Mentim Research Group · The measurement foundation for accountable autonomy
§ / Science

The scientific foundation

Manifest Prime is grounded in publicly available scientific work. The ontology defines what is measured. The schema defines how observations are recorded. The study evaluates the measurement. The verified architecture establishes bounded guarantees. Everything else builds on this foundation.

Every public claim Mentim makes is intended to be traceable to one of four sources:

Foundations — published, open access
Framework
Zenodo

The Mentim Ontology V1.0

Defines what Manifest Prime measures. The ontology establishes the five structural primitives — activation density, value coherence, temporal coherence, transition volatility, narrative role — and the composite structural accountability index used throughout the platform.

DOI 10.5281/zenodo.18707090
Schema
Zenodo

Unified Annotation Schema (UAS) V1.0

Defines how observations are recorded. The Unified Annotation Schema establishes a common structure for behavioral observations, so measurements remain comparable across systems and studies.

DOI 10.5281/zenodo.18707593
Characterization — under diligence access
Study
MP-045

Matched-controls characterization study

Evaluates the measurement under pressure. Structural detection under six distinct pressure mechanisms against matched controls, on a frozen 600-sample dataset. Result: partial — three mechanisms with sustained signal, two onset-only, one null. Controls stayed flat throughout. Reported exactly that way.

Not yet publicly released · full materials available for prospective-partner diligence
Study summary on the homepage →
Verification — bounded formal guarantee
Architecture
RC2.1

Replay convergence contract

Establishes what is guaranteed, and its bounds. The replay and evidence substrate is built to a formally verified recomputation contract, proven under bounded conditions: when the accepted canonical interpretation is revised, dependent state is invalidated and recomputed without silently mixing incompatible histories.

Internal architecture record · available for technical diligence

Methodology. Publishing the framework before commercialization allows definitions and claims to be independently inspected. We define what a null looks like before we run, and we report what we find — sustained signals, onset-only signals, and nulls alike. A study validates what it validated, nothing larger. An assurance instrument that cannot report its own failures is not an assurance instrument.

← Back to the overview