Updated April 2026 · 8 min read · Talogy / Cubiks · Senior-level abstract reasoning
| Provider | Talogy (formerly Cubiks) |
|---|---|
| Test name | Logiks Advanced Abstract Reasoning |
| Format | 30 questions · 12 minutes · advanced inductive only |
| Used by | Senior management, leadership pipelines, partner-track consulting roles |
| Defining feature | Three-rule combinations and conditional patterns absent from Intermediate level |
Logiks Advanced Abstract is the senior tier of Talogy's inductive reasoning suite. Where Intermediate combines 1-2 transformation rules per item, Advanced regularly stacks 3 rules and introduces conditional patterns ("if shape A is present, rule X applies; otherwise rule Y"). The 24-seconds-per-question budget feels longer than Intermediate's 18 seconds, but the cognitive demand per item is higher.
Multi-rule stacking. Advanced items frequently combine three or more transformations: rotation + colour shift + addition, or reflection + size scaling + element substitution. You must identify all three to choose correctly.
Conditional rules. The pattern shifts depending on context — e.g., "rotation rule applies in rows containing a triangle; reflection rule applies elsewhere."
Larger matrices. Some items use 4x4 matrices instead of 3x3, increasing the data to integrate.
Distractor sophistication. Wrong-answer options are constructed to look right if you missed one of the rules. Eliminating distractors requires confirming all rules.
Raw correct, mapped to percentile. Senior graduate cutoffs typically sit at 70th+ percentile (raw score ~20+/30). Partner-track or strategy roles may demand 85th+ percentile.
Used in senior hire pipelines and leadership assessments at Volvo, Pirelli, several European banking groups, and consulting firms screening for senior associates. Less common at graduate-entry level — the test is designed for candidates with several years of professional experience.
It might seem counterintuitive that a senior-level assessment would test abstract pattern recognition rather than something more obviously job-relevant like financial modelling or people-management scenarios — but abstract reasoning is specifically valued at senior levels because it's considered a strong proxy for a kind of fluid intelligence that doesn't depend on prior job-specific knowledge or experience. Where a numerical or verbal test partly measures learned skills (financial literacy, vocabulary built up over a career), abstract reasoning aims to measure raw pattern-recognition and rule-inference ability more directly, which employers use as a general capability signal alongside — not instead of — role-specific interviews and case studies later in the process.
This is also why Logiks Advanced Abstract tends to appear as one component of a broader senior-hire assessment battery rather than a standalone gate — a strong abstract-reasoning score demonstrates general cognitive capability, but employers still separately assess domain expertise, leadership experience, and interpersonal skills through other parts of the process specifically because abstract reasoning alone doesn't capture those dimensions.
Advanced Abstract draws on the same underlying family of transformation rules as Intermediate Abstract — rotation, reflection, size progression, shading/fill change, element addition or subtraction, position shift, and combination — but where Intermediate items typically combine at most two rules applied uniformly, Advanced items regularly layer a rule that applies unconditionally across the whole matrix with a second, conditional rule that only kicks in under a specific condition (only in the top row, only when a particular shape is present, only every other cell). This conditional layering is the real source of Advanced's extra difficulty: the underlying rule vocabulary isn't larger, but the way rules combine and interact is structurally more complex, which is exactly why simply knowing the seven core rules individually isn't sufficient preparation at this level — you additionally need practice specifically at spotting when a second rule is conditional rather than universal.
Triple-rule recognition. Practice items where you must identify three independent rules. The discipline is to keep listing rules until no rule is left over by the row/column constraints. If a 3x3 matrix has only two rules identified, you've missed one.
Distractor checking. Before clicking, verify your chosen answer satisfies all identified rules. Most wrong answers satisfy 2 of 3 — the one that satisfies all 3 is correct.
Time budgeting. 24 seconds per item is generous compared to Intermediate, but Advanced items take more thought. Aim for 30 seconds on hardest items, 15 seconds on items where the rules click immediately, and use the budget gracefully.
Build a rule-identification checklist. Rather than scanning a matrix and hoping a pattern jumps out, work through a consistent mental checklist on every item: shape identity (does the shape itself change?), orientation (rotation or reflection?), size (does scale change across the sequence?), shading or colour (does fill pattern shift?), and count (does the number of elements change?). Running through this list systematically, rather than relying on whichever transformation happens to be most visually obvious first, is what catches the second, conditional rule that Advanced-level items are specifically designed to hide behind an easier first rule.
Practise identifying conditional triggers specifically. The skill that separates Advanced from Intermediate performance isn't spotting a single transformation faster — it's noticing when a rule only applies under a specific condition (a particular row, a specific shape present, an odd-numbered column). Dedicated practice on conditional-rule items specifically, rather than general abstract-reasoning practice, builds this specific pattern-recognition skill faster than broader practice does.
Work backward from wrong answers during practice. When reviewing practice items you got wrong, don't just note the correct answer — identify exactly which rule you missed or misapplied. Over enough practice items, a personal pattern usually emerges (consistently missing conditional rules, or consistently misjudging rotation direction), and once you know your specific blind spot, you can deliberately check for it on every subsequent item.
Logiks Advanced Abstract has no numerical or verbal content at all — every item is a matrix or sequence of shapes, and the challenge is identifying every transformation rule at work, not performing a calculation.
Q. A 3x3 matrix of shapes is shown, with the bottom-right cell blank. Across each row, the shape rotates 90° clockwise. In rows containing a shaded triangle, the shape's size also increases by one step; in rows without a triangle, size stays constant. Which of the four answer options completes the missing cell?
Approach: first identify the base rule (rotation across each row) that applies everywhere. Then check the conditional rule (size change) against this specific row — does it contain a triangle? Only after confirming both rules should you evaluate the answer options, since a distractor will typically satisfy the rotation rule alone while getting the size condition wrong.
The correct answer is whichever option satisfies both the unconditional rotation rule AND the row-specific size condition — checking only one of the two rules is exactly how well-prepared candidates still pick a distractor.
This layered structure — one rule that always applies, plus a second rule that depends on context — is the defining feature that separates Advanced from Intermediate-level abstract reasoning, and it's exactly why simply "spotting a pattern quickly" isn't enough at this level; you need a habit of checking for a second, conditional layer even after finding an obvious first rule.
Logiks Advanced Abstract gives roughly 30 questions in 12 minutes — about 24 seconds per question, noticeably more generous than Intermediate's pacing, but that extra time exists precisely because each item demands identifying multiple rules rather than one.
If you're past 35 seconds and still can't isolate a second rule, flag and move on — you can't recover four lost minutes from one stubborn question.
Use TestSolve to crack the Logiks Advanced Abstract test
AI-powered answers and reasoning sent to your phone in seconds. Free first solve, no signup.
Try it free →See pricing