Krupnik, Valery (2025) Bayes Meets Hegel: The Dialectics of Belief Space and the Active Inference of Suffering. [Preprint]
![[img]](https://philsci-archive.pitt.edu/style/images/fileicons/text.png) |
Text
Bayes Meets Hegel_final_complete with figs.docx - Accepted Version Download (546kB) |
Abstract
The Bayesian brain hypothesis conceives of the brain as a generative model (GM) of its environment, where the model improves its accuracy by updating itself via Bayesian inferential statistics. This is accomplished by adjusting the model’s prior beliefs into posterior beliefs based on the sensory input from the environment. Thus, the Bayesian brain learns the causal structure of world events and refines its belief space. The Bayesian brain is thought to learn by updating its beliefs and the parameters of its GM. It is less clear whether Bayesian updating alone is sufficient to explain the genesis and evolution of belief space, as it appears challenging to explain the generation of truly novel original priors (structure learning) or the transition to the direct opposite of the initial prior through Bayesian updating. To address this challenge, we suggest integrating Bayesian updating with the principles of the dialectical development of thought as conceived in Hegel’s work. We argue that applying dialectics and the free-energy principle to Bayesian inference makes the evolution of belief space both its emergent and imperative property. We naturalize this idea by illustrating how behaviors of different complexity from the molecular mechanisms of the simplest biological behavior, such as bacterial chemotaxis, through psychopathology can be viewed in the ‘dialectical Bayes’ framework. This framework is used to explore the cognitive dynamics of physical and emotional pain and to propose a mechanism of chronic suffering.
| Export/Citation: | EndNote | BibTeX | Dublin Core | ASCII/Text Citation (Chicago) | HTML Citation | OpenURL |
| Social Networking: |
Monthly Views for the past 3 years
Monthly Downloads for the past 3 years
Plum Analytics
Actions (login required)
 |
View Item |
![[feed]](http://philsci-archive.pitt.edu/style/images/feed-icon-32x32.png){kind=icon}