This paper presents an original, information-theoretic model of consciousness based on belief alignment with objective descriptions. Consciousness is formally defined as the ratio between the complexity of true beliefs held by an observer and the complexity of the full inherent description of an object. The model distinguishes between three epistemic states: Consciousness (true beliefs), Schizo-Consciousness (false or misaligned beliefs), and Unconsciousness (absence of belief). Descriptions are represented as sets of statements in the form O(x)-Q(y), where objects and their qualities are mapped onto a structured semantic space.

The paper introduces both scalar and vectorial representations of belief states, allowing comparisons between observers in terms of both the amount and content of their conscious knowledge. Complexity can be quantified using bit-length encoding, programmatic code length, or entropy-based measures such as Kolmogorov or Shannon entropy. The model also incorporates a dynamic component: a belief-updating mechanism driven by external stimuli, analogous to neural or cognitive processes.

This formalism offers potential applications in comparative epistemology, cognitive modeling, and AI alignment by enabling simulations of belief evolution and mapping of conscious, unconscious, and misbelief states. It aims to contribute a rigorous framework for understanding consciousness as a structured, quantifiable process rather than a purely subjective phenomenon.