Abstract
Throughout its forty year history, the conceptual change literature assumed student misconceptions of force are formed prior to instruction. We argue that it may well be the case that misconceptions are not formed until the student considers force and motion in a scientific context for the first time. This has obvious implications for research methods. We are in the early stages of developing a research method for investigating conceptual change in mechanics. To illustrate this method, we have taken examples from one-to-one Socratic tutoring. We conclude by outlining the next step of the research, which is to build a model that will enable the Socratic method to reveal the characteristics of misconceptions.