The World Engine: A New Computational Framework for Predictive Discovery and Latent Behavioral Dynamics

The World Engine proposes a computational framework for discovery systems that model evolving behavioral states rather than relying solely on interaction history, language, and fixed user personas. Drawing on research across artificial intelligence, neuroscience, decision science, and recommender systems, the paper introduces Signal Intelligence, Archetypes, Digital Minds, identities, and World State as testable components of a predictive discovery architecture. The framework is presented as a scientific hypothesis, a working synthetic-world prototype, and an experimental program. Its central claims remain subject to simulation, controlled evaluation, and prospective testing with real users. The proposed evaluation compares the World Engine with random delivery and conventional history, language, and persona-based recommendation systems. Project and research page: https://www.voopacorp.com/research