Loading page content
Original practice bank · All explanations and revision guides free
Try 24 original symbol tasks: sequences, analogies, classification and transformations. All answers and four full guides free. Untimed, no signup or IQ band.
24 questions · about 16 minutes · no signup
Complete 24 original four-option questions, six per section. Each has one checked answer under its stated rules. In inference items, choose within the finite candidate operations or criteria supplied; uniqueness is not claimed among every imaginable pattern. A correct choice adds one, another valid choice zero; no partial credit, timer, guessing penalty or speed adjustment. Finish all items to see every answer and four full guides. Sixteen minutes is an editorial estimate.
Preparing the test…
All questions and full results are free, with no signup. Saving is optional; only saved attempts can be resumed after a refresh or language change. Tab drafts expire after 24 hours and clear on completion or retake.
Work with original letter rows, bit strings, small grids described by rows and finite sets. The complete practice covers continuing a symbolic sequence, transferring a demonstrated relation, classifying positive and negative examples, and executing specified transformations. Every option key is checked by a separate literal computation; the 18 inference items each have one surviving rule within their stated candidate family. The other six give explicit operations to execute. This checks this bank’s arithmetic and symbolic rules, not its psychometric validity. No Raven, SHL or JobCannon questions, scoring norms or result bands are used. It is not a nonverbal matrix battery, IQ measure, clinical instrument, employer certification or prediction of job performance. Language and notation familiarity may affect performance.
Every section has a public revision guide. Completing the practice shows every guide and item explanation without assigning a type by score. Public guides contain no personal responses or counts; studying before practice is welcome. This is not a secure exam.
Review rotations, repeating states and simultaneous symbol shifts with worked examples. Six-item counts describe this original practice bank, not intelligence.
Compare two demonstrations, transfer the same relation, and distinguish sets from ordered strings. Includes fresh worked analogies and explicit practice limits.
Infer a candidate criterion using positive and negative rows. Review exact counts, positional roles and counterexamples without personality or ability labels.
Review ordered changes, conditional branches, simultaneous replacement, grid transposition and set operations, with new examples and clear score limits.
Yes. The full 24-item practice, selected choices, keys, rationales and four complete revision guides are free. No account, email, payment or certificate purchase is required. Public guides can also be read before you answer.
No. Counts describe this fixed original bank. No norms, percentile, IQ conversion, diagnostic interpretation or hiring threshold is established. The written-symbol format also differs from a nonverbal visual matrix test.
Finite examples can be extended by many imagined rules. Listing a finite family makes the task explicit and lets us check that one candidate fits all evidence and one answer. It is still necessary to compare and apply those candidates; the list does not identify which is correct.
Yes. There is no timer or speed score. Around sixteen minutes is an editorial estimate. Paper can help track intermediate rows, sets or candidate-rule checks. Record whether you used help if comparing your own attempts; no standardized comparison is supplied.
You can go back to revise choices and opt into a browser-tab draft that expires after 24 hours. Completion, retake or exit-and-clear removes it. Without a saved draft, refresh loses progress. Repeating known keys can reflect memory, not a change in general reasoning.
Only the public title and test link. It excludes choices, correctness counts, keys, draft data and private identifiers. Practice answers stay out of the shared URL; no inferred ability label is generated.
Each of four sections contributes a count out of six, for a total out of 24. All questions have four options and one checked key. All 24 answers must be valid before the report appears. Missing answers are not silently counted as wrong. Every rationale and all four full guides appear for zero, partial and perfect counts; they are revision topics, not result tiers.
Finite examples can fit many invented functions. Each inference question therefore states a limited family of possible operations or criteria. Separate checks establish exactly one candidate fitting every supplied transition, demonstration or positive/negative label and exactly one matching option. This is a bounded exercise, not a proof of the only conceivable pattern or an unrestricted induction benchmark.
Suppose the permitted rules are reverse, rotate left or leave unchanged. UVWX → XWVU and JKLM → MLKJ select reversal. Applied to PQST it produces TSQP. Rotating left would give QSTP, which fails the demonstrations. Check the full row, not just its outside symbols, and keep the same rule when the input changes.
FermatMind independently wrote the 24 tasks, 96 alternatives, 24 rationales and four guides. Literal computations check answer keys and new worked examples. No licensed vendor bank, population calibration, reliability estimate, cross-language equivalence study or predictive-validation study has been used. References describe their own products and practice limits, and neither endorse nor validate this bank.
These sources provide background, not validation of this questionnaire or an individual diagnosis.
Website content edition 1.0.0 ·
Source question / scoring / report versions: fermatmind_abstract_finite_rules_24_v1 / practice_equal_correct_counts_v1 / fm_abstract_four_revision_guides_v1