Bennett Mechanical Test: All Versions Explained (BMCT, BMCT-II)

The Bennett Mechanical Comprehension Test exists in three live versions in 2026: the original BMCT (68 questions, 30 minutes, still used in some legacy industrial and apprenticeship contexts), the BMCT-II (55 questions, 25 minutes, the default at most current Pearson TalentLens deployments), and the BMCT-II Form S and Form T pair (the same 55-question paper split across an unproctored screening sit and a proctored verification sit). All three cover the same 10 physics topic areas, but the pacing, the visual style of the items, and the cutoff bands differ. This article works through each version with the practical detail candidates need.

Quick takeaways

• The original BMCT is 68 questions in 30 minutes, which works out to roughly 26 seconds per question. It is still used in older industrial hiring pipelines and some military-adjacent roles, but most current employers have migrated to BMCT-II.
• The BMCT-II is 55 questions in 25 minutes, around 27 seconds per question. It is the default product Pearson TalentLens sells today.
• Form S and Form T are not different tests. They are two equivalent BMCT-II papers, with Form S typically run unproctored for screening and Form T run proctored for verification.
• All four versions cover the same 10 physics topic areas: pulleys, gears, levers, springs, hydraulics, structures, electricity, heat, motion, and sound and optics.
• A calculator is not allowed in any Bennett version. Items are visual schematics with multiple-choice answers.
• Typical employer cutoffs cluster at the 70th percentile for manufacturing and industrial, and at the 75th to 80th percentile for engineering apprenticeship programs.

The format card below pins the version differences side-by-side. The full version comparison and topic coverage are in the infographic further down.

The original BMCT (1940, still in circulation)

The original BMCT was published by George K. Bennett in 1940 and is still licensed in some legacy assessment contracts. It runs 68 multiple-choice items in 30 minutes. The items are simple line drawings showing pulleys, levers, gears, and other mechanisms, each paired with a short question and three answer choices. The visual style is deliberately old-school, which throws candidates who only practised on cleaner modern test papers.

In 2026, the original BMCT shows up mostly in three places. Some UK and Commonwealth apprenticeship programs use the paper form for in-person assessment days. A handful of US manufacturing employers (typically older Fortune 1000 industrials with long-running hiring pipelines) still order the original. And it appears in some military adjacent programs in countries that licensed it before the BMCT-II rolled out. If your invitation email mentions "Bennett Mechanical Comprehension Test" without a "-II" or "Form" suffix, you are almost certainly on the original.

The pacing is 26 seconds per question, which is faster than most candidates assume from looking at the page. That pacing only works if you read the diagram before the question text on roughly 60 percent of items, because the diagram tells you immediately what physical principle is involved.

The BMCT-II (2008, the modern default)

Pearson published the BMCT-II in 2008, and it is the version most candidates meet today. It is 55 questions in 25 minutes, which is 27 seconds per question. The topic coverage is identical to the original, but the items were rewritten to remove a few outdated references (one famous item in the original involves a streetcar) and to update the visual style to a cleaner line-drawing standard. The available formats are online (the default) and paper (less common, used mostly when an assessment day is hosted in person).

The BMCT-II is what Pearson TalentLens sells when an employer says "we use Bennett." It comes scored against an industry norm group chosen by the employer at purchase. Most manufacturing employers pick a "Production and Maintenance" norm; engineering apprenticeships pick an "Engineering Technician" norm. The same raw score can land at different percentiles against those two norms.

For the deeper format and practice walkthrough at the BMCT-II level, the Bennett Mechanical Comprehension Test pillar covers the full sit-day workflow, and the Bennett Mechanical Practice Test shows worked solutions across each topic area.

Form S and Form T: the two-step proctored workflow

Form S and Form T are not separate tests. They are two equivalent BMCT-II papers (same 55 items each, same 25 minutes, same topic coverage) used together when the employer runs a two-step screening. Form S is the initial unproctored screen, taken from home. Form T is the proctored verification sit, run after the candidate clears Form S.

The workflow protects against unproctored cheating without burning the proctoring budget on every applicant. If your Form S score is anomalously high compared to your Form T sit, the employer's report flags it and the application is typically rejected. If the two scores are reasonably close (Pearson defines "reasonably close" via a statistical band, not a fixed number), the Form T result is the official one.

Candidates who do well on Form S and then bomb Form T usually fail because they did not practice under proctored conditions. The Form S sit at home permits informal handicaps (a notebook, a desk calculator, an extra monitor) that some candidates rely on without realising they are leaning on them. Form T is video proctored with one screen and no notes, and the pacing pressure hits harder.

The infographic below maps each version to its question count, time limit, format, and who still uses it.

The 10 topic areas (same across every version)

Every Bennett version covers the same 10 physics topic areas. The relative weighting shifts slightly between the original and the BMCT-II (the BMCT-II tilts a little more toward applied electricity and a little less toward classical mechanics), but the headline topics are the same.

Candidates who prepare for one topic at a time tend to outperform candidates who do mixed practice from the start. The reason is that each topic uses a small number of physical principles, and once you have the principle locked in (mechanical advantage for pulleys, ratio of teeth for gears) most items in that topic resolve in under 15 seconds.

Cutoffs by employer and industry (2026)

The Bennett does not publish a fixed pass score. Employers choose a percentile cutoff against their selected norm group. These are the bands reported by candidates and recruiter sources across 2025 and 2026 cycles.

Two operational notes. The "Production and Maintenance" norm group is more lenient than the "Engineering Technician" norm group, so the same raw score can map to a 75th percentile against one and a 65th against the other. And many employers in the manufacturing tier set a soft cutoff (below which the resume goes straight to the bottom of the stack) rather than a hard reject.

For the cross-publisher view of how Bennett compares to Ramsay MAT-4 and Wiesen WTMA on format, visual style, and cutoffs, see Mechanical Reasoning Test: How the Bennett, Ramsay, and Wiesen Compare.

Why this test trips even strong physics candidates

The Bennett is a comprehension test, not a calculation test, and the failure modes are different from a math-heavy aptitude test.

The first failure mode is mis-reading the diagram. The visual schematics are deliberately simple, and candidates who try to mentally re-build the diagram (rotating it, tracking every pulley line) lose 30 seconds per item. Candidates who look at the diagram and immediately ask "what is the physical principle here" finish on time. Practice rewires this.

The second is over-fitting on classical mechanics. Most candidates over-prep on pulleys, gears, and levers because those are the obvious items. The result is that they bank time on those topics but lose disproportionate points on electricity, heat, and the sound-and-optics tail. Splitting practice evenly across all 10 topics is what closes the gap.

The third is calculator instinct. Candidates from STEM degrees often hesitate on items that look numerical (a pulley with 4:1 advantage, a gear train with two stages) because they want to verify with a calculator. There is no calculator. The arithmetic is intentionally small enough to do in your head, but only if you trust the principle.

A 5-day prep plan for the BMCT-II

Day 1 is diagnostic plus topic mapping. Take a 55-question full paper under timed conditions. Score it. Map your wrong answers to the 10 topic areas and rank the topics by miss count. The top three topics are your focus for Days 2 and 3.

Day 2 covers the weakest topic in 90 minutes. Worked solutions, not just answers. The point is to spot the physical principle behind each item, not to memorise the answer. A candidate who understands the principle behind "two pulleys reduce the input force to one half" can answer 6 to 8 pulley questions in the actual test without re-deriving anything.

Day 3 covers the second-weakest topic. Same 90-minute block.

Day 4 is a full mixed paper plus a topic-specific quiz on the third-weakest topic. The point of Day 4 is to confirm that the topic-specific practice from Days 2 and 3 transferred into mixed-format conditions.

Day 5 is light. One 55-question paper in the morning, then no further test prep. Bennett items reward physical intuition, and physical intuition fades when you are tired and cramming.

What employers actually see in the report

Pearson TalentLens reports show the employer your raw score, your percentile against the chosen norm group, and a topic-level breakdown showing which of the 10 areas you scored above or below the norm. The topic-level breakdown is what some employers use to weight technical interview questions. If your report shows you weak on electricity, expect a follow-up question about a basic circuit.

The employer does not see your time per question or your full answer log; they see the score and the band. This is why "I would have got more if I had another 2 minutes" is not a recoverable excuse. Pacing matters because pacing is what the score reflects.

FAQ

What is the difference between BMCT and BMCT-II?

The original BMCT has 68 questions in 30 minutes; the BMCT-II has 55 questions in 25 minutes. The BMCT-II updated the visual style and modernised a few items, but the 10 physics topic areas covered are the same. Pearson TalentLens has effectively replaced the original with the BMCT-II as the default product.

What does BMCT stand for?

Bennett Mechanical Comprehension Test, named for George K. Bennett who published the original in 1940 under the Psychological Corporation, the predecessor to Pearson TalentLens.

Is a calculator allowed on the Bennett Mechanical Test?

No. Every version of the Bennett bars calculators. The arithmetic in the items is intentionally simple enough to do in your head once you have the physical principle right.

How are Form S and Form T different?

They are two equivalent BMCT-II papers. Form S is typically the unproctored screen, sat from home. Form T is the proctored verification sit. The two papers are statistically equivalent, so a wildly different score across the two flags a cheating concern.

Is the BMCT-II adaptive?

No. The 55 questions are fixed in order, and the difficulty does not adjust based on your answers. Pacing matters more than picking which questions to attempt.

What is a good Bennett score?

For most manufacturing roles, the 70th percentile against the Production and Maintenance norm is a defensible target. For engineering apprenticeships, aim for the 75th to 85th percentile. The cutoffs table earlier in this article shows the typical bands.

Can you retake the Bennett?

Most employers permit one retake in a different hiring cycle. Same-cycle retakes are rare. Pearson recommends a minimum 90-day gap to reduce practice-effect inflation.

Which Bennett version will I be sitting?

Look at the invitation email. "BMCT-II" or "BMCT-II Form S/T" tells you the modern version. A bare "Bennett Mechanical Comprehension Test" with no version suffix usually means the original BMCT, especially if the invitation comes from an older industrial employer.

Related on PrepClubs

• Pillar. Bennett Mechanical pillar page. Full test guide with the free starter set.
• Deep practice. Practice Bennett BMCT-II for $39. 600+ scored items across all 10 topic areas. Pass Guarantee.
• Compare. The Bennett Mechanical Comprehension Test (BMCT) in 2026. The pillar article on the test family.
• Article. Bennett Mechanical Practice Test (Free with Walkthroughs). Five worked items per topic area.
• Guide. What is a good cognitive test score. Cross-test percentile context.

Practice on PrepClubs

Practice all 10 topic areas, calibrated to the BMCT-II norm bands

PrepClubs has a 600-item Bennett bank split by topic area so you can focus the weakest first. Every item shows a worked solution with the physical principle named, the topic it belongs to, and your running percentile against Pearson's Production and Maintenance norm. Form S and Form T-style mock sits run the same paper twice so you can practice the proctored handoff. $39 one time. Pass Guarantee.

Start free Bennett Mechanical practice