This paper explores children's efforts to make sense of graphs by analyzing two students' use of a computer-based motion detector. The analysis focuses on the students' growing understanding of the motion detector which enables them to plan their movements in order to create graphs and interpret them in terms of kinesthetic actions. Students gradually began to talk about the tool in ways that did not distinguish between the symbol and the referent. They also changed how they used the motion detector. This paper contributes to a reconceptualization of the nature of symbolizing, the learning of graphing, and the links between children's and scientists' graphing. (Contains 52 references.) (MM)