LOGIC AND COGNITION: Human Performance in Default Reasoning

There has long been a history of studies investigating how people (" ordinary people ") perform on tasks that involve deductive reasoning. The upshot of these studies is that people characteristically perform some deductive tasks well but others badly. For instance, studies show that people will typically perform MP (" modus ponens " : from 'If A then B' and 'A', infer 'B') and bi-conditional MP (from: 'A if and only if B' and 'A', infer 'B') correctly when invited to make the inference and additionally can discover of their own accord when such inferences are appropriate. On the other hand, the same studies show that people typically perform MT (" modus tollens " : from 'If A then B' and 'not-B', infer 'not-A') and biconditional MT badly. They not only do not recognize when it is appropriate to draw such inferences, but also they will balk at doing them even when they are told that they can make it. Related to these shortcomings seems to be the inability of people to understand that contrapositives are equivalent (that 'If A then B' is equivalent to 'If not-B then not-A'). [Studies of people's deductive inference-drawing abilities have a long history, involving many studies in the early 20 th century concerning Aristotelian syllogisms. But the current spate of studies draws much of its impetus from Wason (Wason, 1968; see also Wason & Johnson-Laird, 1972). ] The general conclusion seems to be that there are specific areas where " ordinary people " do not perform very logically. This conclusion will not come as a surprise to teachers of elementary logic, who have long thought that the majority of " ordinary people " are inherently illogical and need deep and forceful schooling in order to overcome this flaw. There has been a similar history–although somewhat shorter–of studies investigating how people perform on tasks that involve probabilistic reasoning. The upshot of these studies is that people characteristically make a number of errors because they do not pay attention to certain relevant information and because they give undue importance to certain kinds of irrelevant information. The study of the types of errors people make on this probabilistic reasoning has come to be called " heuristics and biases " because the underlying assumption is that people have inbuilt or inherent biases as part of their mental equipment that causes them to ignore or give undue status to …