But the test would involve some small numbers, shapes, coins, etc.; not real-life examples.
I dunno… people who do fine at the Wason selection task with ages and drinks get it wrong with numbers and colours. (I’m not sure whether that’s a bug or a feature.)
That seems to me like a reason not to test the skill on real-life examples.
We wouldn’t want a rationality test that a person can pass with original wording, but will fail if we replace “Republicans” by “Democrats”… or by Green aliens. We wouldn’t want the person to merely recognize logical fallacies when spoken by Republicans. This is in my opinion a risk with real-life examples. Is the example with drinking age easier because it is easier to imagine, or because it is something we already agree with?
Okay, I am curious here… what exactly would happen if we replaced the Wason selection task with something that uses words from real life (is less abstract), but is not an actual rule (therefore it cannot be answered using only previous experience)? For example: “Only dogs are allowed at jumping competitions, cats are not allowed. We have a) a dog going to unknown competition; b) a cat going to unknown competition; c) an unknown animal going to swimming competition, and d) an unknown animal going to jumping competition—which of these cases do you have to check thoroughly to make sure the rule is not broken?”
I dunno… people who do fine at the Wason selection task with ages and drinks get it wrong with numbers and colours. (I’m not sure whether that’s a bug or a feature.)
That seems to me like a reason not to test the skill on real-life examples.
We wouldn’t want a rationality test that a person can pass with original wording, but will fail if we replace “Republicans” by “Democrats”… or by Green aliens. We wouldn’t want the person to merely recognize logical fallacies when spoken by Republicans. This is in my opinion a risk with real-life examples. Is the example with drinking age easier because it is easier to imagine, or because it is something we already agree with?
Okay, I am curious here… what exactly would happen if we replaced the Wason selection task with something that uses words from real life (is less abstract), but is not an actual rule (therefore it cannot be answered using only previous experience)? For example: “Only dogs are allowed at jumping competitions, cats are not allowed. We have a) a dog going to unknown competition; b) a cat going to unknown competition; c) an unknown animal going to swimming competition, and d) an unknown animal going to jumping competition—which of these cases do you have to check thoroughly to make sure the rule is not broken?”