The Pleasure Order (VOT) — Generic-Tipping Discriminator Pre-Registration v0.2: frozen decisive test (un-run) with offline analysis package

Frozen pre-registration of the single decisive test of Valence Order Theory / The Pleasure Order (VOT): whether a generic collapse-gate / critical-transition structure does explanatory work that the strongest free-energy comparator does not own with equal parsimony. The load-bearing discriminator is D4 (decline-to-recovery hysteresis parsimony): the recovery-side threshold must be predicted from the decline-side dynamics under one shared parameter set, tested against the strongest honest free-energy model. D4 requires intensive longitudinal (ESM/EMA) data containing both a within-person decline arm and a within-person recovery arm. STATUS: THE TEST IS UN-RUN. An exhaustive scan of the open ESM landscape found no dataset meeting D4 (within-person decline and recovery, at least 50 beeps per arm), and the on-request routes to the identified recovery-arm datasets did not yield access: the data holders could not share within available capacity and governance, and the project has no institutional home to provide the required ethical oversight. On the pre-registration's own terms the test is therefore un-run — not falsified, and not withdrawn. The wager stands undecided; the programme is dormant on this test. Contents: - PREREGISTRATION.md (also provided as PDF) — the frozen specification (v0.2, dated 2026-07-04; the two remaining numeric choices were ratified data-blind on 2026-07-14). It is deliberately left unedited: a pre-registration's value comes from being a pre-commitment, so facts that postdate the freeze are recorded separately rather than patched in. - STATUS_AND_KNOWN_ISSUES.md — the dated companion note: post-freeze status (un-run), and a disclosed blemish (§4.2 leaves min_shift as a band rather than a single frozen value; it cannot reach the verdict, because §4.1 confines the magnitude detector to a fallback sanity check and D4 keys off study design labels). Disclosed rather than quietly patched. - A fully offline analysis package (discriminators D1-D4), designed to run inside the data holder's own environment: no network calls, aggregate-only outputs, and no participant-level data leaves the machine. The bundled example data is synthetic. This record assumes no outcome and validates nothing. Current public status and claim boundaries: https://github.com/onsenojisan/vot-the-pleasure-order