Fix effect, conditions, preparation, reachable alternatives, cost components, time window and uncertainty before examining the outcome. Calibrate parameters independently. Then predict a different load or perturbation and compare with an alternative model. An inconclusive outcome is legitimate.
| Item | Question | Status |
|---|---|---|
| D1 | Conditional elimination or persistence under independently measured process conditions. | Model checks and test design; no new Physarum experiment. |
| H_R | No intrinsic irreversible loss of fine-grained information. Not a demand for time-reversal symmetry. | Additional hypothesis. Collapse-model comparisons are a possible test; unexplained heating alone is insufficient. |
| P1 | Adaptation and recovery of three routes after a blockage. | Design with independent calibration and alternative models; no new empirical verdict. |
| EEML-C1 | Track distance to a specified proxy for the vacuum-stability boundary. | Monitoring proposal with a checked formula, no predicted band or decision rule. |
| CPH-ν | Majorana neutrinos under an additional closure-preference hypothesis. | Candidate hypothesis; exclusion depends on ordering, sensitivity and the assumed decay mechanism. |
| EEML-PT 0.1 | A line-to-loop crossing in a specified finite Hamiltonian. | Synthetic calculation; no inferred cosmic topology. |
| Hawking–Page | Convert a relative free-energy correction into a shift of the crossing. | Formula and local log coefficient checked; complete one-loop free energy remains open. |
These are dossier records and proposals, not external preregistrations. A published script and a dated proposal are not proof that an empirical test was performed.
Selection, innovation and circulation are separate properties. Measure growth of used alternatives, inflow into unused alternatives and directed transitions independently. Persistence of a dominated option can arise without a cycle. Fix a survival time or probability; “it disappears eventually” is not a finite test.
For a suitable weak-noise escape problem, a Kramers-type time can scale exponentially with barrier divided by noise. This is a conditional timescale, not a universal upper bound.
| Quantity | Result |
|---|---|
| Yukawa | Not fixed by the consistency package. |
| Mixing angles | Not fixed by charge and anomaly constraints. |
| θ_QCD | Remains free without an extra mechanism. |
| Λ | Consistency does not select its observed magnitude. |
These are structural non-determination results, not new measurements. An extra hypothesis may turn one into a test case only with a visible new specification.
Known water and current laws, Kummer geometry and exact hidden-variable reductions are benchmarks. Their successful reproduction checks the model and computation. An EEML-specific success requires a consequence not already built into the input, with competing explanations and uncertainty accounted for.
Reconstruction protocol →Earlier neutrino, axion and gravitational-wave claims have not been reinstated as consequences of the present framework. Changing the framework does not erase a failed earlier prediction. This update does not report a new retina analysis or other empirical success.
Reproduction and scope · Corrections →
v0.26 · 28 September 2026