Leggett--Garg Tests in Neural Dynamics: Probing Non-Diffusive Stochastic Structure in Single Neurons

TL;DR

This paper proposes testing Leggett--Garg inequalities in single-neuron dynamics to distinguish diffusive from non-diffusive stochastic models, revealing persistent temporal correlations without implying quantum coherence.

Contribution

It introduces an experimental framework to identify non-diffusive stochastic behavior in neurons via Leggett--Garg tests, linking neural dynamics to quantum-like temporal correlations.

Findings

Diffusive models satisfy Leggett--Garg inequalities.

Persistent stochastic models can violate these inequalities.

Violations indicate non-Markovian, memory, and contextual temporal structures.

Abstract

We propose an experimental programme to test Leggett--Garg-type temporal correlations in single-neuron dynamics. The goal is to distinguish between diffusive (Wiener/cable-equation) models and non-diffusive persistent stochastic models based on Kac-type finite-velocity processes leading to the Telegrapher's equation. We show that while purely diffusive dynamics satisfies Leggett--Garg inequalities, persistent stochastic dynamics can produce oscillatory temporal correlations capable of violating these inequalities. The Leggett--Garg inequality may be viewed as a temporal analogue of Bell-type constraints. In the present context, however, violation is interpreted conservatively not as evidence of microscopic quantum coherence, but as evidence against a simple trajectory-based diffusive description. The resulting temporal correlations indicate persistence, memory, and contextual temporal structure mathematically analogous to that encountered in quantum systems. Using the analytic continuation connecting Kac processes to Dirac-like envelope equations, we argue that finite-velocity persistent stochastic transport provides a natural mechanism for such non-diffusive temporal correlations. These tests therefore offer a possible experimental probe of contextual and non-Markovian structure in neural dynamics without requiring claims of microscopic quantum coherence in the brain.