ABSTRACT
Experimentation in child psychology seems to go through a cycle of methods whenever a new problem-domain is tackled. Typically, tests are given to age-matched groups of children, then, once the tools of the trade have been polished, a battery of tests is given to groups of children; and only then do researchers move on to longitudinal studies and to recombining parts of tests to see whether the standard indices of transfer and of dissociation of functions can be demonstrated within individuals. Such an agenda starts with a concentration on valid tests of whatever competence is under scrutiny and, hopefully, ends with a process model that is truly developmental. The bulk of the evidence on the child’s theory of mind is still at the first stages, working out agreement over tests and the reasons for an association between tests in a battery: the lack of a process model has succinctly been noted by Perner (1991) and Wellman (1990).
