Free Deductive Reasoning Test with Figures

Reasoning from stated rules instead of guessed patterns

Most visual reasoning tests hide their rule. You see a grid of shapes, you infer the pattern that produced it, and you extend that pattern one step further. That is induction, and it rewards a particular kind of imagination — the willingness to try several explanations quickly and commit to the neatest one. Deduction works in the opposite direction. The rule is handed to you in plain language, and the only question is whether a given diagram satisfies it. There is nothing to imagine and nothing to invent. What the test measures instead is discipline: can you hold two or three constraints in mind at once, check them against four candidates without losing your place, and resist the very strong pull to add a fourth rule that nobody wrote down?

The figures on this page are deliberately meaningless. A square is a square, it is large or small, it is shaded or white, and that is the entire vocabulary. Stripping the content out is not laziness; it is the point. As soon as a logic item talks about doctors, cities or animals, your knowledge of doctors, cities and animals starts to interfere with your reading of the rule, and the test quietly becomes a test of background beliefs. Abstract figures remove that interference, which is also why this format travels well across languages and school systems.

How an item is built

Every item begins with two or three rules. Some are universal statements — every square is shaded, no triangle is large — which constrain each figure of that kind without saying any such figure exists. Some are existential — there is at least one square, exactly one figure is small — which demand that something be present, or present a precise number of times. Some are conditional — if a figure is large, it is a circle — which bind only when their trigger fires. Four diagrams follow. Three of them break at least one printed rule, and exactly one breaks none. Because the rules are explicit, the key is not a matter of opinion: you can prove your answer by pointing at the rule each rejected diagram violates.

Working through an item efficiently is a mechanical process, and it is worth doing the same way every time. Take the first rule alone and sweep across all four diagrams, crossing out every one that breaks it. Then take the second rule and sweep across whatever survives. Sorting by rule rather than by diagram is faster because you only have to hold one constraint in working memory at a time, and it is more accurate because you never accidentally check a diagram against a rule you have already half-forgotten.

The mistakes almost everyone makes

The first and largest is treating a universal rule as though it promised existence. Every square is shaded says nothing at all about a diagram that contains no squares, and such a diagram is perfectly legal. Logicians call this vacuous truth, and it is the single most common reason a confident answer turns out to be wrong. The second mistake is reversing a conditional. If a figure is large, it is a circle does not tell you that circles are large; a small circle is untouched by that rule. The third is importing symmetry that was never stated. If one rule concerns squares and another concerns triangles, nothing follows about circles, and a diagram may do whatever it likes with them. The fourth is fatigue: by item twelve the rules start to look familiar, and people begin answering from the shape of the last item rather than reading the one in front of them. Each item is written independently, and rules that appeared earlier do not carry forward.

Reading your score

Your raw total out of sixteen is mapped onto the familiar scale where 100 is average and fifteen points is one standard deviation. Deductive items behave differently from pattern items in one important way: because the answer is verifiable rather than intuited, scores tend to be more stable across repeat attempts and less affected by mood or fatigue. A low score here usually points at a specific, fixable habit rather than a general limit.

Getting a more honest result

• Read all the rules before you look at a single figure. Looking first plants an expectation you will then defend.
• Eliminate rather than select. Finding one broken rule kills a diagram instantly; confirming a diagram takes every rule.
• Say each rule to yourself in the negative — "what would break this?" — before scanning.
• Answer every item. Nothing is deducted for a wrong answer, so a blank is always the worst option, and the Back button lets you revisit anything the clock allows.
• Take it once, rested, on a screen large enough that shading is unmistakable.

Where this test fits

Deductive figure items are close relatives of the abstract reasoning sections used in graduate admissions and in employer aptitude batteries, where the goal is to predict how someone handles written specifications, contracts, eligibility rules and edge cases. If you want the inductive counterpart, the culture fair IQ test asks you to discover rules from matrices instead of applying rules that are given, and the quick IQ test mixes several families into a five minute estimate. Taking one of each is more informative than taking either twice, because a large gap between your inductive and deductive scores says something specific: people who read carefully but pattern-match slowly tend to be strongest right here.

What this test does not claim

A sixteen item screen is not a clinical assessment and does not diagnose anything. It is not normed on a stratified sample, it cannot see how you approached a problem, and it cannot separate a genuine reasoning limit from a bad night's sleep or an unfamiliar interface. Use the number as feedback on a specific skill — applying explicit rules without embroidering them — and treat any site that promises more from a free web page with the suspicion it deserves.