Abstract
Richard Lamb
The purpose of this paper is to outline the creation of a computational model making use of an underlying processing element in the form of an artificial neural network (ANN). Within the study, the ANN models multiple conservation tasks as inputs from video game play during a high school science content learning game. This model is based upon the identification of cognitive attributes and integration of two advanced psychological and educational measurement theories. Using the approach of cognitive diagnostics, and item response theory (IRT) data was examined for computational suitability. Once initial task response patterns are identified via IRT; the patterns are developed and presented to an artificial neural network (ANN) as probabilistic test data. Using the ANN derived Student Task and Cognition Model (STAC-M); the study author simulated a cognitive retraining intervention using 100,000 modeled students in science classrooms. Results of the simulation suggest that it is possible to increase levels of student success as measured through positive changes in underlying cognitive measures using a targeted cognitive attribute approach. The use of computational modeling in this educational context provides a means to develop future science education research and is a means to test current educational theory.
PDFShare this article