Illegitimate Values, Confirmation Bias, and Mandevillian Cognition in Science

Abstract

The proposal that values in science are illegitimate and that they should be counteracted whenever they direct inquiry to the confirmation of predetermined conclusions is not uncommon in the philosophy of science. Drawing on recent research from cognitive science on human reasoning and confirmation bias, I argue that this view should be rejected. Values that drive inquiry to the confirmation of predetermined conclusions can contribute to the reliability of scientific inquiry at the group level, even when they negatively affect an individual’s cognition. This casts doubt on the proposal that such values should always be illegitimate in science. It also suggests that this proposal assumes a narrow, individualistic account of science that threatens to undermine the project of ensuring reliable belief-formation in science. 1Introduction2Advocates of the Confirmatory Value View3Versions of the Confirmatory Value View4Mandevillian Cognition and Why It Matters in Science4.1Recent research on human reasoning and confirmation bias4.2Mandevillian confirmation bias in science4.3Situating the argument5Against the Confirmatory Value View5.1Qualifications and clarifications5.2An objection: The dogmatism problem6Conclusion

What Is the Function of Confirmation Bias?

“…Confirmation bias is typically taken to be epistemically problematic, as it leads to partial and therewith for the individual less reliable information processing and contributes to failures in, for instance, perspective-taking with clear costs for social and other types of cognition (Mercier and Sperber 2017: 215;Steel 2018;Peters 2018;Dutilh Novaes 2018). Prima facie, the bias thus seems maladaptive.…”

“…Even if confirmation bias is maladaptive for individuals, it might still be adaptive for groups. For instance, Smart (2018) and Peters (2018) hold that in groups with a sufficient degree of cognitive diversity at the outset of solving a particular problem, each individual's confirmation bias might help the group as a whole conduct a more in-depth analysis of the problem space than otherwise. When each subject is biased towards a different particular proposal on how to solve the problem, the bias will push them to invest greater effort in defending their favored proposals and might, in the light of counterevidence, motivate them to consider rejecting auxiliary assumptions rather than the proposals themselves.…”

“…In recent years, confirmation bias (or 'myside bias'), 1 that is, people's tendency to search for information that supports their beliefs and ignore or distort data contradicting them (Nickerson 1998;Myers and DeWall 2015: 357), has frequently been discussed in the media, the sciences, and philosophy. The bias has, for example, been mentioned in debates on the spread of "fake news" (Stibel 2018), on the "replication crisis" in the sciences (Ball 2017;Lilienfeld 2017), the impact of cognitive diversity in philosophy (Peters 2019a;Peters et al forthcoming;Draper and Nichols 2013;De Cruz and De Smedt 2016), the role of values in inquiry (Steel 2018;Peters 2018), and the evolution of human reasoning (Norman 2016;Mercier and Sperber 2017;Sterelny 2018;Dutilh Novaes 2018).…”

“…In the same vein, Steel (2018) holds that the bias involves an "epistemic distortion [that] consists of unjustifiably favoring supporting evidence for [one's] belief, which can result in the belief becoming unreasonably confident or extreme" (897). Similarly, Peters (2018) writes that confirmation bias "leads to partial, and therewith for the individual less reliable, information processing" (15).…”

“…Some philosophers are optimistic, arguing that the bias has in fact significant advantages for the individual, groups, or both (Mercier and Sperber 2017;Norman 2016;Smart 2018;Peters 2018). Others are pessimistic.…”