Work

Robustness in the Life Sciences: Issues in Modeling and Explanation

Year: 2020
Type: article
Abstract: My dissertation introduces two new accounts of how robustness can be used to identify epistemically trustworthy claims. Through an analysis of research practices in the life sciences, I focus on two main senses of robustness: robust reasoning in knowledge generating inferences and explanatory strategies for phenomena that are themselves robust. First, I provide a new account of robustness analysis (called ‘scope robustness analysis’), in which researchers use empirical knowledge to constrain their search for possible models of the system. Scope robustness analysis is useful for scientific discovery and pursuit whereas current accounts of robustness analysis are useful for confirmation. Second, I provide a new account of how researchers use different methods to produce the same result (a research strategy called ‘triangulation’). My account makes two contributions: I criticize a prominent account of the diversity criterion for methods because it analyzes an inferential strategy (i.e., eliminative inference) distinct from the inferential strategy underlying triangulation (i.e., common cause inductive inferences). My account also better explains how triangulation can fail in practice by assessing points of epistemic risk, which I demonstrate by applying it to implicit attitude research. Finally, I contribute to a debate about another sense of robustness: phenomena that occur regardless of changes in their component parts and activities. I argue that some robust phenomena in network neuroscience are not best explained mechanistically by citing their constituent parts (e.g. individual neurons) and their activities, but rather by appealing to features of the connectivity among brain areas.
Language: English
Open Access status: green
OpenAlex ID: W6996203318
FWCI:
Cites: 164
Cited by:
Related to: 10