Extrapolation in Biomedical Research: A Multi-Model Perspective

Biomedical research is built upon inferences transposing knowledge across systems – be it across species, between an experimental system and another, or from controlled clinical studies to routine healthcare contexts. While the notion of extrapolation received increasing attention in the recent philosophical literature, extrapolative inferences having distinct epistemic aims are often conflated together, thus making it difficult to assess the validity of such inferences. In this paper, we begin by characterizing the general form of extrapolations, whose structure and components allow a careful and systematic dissection of distinct types of extrapolations, highlighting the specific aims, methods and challenges associated with each type of extrapolation. Finally, we show how some contemporary research practices can challenge the boundaries of this classification, pointing to the need to consider extrapolations in their general form.