Wason selection task 2.0
About this app
Rebuild 2.0 (preview): five scenarios from abstract to invoice audit, flipped cards land in the truth table, resolution with your own choice pattern and an SQL query.
If P, then Q: only one cell counts
A rule “if P, then Q” is violated in only one case: P holds and Q does not. In a truth table with rows P and not P and columns Q and not Q, that is a single cell. This is why the Wason selection task requires turning over the P card and the not-Q card. The Q card and the not-P card can never refute the rule. Following Karl Popper, universal statements cannot be proven by confirmations, but a single counterexample refutes them.
What the app shows
The app offers five scenarios: letters and numbers, a negated rule, a drinking age rule, a discount hypothesis and an invoice audit. After you check, your cards flip first, then the rest, and each lands in the truth table cell described by its two sides. The solution names your choice pattern, for example “P and Q”, and shows the same test as an SQL query for “P and not Q”. The studies panel shows rounded figures on typical answers from research.
What to look for
Compare the abstract rule with the negated rule “if vowel, then not an even number”. Here, picking the cards whose values appear in the rule happens to give the right answer. This exposes the matching bias described by Jonathan Evans. In the discount scenario you see that a hundred discount customers who bought more do not prove the rule. With the thinking aid on, tapping a card highlights the two cells it can land in.
Frequently asked questions
- Why do so many people fail the Wason task?
- The most common choice is P and Q, the cards that seem to fit the rule. Two effects are at work: the search for confirmation and matching bias, picking cards whose values are named in the rule. The necessary not-Q card gets overlooked.
- Why is the drinking age version so much easier?
- In the drinking age version the rule is a regulation, and you are looking for rule breakers. Most people find that easy. Leda Cosmides explained the effect with a sense for detecting cheaters in social contracts, Patricia Cheng and Keith Holyoak with permission schemas.
- What does the Wason task have to do with data analysis?
- A hypothesis such as “all customers with a discount bought more” is tested by looking for counterexamples: customers with a discount who did not buy more. As a database query that is “P AND NOT Q”. If the result is empty, the rule holds for this data.
- What does this version add to the first Wason app?
- It has five scenarios instead of two, places each flipped card in a truth table and interprets your choice pattern. It also adds an SQL query per scenario and a panel with findings from studies.
Subject: Data & decisions
More from Data & decisions
- Anchoring effect
- Benford's law
- Berkson's paradox
- Cherry-picking: base period
- Cherry-picking: base period 2.0
- The three-cowboy duel (truel)
- The three-cowboy duel 2.0
- The three means
A VisuApp by heyprof: interactive, free in your browser, no sign-up. What is a VisuApp? · All apps