Neural Nations

Civilisational complexity research — open data, open code.

Human systems do not fail at random. Their stress builds in patterns.

Neural Nations is an open research platform built around CAMS — the Complex Adaptive Model State — a scale-invariant framework for reading societies as living coordination systems rather than slogans, factions, or headlines. Across 45 historical series, 38 societies, and 39,351 records, CAMS tracks how cohesion, capacity, stress, and abstraction interact across eight institutional nodes, making structural drift visible before crisis becomes obvious.

Most commentary tells you what a society says about itself. CAMS asks a different question: how well is it actually coordinating?

Every durable society has to solve the same functional problems. It needs leadership, defence, memory, legitimacy, production, labour, circulation, and skilled execution. Neural Nations maps those functions across time, then measures how they reinforce each other or drift apart. The point is not ideology. The point is observability. A system under strain leaves a signature.

Why this matters

When a society is healthy, its institutions remain coupled even under pressure. When it begins to fail, the warning signs usually appear well before the spectacle: stress rises faster than capacity, coordination weakens across layers, and the system becomes more reactive, brittle, and prone to shock. CAMS is designed to make those transitions legible.

A note on method

CAMS is a sampling instrument calibrated empirically, not a theory derived from first principles. Kepler mapped planetary orbits with extraordinary accuracy before Newton explained why they moved that way. CAMS works analogously: it fits observed societal behaviour with consistent retrodictive accuracy — but it does not yet have the underlying derivation a mature scientific theory requires.

The author is an independent researcher. CAMS has not yet been formally peer-reviewed. This platform publishes the work openly — datasets, formulas, scoring protocols — precisely so that scrutiny is possible. Rigorous scepticism is not a challenge to this project. It is the point.

75–90%
Ensemble hindcast accuracy across tested societies
r = −0.958
Entropy–health correlation in validated datasets
45 series
38 societies · 39,351 records · 1800–2025
r = 0.78
Seshat cross-validation · Latium/Rome · Apr 2026
Explore the model → Browse the datasets
GitHub Stars Last Commit Corpus Streamlit AI scraping permitted
Read the framework See the full eight-node, four-metric structure, the core formulas, and the logic of CAMS as a comparative instrument. Inspect the evidence Download open datasets, review the scoring method, and follow the validation pathway from raw scores to system-level patterns. Test the tools Use Advanced Analysis, Failure Modes, Escalation Archetypes, and Mindscapes to see how the model behaves across real cases. Follow the research diary Read the work in plain language as it develops: results, revisions, blind tests, and theoretical extensions.

Where to start

📖
For the curious reader Start with the Research Diary — real societies, real data, plain language. No prior knowledge of complexity science required.
📊
For practitioners and strategists Explore the live Streamlit dashboard — 8-node scores, bond strength, phase-space trajectories, Failure Modes and Escalation Archetypes. No login required.
🔬
For researchers and scholars Full methodology documentation, open datasets (39,351 records, GitHub), Seshat cross-validation, Granger causality tests, and scoring protocols. Peer review and co-authorship enquiries welcome.

What's New — April 2026

🏛️ Seshat × CAMS — Cross-Dataset Validation First independent cross-validation of CAMS Λ(t) against Seshat SPC. 35 NGAs, 5,500 years. Latium/Rome overlap: r = 0.78. 📊 Granger Causality Validator In-browser synthetic validation suite — three-scenario stress test with adjustable lag, noise, and series length. 🌐 Mindscapes — National Mood Reports Claude-generated Vision–Affect & Mythic Attractor reports. France and Norway live. Pipeline open. 🕸️ CAMS Failure Modes — Causal Loop Diagram Interactive CLD: R1 Flow-Hands death spiral · R2 Rent-seeking trap · R3 Shield capture · R4 Institutional amnesia.
Kari McKern — researcher, former public servant, IT specialist, and geopolitical analyst based in Sydney. CAMS began as a question about modern geopolitics and became a framework for civilisational dynamics. The data are open. The model is documented. The work continues.

[email protected]
"Patrolling the event horizon between the known and the unknown."