Analogy as categorization: A support for model-based reasoning

Abstract

Generally speaking, model-based reasoning refers to every reasoning that involves model of reality or physical world, and it is especially involved in scientific discovery. Analogy is a cognitive process involved in scientific discovery as well as in everyday thinking. I suggest to consider analogy as a type of model-based reasoning and in relation with models. Analogy requires models in order to connect a source situation and a target situation. A model in an analogy is required to establish salient properties and, mostly, relations that allow transfer of knowledge from the source domain to the target domain. In another sense, analogy is the model itself, or better, analogy provides the elements of model of reality that enable the processes of scientific discovery or knowledge increase. My suggestion is that some insight on how an analogy is a model and is connected to model-based reasoning is provided by recently proposed theories about analogy as a categorization phenomenon. Seeing analogy as a categorization phenomenon is a fruitful attempt to solve the problem of feature relevance in analogies, especially in the case of conceptual innovation and knowledge increase in scientific domain.