A new composite index that combines three properties of brain activity—scale-free temporal organization, cross-frequency organization, and metastable flexibility in large-scale synchronization—separates conscious from non-conscious synthetic brain states without overlap and distinguishes wakefulness from N2 and REM sleep in 30 healthy adults using two-channel EEG recordings. The index captures organized dynamical complexity rather than raw signal complexity alone and remains stable across sensitivity and Monte Carlo analyses. The framework is not tied to any single theory of consciousness and may be applicable to anesthesia, disorders of consciousness, and basic consciousness research.
A theory-neutral framework quantifies consciousness-related neural dynamics using three properties: hierarchical integration, cross-frequency complexity, and metastability. Implemented in a generative model of synthetic EEG, the composite index reliably separates high-consciousness states (psychedelic, wakeful) from impaired or non-conscious states (dreaming, non-REM sleep, minimally conscious, anaesthetised, seizure-like) across single trials and Monte Carlo ensembles. Validation with real EEG from the Sleep-EDF dataset alongside matched synthetic EEG shows that synthetic data recapitulate the empirical ordering and magnitude of the index across Wake, N2, and REM sleep, indicating the index captures stable and biologically meaningful distinctions. The approach offers a principled tool for quantifying consciousness-related neural organisation applicable to biological and artificial systems.