Biomemetic Consciousness
A unified theory of consciousness — and a computational engine that applies it to drug discovery, financial markets, and beyond.
What the Engine Does
One engine. Multiple industries. Zero training data.
Why should I care?
The gap is real.
We have 5+ competing theories of consciousness — none explains more than a fragment. IIT handles complexity but not emotion. GNW handles attention but not dreams. Predictive Processing handles perception but not identity.
BMC unifies them.
All five theories emerge as special cases of a single framework: replicator competition on a physical substrate. One model, many perspectives — formally proven, not just claimed.
It's testable.
BMC catalogues 149 predictions across three epistemic tiers, and says plainly which tier carries weight: 40 are prospective, and 11 of those are committed as separating tests against a named rival. Many can be checked against existing data — and we publish the screen that says which of them are not yet real tests.
Dual-Replicator Model
Genes (G) and memes (M) compete for the same neural substrate. Consciousness emerges from their interaction — not as an epiphenomenon, but as a functional necessity of the conflict.
11 Separating Tests
Of 149 catalogued predictions, 40 are prospective and 11 commit against a named rival making the opposite prediction. Commitments, not results: 8 of the 11 still need a numeric threshold. One has been run outside the engine; two are in-engine implementation checks.
Subsumes 5 Rival Theories
IIT, GNW, HOT, AST, and Predictive Processing emerge as special cases within the BMC framework. One theory, multiple perspectives unified.
Computational Engine
A Rust engine with 103 gate checks and 300+ tests. Emergent language (97.5% Lewis signaling, 533 concepts), reproduced WM capacity (Cowan's 4±1), and a survival-neutrality result for language that holds in the tested regime only — all from first principles.
Theory at a Glance
BMC = (G, M, I, S) — four components, any scale