Details of Research Outputs

In a recent study (Pascual-Leone, Baillargeon, Lee, &Ho, 1991), the growth of performance scores in four M-tasks of Pascual-Leone showed only uncertain traces of steps with age corresponding to the theory-predicted stages. The present study is an attempt to investigate more deeply the same data in a theory-guided manner, by selecting a core-sample of subjects from the original grand sample, hoping to facilitate emergence of the predicted developmental growth of M-task performance level in the manner of a stage-bound step function. Other positive contributions of this thesis, would be twofold: (1) Introduce a methodology of Categorical Data Analysis (or Contingency Table M-Analysis (CTM analysis)), never before used in Developmental analysis, that can unambiguously decide whether step functions are found in children's developmental performance, in any task. (2) Evaluate, using the same analytical tools, which are the relations of relative difficulty that exist among the tasks, and between each of the tasks and the theoretically predicted developmental M-growth step function.

The CTM analysis results obtained allow us to claim that the data are not inconsistent with the step function M prediction. Moreover, when the contingency tables are examined in their inter-relations in search of transitivity of equivalence or implication transitivity holds over the whole set of tasks. The fact that transitivity holds in these relations across tests suggests that all the tests share the semantic features (sources of variance) over which the relations are tacitly defined. This finding is consistent with the theory because all these tests are designed to be M-capacity estimators.