Original questions written to match the format and difficulty of the real assessments — never copied from them. Full answer key and a worked explanation on every question.
Five full-length tests covering all 10 interactive question types — charts, scheduling, allocation, ranking, sequences and more.
Deduction, Interpretation, Recognition of Assumptions and Inference questions in the critical-thinking format used by law firms and professional-services graduate schemes. Every answer key is derived from the logic itself.
Six subtests — the three Challenge tests, plus scales numerical, scales verbal and Understanding of Instructions on Aon's tabbed data-sheet screen. Speed-timed and scored +1 right, −½ wrong, the way Aon marks.
Sequences, odd one out and matrix completion — the three formats abstract and inductive tests are built from. Every key derived from the rule that built the figures.
Rotation against reflection, and operator panels. Built on genuinely chiral shapes, so the mirror-image trap these tests turn on actually works.
All five reasoning modules Cappfinity sets — numerical, verbal, critical reasoning, logical and diagrammatic, and error checking. Critical reasoning is the high-ceiling one and where graduate schemes sort candidates.
All three PATH cognitive strands, each separately clocked the way Talegent runs them — numerical at 15 minutes, verbal at 14, logical at 18. Adaptive, so the questions get harder as you get them right.
Both answer formats employers set — multiple-choice calculation, and true, false or cannot say against a data sheet. Every wrong option is a named miscalculation, not a nearby number.
True, false or cannot say — and the two-option version that removes cannot say, where a passage that settles nothing still has to be answered.
Most candidates lose marks to the clock and to surprise, not to the reasoning itself. Sitting a test laid out the way the real one is laid out — same question shapes, same timer pressure, same answer mechanics — takes most of that surprise away before it counts for anything.
Every question is generated from scratch by TestSolve. Nothing here reproduces real test content: we research the published format of each assessment, then write original questions to that structure. On the reasoning subtests the answer key is derived symbolically from each question's own logic and re-checked before the question ships, so the marked answer is provably right rather than merely intended.
The generated sets are also audited for answer leaks — surface patterns like a particular opening word or item position that would let someone score well by spotting a cue instead of reasoning.
© 2026 TestSolve. Independent and not affiliated with any test publisher.