Theory of learning mathematics with computer-assisted instruction: Hierarchical and nonhierarchical options

摘要: The learning of arithmetic problems is assumed to be a Markov process involving conditioning set k subskills, each consisting one or more productions. An axiom provided, with the choice between two options for controlling which models results. Model 1, assumes that every subskill attempted on occurrence given problem, nonhierarchical model. 2, in temporal sequence problem until failed, somewhat unconventional hierarchical model: It sense conditioningor chance success at level prerequisite performance next (next level) problem. As value guessing parameter, g, declines, 2 leads less efficient than 1.