Recursive Self-Modeling Cosmology: A Research Framework for Cosmological Reproduction, Inheritance, and Predictive Self-Modeling

This preprint introduces Recursive Self-Modeling Cosmology (RSMC), a speculative research framework that explores whether cosmological reproduction, inheritance, selection, and predictive self-modeling can be described within a single recursive physical architecture. RSMC connects several existing lines of research—including cosmological natural selection, black-hole/baby-universe scenarios, quantum-gravitational bounce models, relational and emergent time, predictive information, Markov-blanket formalisms, and learning-like dynamics in fundamental physics—while explicitly distinguishing the proposed synthesis from these antecedents. The framework investigates four central questions: whether black-hole-mediated cosmological reproduction can generate an evolving population of descendant universes; whether effective physical parameters can exhibit statistically defined inheritance without violating quantum unitarity or the no-cloning principle; whether predictive self-modeling can emerge as a physically identifiable process rather than merely as correlation or passive information storage; and whether population-level cosmological selection can be empirically distinguished from anthropic selection and competing reproduction mechanisms such as eternal inflation. The paper develops a minimal mathematical structure for inheritance kernels, branching dynamics, predictive self-models, and information-accounting constraints, while emphasizing that RSMC is currently a research hypothesis rather than an established physical theory. Its purpose is to identify a falsifiable research program and clarify the theoretical consistency conditions that any recursive, self-modeling cosmology would need to satisfy. Author: Barna Horváth, Independent ResearcherVersion: 1.0DOI: 10.5281/zenodo.21985764License: CC BY 4.0