The Modeling Instruction pedagogy for the teaching of physics has been proven to be quite effective at increasing the conceptual understanding and problem-solving abilities of students, as measured by the Force Concept Inventory and the Mechanics Baseline Test, to a much greater extent than that of non-modeling students. Little research has been conducted concerning the cognitive and metacognitive skills that modeling students develop that allow for these increases. In this thesis, two studies were designed to answer the following question: In what ways do the knowledge structures, metacognitive skills, and problem-solving abilities differ between modeling and non-modeling students?

In Study 1, the knowledge structures developed by both modeling and non-modeling students were determined using a card sort task. The type of knowledge structure developed by the students was quantified via expert, surface feature and "questions asked" scores. The student's knowledge structures were then correlated to the scores they obtained on two measures: FCI and a problem solving task (PS Task).

In Study 2, a verbal protocol design allowed for a detailed study of the problem-solving and metacognitive skills utilized by the two groups. It was determined that the skills developed in both of these areas by the modeling students were more expert-like as based upon prior research. In addition, the modeling students produced significantly fewer physics errors while catching and repairing a greater percentage of their errors.