Logical reasoning is the cognitive backbone of mathematics, law, programming, and any kind of structured argument. It is also the cognitive ability with the cleanest right-or-wrong test in psychometrics: every logic item has an answer the rules of inference produce, and either you reach it or you don't. There is no "I can see why someone might think that" in formal logic.
This guide walks through what a logical reasoning test actually measures, why even highly educated adults fail surprisingly basic logic tasks, and how to read your score honestly. If you'd rather see your own score first, our free 10-question logical reasoning test takes about six minutes.
What logical reasoning is
Logical reasoning is the ability to draw valid conclusions from given premises — to follow a chain of inference without losing the thread, and to recognize when an argument looks sound but isn't. The standard distinction in psychometrics is between two modes of reasoning:
- Inductive reasoning. Generalizing from observations. "Every swan I've seen is white, therefore all swans are white." Probabilistically useful, formally invalid.
- Deductive reasoning. Applying rules of inference to premises. "All A are B; this is an A; therefore this is a B." If the premises are true and the rules are followed, the conclusion is guaranteed.
A logical reasoning test focuses primarily on deductive reasoning — on whether you can apply the rules even when the conclusion clashes with intuition or background knowledge. Pattern recognition tests measure your ability to find the rule. Logic tests measure your ability to follow one.
What a logical reasoning test measures
A modern logic test is built from a small set of canonical item types, each isolating a specific deductive move:
- Modus ponens. If P then Q; P; therefore Q. The most basic deductive move — almost everyone gets this right when stated clearly.
- Modus tollens. If P then Q; not Q; therefore not P. Formally identical in difficulty, but error rates double or triple compared to modus ponens (Evans, 2003).
- Categorical syllogisms. Reasoning with "all", "some", and "no". "All A are B; some B are C; therefore..." — deceptively hard, especially when conclusions feel intuitively true but don't follow.
- Logical equivalence. Recognizing that "If P then Q" is equivalent to "If not-Q then not-P" (the contrapositive). Educated adults regularly miss this.
- Necessary vs. sufficient conditions. Distinguishing what must be true from what guarantees a result. The single most common source of legal and engineering errors.
- Pigeonhole / minimum-case reasoning. "How many socks must I draw to guarantee a matching pair?" Counting the worst case rather than the average.
- Knights-and-knaves puzzles. Multi-person scenarios where some always tell the truth, others always lie. Requires case analysis — a key marker of strong logical reasoning.
- Truth-table evaluation. Formally evaluating compound statements with AND, OR, NOT, and conditional operators.
Our logical reasoning mini-test samples across these item types in 10 questions, with a deliberate difficulty ramp: gentle at first, then ceiling items at Q9–Q10 designed to differentiate the top decile.
"If it rains, the picnic is canceled." The picnic was canceled. Therefore it rained? No. The conditional only goes one direction. The picnic could have been canceled for many other reasons. This is "affirming the consequent," one of the most common formal fallacies in everyday reasoning.
The Wason result and why it matters
In 1966, Peter Wason published what became one of the most-cited results in cognitive psychology. Participants are shown four cards, each with a number on one side and a color on the other:
3 8 brown red
The rule to test: If a card shows an even number on one face, the opposite face is red. Which cards must you turn over to test the rule?
Most people pick "8" and "red." The correct answer is "8" and "brown" — the brown card could falsify the rule (an even number on the back would break it), while the red card cannot (whatever's on the back, the rule still holds). Fewer than 10 percent of college-educated adults solve the abstract version correctly. The same participants solve a structurally identical problem framed in social terms ("If you drink alcohol, you must be over 21") at over 75 percent accuracy.
What this tells us: formal logical competence is partly decoupled from everyday reasoning. People reason well when content matches familiar social rules, and poorly when the same logic is presented abstractly. A logical reasoning test measures the abstract capacity — the harder, more transferable kind.
Wason’s card task shows people fail abstract logic. Evans, Barston, and Pollard’s 1983 belief-bias studies show something more uncomfortable: people actively misjudge validity based on whether they like the conclusion. Given a syllogism with a believable conclusion, participants accept it even when the logic is invalid. Given the same logical structure with an unbelievable conclusion, they reject it even when the logic is valid. The effect is large (around 25–30 percentage points), survives instructions to ignore content, and barely diminishes with intelligence: high-IQ adults are still vulnerable to it, just at slightly lower rates. This is why logic tests matter: they measure your ability to follow the rules against what you want to believe, which is the cognitive move most adults never deliberately train.
Logic vs. pattern recognition
Both correlate with general intelligence (g), but they tap different cognitive operations:
| Pattern recognition | Logical reasoning |
|---|---|
| Bottom-up: spot regularity | Top-down: apply rules |
| Rule is hidden, must be inferred | Rule is given, must be followed |
| Visual / abstract figures | Verbal / symbolic premises |
| Loads heavily on Gf | Loads on Gf and verbal-symbolic ability |
| Resists practice | Highly trainable |
The dissociation matters in practice. Some people are excellent at spotting hidden regularities (high pattern score) but average at multi-step deductive chains. Others are the reverse. A full IQ test scores both because each captures something the other misses.
See where you stand on logical reasoning
All five domains, on the standard IQ scale.
What your logic score predicts
Logical reasoning predicts performance in any field where structured argument matters: law, mathematics, philosophy, programming, the natural sciences, and analytic finance. The strongest single piece of evidence comes from LSAT validity studies, where logical-reasoning subsection performance correlates with first-year law school grades at approximately r = 0.4 — a substantial effect even after controlling for undergraduate GPA (Wightman, 1998).
Outside formal academic settings, logical reasoning predicts the ability to:
- Spot logical inconsistencies in arguments and contracts.
- Debug code — tracing why an unexpected output occurred.
- Navigate complex regulations and tax codes.
- Construct mathematical proofs.
- Distinguish correlation from causation in data.
What it doesn't predict: persuasiveness, emotional intelligence, or the ability to win arguments. Persuasion runs on rhetoric and audience; logic runs on validity. The two correlate weakly.
How to read your score
On a 10-question logical reasoning mini-test with a difficulty ramp, here's how raw scores typically map to bands:
| Score | Band | What it means |
|---|---|---|
| 9–10 | Exceptional | Top few percent. Solves multi-step deductions including knights-and-knaves. |
| 7–8 | Strong | Clear above-average. Reliable on syllogisms and contrapositives. |
| 5–6 | Average | Reliable on modus ponens; struggles with modus tollens and quantifiers. |
| 3–4 | Below average | Likely confused by abstract phrasing; would benefit from explicit logic training. |
| 0–2 | Significantly below | Worth retesting fresh; if consistent, suggests verbal-symbolic strengths in other domains. |
Two caveats. First, logical reasoning is the most trainable of the cognitive domains — an introductory logic course or a few weeks of LSAT prep can shift scores by a full band. The score reflects your current capability, not a fixed ceiling. Second, fatigue and time pressure both degrade logical performance disproportionately. A score taken at the end of a long day may understate your actual ability.
How to take a logical reasoning test fairly
- Read each premise twice. Most logic errors come from misreading the conditional, not from invalid inference. "If P then Q" and "P only if Q" mean different things.
- Translate to symbols when stuck. Replace nouns with letters: "all dogs are mammals" becomes "all D are M". Symbol manipulation is what your brain is being asked to do; nouns can mislead.
- Watch for the converse and inverse. "If P then Q" does not imply "If Q then P" (the converse) or "If not-P then not-Q" (the inverse). Only the contrapositive ("If not-Q then not-P") is equivalent.
- For syllogisms with "some", draw a Venn diagram mentally. "Some A are B; all B are C" lets you conclude "some A are C" — but "Some A are B; some B are C" lets you conclude nothing.
- Don't let intuition override the form. An argument can be valid (the conclusion follows from the premises) even when the premises are false, and invalid even when the conclusion happens to be true.
Frequently asked questions
What does a logical reasoning test measure?
It measures your ability to draw valid conclusions from given premises. Item types typically include modus ponens and modus tollens (the two foundational deductive moves), categorical syllogisms with quantifiers, conditional reasoning, and multi-step puzzles like knights-and-knaves that require case analysis.
How is logical reasoning different from pattern recognition?
Pattern recognition is bottom-up: you spot regularity in unfamiliar material and infer the rule. Logical reasoning is top-down: you start with explicit premises and apply rules of inference. Both correlate with general intelligence but use different cognitive subsystems — some people score high on one and only average on the other.
What is the Wason selection task and why does it matter?
The Wason selection task, designed in 1966, asks participants to test a conditional rule by selecting cards. Fewer than 10 percent of college-educated adults solve the abstract version correctly, even though the underlying logic is simple. It became a foundational result in cognitive psychology — it shows that formal logical competence is partly decoupled from everyday reasoning.
Can logical reasoning be improved with practice?
Yes, more reliably than most cognitive abilities. Studies on critical-thinking instruction (Halpern, 1998; Abrami et al., 2015) show explicit teaching of logical principles produces measurable transfer to novel problems. Law-school training in particular improves performance on conditional-reasoning tasks.
Why do logic items appear on the LSAT and GRE?
Logical reasoning predicts performance in fields where structured argument matters — law, mathematics, philosophy, programming, science. The LSAT uses logic items because they are a strong predictor of first-year law school grades, with validity coefficients around r = 0.4 even after controlling for undergraduate GPA.
Related reading
- What Your IQ Score Actually Means — how the standard IQ scale works, and what your number does and doesn't predict.
- Pattern Recognition Test — the bottom-up companion to deductive reasoning.
- Numerical Reasoning Test — reasoning with quantities, sequences, and proportions.
- Verbal Reasoning Test — reasoning with language, vocabulary, and analogy.
References
- Wason, P. C. (1966). Reasoning. In B. M. Foss (Ed.), New Horizons in Psychology. Penguin Books.
- Evans, J. St. B. T. (2003). In two minds: Dual-process accounts of reasoning. Trends in Cognitive Sciences, 7(10), 454–459.
- Cosmides, L. (1989). The logic of social exchange: Has natural selection shaped how humans reason? Cognition, 31(3), 187–276.
- Halpern, D. F. (1998). Teaching critical thinking for transfer across domains. American Psychologist, 53(4), 449–455.
- Abrami, P. C., et al. (2015). Strategies for teaching students to think critically: A meta-analysis. Review of Educational Research, 85(2), 275–314.
- Wightman, L. F. (1998). LSAC National Longitudinal Bar Passage Study. Law School Admission Council.
- Johnson-Laird, P. N. (1983). Mental Models: Towards a Cognitive Science of Language, Inference, and Consciousness. Harvard University Press.
- Evans, J. St. B. T., Barston, J. L., & Pollard, P. (1983). On the conflict between logic and belief in syllogistic reasoning. Memory & Cognition, 11(3), 295–306.