The paper introduces NEST (Nested Episodic State Topology), a foundational graph-theoretic representational ontology designed to model cognition through structured state formation and transformation. It represents concepts, episodes, percepts, and task contexts as typed, weighted graphs with nodes carrying internal subgraph payloads and edges classified into six relation types.
- Durable belief graphs are separated from capacity-limited working-memory graphs that host transient content.
- WM-belief grounding, conflict catalogs, and belief-update operators define how transient structures interact with stored knowledge.
- A reusable operator toolkit includes activation, graph-property functionals, and awareness trajectory functionals.
- Derived diagnostics such as fragmentation, involvement, coherence, and active conflict define familiar cognitive phenomena within the ontology.
- Compatibility mappings embed ACT-R, Soar, Sigma, the Common Model of Cognition, Global Workspace Theory, semantic networks, Theory-Theory, and chunking as constrained regions of this language.
The contribution serves as a transparent representational substrate intended for later empirical, computational, and domain-specific work.