"The Wason selection task
The Wason selection task provides important evidence for dual-process accounts of reasoning, because performance with the task is so sensitive to the content and context with which it is presented. The abstract, indicative selection task is illustrated in Figure Ia. The correct answer is generally agreed to be A and 7, although this has been disputed by some theorists. The statement can only be falsified by finding a case of a card that has an A on one side and does not have a 3 on the other. Only turning the A and 7 (not a 3) can lead to discovery of such a case. However, few people (10–20%) give this correct answer when tested.
Performance on the abstract selection task is thought to be strongly influenced by a System 1 heuristic known as ‘matching bias’. The effect is demonstrated by introducing negative components into the conditional statement. Imagine that the rule presented in Figure Ia was ‘If the there is an A one side of the card, then there is not a 3 on the other side’. The A and 3 cards, which have a perceptual match to the items named in the rule, are now also the logically correct cards, because discovery of an A3 card would falsify the rule. There is hence a strong tendency to choose matching cards, regardless of their logical status. This is one of the most reliable and robust biases in the psychological of reasoning. When the difficult affirmative form of the Wason selection task is couched in realistic terms it is very much easier to solve. For example, Griggs and Cox reported around 75% successful solutions of a problem in which they were instructed to imagine that they were police officers observing people drinking in a bar (Figure Ib). Most participants selected the cards showing ‘Drinking beer’ and ‘16 years of age’, the equivalent of the A and 7 cards on the abstract task described above. These choices could lead to discovery of an under-age drinker who is breaking the rule. This kind of problem is often described as a deontic selection task, as it concerns following rules rather than truth and falsity.
Figure I.
(a) In the abstract, indicative version of the Wason selection task, participants are given a conditional statement and shown four cards, each of which is known to have a letter on one side and a number on the other. The task is to decide which cards need to be turned over to find out whether the statement is true or false. In the following example, typical choices are A or A and 3, but the logically correct choices is A and 7.
(b) In a realistic and deontic version of the selection task, the participants are given a rule or regulation and need to check whether it is being obeyed. In the following example, they play a police officer checking people drinking in a bar. The cards represent different drinkers with the beverage on one side and the age on the other. Most people correctly choose the beer drinker and the 16 year old."
© Jonathan St.B.T. Evans
"In Two Minds: Dual-process Accounts of Reasoning" (2003)
[link]
"So, for example, I might give you a statement, “if there's an A on the one side, there's a three on the other.” And I might present you with four cards, and ask you which ones you need to turn over. [image of 4 cards] Obviously, you need to turn over the A, and everybody recognizes that. You need to check whether there's a three on the other side. Obviously, you don't need to turn over the D. You know that the cards have a letter on one side, and a number on the other. So there's no worry that there's an A on the other side of here. But people have a tendency to think that you need to turn over the three, and that you don't need to turn over the seven. But look out. Right on that other side of the seven was an A, and the statement turns out to be false. The ones you need to check are the A and the seven.
People find this task relatively difficult. But here's a structurally identical task that people find relatively easy. “If a person is drinking beer, the person must be over 21.” I show you four cards. The beer drinker, the soda drinker, the over-21 year old, and the 17 year old. And every single one of you could I take it get a job as a bouncer, walk in, discover that the 17 year old is drinking a beer. And thereby, learn exactly what was hard to see in earlier case, that the cards you need to turn over are the first and the last, rather than the first and the third.
So you notice that the second one is different from the first in two ways. One is that it's socially embedded, whereas the first is purely abstract. And we'll discover, as we continue our readings this semester, that social embedding awakens reasoning processes that aren't present otherwise. In fact, we learned that last week, with the eyes study when we learned that people are more pro-social, more likely to engage in morally normative behavior, when they're presented visually, with eyes, in part, because it awakens and activates a kind of social understanding which all of us have.And the second difference between the bottom and top is, of course, that this is a normative rule. It's about how things ought to be. Whereas this is a descriptive rule. It's about how things are."
© Tamar Gendler
Yale Lectures on Philosophy and the Science of Human Nature, 2011
[link]
The Wason selection task provides important evidence for dual-process accounts of reasoning, because performance with the task is so sensitive to the content and context with which it is presented. The abstract, indicative selection task is illustrated in Figure Ia. The correct answer is generally agreed to be A and 7, although this has been disputed by some theorists. The statement can only be falsified by finding a case of a card that has an A on one side and does not have a 3 on the other. Only turning the A and 7 (not a 3) can lead to discovery of such a case. However, few people (10–20%) give this correct answer when tested.
Performance on the abstract selection task is thought to be strongly influenced by a System 1 heuristic known as ‘matching bias’. The effect is demonstrated by introducing negative components into the conditional statement. Imagine that the rule presented in Figure Ia was ‘If the there is an A one side of the card, then there is not a 3 on the other side’. The A and 3 cards, which have a perceptual match to the items named in the rule, are now also the logically correct cards, because discovery of an A3 card would falsify the rule. There is hence a strong tendency to choose matching cards, regardless of their logical status. This is one of the most reliable and robust biases in the psychological of reasoning. When the difficult affirmative form of the Wason selection task is couched in realistic terms it is very much easier to solve. For example, Griggs and Cox reported around 75% successful solutions of a problem in which they were instructed to imagine that they were police officers observing people drinking in a bar (Figure Ib). Most participants selected the cards showing ‘Drinking beer’ and ‘16 years of age’, the equivalent of the A and 7 cards on the abstract task described above. These choices could lead to discovery of an under-age drinker who is breaking the rule. This kind of problem is often described as a deontic selection task, as it concerns following rules rather than truth and falsity.
Figure I.
(a) In the abstract, indicative version of the Wason selection task, participants are given a conditional statement and shown four cards, each of which is known to have a letter on one side and a number on the other. The task is to decide which cards need to be turned over to find out whether the statement is true or false. In the following example, typical choices are A or A and 3, but the logically correct choices is A and 7.
(b) In a realistic and deontic version of the selection task, the participants are given a rule or regulation and need to check whether it is being obeyed. In the following example, they play a police officer checking people drinking in a bar. The cards represent different drinkers with the beverage on one side and the age on the other. Most people correctly choose the beer drinker and the 16 year old."
© Jonathan St.B.T. Evans
"In Two Minds: Dual-process Accounts of Reasoning" (2003)
[link]
"So, for example, I might give you a statement, “if there's an A on the one side, there's a three on the other.” And I might present you with four cards, and ask you which ones you need to turn over. [image of 4 cards] Obviously, you need to turn over the A, and everybody recognizes that. You need to check whether there's a three on the other side. Obviously, you don't need to turn over the D. You know that the cards have a letter on one side, and a number on the other. So there's no worry that there's an A on the other side of here. But people have a tendency to think that you need to turn over the three, and that you don't need to turn over the seven. But look out. Right on that other side of the seven was an A, and the statement turns out to be false. The ones you need to check are the A and the seven.
People find this task relatively difficult. But here's a structurally identical task that people find relatively easy. “If a person is drinking beer, the person must be over 21.” I show you four cards. The beer drinker, the soda drinker, the over-21 year old, and the 17 year old. And every single one of you could I take it get a job as a bouncer, walk in, discover that the 17 year old is drinking a beer. And thereby, learn exactly what was hard to see in earlier case, that the cards you need to turn over are the first and the last, rather than the first and the third.
So you notice that the second one is different from the first in two ways. One is that it's socially embedded, whereas the first is purely abstract. And we'll discover, as we continue our readings this semester, that social embedding awakens reasoning processes that aren't present otherwise. In fact, we learned that last week, with the eyes study when we learned that people are more pro-social, more likely to engage in morally normative behavior, when they're presented visually, with eyes, in part, because it awakens and activates a kind of social understanding which all of us have.And the second difference between the bottom and top is, of course, that this is a normative rule. It's about how things ought to be. Whereas this is a descriptive rule. It's about how things are."
© Tamar Gendler
Yale Lectures on Philosophy and the Science of Human Nature, 2011
[link]
No comments:
Post a Comment