"The occurrence of both sound and unsound inferences in reasoning experiments and more generally in everyday human thinking can be explained by the roles played by these two kinds of processes." . . . .