A critical problem with the examination of learning in education is that there is an underlying assumption that the dynamic systems associated with student information processing can be measured using static linear assessments. This static linear approach does not provide sufficient ability to characterize learning. Much of the modern research within fields adjacent to education has embraced the idea that information processing and cognitive states are dramatically dynamic and nonlinear. The dynamic and nonlinear aspects of information processing makes current educational research methods ill equipped to examine cognitive processing. This is because many quantitative and qualitative methods of examination tacitly assume that "snap-shots" provide an appropriate view of student information processing during learning. This study examines the role of nonlinear dynamics as a descriptor of student learning in the science classroom. The purpose of this study is to establish explanatory mechanisms for understanding the variable outputs seen within student cognitive processing of science-based tasks. Initial parameterization of the subject response to the task sets presented in a serious educational game occurred using the DINA-N model. Drawing upon the tools from other disciplines such as computational modeling and artificial neural networks, it is possible for science education researchers to understand the interplay of multiple processing units known as a cognitive attributes. At this point, a limitation of this approach in education is the accuracy and lack of the model's ability to represent the fractal dimensionality, emergent properties associated with cognition, and aspects of multiple interaction systems of the brain.