The Internal Dynamics of Decisions: From rats and primates to AI

The finding, in brief

For thirty years the argument has been whether it matters how a system computes or only what it computes. The usual objection is that behaviour can never answer it, since a feedforward network can copy any brain’s behaviour. Six networks wired two ways, trained to the same score on one task, and their dynamics still tell them apart.

How. A subspace state-space demix splits each next state into input drive and the population’s own recurrence, validated on synthetic systems with known answers before any real data. The six architectures were matched on size as well as score, and none of the 144 networks was dropped, so the signature tracks architecture, not task. The estimator is N4SID (Van Overschee & De Moor 1994) and the input-aware demix is InputDSA (Huang et al. 2025), on Ostrow et al. 2023.

Bio

Nathan M. Thornhill is a complexity science researcher in Fort Wayne, Indiana, working in information theory and pattern persistence. He is self-taught, and his earlier career was in healthcare: nursing assistant, then nursing-home administration, then ICU admissions.

What he studies is whether a system holds together and how you would measure that: pattern persistence in simple discrete systems, sleep stages from EEG, task success read off a model’s internal geometry. The analysis plans get written down and frozen first, and the measures that fail are reported next to the ones that work. The results so far are out as preprints and one book.

When he isn’t researching he’s playing guitar, gardening, or with his wife and daughter.

How this work gets madeWho pays for it

No university and no grant. A web design business pays for the research, ICSAC publishes it, and two of the tools the research needed have since turned into products that bring in money of their own.

3Rivers WebTech 3riverswebtech.com

Websites, point of sale, and the ordinary technology work small businesses in Fort Wayne need. Client work buys the time this research takes.

ICSAC icsacinstitute.org

The Institute for Complexity Science and Advanced Computing, which publishes this work. Submissions go to an AI review panel with citation verification and double-blind preprocessing, and a curator checks every recommendation. The review stack is open source, so you can read the rubrics that evaluated your paper.

CiteStamp citestamp.com/claim

Catch the citation your AI hallucinated. CiteStamp checks references where you write, in Overleaf, Word, Zotero and the browser, against a graph that keeps human-signed claims apart from machine-inferred ones. Drop in a DOI and walk its citation neighbourhood.

PeerSigned peersigned.com

Signed peer review that pays. Authors post a paper, a professional in the field takes the match, and the review goes online permanently as citable work.

Follow the workElsewhere

Data and pre-registrations

Every recording analysed here is public and was collected by other laboratories. Nothing on this poster required new animal work.

1. Luo TZ et al. 2025. Data from: Transitions in dynamical regime and neural mode during perceptual decisions. Dryad. doi:10.5061/dryad.sj3tx96dm
2. Neupane S, Fiete I, Jazayeri M. 2024. Mental navigation in the primate entorhinal cortex. DANDI Archive, dandiset 000897. dandiarchive.org/dandiset/000897
3. Schiereck SS et al. 2026. Data from: The orbitofrontal cortex updates beliefs for state inference. Zenodo. doi:10.5281/zenodo.16997337

Pre-registrations. Both analysis plans were written down and frozen before a single dynamical measure was computed, and both are carried in the preprint: doi.org/10.20944/preprints202608.1095.v1. The measure that separates the six networks and the one that does not were both named in advance, and both are reported.

References

The sources the poster draws on. The preprint carries the full text and both pre-registrations.