Recognition Dynamics in the Brain under the Free Energy Principle

TL;DR

This paper models perception in the brain using the free energy principle, deriving recognition dynamics as a Hamiltonian system and demonstrating its implementation at cellular and hierarchical levels.

Contribution

It introduces a novel Hamiltonian mechanics framework for brain recognition dynamics based on the free energy principle, linking Bayesian filtering with physical principles.

Findings

Recognition dynamics derived as Hamiltonian system

Implementation demonstrated at single-cell and hierarchical levels

Formal solutions provided for linear regime models

Abstract

We formulate the computational processes of perception in the framework of the principle of least action by postulating the theoretical action as a time integral of the free energy in the brain sciences. The free energy principle is accordingly rephrased as that for autopoietic grounds all viable organisms attempt to minimize the sensory uncertainty about the unpredictable environment over a temporal horizon. By varying the informational action, we derive the brain's recognition dynamics (RD) which conducts Bayesian filtering of the external causes from noisy sensory inputs. Consequently, we effectively cast the gradient-descent scheme of minimizing the free energy into Hamiltonian mechanics by addressing only positions and momenta of the organisms' representations of the causal environment. To manifest the utility of our theory, we show how the RD may be implemented in a neuronally based biophysical model at a single-cell level and subsequently in a coarse-grained, hierarchical architecture of the brain. We also present formal solutions to the RD for a model brain in linear regime and analyze the perceptual trajectories around attractors in neural state space.