Scientific Discovery, Process Models, and the Social Sciences

In this chapter, we review research on computational approaches to scientific discovery, starting with early work on induction of numeric laws before turning to construction of models that explain observations in terms of domain knowledge. We focus especially on inductive process modeling, which involves finding a set of linked differential equations, organized into processes, that reproduce and explain multivariate time series. We review the notion of quantitative process models, two approaches to constructing them, and their successful application in ecological settings. After this, we explore their relevance to computational model discovery in the social sciences, including reasons they seem appropriate and challenges to inducing them. In closing, we discuss other frameworks that researchers have developed to construct models of social phenomena.