# Wason Selection Task: The 4-Card Puzzle 90% Fail

*Society, Psychology & Life · Daniel Foss · 2026-10-04*

## TL;DR

The Wason Selection Task is a four-card reasoning test introduced by Peter Wason in the 1960s. Most people fail it, and the task reveals that we rarely search for evidence against our own assumptions. In 1982, Richard Griggs and James Cox showed that an alcohol-and-age version made the logic far easier.

## Key facts

- Peter Wason developed the four-card selection task in the 1960s to test how people evaluate conditional rules.
- In the standard version, participants see four cards showing a vowel, a consonant, an even number and an odd number, and must decide which cards to turn over to test the rule "if a card has a vowel on one side, then it has an even number on the other."
- The logically correct choices are the vowel card and the odd-number card, because only those two could expose a counterexample to the rule.
- Most participants choose the vowel card and the even-number card, a pattern researchers call matching bias.
- A 1982 study by Richard Griggs and James Cox found that framing the same rule around drinking age and alcohol sharply raised the proportion of correct answers.

## The Four Cards That Reveal How We Really Think

In the 1960s, a British psychologist named Peter Wason built an experiment that many researchers still rank among the most important experimental models in the psychology of reasoning. It became known as the Wason Selection Task, and it is disarmingly simple: four cards, one rule, and a question about which cards you would flip over to test whether that rule holds.

Wason himself was not a dry academic type. Colleagues and students remember him as witty, original, and a little stubborn about how science should be done. He treated reasoning less like a skill to be measured and more like a puzzle to be solved. That distinction shaped everything about his work.

One habit of his says a lot. He argued that researchers should not know exactly what they were looking for before they ran an experiment. The point was not to confirm a theory you already believed. The point was to go find out what human thought actually does when you corner it. As his students put it, his real aim was to show that people are not nearly as rational as they believe themselves to be.

It is a strange thing to spend a career on, and it is exactly why those four little cards still make people squirm today.

## How the Wason Selection Task Works, and Why Almost Everyone Gets It Wrong

Picture four cards lying flat on a table in front of you. One shows the letter A, one shows D, one shows the number 4, and one shows the number 7. Each card carries a single digit on one side and a single vowel or consonant on the other, and you can only see one face of each.

Now the rule: *If a card has a vowel on one side, it must have an even number on the other side.* Your job is to decide which cards you would need to turn over to test whether that rule is true, without flipping any card you do not actually need. That is the **Wason Selection Task** in its standard form, the version Peter Wason and Philip Johnson-Laird laid out in *Psychology of Reasoning: Structure and Content* in 1972.

Take a second and commit to an answer before you read on. Most people land on the A card alone, or on A and 4 together. The correct answer is A and 7.

### Why the 7 card matters

The logic is narrower than it first looks. Turning over the A card is obvious: if an odd number sits behind it, the rule collapses on the spot. The 4 card feels equally obvious, but it cannot break anything. The rule only says a vowel needs an even number on the other side. It never says what has to sit behind an even number, so a consonant behind the 4 would leave the rule perfectly intact. The 7 card is the one people skip, and it is the one that does the work. If a vowel is hiding behind that 7, you have a vowel paired with an odd number, and the rule is dead.

Testing a rule means hunting for the cases that could falsify it, not the cases that flatter it. That single shift in framing is what separates the A-and-7 answer from the A-and-4 answer, and it is exactly the shift most people fail to make.

### The same wrong answer, over and over

What struck Wason was not that people erred. It was that they erred identically. Accounts of the original work often claim that more than 90 percent of his participants made mistakes, though that exact figure is not reliably documented, and the errors clustered around the same handful of choices rather than scattering randomly across the options. A participant later told Wason something that has the ring of a confession: "I'm not happy with my first choice, but if I did the task again I'd still pick the same cards."

That line unsettled him. The prevailing view in the 1960s held that human beings reason logically and analytically by default, and here were people who could see their own error clearly and still walk straight back into it. Wason was left with a question he could not shake: were people genuinely bad at logic, or was the problem simply wearing a costume that made it look harder than it was?

## The 1982 Alcohol Scenario That Changed Everything

In 1982, two psychologists at the University of Florida, Richard Griggs and James Cox, did something almost cheeky with the Wason Selection Task. They kept the logic exactly the same and threw out the letters and numbers. Instead, they asked people to imagine they were police officers checking whether alcohol was being sold to minors. The rule on the table: "If someone is drinking beer, they must be over 21."

The result turned the experiment upside down. In this version, a large majority of participants found the right answer, a jump so steep that researchers gave it a name: the content effect. Widely repeated figures put the success rate around 75 percent, though that number gets cited more confidently than the original data can support. What is not in doubt is the direction of the shift. The same people who stumbled over A, D, 4, and 7 suddenly became sharp-eyed detectives when the cards turned into drinkers and ages.

Here is the part that stings. Both problems share the identical logical skeleton: "If P, then Q." The abstract version demands you check the vowel card and the odd card. The beer version demands you check the beer drinker and the under-21 drinker. Same move, different costume. Yet the abstract task left most people flailing, while the barroom scenario had them spotting rule violations almost effortlessly.

Same logic**Abstract:** "If a card has a vowel, it must have an even number."

Same logic**Real life:** "If someone is drinking beer, they must be over 21."

Griggs and Cox had not made the puzzle easier. They had made it *mean* something. And that small act of reframing exposed a crack in the assumption that reasoning is a single, context-free muscle. People were not bad at logic. They were bad at logic stripped of any human stake.

The finding did not settle the debate. It lit it on fire. If a problem's difficulty depends not just on its logical structure but on how it is presented, then what exactly are we measuring when we test "reasoning"? The answer would take decades to untangle, and it would eventually pull in one of the most famous psychologists of the century.

## Kahneman, System 1 and System 2, and the Lazy Brain Behind the Error

Daniel Kahneman, the psychologist who won the 2002 Nobel Prize in Economic Sciences, published *Thinking, Fast and Slow* in 2011. That book carried the dual-process theory, the idea Wason had been circling decades earlier, to millions of readers. The mind, Kahneman argued, runs on two different cognitive systems, and they do not take turns politely.

System 1 is fast, intuitive, and automatic. System 2 is slower, analytical, and deliberate. In the Wason Selection Task the two are effectively competing, and most people answer under the influence of fast System 1. That is why so many hands reach for the A and 4 cards. The rule mentions "vowel" and "even number," so the mind automatically gravitates toward the cards that match those words. Psychologists call this matching bias, documented by Jonathan Evans in 1972.

Here is the trap. Turning the 4 card is logically useless. If a consonant sits on the other side, the rule is not broken at all. System 1 does not run that check, because running it costs effort. Intuitive thinking is quick and demands less mental energy. Abstract reasoning is more tiring, and System 2 often acts lazily, leaving decisions to System 1. Kahneman is often cited as describing this tendency as "cognitive laziness," though the exact phrasing as his signature characterization is not firmly confirmed.

Two further claims from the source deserve a careful label. The first is that people who solve the Wason problem correctly tend to show higher academic achievement. The second is that System 2, linked to analytical thinking, develops through childhood and weakens in old age. Both are reported but contested, and neither should be treated as settled science. What is not in dispute is the architecture itself: one system that answers before you notice, and one that could check the work but usually would rather not.

## What the Wason Selection Task Still Can't Settle

Six decades after Peter Wason first put four cards in front of volunteers, researchers still cannot agree on what the Wason Selection Task actually measures. The puzzle that produced those stubborn errors is not a closed case. It remains one of the most argued-over open questions in the psychology of reasoning, and competing labs keep reading the same data in different ways.

Part of the trouble is that the task refuses to behave like a single experiment. In the abstract version, people face A, D, 4 and 7 and mostly fail. In the 1982 drinking-age version from Richard Griggs and James Cox, performance jumps, and the researchers themselves labeled that shift the "content effect." One camp reads the gap as proof that human reasoning is deeply context-bound, that we are built to catch rule-breakers in social situations and stumble over pure logic. Another camp insists the two versions are not testing the same mental operation at all, so comparing them proves less than it seems.

Kahneman's dual-process picture, popularized in *Thinking, Fast and Slow* in 2011, gave the debate a shared vocabulary, but it did not settle anything. System 1 is fast, intuitive, and automatic; System 2 is slow, analytical, and deliberate. But a label is not an explanation. Saying that System 1 answers the cards and System 2 stays in its chair describes the pattern without telling us why the analytical system declines to show up, or whether it is even able to override the intuitive pull on demand.

Then there is the question of who gets it right. Some work links correct solutions to stronger academic performance, and there is a suggestion that analytical reasoning sharpens through childhood and dulls with age. Treat these as leads rather than verdicts. The evidence is suggestive, not airtight, and a clean story about System 2 rising and falling across a lifetime outruns what the studies can currently support.

What survives all the disagreement is the core finding Wason landed on in the 1960s: people are not the reliably logical creatures they imagine themselves to be. The correct answer is still A and 7, the matching bias toward the cards named in the rule is still real, and the puzzle still catches almost everyone. What stays unresolved is the deeper why. Is the failure a flaw in logic itself, or a trick of how the problem is dressed? On that, the field is still arguing.


## Frequently asked questions

**What is the Wason Selection Task and why is it famous?**

The Wason Selection Task is a four-card reasoning experiment created by British psychologist Peter Wason in the 1960s. In the standard version, cards show A, D, 4, and 7, and you must test the rule "If a card has a vowel on one side, it must have an even number on the other." It became one of the most important experimental models in the psychology of reasoning.

**Why do most people fail the Wason Selection Task?**

People tend to hunt for evidence that confirms the rule instead of evidence that could break it, so they rarely think to check the 7. A second force, matching bias, pulls them toward the cards whose values are actually named in the rule, A and 4, even though turning the 4 tells you nothing useful.

**How did Griggs and Cox change the task in 1982?**

In 1982, psychologists Richard Griggs and James Cox kept the identical logic but wrapped it in a real-life scenario. Participants imagined checking whether minors were being sold alcohol, under the rule "If someone is drinking beer, they must be over 21." Same structure, familiar content, very different results.

**What is the content effect?**

The content effect is the finding that performance jumps dramatically when a problem is dressed in familiar real-world material, even though the underlying logic of "If P, then Q" never changes. The abstract version and the drinking-age version demand exactly the same reasoning, yet people handle the concrete one far more successfully.

**Was the Wason Selection Task proof that people are irrational?**

Not quite. It showed that people struggle with abstract logic but do much better when a problem connects to real life. That complicated the assumption of Wason's era, when many researchers believed humans simply reason in a naturally logical, analytical way. The task revealed a gap between abstract and everyday reasoning, not a blanket verdict of irrationality.

**How does Kahneman explain the task with System 1 and System 2?**

Nobel Prize-winning psychologist Daniel Kahneman laid out dual-process theory in his 2011 book Thinking, Fast and Slow . System 1 is fast, intuitive, and automatic; System 2 is slow, analytical, and deliberate. In the Wason Selection Task the two compete, and the quick System 1 usually wins.

**What is matching bias?**

Matching bias is the tendency to select the cards whose values are explicitly named in the rule, here A and 4, rather than the cards that could actually falsify it. Psychologist Jonathan St. B. T. Evans documented the pattern in 1972, and it helps explain why so many people land on the same wrong answer. Editorial note: Details that remain unconfirmed, including exact figures and dates, are presented here as approximations rather than settled facts.

Canonical: https://curio.madeda.com/wason-selection-task-the-4-card-puzzle-90-fail/