| Recently, a large variety of new instructional methods have come to complement or replace sometimes the traditional ways of teaching physics. Many of these methods bring along a very convincing array of evidence testifying for their improved performance in classrooms and outside. All this has been made possible by the efforts of the Physics Education Research community, and it is what one would expect as the first step for any developing field. However, we believe that the time has come for a shift of interest from "what might we do to improve our instruction?" to "why the methods that we already have seem to work well?" We believe that theoretical sophistication of the research methods could and will support experimental work. Formal theoretical models applied to educational systems could guide the way we build experiments and enlarge our understanding of the educational process. This thesis points out possible ways of constructing these models, their limitations as well as their power. To this avail, we build computational models that reproduce many of the students' mistakes, that one may use to test different learning strategies, and that simulate aspects of the students interaction within a study group, such as the relation between communication and group and individual performance. |
