Realism and rationality in the selection task
Peter C. Wason · 2013
I t is not ju s t that it is difficult - a five-move chess problem can be very difficult for even a com petent chess problemist. And it is not ju s t that it provokes irrationality - novel ideas in the arts (surrealism) and in the sciences (psychoanalysis) have in the past m ade rational individuals behave in an irrational way. By any standards, the selection task ought not to be difficult, and it certainly ought not to ensnare people in a delusion of their own making. For it possesses a simple logical structure which can readily be grasped if you don’t try to solve it, and its content would hardly be expected to pose a threat to conventional wisdom. So it is something of a paradox that w hat looks like a simple formal problem should not only prove difficult, bu t should often arouse a rigid pattern of thought which is typically resistant to correction by the experim en ter’s intervention. Some people (not so m uch the subjects) seem to be worried about this. O n the other hand, certain modifications may make the problem smoothly easy (almost a different problem ), and hence can be said to elicit reasonable behaviour. In principle, this provides an opportunity to dem arcate the conditions for rational thought, not of course that we have achieved this on any grand scale, but a m ap of the territory begins to become visible. T h a t is the crux of the problem, and perhaps it is appropriate in this reappraisal to say first a few words about its origin.