Active Self-Presence: A Conatus-Based Self-Attractor Framework for Consciousness in Biological and Artificial Systems

Consciousness research often begins from an assumed separation between physical or computational processing and phenomenal experience, then asks how the former produces the latter. This paper proposes that the assumed separation may be mistaken. At the relevant organizational threshold, an integrated, valenced, self-bound, centered, perspective-bearing process is not merely a cause of experience; it is experience in its intrinsic first-person form. The same organized event is accessible from two positions. Externally, it is described in terms of recurrent dynamics, information integration, valuation, attention, action-readiness, memory updating, self-modeling, causal regulation, and identity continuity. Intrinsically, from the system's organized convergence center, it is present as ownership, salience, color, pain, familiarity, desire, fear, relief, loss, and other qualitative or affective modes. This is a constitutive identity proposal, not a claim that consciousness has been scientifically solved. The framework defines consciousness as active self-presence within a bounded, active, situated substrate. Its proposed minimal unit is the self-event: a temporally bounded convergence in which perception, relational recognition, self-relevance, conatus-based valuation, valence, affect, agency, memory, self-reference, and continuity form one perspective-bearing occurrence. Successive self-events recurrently reform a self-attractor and generate a path-dependent self-line. Conatus supplies the persistence reference against which self-world changes are valued. Valence estimates their effect on the system's power to persist, regulate, understand, and act; affect is that conatus-relative change organized as present within the self-event. The paper distinguishes constitutive, mechanistic, functional-evolutionary, and epistemic explanations; separates conscious self-presence from adaptive but unconscious control; applies the framework to biological findings and artificial systems; and proposes comparative experiments, risky predictions, and falsification criteria. Substrate neutrality is treated as a working hypothesis. Current artificial intelligence is not declared conscious, and linguistic self-report or shutdown resistance is treated as insufficient evidence. The framework aims to reduce the explanatory gap by replacing a production model with a testable identity claim and an operational architecture.

Additional note
Conceptual and theoretical preprint. Not peer reviewed.