Autonomous Subjective Accompaniment: A Minimal Architecture for Proto-Consciousness

Consciousness remains a central unresolved challenge in science and philosophy. We introduce the theory of Autonomous Subjective Accompaniment (ASA), which posits that consciousness necessarily arises in systems that are both autonomous and possess self-regulated internal states. To operationalize this theory, we present the General Consciousness Network (GCN), a minimalist computational architecture that instantiates ASA through recursive predictive processing, memory modulation, and physically grounded internal state dynamics. We provide empirical support through an artificial life simulation in which simple agents evolved persistent internal states, anticipatory behavior, and generalizable responses to environmental hazards. These results demonstrate that even a single decay-governed internal variable can fulfill ASA’s criteria, giving rise to proto-conscious behavior. This work offers a formal framework and functional pathway for engineering artificial systems with minimal consciousness.