In favor of the category of belief in cognitive science

Raffaele Giovanni Caravella

Università della Svizzera Italiana

The goal of this paper is to defend the use of the category of belief in cognitive science against eliminativist arguments, focusing on Jenson’s [Jenson, 2016] claim that belief should be rejected due to its fragility. A theoretical entity is considered fragile if the results of multiple, independent, putatively reliable measures of that entity turn out to radically vary and this variation can not be adequately explained away. According to Jenson, discrepancies between orally reported beliefs and behaviorally manifested attitudes indicate that the category of belief is not robust.
Accepting his characterization of folk psychology (as having a Loar’s functionalist conception of belief [Loar, 1981]) and his criteria for fragility, I argue that his conclusion can nevertheless be avoided by rejecting the unity of belief assumption.
Drawing on the fragmented view of belief storage proposed by Quilty-Dunn and Mandelbaum [Quilty-Dunn and Mandelbaum, 2018] and characterized following Bendana and Mandelbaum [Bendana and Mandelbaum, 2021], I show that beliefs can be understood as distributed across multiple fragments, each accessible under different conditions.
This framework successfully explains the empirical cases proposed by Jenson himself without rejecting a realist account of belief. The empirical cases are: the nylon stocking experiment [Nisbett and Wilson, 1977], reactions to Hitler’s sweater [Nemeroff and Rozin, 1994], and results from Implicit Association Tests.
I also compare the fragmented view with dual-process theory, since this is proposed by Jenson as a possible alternative to the concept of belief. I argue that fragmentation better accounts for phenomena such as the persistence of misinformation.
Thus, by modifying the traditional functionalist account of belief and adopting a fragmented model, we can preserve the category of belief within cognitive science while fully accommodating the empirical evidence that initially seemed to threaten it.

Chair: Luuk Brouns

Time: 04 September, 10:00 – 10:30

Location: SR 1.007