- Subjects: Biophysics; Thermodynamics
- |
- Contributors:
- Luca Salasnich ,
- Luca Salasnich ,
- Dario Lucente ,
- Giacomo Gradenigo
- entropy production
- neural dynamics
- stochastic processes
This video is adapted from 10.3390/e27111104
Brain dynamics, when viewed through statistical mechanics, can be understood as inherently out of equilibrium. In non-equilibrium statistical mechanics and stochastic processes, entropy production along trajectories serves as the key observable for distinguishing equilibrium from non-equilibrium behavior. This work analyzes entropy production in the stochastic Amari model, a coarse-grained integro-differential description of neural activity. Because stochasticity can be introduced in multiple non-unique ways in coarse-grained systems, the role of different noise structures is examined in relation to the model’s intrinsic dynamics. Conditions are identified under which the stationary state corresponds to thermal equilibrium or, alternatively, to a genuinely non-equilibrium regime, with explicit and simplified expressions provided. The derivation further shows that the entropy production rate can be expressed in terms of the time variation of the system’s Shannon entropy.