the current specimen environment to practise navigation and answer recording.

These activities are related but not interchangeable:

• reading establishes the source;
• free recall tests access without cues;
• comparison tests organisation; and
• the specimen paper rehearses the target interface and response process.

Repeating the specimen page alone may make the layout familiar without proving that the structure can be recalled or that first-seen questions can be handled. Conversely, reciting the structure on paper does not show that computer navigation and pacing are ready.

This example contains exam logistics only. It does not provide clinical instruction or reproduce protected examination content.

Build rehearsal in phases

Foundation phase: vary to diagnose

Use explanations, short-answer prompts, flashcards, untimed first-seen questions and source checks. The purpose is accurate retrieval and visible feedback, not a realistic score.

Integration phase: add SBA discrimination

Increase first-seen mixed SBA practice while retaining occasional cue-free retrieval. Compare why an option was chosen, not merely whether the letter matched. Use changed wording and mixed topics so performance is not tied to one chapter or cue.

Rehearsal phase: increase target similarity

Add timed computer-based blocks, current specimen-paper orientation and longer uninterrupted sessions. Keep some varied retrieval between rehearsals to repair errors rather than repeatedly simulating them.

Final checks: rehearse the process, not predictions

Use representative blocks to test start routine, pacing checkpoints, moving-on rules and review behaviour. The time-management guide can help translate observed pacing into a personal plan.

A rehearsal score is one sample under one set of conditions. Do not convert it into a promised result.

Common mistakes

Turning every learning task into multiple choice

Options can cue recognition and conceal weak generation. Keep some cue-free prompts, especially for recurring guesses and high-confidence errors.

Mistaking the same device for the same processing

Reading notes on a laptop does not rehearse single-best-answer discrimination. Device matching is useful only when the target involves screen navigation, layout, scrolling or sustained computer work. The paper-versus-screen guide addresses the medium separately.

Repeating one official sample until it feels easy

Familiarity changes the task. Record the first attempt, review it thoroughly and later use the sample for interface rehearsal rather than treating a memorised score as new evidence.

Starting full-length timed work too early

Long simulations can identify broad problems, but they are expensive. If most errors arise from unstable knowledge, use targeted repair before another full rehearsal.

Practising only in one perfect environment

A stable study base can improve focus, but knowledge should not depend entirely on one desk, soundtrack or ritual. The study-location guide explains when controlled variation can test portability.

Copying the exam's stress rather than its task

Deliberate sleep loss, unsafe restriction of fluids, excessive sessions or denial of approved adjustments are not useful realism. Match legitimate task demands and current official rules, not preventable harm.

Use PaedVault without confusing learning and rehearsal

PaedVault question sessions can serve different purposes when they are labelled in advance.

For learning:

• use first-seen untimed questions;
• review feedback and source boundaries;
• retrieve the correction without reopening it; and
• schedule a changed return.

For integration:

• use mixed first-seen sets;
• explain the discriminating clue or decision rule;
• track low-confidence correct answers; and
• compare error causes across topics.

For rehearsal:

• choose a representative computer-based block;
• set the time boundary before starting;
• defer feedback until the block ends; and
• review pacing, unanswered items and change-of-answer decisions separately from content errors.

Do not copy, extract or share PaedVault questions, options, explanations or notes. Protected content remains licensed for personal use under the PaedVault Acceptable Use Policy.

A seven-day format audit

Use this personal comparison to decide what each format contributes.

1. Select two comparable non-clinical knowledge sets or exam-logistics topics.
2. For the first, use cue-free recall followed by feedback and a first-seen SBA check.
3. For the second, use SBA practice followed by feedback and a cue-free check.
4. Keep total time similar.
5. After two days, test both with changed wording.
6. After seven days, use one first-seen mixed block.
7. Record retrieval, discrimination, time, confidence and error cause separately.

This is not a controlled trial and cannot predict an examination result. It is a structured way to replace format preference with evidence about your own failure modes.

Conclusion

MRCPCH preparation should become more exam-like as the target shifts from building knowledge to rehearsing performance, but exact simulation should not replace varied retrieval. Match the important processes: retrieving from a stem, discriminating between plausible options, navigating the computer interface, managing time and sustaining attention.

Use MATCH: Map the target, Acquire accurately, Test across formats, Calibrate in the official format and Hold separate evidence. The strongest plan does not choose between "realistic" and "varied" practice. It uses each for the job it can actually do.

Frequently asked questions

Should all MRCPCH revision use single-best-answer questions?

No. Single-best-answer practice is essential for learning option discrimination and rehearsing the current examination process, but options can cue recognition and hide weak recall. Combine first-seen SBAs with cue-free retrieval, explanation and delayed checks. Use the format that exposes the current learning problem, then confirm performance in the official format.

When should I start timed computer-based MRCPCH practice?

Begin small timed blocks once you can complete representative questions with reasonably stable reasoning and feedback review. Increase duration as the examination approaches and use current RCPCH specimen material for interface orientation. There is no universal switch date or score; base progression on first-seen performance, pacing and error patterns.

Is practising on paper a waste of time for a computer-based exam?

No. Paper can support planning, cue-free recall or focused explanation. It is insufficient for rehearsing computer navigation, scrolling and on-screen answer recording, so add computer-based practice separately. Judge the medium by the process being trained rather than assuming every study task must copy the final interface.

Do practice questions need to use the same format as the final exam?

Not always. Meta-analytic evidence shows that retrieval practice can transfer across test formats and to application or inference tasks, although transfer varies. Use some matched SBAs for target discrimination and some changed formats to test flexible access. Feedback and successful initial learning are important conditions.

How often should I complete a full MRCPCH mock?

There is no evidence-based universal frequency. Full rehearsals are expensive, so use them when the result will change a pacing, endurance or exam-process plan. Repair identified problems before repeating. Between full mocks, use shorter representative blocks and varied retrieval so simulation does not become repeated exposure without learning.

Sources

1. Royal College of Paediatrics and Child Health. Theory exams: structure and syllabi. Updated 27 August 2026.
2. Royal College of Paediatrics and Child Health. Theory exams: specimen papers. Accessed 9 November 2026.
3. Royal College of Paediatrics and Child Health. Theory exams: how you sit your exam. Updated 27 August 2025.
4. Morris CD, Bransford JD, Franks JJ. Levels of processing versus transfer appropriate processing. Journal of Verbal Learning and Verbal Behavior. 1977;16(5):519 - 533.
5. Adesope OO, Trevisan DA, Sundararajan N. Rethinking the use of tests: a meta-analysis of practice testing. Review of Educational Research. 2017;87(3):659 - 701.
6. Pan SC, Rickard TC. Transfer of test-enhanced learning: meta-analytic review and synthesis. Psychological Bulletin. 2018;144(7):710 - 756. doi:10.1037/bul0000151.
7. Rowland CA. The effect of testing versus restudy on retention: a meta-analytic review of the testing effect. Psychological Bulletin. 2014;140(6):1432 - 1463.
8. Roediger HL, Butler AC. The critical role of retrieval practice in long-term retention. Trends in Cognitive Sciences. 2011;15(1):20 - 27.
9. Agarwal PK, Nunes LD, Blunt JR. Retrieval practice consistently benefits student learning: a systematic review of applied research in schools and classrooms. Educational Psychology Review. 2021;33:1409 - 1453.
10. Larsen DP, Butler AC, Roediger HL. Test-enhanced learning in health professions education: a systematic review. Medical Teacher. 2018;40(4):337 - 350.
11. PaedVault. Acceptable Use Policy. Updated 28 July 2026.