Detecting the Earliest Recursive Regime Transition in Neural Systems

Abstract

Current empirical approaches to consciousness rely on measures of integration, responsiveness, and reportability, all of which track relatively late-stage phenomena. This paper introduces a minimal operational criterion for detecting an earlier event: the first sustained instance of recursive closure in neural dynamics. We argue that recursion functions as a constraint-satisfying mechanism whose irreducible role is identity maintenance — the preservation of a system’s configuration through time via self-referential closure. The proposed “spark” is defined as the earliest time interval in which three conditions co-occur: re-entry (state dependence on immediate past), persistence (non-fragmenting continuity), and configuration-relative comparison (path-dependent evaluation of current state against prior internal configuration). Recursive closure is treated as a binary event that may occur transiently before stabilizing into a regime. Two onset markers are distinguished: T₀ (first closure instance, possibly brief) and T* (first sustained closure exceeding stability threshold τ). Crucially, τ is not a free parameter but is anchored to intrinsic system timescales. The resulting onset timestamps are evaluated using lead–lag tests against established markers such as large-scale integration and behavioral responsiveness. The framework includes a pre-registered analysis protocol specifying all parameter choices, proxy thresholds, and decision rules in advance, together with a practical application guide addressing use cases from clinical anesthesiology to computational model validation. This framework is explicitly agnostic with respect to phenomenological interpretation and provides a constrained, falsifiable method for identifying the earliest detectable recursive regime in neural systems.