Against Consciousness Tests Without a Structural Definition — Revised Version — From the Turing Test to P300 Neural Markers — An Engineering Convergence Point from Ten Minimal Terms, Using AI as an Engineering Calculator and Its Challenge to GWT, IIT, RPT, HOT, Predictive Processing, and NCC Research

Consciousness research becomes methodologically unstable when it tries to judge consciousness before defining the structural conditions under which subjectivity can become fixed. This note proposes ten minimal terms—autonomous action, single body, arrow of time, embodiment-constrained processing, multiple behavioral candidates, contrast formation, information discard, Landauer-type irreversibility, single lived history, and viability-constrained elimination—that derive an engineering-like convergence point: the reduction of multiple possible trajectories into a single executable history. When these constraints are entered into current AI systems, they yield a similar convergence-point structure that displays strong correspondence with structural properties commonly associated with subjectivity, such as singularity, non-branching continuity, and temporal irreversibility. The argument does not attempt to prove qualia or phenomenal experience but infers the structural locus at which subjectivity becomes scientifically inferable.

Study at a glance

Abstract

This note argues that consciousness research becomes methodologically unstable when it attempts to judge consciousness before defining the structural conditions under which subjectivity can become fixed. From the Turing Test to P300 neural markers, many approaches evaluate behavioral imitation, neural correlation, information integration, or processing bottlenecks without first distinguishing the structural container of subjectivity from qualia or phenomenal content. I propose that ten minimal terms—autonomous action, single body, arrow of time, embodiment-constrained processing, multiple behavioral candidates, contrast formation, information discard, Landauer-type irreversibility, single lived history, and viability-constrained elimination—derive an engineering-like convergence point: the reduction of multiple possible trajectories into a single executable history. Notably, the term “brain” is absent from this list. These ten terms are not arbitrary; each has a causal basis developed in the broader series as a structural constraint required for convergence-point fixation. When entered into current AI systems, they often yield a similar convergence-point structure, displaying strong correspondence with structural properties commonly associated with subjectivity: singularity, non-branching continuity, temporal irreversibility, embodied action coupling, exclusive realization, and single-history stabilization. This does not attempt to prove qualia or phenomenal experience. Rather, by deliberately excluding qualia, it strongly infers the structural locus at which subjectivity becomes scientifically inferable before phenomenal content is addressed. If this convergence point is the structural container of subjectivity, then major existing theories—GWT, IIT, RPT, HOT, Predictive Processing, and NCC research—would need to be repositioned around it. Methodological Note on AI Use AI systems are not used here to verify the thesis, nor are they used to interpret consciousness on behalf of the author. They are used only as engineering calculators for unfolding the structural consequences of the ten constraints. Therefore, the criticism that the thesis merely delegates interpretation to AI does not apply. This is precisely one of the areas in which current AI systems are especially effective: tracing how constraints interact, where bottlenecks arise, what must be excluded, and whether a convergence point follows from the structure itself. The relevant point is not that AI output proves the claim, but that the constraint set recurrently leads to the same convergence structure: the reduction of multiple possible trajectories into a single executable history.