Description
People discover a hidden rule by proposing number triples and receiving yes or no feedback, revealing whether they test hypotheses by confirmation or by trying to falsify them.
The experimenter says the triple 2, 4, 6 fits a rule. Participants propose triples with reasons and are told only whether each fits, announcing the rule when confident. Most hypothesise 'increasing by two' and keep testing triples that fit it, so they never discover the real rule, any three increasing numbers. The few who succeed first time test triples designed to break their own hypothesis, such as 2, 4, 5. The record of proposed tests shows the proportion of confirming versus disconfirming tests and the correctness of the final rule, making the task a direct measure of hypothesis-testing strategy rather than of knowledge.
Attributes
- Discipline
- Behavioral Science & Economics
- Loss type
- Unmigrated
- Stage
- Discovery
- Data
- Mixed
- Effort
- Hours
Steps
- Give the seed instance (2, 4, 6) and explain that it fits a hidden rule.
- Ask participants to propose triples one at a time, writing the hypothesis and reason behind each.
- Give only 'fits' or 'does not fit' feedback.
- Let participants announce the rule when confident; record whether it is correct and how many tests preceded it.
- Code each test as confirming or disconfirming relative to the participant's stated hypothesis.
- Optionally adapt the task to a domain (for example discovering which settings trigger a product behaviour) and compare strategies.
Sources
- Cognitive Psychology (Goldstein, 2019)
- Wason, P. C. (1960). On the failure to eliminate hypotheses in a conceptual task. Quarterly Journal of Experimental Psychology, 12
- Handbook of Cognition (Lamberts & Goldstone, 2005)
- MIT Psychology (Stangor, 2012)
Also known as
- Rule discovery task
- 2-4-6 task
- Wason rule discovery
Categories
- Analytics & Behaviour
- Behavioral Science & Economics