There is no single best MRCPsych Paper A resource, and any page that names one is guessing at your situation. The right choice depends on how many weeks you have, what you can spend, and where your knowledge is genuinely thin. This decision tree matches question banks, courses and official material to five candidate profiles, then adds budget and time branches. Throughout, iatroX is used as the unseen-measurement layer that tells you which branch you are actually on — not as the one bank to rule them all.
Start with the exam, not the shopping list
MRCPsych Paper A is a written paper of approximately 150 questions in three hours, roughly two-thirds single-best-answer multiple-choice questions and one-third extended matching items (EMIs), delivered by computer at Pearson VUE. The College reduced the paper from 200 to 150 questions in a recent diet, so confirm the current count and duration on rcpsych.ac.uk before you plan. Paper A tests the scientific and theoretical basis of psychiatry: behavioural science and sociocultural psychiatry, human development, basic neurosciences, clinical psychopharmacology, and classification and assessment. The clinical CASC (an OSCE) and the critical-appraisal-heavy Paper B are separate assessments; nothing in this article prepares you for the CASC, and iatroX is explicitly not a CASC simulator.
Two facts about the format drive every resource decision. First, one-third of the paper is EMIs, so any stack that drills only single-best-answer stems leaves a third of the format under-rehearsed. Second, psychopharmacology and neuroscience carry disproportionate marks relative to how comfortable most trainees feel with them, so a resource that merely lets you coast through behavioural science is not testing you where the paper will.
The five candidate profiles
Before you compare products, place yourself in one of these groups, because the same resource serves them differently.
- First attempt, clean slate. You have not sat Paper A and need breadth, structure and a sense of the format. Your risk is drifting through comfortable topics.
- Retake after a near miss. You have a score report and a specific pattern of failure. Your risk is repeating the same bank and mistaking familiarity for competence.
- Busy trainee revising around a full clinical rota. Time, not money, is your constraint. Your risk is starting resources you cannot finish.
- Weak foundations. Basic sciences — neuroscience, statistics-adjacent psychology, pharmacology mechanisms — were never solid. Your risk is drilling questions you cannot yet reason through.
- Strong knowledge, weak pacing, or strong MCQ and weak EMI. You know the material but run out of time, or you handle stems well and fall apart on matching sets. Your risk is measuring the wrong thing (overall percentage) and missing the real gap.
The primary keyword people search — the best MRCPsych Paper A resources by candidate type — only makes sense once you have named your type. What follows is a decision tree, not a ranking.
Define the minimum stack
Most candidates over-buy. A defensible Paper A stack has four slots, and only two are compulsory:
- One primary question bank (compulsory) — your daily driver for volume and explanations.
- Official calibration material (compulsory) — the free RCPsych sample questions, used to check that your bank's difficulty and style match the real paper.
- One teaching or reference source (optional) — notes, a crash course or a textbook, added only where a domain is genuinely un-taught, not everywhere.
- One modality or measurement tool (optional) — a second, unseen bank for timed mixed blocks, or a spaced-retrieval habit, added when you need to know whether learning is transferring.
If you add a fifth thing, remove something. Two overlapping question banks used in parallel usually means you finish neither and calibrate against neither.
Map the principal platforms to their strongest job
The table below maps the main Paper A resources to the one job each does best. Prices and question counts are vendor-reported and were last checked on 20 July 2026; verify every figure on the product page before you buy, because banks re-price and re-count frequently.
| Resource (verify on the product page) | What it is | The one job it is strongest at | Indicative price, vendor-reported (GBP) |
|---|---|---|---|
| Official RCPsych sample questions | Free specimen items on the College website | Format calibration — the gold standard for question style and difficulty | Free |
| SPMM | SmartRevise notes, HiYield subject question banks, a 20-hour e-Crashclass and reconstructed mock papers | Conceptual teaching plus exam-realistic mock papers | e-Crashclass £199; Starter/HiYield £299; Mock papers £249 |
| PassMRCPsych | A large MCQ bank (5,000+ items), revision notes/online textbook and five 150-question mock exams | High-volume MCQ drilling with structured notes attached | £79 / 3 months; £129 / 6 months |
| BMJ OnExamination (Paper A) | 970+ questions with performance analytics, peer percentiles and mock tests | Performance analytics on a polished, well-supported platform | £49.99–£149.99 by term length |
| MRCPsychMentor | An established Paper A bank and notes | Affordable breadth for early revision | Verify current count and price on the product page |
| iatroX MRCPsych Paper A | Unseen, mixed, timed blocks with a Socratic Tutor for missed items | Independent measurement and transfer as a second bank | UK-core content free; verify current access |
Read the table as "which job," not "which winner." SPMM is a strong option when you need teaching and realistic mocks; PassMRCPsych is a strong option when you want volume with notes in the same place; BMJ OnExamination is worth including when analytics and platform polish matter to you. iatroX earns its slot as the measurement layer precisely because it is not your teaching bank — the second bank in the two-Q-bank rule works only if it holds items you have never seen.
One honest note belongs here. If a colleague points you to "Psychiatry-UK courses," check what you are actually being sent to: Psychiatry-UK is primarily a clinical service, not a verified Paper A course provider. Do not build a plan around a course you cannot confirm exists.
Budget bands
Free / low-cost. The RCPsych sample questions plus a free UK-core bank (iatroX) plus disciplined self-testing will get a well-prepared candidate a long way, and it is the correct band if money is genuinely tight. What you sacrifice is breadth of explanations and reconstructed-mock realism.
One premium resource. Add a single paid bank or course chosen for your profile: SPMM if you need teaching, PassMRCPsych if you need volume, BMJ OnExamination if you need analytics. Keep the free official material and a free measurement bank alongside it. This band suits most candidates.
Comprehensive stack. One primary bank, one teaching source for weak domains, the mock papers, and a measurement bank. This is defensible only if you have the weeks to use all of it; otherwise it is expensive shelfware. Buy the mock papers late, not early.
Time bands — and what to cut
More than twelve weeks. You can run a full loop: teaching for weak domains, a primary bank first pass by topic, then mixed timed blocks, with official material for calibration at the midpoint and the end. Do not buy the reconstructed mock papers yet; they are most useful in the final weeks.
Four to twelve weeks. Drop topic-by-topic teaching except for your one or two weakest domains. Start mixed sooner. Use the bank for volume and reserve a protected pool of unseen questions for the last fortnight. Cut any second course.
Under four weeks. One bank, mixed and timed, plus the official samples and one set of mock papers. Cut all new teaching content, cut note-writing, and cut any resource you have not already started. At this range, the job is calibration and pacing, not learning new frameworks.
State the cut explicitly. The commonest failure in a short run is trying to do the twelve-week plan in three weeks and finishing none of it.
Worked examples
Profile A — first attempt, ten weeks, one premium resource. A CT1 sits Paper A in ten weeks with a modest budget. Stack: PassMRCPsych as the primary bank, iatroX as the unseen measurement bank, RCPsych samples for calibration. Weeks 1–6: 40 questions a day on PassMRCPsych, first pass mixed except for neuroscience and psychopharmacology, which get short topic-filtered blocks. Weeks 7–9: mixed timed blocks of 50, alternating with a weekly unseen iatroX block to check transfer. Week 10: official samples plus one mock paper. Exit criterion: unseen timed accuracy stable at or above the candidate's target across two consecutive mixed blocks, with no domain floor unmet — not "bank finished."
Profile B — retake, six weeks, strong MCQ and weak EMI. A trainee narrowly failed, with a score report showing solid single-best-answer performance and weak extended-matching sets, especially in psychopharmacology. Repeating the previous bank would mainly rehearse remembered answers. Stack: SPMM e-Crashclass for the psychopharmacology conceptual gap, iatroX for unseen EMI-style and mixed transfer practice, and the reconstructed mock papers in the last fortnight. Weeks 1–2: watch bounded segments of the crash course, then immediately test on fresh items. Weeks 3–5: mixed timed blocks weighted toward EMIs and pharmacology, all unseen where possible. Week 6: mocks and official samples. Exit criterion: EMI-set accuracy rising on unseen items and pharmacology no longer the lowest domain.
Profile C — busy trainee, sixteen weeks, forty minutes a day. Time, not money, is the limit. Stack: one primary bank used in short daily blocks, iatroX for a weekly unseen measurement block at the weekend, official samples at weeks 8 and 16. Twenty questions on weekdays with full explanation review; one 50-question timed unseen block on Saturday. No second course; no note-writing beyond one-line error codes. Exit criterion: coverage of every domain reached and the weekly unseen block stable, even if the primary bank is not 100% complete.
The evidence hierarchy you should trust
When resources disagree, rank your sources: the official RCPsych material first for format and difficulty; primary guidance (NICE, CKS, SIGN, and the SmPC/eMC for medicines facts) first for content, never a bank's paraphrase; vendor pages for product facts, treated as claims to verify; and independent, hands-on testing for user experience. A bank's explanation is a teaching aid, not a citation. Your overall Q-bank percentage sits near the bottom of the hierarchy as a signal — see why it is not your exam score.
The cannibalisation guardrail
This page is the hub for one decision: which Paper A resource fits your profile. It deliberately does not reproduce the deep, single-platform audits — the detailed PassMRCPsych workflow, the SPMM conversion method, the BMJ OnExamination analytics teardown — that live as their own articles. When you have chosen a branch, follow the narrow child article for that platform rather than expecting full evidence here, and use the iatroX comparison hub to line the options up. Building a blueprint-coverage matrix for your own attempts will do more than any ranking to show you which resource you actually need.
Bottom line
Choose by profile, budget and weeks-to-exam, not by brand loyalty or a headline question count. Most candidates need exactly one primary bank, the free official material, and a way to measure transfer on unseen items — and only some need a teaching course on top. Pick the resource that does the one job your profile demands, protect a pool of unseen questions to measure yourself honestly, and let the measurement, not the marketing, tell you when to switch or stop.
Frequently asked questions
How do I know whether I have covered the full MRCPsych Paper A blueprint? Completion of a bank is not coverage of the blueprint, because banks are weighted by what is easy to write, not by the College's syllabus. Map your attempted questions against the published Paper A domains — behavioural science and sociocultural psychiatry, human development, basic neurosciences, clinical psychopharmacology, and classification and assessment — and count attempts per domain rather than trusting a single completion figure. A blueprint-coverage matrix, updated weekly, will expose the low-volume domains (often neuroscience and psychopharmacology) that a topic-agnostic feed lets you skip. You have covered the blueprint when every domain has been tested on unseen, timed items, not when a progress bar reads 100%.
Can one question bank be enough for MRCPsych Paper A? One well-chosen bank can carry the bulk of your learning, but a single bank cannot honestly measure itself, because by the second pass you are recognising items rather than reasoning through them. The pragmatic answer is one primary bank for volume and explanations plus a small reserve of genuinely unseen questions for timed mixed assessment, which is the logic of the two-Q-bank rule. If budget forces a single paid resource, keep the free official samples and a free unseen bank as your measurement layer so you are not grading yourself on questions you have already memorised.
What should I measure instead of my overall Q-bank percentage for MRCPsych Paper A? Measure four things your headline percentage hides: coverage (attempts and accuracy per blueprint domain), unseen timed accuracy (performance on first-seen items under exam pacing), pacing (questions completed per minute against the roughly 72-second budget), and retention (accuracy on items revisited after a spacing interval). Your overall percentage rises simply from re-seeing questions, so it drifts upward while true readiness stays flat. Track the domain floors and the unseen trend; those move for the right reasons.
When should I stop doing new MRCPsych Paper A questions? Stop generating volume when new questions stop changing your error pattern — when your misses are careless slips rather than knowledge gaps, every domain floor is met, and your unseen timed accuracy has plateaued across two or three mixed blocks. Beyond that point, additional questions mostly rehearse what you already know while eating revision time you should spend on spaced review of prior misses and one or two full timed papers. The signal to stop is a stable unseen score with even coverage, not an emptied bank.
Which MRCPsych Paper A resource should I use for my weakest component? Match the resource to the type of weakness, not to its brand. For an un-taught conceptual gap — neuroscience or psychopharmacology mechanisms you never learned — a teaching source such as an SPMM crash course or focused notes comes first, then testing. For a format weakness — strong on single-best-answer stems, weak on EMIs — you need a bank with genuine extended-matching volume and then unseen EMI-style practice to confirm transfer. For a pacing weakness, no new content helps; you need timed mixed blocks. Diagnose the weakness with a baseline first, then pick the tool built for that job.
Editorial notes and references
Written by Dr Kolawole Tytler, NHS GP and founder of iatroX. Last checked 20 July 2026. Question counts, prices and features cited here are vendor-reported and change without notice; verify every figure on the relevant product page and confirm the current exam format on rcpsych.ac.uk before you rely on it. Disclosure: iatroX operates a competing MRCPsych Paper A question bank, so its role in this article is confined to the jobs the other products do not claim — independent, unseen, timed measurement and transfer practice, not teaching or CASC simulation. Corrections are welcome via the feedback route on iatrox.com.
References: Royal College of Psychiatrists — exams and preparing-for-exams pages (rcpsych.ac.uk) and the free Paper A sample questions; SPMM Course package pages (spmmcourse.com); PassMRCPsych Paper A (passmrcpsych.com); BMJ OnExamination MRCPsych Paper A (onexamination.com); iatroX internal references — the MRCPsych Paper A bank and why your Q-bank percentage is not your exam score.