A Multi-Scale Dynamical Framework for Conscious State Dynamics
We present a dynamical framework in which conscious experience is modeled as a trajectory through a nested attractor landscape. Rather than treating consciousness as a discrete output of neural processes, we describe it as a continuous path through a state space shaped by stable and metastable attractors, with individual, social and environmental dynamics coupled across scales. The framework’s estimation machinery — how a landscape is defined as an empirical object, reconstructed from data, and reported — is developed and validated separately (Devika, 2026a); this paper takes that protocol as given and asks what follows for a theory of conscious dynamics. Three things do. First, features of subjective experience map onto measurable landscape geometry, and the mapping is stated at the level of robustness the estimator actually supports rather than uniformly. Second, the coupling between individual and collective scales has a computable synchronization threshold: the mean-field model undergoes a Desai–Zwanzig transition at a critical coupling Kc(D) determined by the variance of the single-agent stationary density, so “collective attractor” is a claim with a number attached. Third, the framework yields three predictions that discriminate it from neighbouring accounts, and we withdraw two predictions of earlier drafts that do not. The sharpest test concerns the distribution of dwell times within a state, which no coordinate choice can adjust; published resting-alpha measurements contradict its unconditional form, and we state the conditional form that survives, together with the quantitative reason — smooth landscape drift saturates well short of the observed exponent — that the obvious rescue is insufficient.
Keywords — attractor dynamics, consciousness, metastability, nested systems, collective dynamics, phase transitions, dwell times, falsifiability