COGNITIVE PSYCHOLOGY

Cognitive Psychology

Cognitive Psychology

Question No. 1

It is a fact that young children fail in experimental tasks which are supposed to test their grasp on logic mathematical concepts such as classificatory logic. The traditional explanation for the failure (Inhelder & Piaget, 1964) has been that these children have not yet developed the intellectual structures essential for success. However, this whole approach has recently been challenged. Various studies have been set up to demonstrate that very young children can succeed in certain versions of Piagetian-type tasks. For example, 5 year olds are usually defeated by Piagetian class inclusion questions, yet they succeed when details of the procedure are changed (Donaldson, 1978; McGarrigle & Donaldson, 1974), or when part-whole relations are simplified (Markman & Siebert, 1976). Furthermore, while young children do not construct classifications when asked to sort typical experimental objects, again, it has been recently shown that they are as likely as older children to construct classifications when modifications are made to the object collection (Rosch et al., 1976). In other words, simplifying the procedure, objects or relations involved in a task seems to allow the small child to demonstrate greater competence than hitherto suspected (Donaldson, 1978; Fodor, 1972; Gelman, 1978; Johnson-Laird & Wason, 1977).

However, the emphasis on early competence is as open to challenge as the traditional approach. A major objection can be put forward: the argument for early competence has taken for granted that equivalent task performance is evidence for equivalent underlying processes. In other words, when a 5 year old succeeds in a modified version of a task, it is maintained a priori that she does so by the same process as used by the older child in succeeding in the traditional version of the task. However, (1) this inference is not a necessary one, (2) it is questioned by the fragile nature of younger children's success even when tasks are altered, and (3) modified tasks are, in any case, not strictly equivalent to their traditional versions (Dean, Chabaud, & Bridges, 1981). Thus, in stressing the similarity between older and younger children's competence, it is my view that the recent approach has underestimated developmental changes in processing.

However, phenomena exist which question this conclusion: first, the observation that 5 and 10 year olds produce ostensibly identical sortings only holds true for some collections of objects and not others (Rosch et al., 1976). Second, it is only the case where the child's initial efforts to sort the objects meet with no difficulties. When a first attempt to sort the objects encounters a problem (such as inability to sort the as yet unsorted objects into classes complementary to those already completed, i.e. using the same dimension of variance as that used in forming the earlier classes), the 10 year old aborts her initial sorting and starts afresh, producing a new system of complementary classes. By contrast, the 5 year old does not throw away her initial efforts: she retains them, and abandons similarity as a criterion for grouping the ...