ScienceSeparate Contracts

Scientific Metrics

ECAM-X, AEI-Delta, Human Gate records, and AIJIM TCRE remain separate authorities.

No super-score

AIJIM must not collapse ECAM-X, AEI-Delta, Human Gate records, and AIJIM TCRE into one quality number. They answer different questions and have different authorities. TCRE cannot emit a composite score, traffic light, certificate, recommendation, or global appropriate-reliance result.

Interpretation Matrix

These contracts answer different scientific questions. The most important column is what each does not measure: it prevents scope creep, trust-washing, and accidental authority transfer between protocol records, governance, and research instruments.

Axis	Scientific Question	Measures	Does Not Measure	Authority	Falsifiable Via	Reader Label	Audit Depth	Fail-Closed Form	Paper Section
ECAM-X	Does this evidence support this claim?	NLI label, similarity, evidence-link coverage, negative controls, judge configuration, and attribution confidence.	Causality, source bias, editorial relevance, explanation stability, or human reliance quality.	Publication-required for claim/evidence support metadata.	Canonical JSON, model snapshot/hash, evidence content hash, judge config, and reproduced NLI link output.	Evidence grounding	Run IDs, evidence hashes, thresholds, NLI labels, similarity scores, and judgeConfig.	claim_missing_evidence_links or explicit evidence grounding unavailable.	Attribution Validity
AEI-Delta	Does the explanation remain stable under perturbation?	Seeds, perturbations, delta distributions, confidence intervals, sanity checks, and stability drift.	Truth, evidence quality, causality, bias, or editor trust. A stable explanation can still be wrong.	Publication-strengthening disclosure. Absence is honest not-measured, not failure and not green.	Seeds, perturbation recipe, model snapshot, stability artifact, and deterministic delta recomputation.	Stability evaluated / not evaluated	Perturbation set, delta histograms, CI bounds, failed sanity checks, model snapshot.	stability not evaluated; never synthesize a neutral stability score.	Explanation Stability
AIJIM TCRE	How did item-level decisions change after advice or bound evidence, relative to a preregistered independent item-quality reference?	When all study gates close: beneficial update, missed benefit/under-reliance, harmful deference/over-reliance, resilient self-reliance, and preregistered task-weighted benefit/harm, reported separately.	Authorization quality, aggregate override activity, factual truth, a global trust score, or universally appropriate reliance.	Proposed behavioural research instrument. Current status: PROPOSED / NOT_VALIDATED / NOT_MEASURED.	Preregistered task and adjudication protocol, independent item-quality reference, participant/item/task/condition identities, exact pre/exposure/post decisions, uncertainty rules, HREC, validation evidence, and independent replication.	Reliance evaluation not measured / cannot verify	Task contract, reference provenance and status, item transitions, separated outcomes, uncertainty, missingness, and exclusions.	Missing reference or transitions means NOT_MEASURED; disputed, leaked, or under-reliable reference means CANNOT_VERIFY. Human Gate history is never the reference.	Task-Conditioned Reliance Evaluation

Measurement axes are not protocol loci

Triple Falsifiability asks where a publication record can fail: Producer Conformance (P), Surface Observability (S), or Audit Replicability (A). The measurement axes below ask what scientific property was measured. A measurement artifact can be challenged at any of the three protocol loci.

Bundle integrity

Core hashes, signatures, and canonical JSON bind the bundle. ECAM-X supplies reproducible claim/evidence attribution metadata inside that signed integrity envelope.

Experimental

AEI-Delta exposes seeds and perturbations so stability can be re-run like a reproducibility capsule.

Audit-log

AIJIM TCRE requires participant/item/task/condition observations, exact pre/exposure/post decisions, and an independently assessed preregistered item-quality reference. Ordinary authorization and override logs cannot satisfy that reference gate.

Two-Tier Surface Discipline

Reader UI	Quiet labels: evidence grounding, stability evaluated/not evaluated, and TCRE not measured/cannot verify. No raw run IDs or scores.
Editor UI	Actionable state: which axis is missing, blocked, measured-positive, or measured-negative.
Audit UI	Technical depth: run IDs, hashes, thresholds, model snapshots, verifier outputs, and RFC conformance warnings.
Publication Bundle	Immutable disclosure: first-class publicationId plus axis artifacts that actually exist.
Verifier CLI	Strict publication-integrity PASS/FAIL first; EU AI Act readiness as PASS/PARTIAL/FAIL; optional quality report second. No quality score may weaken cryptographic verification.
Protocol Paper	P/S/A protocol loci are reported separately from ECAM-X, AEI-Delta, Human Gate governance, and AIJIM TCRE.

Honest absence

Honest Absence is not a score or a negative result. For current TCRE, missing study inputs yield NOT_MEASURED and a disputed, leaked, or under-reliable reference yields CANNOT_VERIFY. The current line must never reinterpret ordinary Human Gate history as AIJIM TCRE. Protocol conformance may verify the artifact identity, bindings, status, and reason codes; it cannot validate the scientific reference or behavioural interpretation.

EU AI Act readiness

The verifier reports EU AI Act readiness as PASS, PARTIAL, or FAIL from three artifacts:model_card.v1,risk_register.v1, andevaluation.v1. PARTIAL is an honest-absence state, not a score and not a cryptographic publication failure.

Verifier Stance

aijim-verify remains strict about publication integrity: publication identity, body hash, bundle hash, Merkle root, schema version, and mandatory ECAM-X support metadata must verify or the publication does not pass. AEI-Delta and AIJIM TCRE remain separate optional artifacts. A verifier may check their schemas, bindings, status, and Honest Absence, but cannot claim scientific validity or turn a cryptographic failure into a pass.

Protocol Paper Framing

The Protocol Paper reports bounded Triple Falsifiability through Producer, Surface, and Audit loci. ECAM-X and AEI-Delta remain bounded evaluation disclosures; Human Gate records remain governance evidence; AIJIM TCRE remains a proposed research instrument with its own validity and Honest Absence gates. None is a universal truth or quality score.