A computational perspective on faith: religious reasoning and Bayesian decision

ABSTRACT

Religious reasoning (the processes through which religious beliefs are formed) has been investigated by two different approaches. First, explanation theories portray religious reasoning as arising for explaining salient aspects of reality. Second, motivation theories interpret religious reasoning as driven by other motives, for example fostering community bonding. Both approaches have provided fundamental insight, yet whether they can be reconciled remains unclear. To address this, I propose a unifying computational theory of religious reasoning expressed in mathematical terms. Although a mathematical approach has been rarely applied to study religion, its advantage is describing a phenomenon clearly and formally. Relying on a Bayesian decision framework, the model comprises three key elements: prior beliefs, novel evidence, and utility. The first two describe the processes classically interpreted by explanation theories, while utility captures phenomena highlighted by motivation theories. By reconciling explanation and motivation theories, this proposal offers a unifying computational picture of religious reasoning.

KEYWORDS:

• Bayesian
• computational modeling
• motivation
• religion
• decision theory

Disclosure statement

No potential conflict of interest was reported by the author(s).