This paper reviews several attempts to examine the structure of a given subject mattter (content structure) and its representation in student's memory(cognitive structure) within a general framework of current models of semantic memory. A new method i...
This paper reviews several attempts to examine the structure of a given subject mattter (content structure) and its representation in student's memory(cognitive structure) within a general framework of current models of semantic memory. A new method is then presented for extracting representation of content structure and cognitive structure and for examining correspondence between them. The method is based on the target partition analysis which is an incorporation and extension of quadratic assignment paradigm, latent partition analysis, and hierachical clustering scheme.
It is shown that this method can be embedded within the general models of semantic memory, network models in particular, to produce quantitative and graphic representations of cognitive structures. These representations can be obtained for subject matters or for individuals and in some cases the nature and extent of individual differences in cognitive structure can be also characterized. The findings of related studies reported in the present, paper demonstrate validity of the method. There was a convergence between different measures of cognitive structure. A comparision of cognitive structure, cognitive abilities, and achievement data indicated that learning of structure may differ from learning mapped by traditional achivement tests. Cognitive structure corresponded more closely to content structure at the end of instruction than at beginning, suggesting that the measure may be sensitive to instructional treatment effect as well as individual variation.