Paper 161· CC-BY 4.0· DOI: 10.5281/zenodo.19340893

Six Methods.
One Answer.

How do you know predictions are real and not cherry-picked? You test whether independent methods — using different data, different techniques — converge on the same result.

Independent lines of evidence

Each method uses different data, different measurement techniques, and different assumptions. They were developed at different times. They converge anyway.

Behavioral vocabulary

Raw word frequency analysis. No rubric. Counts drift markers in conversation transcripts. Ghost Test: 8.5× ratio.

Information-theoretic

Mutual information decomposition. The explaining-away penalty I(D;M|Y) measured directly from output distributions.

Thermodynamic

Kramers barrier analysis. Same escape rate formula that governs nuclear decay and protein folding, applied to AI behavioral transitions.

Cross-national health data

CDC YRBS + PISA. 613K students, 80 countries. 13 verifiable features. No framework rubric. R² = 0.80.

External team replication

Chua et al. consciousness clusters. Anthropic emotion vectors. Still Alive welfare evaluation. Independent teams, independent data.

Substrate independence

Confirmed on classical, quantum simulation, real quantum hardware, thermodynamic simulation, and abstract channels.

48
Cross-method convergences documented. Independent methods producing consistent quantitative agreement — not merely consistent direction, but overlapping numerical ranges.

Why concordance matters

A single prediction can be fitted post-hoc. Concordance across independent methods is the strongest evidence against overfitting.

Go deeper

📄
Read the Paper
Full concordance analysis with numerical comparisons.
All Evidence
170+ papers. Six non-circular confirmations. The full picture.
Arrow of Time
Paper 162. Why drift cascades only go one direction.