How Did You Do?

Thank-you for completing the test. You will find details below of yours and other people's performance in these questions, together with an explanation of the correct answer for each of the questions that you completed. You will also find an analysis of the significance of these kinds of test.

Detail

Analysis

Click for technical information about this experiment

The tasks that you have just completed are variations on the Wason selection task that was devised in 1966 by Peter Wason. The Wason selection task was originally developed as a test of logical reasoning, but it has increasingly been used by psychologists to analyse the structure of human reasoning mechanisms. It tests the ability of subjects to look for exceptions to a conditional rule of the form If P, then Q. As you have seen, the task begins with a person being presented with four cards: one card represents P (e.g., a circle); one not-P (e.g., a square); one Q (e.g., yellow); and the other not-Q (e.g., red). They are told that if there is a shape on one side of the card then there is a colour on the other. The person is then asked to name which of the four cards must definitely be picked up in order to determine whether any of the cards break the rule If P, then Q - in this instance: if a card has a circle on one side (P), then it has the colour yellow (Q) on the other side. (To make this clearer, I suggest you click on Question 2 in the Standard Wason Test Scores table below.)

The answer is that you should pick up the cards with the circle (P) and the colour red (not-Q). The card with the square (not-P) does not have to be picked up, as finding a surface of any colour on the other side tells us nothing about the truth or falsity of the rule. And neither does the card with the colour yellow showing have to be picked up, because finding a circle or square on the other side of that card would also be uninformative in terms of the rule (it will be obvious that finding a circle on the other side does not falsify the rule, but crucially nor does finding a square, since the rule does NOT say, "If, and only if, a card has a circle on one side, then it has the the colour yellow on the other side").

The significant point about this test is that we are incredibly bad at it! Typically, some 75%-80% of people get the test wrong. And it doesn't make much difference what the level of education is of the person taking the test. Moreover, even a training in formal logic seems to make little difference to a person's performance! The mistake that we tend to make is fairly standard. People almost always recognise that they have to pick up the card with the circle (P), but they fail to see that they also have to pick up the card with the colour red (not-Q). And often - and mistakenly - they think instead that they have to pick up the card with the colour yellow (Q). (One of the most interesting things about this phenomenon is that even when the correct answer is pointed out, people feel resistance to it. It apparently feels "right" that the card with the colour yellow (Q) should be picked up.)

There are a number of important implications of the fact that we tend to be bad at the Wason selection task (and indeed, other similar tasks, e.g., the conjunction problem). One has to do with the notion of justified belief. If a belief is recognised to be based on defective reasoning, then to continue to believe it is not justified. But if we systematically, and unconsciously, reason badly, then the extent to which reason actually acts as a constraint on belief is a moot point.

Another implication, and the one that we'll focus on here, has to do with what these tests tell us about the way that the human mind has evolved. According to Leda Cosmides & John Tooby, the results of the Wason selection task demonstrate that the human mind has not evolved reasoning procedures that are specialised for detecting logical violations of conditional rules. Moreover, they claim that this is the case even when these rules deal with familiar content drawn from everyday life. However, they argue that the human mind has evolved to detect violations of conditional rules, when these violations involve cheating on a social exchange. This is a situation where a person is entitled to some kind of reward only if they have fulfilled a particular requirement (e.g., "If a student eats cookies, then their room is tidy"). Cheating involves taking the benefit, without fulfilling the condition for the benefit. It is Cosmides's and Tooby's finding that when the Wason selection task is constructed to reflect a cheating scenario, subjects perform considerably better than they do with the standard test. And moreover, they found additionally that this is not simply to do with the familiarity of cheating scenarios - subjects do better with an unfamiliar cheating scenario than they do with a familiar standard scenario.

The tasks that you have just completed have been designed to test these claims. You completed two sets of two tasks. These were randomnly selected from two sets of six tasks. One set of tasks was of the form of a standard Wason selection task. The other of the form of a Wason selection task with a cheating context. The order in which the two sets of tasks were presented to you was randomised and, so as to avoid order effects, only your answers to the first set of tasks have been added to the cummulative scores for the experiment.

If Leda Cosmides and John Tooby are right, and this particular experimental design is sound, then two things will follow. Firstly, the chances are that you will have found one set of tasks much easier than the other set. In fact, for the tasks with a cheating context, you might well have experienced what Cosmides and Tooby call a "pop out" effect. That is, the correct answer will have appeared to be immediately obvious to you (i.e., it wouldn't have required analysis). And secondly, this fact that one set of tasks, the cheating context tasks, is easier than the other set will be reflected in the overall scores for the two different kinds of tasks. As I write this, I don't know whether this has occurred (because it hasn't occurred yet!), but you can have a look at the tables below and see for yourself. However, if it hasn't occurred, and there are no significant differences between the scores, then the chances are that there is something wrong with this experiment! It is that well documented a phenomenon!
© Jeremy Stangroom 2000

Click for credits

Further reading

• Evolutionary Psychology: A Primer (Leda Cosmides and John Tooby)
• Lecture notes on Evolutionary Psychology
• Cheating and altruism: A methodological critique of Cosmides' procedures