A framework for organizing what a theory of consciousness should explain is proposed, based on seven theoretically neutral questions that are causally and functionally relevant and applicable across different systems. The framework emphasizes causation, arguing that causal relations cannot be fully captured by standard physical descriptions alone. An asymmetric causal structure is introduced to explicitly represent internal mechanisms and distinguish between variable- and structure-level causation. The framework is applied to analyze the Dual-Laws Model. Its aim is not to offer a definitive theory but to provide a common basis for analyzing and developing theories of consciousness.
The mind-brain problem asks how mental events relate to neural events. Mathematical models have tried to explain how the brain represents the discriminative structure of qualia (subjective experiences), but lack validation. In unsupervised learning, independence between axes in a latent space cannot distinguish between different qualia types (e.g., vision vs. touch) and different instances within the same type (e.g., green vs. red). The authors hypothesize that weakening inter-axis independence is necessary to discriminate qualia types. They formulate an algebraic independence linked to other-qualia-type invariant transformations, where the transformation value is a vector space. A brain model learning this algebraic independence separates the latent space into multiple metric spaces corresponding to qualia types, suggesting a contribution to the mathematical theory of consciousness.