CLEAR: Consciousness as Local Effective Access to Representations

This paper develops CLEAR (Consciousness as Local Effective Access to Representations), an architecture-based account of consciousness that originated in a practical problem of artificial-agent design: how can an internal process remain genuinely inaccessible to the agent’s own self-modeling? CLEAR identifies effective access — a process taking a representation as operative input — as the constitutive minimum of consciousness, prior to self-attribution, metaconsciousness, and report.To state that claim precisely, the paper develops a criterion that distinguishes representational systems from fixed causal pipelines; tests it against the cases of a thermostat and a Kalman filter; and explains entities, episodic memory, biography, and self as structures built from accessed representational constituents. It also separates the necessity of access from its sufficiency for consciousness. The former is defended architecturally; the latter remains an explicit open problem. Section 9 defines the theory’s current epistemic limits, while Appendix A provides a compact formal reference.