ABSTRACT
Scientists use models as representations of systems or types of systems in order to gain knowledge about them: the representational relationship between features of the model and features of the system enables scientists to draw inferences concerning the system by reasoning with the model. Theoretical models, such as the simple pendulum, as well as data models, can be displayed in different formats: the equation of the simple pendulum can be written in Cartesian coordinates as well as in polar coordinates; temperature values in function of time can be displayed in a list as well as in a graph or a table. Lists and graphs can contain the same information though they convey it in different formats. This paper intends to show that the particular format in which a model is presented matters to the reasoning processes of its users. Moreover, it aims at clarifying the very notion of “format.”
