The mechanical reasoning test: spatial skill, learned physics and what research says

0 The short answer

A mechanical reasoning test shows gears, pulleys, levers and other devices in pictures and asks what will happen, and it samples spatial visualization together with physical knowledge learned from experience, so a score is not a pure measure of either. The best known publisher version, Pearson TalentLens's Bennett Mechanical Comprehension Test (BMCT-II), has 55 items in 25 minutes according to the publisher page we read on October 29, 2026, and similar sections appear in military and fire service testing. ACIS has no mechanical comprehension subtest and no physics task; the spatial part shows in its Visual Spatial Index, which the Custom Test returns for 15 dollars and the Optimized form (30 dollars, about 110 minutes) partly covers (prices read October 29, 2026). The decisive limit is that an online, unsupervised ACIS result describes your own spatial profile and is not preparation for, or a predictor of, any employer's test.

1 What Is a Mechanical Reasoning Test?

A mechanical reasoning test is a multiple choice test in which a picture of a mechanical situation, such as meshing gears, a pulley system or a lever, is followed by a question about what will happen, and its name varies by publisher because "reasoning," "comprehension" and "aptitude" are used almost interchangeably. The cognitive scientist Mary Hegarty describes the problem type in a 2004 review in Trends in Cognitive Sciences. In her Figure 1 caption, the task is to infer the behavior of a mechanical system from a static diagram that shows the shape of its components and how they connect, and "many of the items in tests of mechanical ability are of this type." We describe the format only, and we reproduce no item from any test.

Three real formats show how much the label varies. Pearson TalentLens, the publisher, lists the BMCT-II as a timed, online, unsupervised test of 55 items in 25 minutes, and its product page (read October 29, 2026, no publication date shown) names "mechanical reasoning and spatial perception" among the abilities covered. The page says the test has been used "for more than 60 years," that it is available in Dutch, English (UK and US), Spanish and French, and that candidates should use a desktop or laptop with a dependable internet connection instead of a phone. The United States military's ASVAB includes a Mechanical Comprehension subtest, and the official CAT-ASVAB page (dated March 8, 2024, read October 29, 2026) lists it with 15 scored questions, 15 possible tryout questions and a 22 minute limit without the tryout questions (42 minutes with them). And in California, the written exam that many fire departments draw on has a section called "Apply Mechanical Reasoning," described in the next section.

The names matter for one practical reason. A candidate who searches for a "mechanical reasoning test" finds pages for several different products with different lengths, different publishers and different cut scores, and a result on one does not transfer to another. Which test you will sit is decided by the employer or program, not by you, and the only reliable way to learn its format is the invitation, the employer page or the publisher page. The page on the pre-employment cognitive test covers the general ability tests that some employers also use, and the page on aptitude versus intelligence explains why a test built to forecast one outcome is a different instrument from an intelligence battery even when the raw material overlaps.

One more boundary. This page is written for the adult who will sit, or has sat, the test. It is not a guide for employers, it does not tell anyone whether a score will qualify them, and it does not contain practice items, answers or score tables for any publisher. Those belong to the publisher and the employer.

2 Who Sets a Mechanical Reasoning Test, and for Which Jobs?

Mechanical reasoning tests are set by employers, training programs and military services, not by candidates, and the publisher pages we read name skilled trades, engineering, emergency services and technical training as the typical settings. Pearson TalentLens lists recruitment, training and risk management as uses of the BMCT-II, and occupations including engineering, airlines, emergency services and skilled trades, with maintenance, production and technical training named as suited. That is the publisher describing its own product, not independent evidence about who benefits.

The fire service is the setting most searchers name. The Long Beach Fire Department's recruitment page (undated, read October 29, 2026) describes a 100 question multiple choice written exam with four sections: recall and comprehend verbal and visual information, apply mechanical reasoning, solve mathematical problems, and recall and comprehend technical information from written materials. It gives candidates 2.5 hours and says the exam is the FCTC Written Exam. The FCTC written testing page (read October 29, 2026) states that "the FCTC Written Test was developed by the California Firefighter Joint Apprenticeship Committee (Cal-JAC)." So in this example the employer side is a joint apprenticeship committee, and mechanical reasoning is one of four sections, not the whole exam. Other departments use other tests, and a candidate should read the page of the department that is hiring.

The table sets the options a reader is likely to meet side by side. It holds only what we could read on an official or publisher page on October 29, 2026. Where the page is silent, the cell says so, because a guessed time, norm group or price would be worse than a blank. It also includes the ACIS spatial route, because the brief of this page is to show the spatial part, not to claim a match.

None of the employer tests in the table returns an IQ. They return a result for one job or program, and the page on where to take an IQ test compares the routes for readers who want a cognitive score for themselves rather than a hiring result. Other formats that candidates meet are covered later on this page, so this section stays with the mechanical one.

3 What Does a Mechanical Reasoning Test Measure?

A mechanical reasoning test samples a blend, in our reading of the CHC framework: visual processing (Gv) for imagining how a system moves, domain specific knowledge (Gkn) for the physical principles a candidate already knows, and some reasoning (Gf) for working out a chain of cause and effect. In the Cattell-Horn-Carroll framework that the page on the CHC model lays out, Gv covers spatial relations, visualization and mental rotation, and Gkn is specialized knowledge in particular domains. The ACIS page names "mechanical knowledge" as an example of Gkn. A mechanical reasoning item can lean on either, and a single score does not say which.

The strongest evidence for the spatial half comes from Hegarty's review. Under the heading "Individual differences," she writes that solving mechanical reasoning problems, such as a pulley problem, "is highly correlated with measures of spatial ability," but "not significantly correlated with verbal ability." A second line of evidence comes from dual task studies: she describes one in which holding a visuospatial memory load interfered with mechanical reasoning more than holding a verbal load, which suggests the reasoning relies on visuospatial working memory (Gwm). The third comes from people thinking aloud, who use imitative hand gestures that precede their verbal descriptions. Hegarty calls the individual differences result preliminary evidence that mechanical inference involves transformations of spatial representations and depends less on verbal representations. We report her review, not a verdict on any one test.

That is why a mechanical reasoning test is a poor stand-in for either of two things people sometimes assume it measures. It is not a general intelligence test: the page on what IQ measures explains that a general score summarizes performance across many abilities, while a mechanical test samples a narrow slice for a narrow purpose. And it is not a measure of spatial ability alone, because a candidate who has studied how levers and gear trains behave brings knowledge that a pure spatial test is built to exclude. Pure spatial formats look different. A mental rotation test asks only whether a shape matches after turning, as the page on the 24 item mental rotation test describes, and a Wechsler examiner gives Block Design with physical blocks, which the page on the Block Design test online explains. Those tasks strip out the physical knowledge on purpose.

The same reasoning explains why the page on cognitive domains matters for reading any aptitude result. A candidate with strong visual processing and little hands-on background, and a candidate with the reverse, can reach similar raw scores for different reasons, and the label "mechanical aptitude" hides the difference.

4 Why Does It Mix Spatial Ability With Learned Physical Knowledge?

The mix is built into the task: to say what a gear train or pulley system will do, a person usually runs a mental simulation of the parts, and how well that simulation runs depends both on spatial processing and on what the person already knows about how such parts behave. Hegarty's review separates the two. Her abstract states that reasoning by mental simulation "can be dissociated from reasoning based on descriptive knowledge in that it depends on different abilities and memory stores," and that it "can result in correct inferences in situations where people do not have correct descriptive knowledge." In plain terms, the two can come apart: a person can sometimes get a mechanical answer right without being able to state the rule behind it.

The review also describes how the simulation unfolds. People do not appear to picture a whole machine moving at once. Hegarty reports that in a pulley experiment people took more time to infer the motion of the lower pulley than the middle one, and more time for the middle one than the upper one, and she reasons that they work piecemeal through the causal chain, starting from the rope being pulled. Eye fixations were consistent with that account: people looked at the component in question and at earlier components in the chain, not at later ones. A timed test therefore rewards a candidate who can hold several steps of a chain in mind quickly, which is one reason working memory (Gwm) matters when interpreting any aptitude result.

What the science does not do is give a share. Hegarty closes with open questions, and one asks how to characterize individual differences in mental simulation ability: whether they reflect differences in the capacity of spatial working memory, differences in executive processes such as task decomposition, or "differences in knowledge of the mechanical situations to be simulated." Another asks how simulation changes with expertise. The review leaves those questions open in 2004, and we do not claim to settle them. A claim that a mechanical reasoning score is "X percent spatial and Y percent knowledge" has no source in the literature we read.

Test publishers describe the blend in their own words. The Pearson TalentLens page lists, among the abilities of the BMCT-II, "mechanical reasoning and spatial perception," problem solving, learning mechanical concepts quickly and knowledge of basic mechanics. The California exam titles its section "Apply Mechanical Reasoning," while a neighboring section asks candidates to "recall and comprehend" technical information. That wording separates applying from recalling, but the sources do not say how a given item divides between the two.

Researchers who used such tests have flagged the same issue. In the study of mechanical engineering students described in the next section, the authors adapted a paper and pencil test from practice material that serves as preparation for civil service, military and trade exams, and they wrote that it "may not be the best test for identifying the physical intuition that is most important for mechanical engineers in the workplace." A test aimed at skilled trades was being used to ask a question about engineers, which is a reminder that a result answers the question the test was built for. The page on how spatial intelligence is tested and trained covers spatial ability by itself, without the physics.

5 What Does Research Say About Experience and Mechanical Reasoning Scores?

The research we could open links hands-on experience to mechanical test scores, but only as correlations from a small number of studies, so experience is a plausible contributor and not a measured cause. The most direct study we found is a conference paper by Miller, Pereira and Mitchell (2011) from the American Society for Engineering Education. It reports a paper and pencil mechanical aptitude test given to mechanical engineering students at one US university across five semesters. The test was adapted from a practice test and has questions about gears, pipes, linkages and other mechanisms. The students also filled in questionnaires about prior experiences, and the authors correlated the two.

The strongest correlations were modest. Using hand tools had r = 0.290, target shooting 0.284, canoeing 0.259 and repairing equipment 0.243, in a subsample of 156 students. The authors note that the coefficients "are all quite low," and they suggest this is likely because many activities contribute to mechanical aptitude and not just one or two. The list also contains canoeing, instrumental music and marching band. That observation is ours, not the authors' conclusion: correlations of this kind cannot say which experiences matter, because people who do one kind of activity often do others, and a questionnaire cannot separate them.

Three limits keep the result in proportion. The students were in an engineering program, so the range of experience and aptitude is not that of the general population or of a fire department applicant pool. The experiences were self reported. And the authors themselves call the test directed at the skilled trades and say their future work includes a new test with high reliability and validity, which tells the reader the instrument was not presented as a settled measure.

Is spatial skill itself changeable? Uttal and colleagues' meta-analysis of 217 training studies in Psychological Bulletin reports an average effect size (Hedges's g) of 0.47 for training relative to control, stable over delays, with transfer to spatial tasks that were not directly trained. They state their aim as determining whether and to what extent "training and experience can improve these skills." We found no study in the material we opened that trained people on spatial tasks and then measured the result on a mechanical comprehension test, so the step from better spatial scores to a better mechanical score is an inference, not a finding. The page on the spatial IQ test walks through that meta-analysis in more detail.

For a candidate the sensible reading is modest. Part of what a mechanical test samples is learned, so a score on a first attempt describes where a person is, not where a person must stay. The Standards for Educational and Psychological Testing make a related point from the employer side. Their comment on Standard 11.2 says the knowledge, skills and abilities in an employment test's content domain "should be those that qualified applicants already possess when being considered for the job in question" (AERA, APA and NCME, 2014, in the open access edition we read). Whether a given mechanical test meets that standard is a question for the employer to answer, not for the candidate to guess.

6 What Does Research Say About Group Differences on Mechanical and Spatial Tests?

Research reports differences between group averages on mechanical and spatial tests, but these are findings about groups, they depend on the test used, and they cannot say where any individual person falls. We report what we could open, and we do not extend it beyond the test, the sample or the year.

The broadest evidence concerns spatial tests in general. Voyer, Voyer and Bryden (1995) meta-analyzed 286 effect sizes from a variety of spatial ability measures. Their abstract reports that sex differences are significant in several tests but that some intertest differences exist, that there was partial support for the notion that the magnitude of the differences has decreased in recent years, and that the age at which sex differences emerge depends on the test. The finding is not one number for "spatial ability": it varies by task, and the study covers spatial measures, not mechanical comprehension tests.

For a mechanical test specifically, the engineering student study above reports that male students scored higher than female students on the mechanical aptitude test and on physics computer games at statistically significant levels. It also reports that male students spent many more hours on the activities with the strongest correlations to test performance, and that women reported lower confidence in and enjoyment of questions about mechanical and tinkering skills. The authors call the experience difference a likely explanation for the lower female scores. That wording is theirs, and it is a hypothesis: the study is correlational, the women in the test sample numbered fewer than 50, and the correlation tables for female students rest on 18 and 23 respondents.

How should a reader hold this? Three points. First, a difference between averages tells you nothing about where one person falls; a candidate of any sex can score at any level. Second, the experience findings point to opportunity as at least part of the story, which is a reason not to read a low first score as a fixed limit. Third, the Standards for Educational and Psychological Testing address the issue in their discussion of validity: they note that test score differences for groups defined by race or ethnicity, gender, age and other characteristics are one domain where unintended negative consequences are sometimes observed, and that it is important to distinguish between evidence directly relevant to validity and evidence that may inform decisions about social policy. Examining that evidence is work for the people who build and use the test. A candidate's task is to read their own result without assigning it a group meaning.

7 Does a Mechanical Reasoning Test Predict Job Performance?

A mechanical comprehension test has predicted supervisor ratings of job performance in at least one documented sample, but one sample from one kind of plant does not show that any given test predicts performance in any given job. Muchinsky's 1993 study in the Journal of Business and Psychology used 193 manufacturing employees who produce electro-mechanical components in a concurrent criterion related validity study, which means current employees took the tests and were rated on the job. They took the Bennett Mechanical Comprehension Test (Form S), the Flanagan Aptitude Classification Test, Mechanics, and the Thurstone Test of Mental Alertness (Form A). Supervisors rated fifteen job dimensions at two points 60 days apart. The abstract reports that the Bennett test was "the best single predictor of job performance," with an uncorrected correlation of r = 0.38, and that the additional tests did not add significantly to prediction.

The author's own discussion matters as much as the number. He argues that conventional mechanical aptitude tests may not be sensitive to the changing nature of manufacturing jobs. We cannot say from the abstract how well the finding holds for fire service work, for apprenticeships or for other trades, and we do not claim it does. It is one study of one setting, and in that setting the mechanical test outpredicted the other two. A correlation of 0.38 is also far from certainty: it leaves most of the variation in ratings unexplained, which is our plain reading of the number and not the author's claim.

Spatial ability has a separate record, on a different outcome. Wai, Lubinski and Benbow's 2009 analysis in the Journal of Educational Psychology drew participants from a stratified random sample of US high schools (about 400,000 students), tracked them for 11 years, and aligned the findings with earlier work. Their abstract reports that spatial ability assessed during adolescence has surfaced as a salient attribute among those who go on to advanced educational credentials and occupations in science, technology, engineering and mathematics (STEM). That is evidence about spatial ability and educational and occupational paths, not about any mechanical comprehension test, and the page on IQ tests for engineers places it in the context of engineering careers. The page on average IQ by profession is a reminder of the same caution: a group average, whether of IQ or of a test score, describes a group of workers and not the next person who applies.

The Standards for Educational and Psychological Testing supply the vocabulary. Standard 1.1 says no test permits interpretations that are valid for all purposes or in all situations, and calls the unqualified phrase "the validity of the test" incorrect. Standard 1.4 says that if a score is interpreted for a use that has not been validated, it is incumbent on the user to justify the new interpretation. Standard 11.1 asks for a clear statement of the intended interpretations of employment test scores for specified uses before validation, with the validation designed to determine how well this has been achieved for all relevant subgroups. A fair summary is that a mechanical test can be a valid tool for a defined use, and that the employer must show it for that use.

8 Does Practice Help, and Where Do Questions and Answers Come From?

Retest research on cognitive ability tests finds that scores rise on a second attempt, more so with coaching and with identical forms, and the safest place to learn a test's format is the employer or publisher, not a practice site or a file claiming to hold the answers. The best summary of retest effects is a meta-analysis by Hausknecht, Halpert, Di Paolo and Moriarty Gerrard (2007) in the Journal of Applied Psychology. Its abstract notes that as many as 25 to 50 percent of applicants in organizational and educational settings are retested with measures of cognitive ability. The authors pooled 50 studies of practice effects, 107 samples and 134,436 participants, and report an adjusted overall effect size of .26, with larger effects when practice was accompanied by test coaching and when identical forms were used. The study covers cognitive ability tests as a group, and the abstract does not single out mechanical comprehension tests, so the number is a general guide and not a prediction for any one test.

Two cautions follow. A gain on a retest does not by itself show how much of the change is ability and how much is familiarity with the format, and the abstract does not claim to separate them. And the finding that identical forms produce larger gains is a reason to treat any claim that a practice site "matches" a real test with care: we cannot verify such claims, and we do not repeat them.

What do official sources offer? The FCTC written testing page (read October 29, 2026) describes a free study guide that covers all four sections with study tips and sample questions, and lists an online practice written test at 25 dollars. It also states that outside vendors selling study materials or programs claiming to provide help with the written test have no affiliation with FCTC. The official ASVAB site has a sample questions page for the Mechanical Comprehension portion, which it describes as "focused on mechanical and physical principles." For the BMCT-II, we did not find a candidate guide on the Pearson TalentLens page we read, so the invitation from the employer is the place to look. Those official materials belong to their owners, and this page does not copy them.

The searches behind this page need a direct answer. People type "mechanical reasoning test questions and answers" and "mechanical reasoning test pdf" hoping for a document. A search for the plain phrase on October 29, 2026 returned mostly commercial practice sites, employer assessment vendors and files hosted on unrelated sites. We do not use or link those, and we do not host items or answer keys for any test. The reason is simple: the items are the publisher's property, and a page that cannot show where its answers came from has no way to show they are right. If the "free" in your search is about a score for yourself rather than a drill, the page on the free IQ test is the place to look.

Practical steps that need no document: read the test name and time limit in your invitation; check whether the test is taken on a computer, as the Pearson page recommends, or in a room; find out from the employer whether and when a retest is allowed; and use the employer's or publisher's own free guide if one exists.

9 How Should You Read a Mechanical Reasoning Result?

A mechanical reasoning result is a count of correct answers turned into a comparison with a reference group, so what it means depends on that group, on the cut score the employer chose, and on how precisely the score is measured, and the candidate usually sees none of the three. The pages we read for the BMCT-II, the ASVAB subtest and the California exam do not state norm groups, score scales or pass marks, and we do not guess them. The Standards for Educational and Psychological Testing say in Standard 1.1 that the population for which a test is intended "should be delimited clearly," and that the construct the test is intended to assess should be described clearly. If you cannot find that information for your test, it is reasonable to ask the employer what the score will be compared with.

Three vocabulary points help with any aptitude report. A raw score is the count of items correct, and it means little alone. A percentile states the share of a reference group scoring at or below a score, and it is not a share of items correct; the page on how IQ scores are normed explains how reference groups are built. A standard score places a result on a scale with a stated mean and spread, and the page on standard deviation 15 explains the most common one. The reason IQ style percentiles cannot be averaged is covered on the page about IQ score versus percentile, and the same logic holds for any test that reports both.

Precision matters as much as the number. Every score has error around it, which is why the page on reliability and validity is the place to learn what a confidence interval says. A single short test generally gives a less precise picture than a battery that samples several abilities, and it is not meant to do more: it was built to inform a defined decision.

Last, resist converting a mechanical score into an IQ, or an IQ into a mechanical score. The two are different instruments with different reference groups. The Standards (1.4) say a score interpreted for a use that has not been validated needs justification from the user, and no source we read supports a conversion table between a mechanical comprehension test and an IQ scale. If someone offers one, ask for its source.

10 What Can an Adult Learn About the Spatial Part on Adult Norms?

ACIS cannot measure mechanical reasoning, because it has no mechanical comprehension subtest and no physics knowledge task, but its three visual spatial subtests report the spatial component (Gv) on adult norms as a profile of your own, and nothing more. The three are Visual Puzzles, Layer Rotation and Spatial Comprehension, and they build the Visual Spatial Index (VSI). The page on Visual Puzzles describes a task of mental part-whole assembly, in which the person imagines how separate pieces combine into a finished form without moving them. The page on Layer Rotation describes mental rotation and figural integration: each part is turned in the mind to a common orientation, combined into one figure, and matched to an option. The technical manual says Layer Rotation "carries no words at all." The page on Spatial Comprehension describes understanding orientation, position and imagined movement, including perspective taking.

None of those tasks asks about forces, gears, pulleys or levers, and none asks a physics question. That is the point of the table earlier: the spatial half of a mechanical test is the part ACIS can show, and the learned knowledge half is not on the site.

The forms differ in how much of the spatial part they cover (prices and minutes read October 29, 2026, and they can change). The Quick form (15 dollars, about 45 minutes) has no spatial subtest. The Optimized form (30 dollars, 13 subtests, about 110 minutes) includes Spatial Comprehension and Layer Rotation. The Full Scale form (50 dollars, all 20 subtests, about 175 minutes) includes all three. The Custom Test costs 5 dollars per subtest, so the three together cost 15 dollars, and the site states that "any domain you complete returns its full index score, and every subtest returns its scaled score." The technical manual also lists a composite, the CVSI, which "expands visual spatial coverage beyond the primary VSI by including Layer Rotation, Visual Puzzles, Spatial Comprehension, and Visual Sequence." The page on the Visual Spatial Index explains what an index is, and the technical manual lists the documentation.

What does the report give? Scaled scores on the 1 to 19 scale (mean 10, standard deviation 3), an index on the scale with mean 100 and standard deviation 15, a percentile and a 95 percent confidence interval, against adult norms for ages 16 to 90. That is information about your own spatial profile. It does not tell you how you would score on a Bennett, ASVAB or FCTC section, which are different tests with different content, and it should not be used that way.

Practical conditions for an unsupervised session are short. Use a computer in a quiet room. Sound is needed only for the three audio subtests (Digit Span, Alphanumeric Sequencing and Arithmetic), none of which is in the spatial index. Breaks are allowed, progress is saved between subtests, and a form must be completed within 30 days. A free trial with no card is available, once per person.

11 Which Other Aptitude Tests Sit Next to a Mechanical Section?

Candidates rarely meet a mechanical section alone: employers and programs combine formats, so the useful question is which ability each section samples, and each of the formats has its own page. The California exam is an example from this page. Mechanical reasoning is one of four sections, with the others covering verbal and visual information, mathematical problems and technical information from written materials. A candidate who prepares only for the mechanical section has prepared for one of four sections, and the pages we read do not say how sections are weighted.

The sibling pages cover the other formats. The page on the spatial reasoning test covers spatial reasoning tests as a format of their own, and the page on the diagrammatic reasoning test does the same for diagrammatic formats. The error checking test and the GATB aptitude test have their own pages as well. We do not repeat them here, because each deserves its own treatment and mixing formats is how candidates end up studying the wrong thing.

If what you want is a broader picture of your own abilities, and not a hiring result, two entry pages are relevant. The cognitive ability test page is the ACIS entry point for an adult normed look at the broad abilities, and the IQ test for adults page is the entry point for an IQ score. Neither is a way to prepare for an employer's test, and neither is accepted by employers as a substitute.

If you do take a profile, read gaps between indices with care. The page on index score differences explains how to read them. The Standards (Standard 1.14) say that when interpretation of subscores, score differences or profiles is suggested, the rationale and relevant evidence for it should be provided, and that evidence for two separate scores would not necessarily justify an interpretation of the difference between them. A high spatial index beside a lower verbal index is a description of performance on those tasks, not a prescription for a career.

12 Where ACIS Sits, and What a Buyer Should Do

ACIS is a normed online battery that can show you the spatial part of your own profile, it is not a mechanical reasoning test, and it should not be used to prepare for, predict or replace one. The limits are plain: it is online and unsupervised, not a clinical or diagnostic instrument, and English only. The site's page for researchers states that ACIS "is not intended for clinical diagnosis, hiring decisions, accommodation determinations or admission to high IQ societies." A result belongs to you, as information about yourself, and this page does not suggest sending it to an employer.

The Standards for Educational and Psychological Testing (AERA, APA and NCME, 2014), the joint publication of the American Educational Research Association, the American Psychological Association and the National Council on Measurement in Education, frame how any score should be read. Standard 1.1 asks for a clear statement of how scores are intended to be interpreted and used, and Standard 1.4 puts the burden on the user who applies a score to a use that has not been validated. For ACIS, the intended interpretation is a description of performance on its tasks against adult norms. An employer's decision about a candidate is a different use, and nothing on this page validates it. We could not open the APA's 2020 Guidelines for Psychological Assessment and Evaluation on October 29, 2026, so we cite the APA here only as a co-author of the Standards.

The strengths are real and modest. Adult norms cover ages 16 to 90, each score comes with a 95 percent confidence interval, and the index reports a single ability (Gv) rather than a blend. The guarantee is stated literally on the site: "Full refund within 5 days if the report does not deliver the features described at checkout or a technical issue prevents you from accessing your results."

What should a buyer do? If a test date is close, read the invitation, the employer's page and the publisher's page, and learn the format from them. If the question is about your own spatial profile, the free trial costs nothing and needs no card, the Custom Test returns the Visual Spatial Index for 15 dollars, and the technical manual documents the measurement. Keep any result as information for yourself.

13 Sources Behind This Page

Every figure, quotation and product fact on this page comes from one of the sources below, opened on October 29, 2026 or checked by its DOI.

• Pearson TalentLens. Bennett Mechanical Comprehension Test (BMCT-II) product page. Item count, time limit, delivery, languages, uses and occupations, undated page, read October 29, 2026.
• Official ASVAB program. The CAT-ASVAB, page dated March 8, 2024, read October 29, 2026, and Mechanical Comprehension (MC) sample questions page, read October 29, 2026.
• Long Beach Fire Department. Fire recruits, step 2: the written exam. Sections, question count and time, undated page, read October 29, 2026.
• Firefighter Candidate Testing Center (FCTC). FCTC Written Test. Developer of the test, study guide, practice test and statement on outside vendors, read October 29, 2026.
• Hegarty M. Mechanical reasoning by mental simulation. Trends in Cognitive Sciences, 2004, volume 8, issue 6, pages 280 to 285.
• Miller M, Pereira A, Mitchell B. Comparison of Mechanical Aptitude, Prior Experiences, and Engineering Attitude for Male and Female Mechanical Engineering Students. 2011 ASEE Annual Conference and Exposition Proceedings, 2011, pages 22.352.1 to 22.352.12.
• Uttal D, Meadow N, Tipton E, Hand L, Alden A, Warren C, Newcombe N. The malleability of spatial skills: A meta-analysis of training studies. Psychological Bulletin, 2013, volume 139, issue 2, pages 352 to 402.
• Voyer D, Voyer S, Bryden M. Magnitude of sex differences in spatial abilities: A meta-analysis and consideration of critical variables. Psychological Bulletin, 1995, volume 117, issue 2, pages 250 to 270.
• Muchinsky P. Validation of intelligence and mechanical aptitude tests in selecting employees for manufacturing jobs. Journal of Business and Psychology, 1993, volume 7, issue 4, pages 373 to 382.
• Wai J, Lubinski D, Benbow C. Spatial ability for STEM domains: Aligning over 50 years of cumulative psychological knowledge solidifies its importance. Journal of Educational Psychology, 2009, volume 101, issue 4, pages 817 to 835.
• Hausknecht J, Halpert J, Di Paolo N, Moriarty Gerrard M. Retesting in selection: A meta-analysis of coaching and practice effects for tests of cognitive ability. Journal of Applied Psychology, 2007, volume 92, issue 2, pages 373 to 385.
• American Educational Research Association, American Psychological Association and National Council on Measurement in Education. Standards for Educational and Psychological Testing, 2014. Standards 1.1, 1.4, 1.14, 11.1 and the comment on 11.2, and the discussion of group score differences, read in the open access edition.

14 Frequently Asked Questions