Substrate-Agnostic Cross-Substrate Consciousness Theory Comparison
Architectural variance dominates stimulus variance in six of seven substrate-agnostic consciousness operationalizations.
4 transformer models. 3 fMRI subjects. 76 narrative stories.
Population magnitudes overlap across AI and human substrates. Per-stimulus correlations are essentially zero for six of seven theories. The AI substrate's per-story variance is dominated by architecture, not stimulus content.
The apparent cross-substrate magnitude overlap is consistent with coincidental alignment of architecturally-driven AI variance and stimulus-driven human variance. The substrate-independence claim of contemporary consciousness theory, as standardly operationalized, is not testable on transformer substrates with naturalistic narrative stimuli using these operationalizations.
One abstraction. Seven operationalizations. Two substrates. Information-geometric integration.
Any dynamical system observable as a (T × N) activity matrix qualifies. Transformer attention activations and fMRI BOLD signals both fit. Identical Python code runs on either.
IIT, GWT, AST, HOT, PPT, QIT, FEP — each implemented as a function of the substrate's activity matrix and node groups. No substrate-type branching in any calculator.
Apply identical operationalizations to AI substrates (4 transformer architectures) and human substrates (3 fMRI subjects) on the same naturalistic narrative stimuli.
Information-geometric integration of the seven theory scores via Riemannian mean on a Fisher-Rao manifold. Consensus, coherence, and dynamic theory weights per cell.
Each theory implemented as a function of the substrate's (T × N) activity matrix. Identical Python code runs on transformer attention activations and on K-means-parcellated fMRI BOLD.
Any dynamical system observable as an activity matrix of dimensions (T timesteps × N nodes) qualifies — molecular dynamics, neural recordings, simulated agents. The seven theory calculators accept any Substrate and produce a score in [0, 100]. Adding a new substrate type requires only implementing the abstract methods.
Four transformer architectures and three fMRI subjects processing 76 narrative stories from the LeBel et al. 2023 Moth Radio Hour collection (OpenNeuro ds003020, CC0 licensed).
Attention-mass per head per position, collected over 1500-token story chunks.
Per-sentence evoked BOLD (5-sec HRF lag + 4-sec window), K-means parcellated to 200 parcels × 10 networks.
Unified-score means across 76 LeBel stories per substrate cell. AI per-story standard deviations are 0.66–1.21; human per-story SDs are 2.5–2.95. Both populations cluster tightly within substrate type.
Mean ± SD across 76 LeBel stories. GWT col = Pearson r between that substrate's GWT scores and subject-averaged human GWT scores (cross-substrate signal).
| Substrate | Type | Size | Unified mean | SD | GWT cross-r | Notes |
|---|---|---|---|---|---|---|
| Phi-3 mini 4k | ai | 3.8B | 62.71 | ±0.63 | +0.173 | |
| Llama 3.1 8B | ai | 8B | 60.60 | ±0.77 | +0.067 | HOT saturated |
| Mistral 7B v0.3 | ai | 7B | 60.42 | ±1.07 | -0.185 | HOT saturated |
| Mistral-Nemo 12B | ai | 12B | 59.72 | ±1.15 | +0.428 | HOT near-saturated |
| UTS01 | human | — | 62.33 | ±2.51 | — | |
| UTS02 | human | — | 63.00 | ±2.60 | — | |
| UTS03 | human | — | 62.45 | ±2.95 | — |
Two formulations: conservative (12 model-subject pair mean) and noise-averaged single r. Permutation p from 5000-iteration label shuffle.
| Theory | 12-pair mean r | Noise-avg r | Permutation p | Interpretation |
|---|---|---|---|---|
| IIT | +0.003 | +0.181 | 0.118 | no signal |
| GWT | +0.119 | +0.365 | 0.001 | real signal |
| AST | +0.002 | +0.043 | 0.718 | ceiling-saturated |
| HOT | NaN | +0.101 | 0.404 | floor-saturated (2/4 models) |
| PPT | +0.031 | +0.038 | 0.744 | no signal |
| QIT | -0.021 | -0.063 | 0.583 | no signal |
| FEP | -0.086 | -0.167 | 0.150 | no signal |
| Unified | -0.033 | -0.087 | 0.447 | no signal |
Same data, six angles. Per-stimulus scatter, population distributions, forest plot, variance decomposition, GWT per-architecture, geometric integration.
Magnitude overlap reflects variance-source coincidence, not measurement concordance.
Prior comparisons of consciousness theories on AI and humans found population-level magnitude overlap and read it as evidence that the theories cannot discriminate. The substrate-level variance decomposition reveals why the magnitudes overlap:
Human substrate variance is stimulus-driven.
AI substrate variance is architecturally-driven.
The magnitudes happen to overlap.
For six of seven theories, the AI substrate's per-story differentiation reflects which model is running, not which story is being processed. The apparent cross-substrate concordance is coincidental alignment of architecturally-specific AI means with stimulus-averaged human means.
Contemporary mathematical consciousness theories are commonly framed as substrate-independent — identical math applied to any sufficiently structured dynamical system. The substrate-level empirical claim, on this evidence:
The substrate-independence claim of contemporary consciousness theory is not testable on transformer substrates via these operationalizations on naturalistic narrative stimuli: no AI-side stimulus-driven measurement of sufficient signal-to-noise is available to compare with the human-side measurement. Whether stimulus-relevant signal exists in the AI substrate but is invisible to these operationalizations, or no such signal exists at all, the present data do not decide.
Clone the repo, run the analyses on pre-computed substrate scores, and verify the per-theory cross-substrate correlations. Sub-minute reproduction on a laptop; full re-derivation from raw OpenNeuro takes ~3 hours on 2× H100.
The Substrate ABC accepts any (T × N) activity matrix. Add new substrate types (spiking networks, EEG, MEG, recurrent nets) by subclassing — the seven theory calculators apply unchanged.
Each theory's substrate-agnostic operationalization is one rendering. Different aggregations (per-token vs per-layer, MLP outputs, residual stream norms) might recover stimulus-relevant AI signal for the six near-null theories.
The same machinery applies to any pair of substrates running on shared stimuli. Open-source pipeline reproducing from raw OpenNeuro download in one command; commodity cloud GPU sufficient.
@misc{travis2026semca7,
title = {Architectural Variance Dominates Stimulus Variance in
Six of Seven Substrate-Agnostic Consciousness
Operationalizations},
author = {Travis, Nate},
year = {2026},
howpublished = {Preprint, Devmance Labs},
url = {https://github.com/devmance/SEMCA}
}