Abstract Reasoning Test: Pattern Rules and Preparation

Abstract reasoning tests can feel strange because they use shapes and patterns instead of ordinary workplace information. But the questions are not random. They are built from repeatable visual rules. This guide explains what abstract reasoning tests measure, the common pattern families, and how to practise so the rules become easier to recognize under time pressure.

What is an abstract reasoning test?

An abstract reasoning test measures how well you can identify patterns and relationships in unfamiliar visual information. Instead of words or numbers, the test usually uses shapes, symbols, grids, matrices, sequences, or diagrams. The candidate must infer the rule and select the option that completes the pattern. Thomas describes abstract reasoning as a type of aptitude test connected to lateral thinking and fluid intelligence, while prep-market sources commonly connect it to pattern recognition, non-verbal reasoning, and adaptability.

Abstract reasoning is popular because it reduces dependence on job-specific knowledge. A candidate does not need to know accounting, law, or engineering to attempt a shape pattern. However, that does not make the test easy. The difficulty comes from hidden rule combinations, strict timing, and distractor options that satisfy only part of the pattern.

Where abstract reasoning appears

Abstract reasoning can appear as a standalone test or as part of a broader cognitive ability assessment. It is common in graduate recruitment, technology, consulting, engineering, operations, and roles where employers want to assess problem-solving with unfamiliar information. It also overlaps with inductive reasoning, diagrammatic reasoning, and spatial reasoning. Some providers use the term abstract; others use inductive, non-verbal, matrices, logical, or pattern recognition.

Common pattern rules

Most abstract reasoning questions are built from a manageable set of rule families. Look for changes in number, shape, size, shading, colour, position, rotation, direction, symmetry, overlap, sequence, addition, subtraction, movement, and alternation. In matrix questions, rules may operate across rows, down columns, or both. In sequence questions, one feature may change every step while another changes every second step. In odd-one-out questions, four items may share a hidden relationship while one breaks it.

The key is to scan systematically rather than stare. A practical scan order is: count items, compare shape type, compare shading, check rotation, check position, inspect movement, then test row/column relationships. If one rule explains only part of the pattern, keep looking. Many harder items require two rules, such as “the black square rotates clockwise while the number of circles increases by one.”

How to prepare for abstract reasoning

Preparation should focus on rule recognition. Start untimed and verbalize the rule for each question. Do not accept “I guessed right” as a successful review. If you cannot state the rule clearly, you did not learn the pattern. Next, practise with timing and develop a skip threshold. Abstract reasoning tests can punish over-investment in one puzzle. If no rule appears after a reasonable scan, move on and return if the interface allows it.

Build a personal rule bank. After each practice set, write down the rules you missed: alternating shading, mirrored rotation, diagonal addition, shape count, nested shape movement, row arithmetic, or opposite-corner logic. Over time, you will recognize families faster. TestSolve can support this by taking practice screenshots and turning the visual pattern into a written explanation that the candidate can remember.

Worked example: a 3x3 matrix question

Most abstract reasoning tests present a 3x3 grid with eight cells filled and the ninth left blank, then ask you to choose which of five or six answer options completes the grid. Here is an original, illustrative example — not a reproduction of any real test provider's actual item — that shows how to work through one systematically.

This example is intentionally simple to demonstrate the scanning method. Real test items typically combine two active rules simultaneously — for instance, size increasing left to right AND shading alternating top to bottom — which is exactly why the systematic scan order matters more than raw pattern-spotting instinct. A candidate who scans for one feature at a time (count, then shape, then shading, then rotation, then position) catches combined rules that a candidate who just "looks for the pattern" holistically often misses under time pressure.

Scoring and what counts as a good result

Abstract reasoning tests are scored against a comparison or norm group — typically a general working-population sample, though some providers use role-specific or graduate-specific norms. Your raw score (number correct, sometimes adjusted for guessing on some platforms) converts to a percentile or a standardised score band relative to that group, rather than being reported as a flat pass/fail cutoff. Employers then set their own minimum threshold for the specific role — a threshold in the 50th-60th percentile range is common for general graduate screening, while more analytically demanding roles (data, engineering, quantitative finance) sometimes set the bar at the 70th percentile or higher. Because abstract reasoning specifically measures fluid pattern-recognition rather than acquired knowledge, providers generally treat it as a stable, coachable-only-to-a-point measure — practice improves your speed and familiarity with rule types, but it doesn't change the underlying skill the way memorising vocabulary might for a verbal test.

Common mistakes on abstract reasoning tests

Fixating on the first rule you spot. Many items look solvable after finding one obvious rule (say, shape count increasing), but the correct answer also depends on a second, less obvious rule (shading alternating) that only becomes visible once you check every answer option against the first rule and find more than one seems to fit.

Ignoring rotation and reflection. Candidates who are comfortable with size, shading, and position changes often under-practice rotation and mirror-reflection rules specifically, since they're visually subtler and harder to verbalise quickly — these are worth deliberately including in practice sessions rather than skipping past.

Checking only rows, not columns (or vice versa). In matrix questions, the governing rule can run across rows, down columns, or both simultaneously. A scan that only checks one direction will occasionally land on a plausible-looking wrong answer that happens to satisfy the row logic but breaks the column logic.

Treating unfamiliarity as difficulty. Because abstract reasoning uses unfamiliar visual material by design, it's easy to feel like a question is unusually hard when it's actually a standard rule type you simply haven't drilled yet. Building a personal rule bank (described above) directly addresses this by converting "unfamiliar" into "recognised" over repeated practice.

Frequently asked questions

TestSolve is independent and not affiliated with any test provider or employer named on this page. All product names and trademarks belong to their respective owners.