The problem of how to describe the structure of a student's current conceptions in a given area is a fundamental one for researchers who seek to develop a theoretical foundation for cognitively oriented instruction. Recent developments in the physics teaching community, for example, have emphasized cognitively oriented approaches to teaching and the need for understanding the cognitive processes that underlie one's ability to "do" physics. Ideally, we would like to have a picture of the kinds of knowledge structures and reasoning processes that are present in beginning students and that are present in experts. Detailed descriptions of experts' knowledge structures, including those tacit knowledge structures not represented explicitly in the curriculum, would presumably help to define more clearly what is to be learned by the student. Detailed descriptions of beginning students' preconceptions and misconceptions would have value not only as a sophisticated evaluation tool, but would also make it more possible to take common preconceptions and misconceptions into account during instruction.
This paper attempts to show that it is possible to study systematically certain types of beginning students' conceptions in physics; specifically, causal conceptions in mechanics. The paper examines the conceptions a freshman student uses to understand a simple physical system involving the horizontal motion of a cart launched across a table. The task given to the student does not ask him to find a long series of actions which will solve a problem. Rather he is asked for a prediction and explanation of the effects on the system resulting from a single action. Protocols of such explanations are particularly interesting because they tend not to be limited to formal, deductive arguments, but to include informal arguments that reflect the structuring of the subject's physical intuitions.
The methodology used in this study involves two phases: obtaining problem- solving protocols via taped interviews, and analyzing these protocols to produce a model of the conceptions that underlie the student's responses in the interview. Several considerations are important to the success of this technique. An important consideration during the interview process is the attempt to encourage the student to express himself verbally as he thinks through a problem. The interviewer must also search for questions which match the level of the student's conceptions. An important consideration during the analysis phase is the attempt to model the student's conceptions at a level that is neither simpler nor more complex than the level reflected in the student's comments. In addition, the analyst must be ready to encounter conceptions that are qualitatively