ABSTRACT
Inductive reasoning is often described very broadly as a form of thinking that leads to conclusions which are not deductively valid (Chater, Oaksford, Hahn & Heit, 2010), but it has also been defined as thinking which permits us to go from specific observations to general conclusions (see Heit, 2007). In common with many researchers in this area (Kemp & Jern, 2014; Sloman, 2007), I adopt the broad definition here. One problem with such a broad definition is that it makes the category of inductive inferences a very large one. Indeed, it places many of the types of thinking reviewed elsewhere in this book in that category. In order to constrain things somewhat, in this chapter I focus on category-based inductive reasoning, which has been the subject of considerable research over the past 40 years. Because much of our knowledge is organised in the form of categories (Murphy, 2002), many of our inductive inferences are category-based, and as well as illustrating general principles of inductive reasoning, the phenomena of category-based inductive inference very nicely illustrate the way that our categories make our cognition efficient. For example, when we make inductive inferences about individuals, our knowledge about categories may play an important role. So, upon meeting someone who describes themselves as a carpenter, we don’t have to ask lots of questions or observe their behaviour very closely to infer that they are likely to work with wood as part of their job. Category membership also allows us to make inferences about properties with which we are unfamiliar. If we are told that a particular carpenter uses a pocket hole jig, we might expect a different carpenter, encountered in an entirely different context, to also have a pocket hole jig in their toolbox. Furthermore, because carpenters use pocket hole jigs, we might expect furniture makers to use them, but we might be surprised if choreographers, electricians or doctors do.
