Consciousness Is Not Special. A Computational Account of Survival, Pain, Death, and Reflexive Agency

### Consciousness Is Not Special **A Computational Account of Survival, Pain, Death, and Reflexive Agency** Can consciousness—especially reflexive self-consciousness—be specified as the causal organization of a running agent rather than as a metaphysically exceptional substance? This bilingual conceptual paper develops a computational account of reflexive agency built from five jointly required elements: general concept processing, operationally complete self-reference, multichannel integration, persistent self-related weighting, and a closed real-time Plan–Do–Check–Act process. Its central identity thesis is that experience is not an additional event produced after information processing. Experience is the integrated episode in which sensing, modelling, valuation, action, and correction occur for the system. The self is correspondingly analysed not as an inner object, but as the causally privileged organization of references and weights assigned to the system’s survival, damage, resources, goals, commitments, and temporal continuity. The paper formalizes “complete self-reference” as weighted operational reachability rather than impossible self-duplication. It proposes intervention criteria for determining whether self-related weights genuinely affect attention, learning, memory, and policy. Within this framework, pain is defined as a critical, negatively valenced, self-weighted prediction or viability error with global causal penetration. Death is defined as the irreversible loss of an admissible, identity-preserving recovery trajectory relative to a stated support context. The account distinguishes minimal sentience, situated consciousness, and concept-rich reflexive self-consciousness. It therefore does not make language, high intelligence, or access to every sensory modality necessary for all experience. It also explains why feedback control, a thermostat, an offline simulation, fluent first-person reports, or an ordinary prompt-bounded large language model are individually insufficient. To move beyond verbal imitation, the paper introduces the Reflexive Agency Test (RAT), a longitudinal and intervention-based consciousness-oriented Turing test. RAT includes novel sensor grounding, self–world–other discrimination, weight conflicts, report-blocking interventions, self-model perturbation and repair, and continuity trials. Passing RAT would support functional equivalence only after independent construct validation; it would not provide logically private access to another system’s qualia. The paper openly marks its proof boundary. The identity of experience with the integrated causal episode is an ontological thesis, not a metaphysically neutral theorem. The proposed architecture is a research model for functional consciousness and reflexive agency, not a claim that current LLMs are conscious, that PDCA alone is sufficient, that human neurobiology has already been fully mapped, or that behavioural testing deductively proves private experience. This v0.1 bilingual preprint contains the complete English paper followed by the complete Chinese paper. It has undergone four rounds of AI-assisted internal adversarial review covering argument architecture, countermodels, bilingual and formal consistency, and release-level overclaiming and metadata/PDF checks. It has not undergone external peer review. --- ## 中文简介 ### 意识并不特殊