ABSTRACT
This chapter presents an application of Markov models in a learning experiment in which children of different ages and adults are compared as to the learning strategies they bring to bear on the task. It describes the conceptual background of discrete state models and the transition dynamics that are used to model processes of change. The chapter also presents a formal treatment of the Markov model and extensions. It explores the basics of computing likelihoods for such models and how to estimate and optimize parameters of Markov models. The chapter provides a discrimination learning experiment that is used to illustrate a number of possibilities in applying Markov models to longitudinal data. It discusses the results, possible extensions, and suggestions for further research. To further characterize the component with slow learners, a model with covariates was used to test the hypothesis that the slow learners may be learning incrementally.
