A co-measurement gap: no standard paradigm jointly tracks a felt valence signal and an independent viability endpoint under within-individual perturbation (position note)
POSITION / RESEARCH-GAP NOTE. Version 1.1 (2026-07-26); first published 2026-07-18. This note establishes no empirical result and must not be cited as one. It argues a structural gap and gives the design specification of the study that would close it. — The gap — A family of accounts holds that a felt or experienced signal (valence, coherence) tracks an organism's viability — its future capacity to persist. Testing any such claim requires one measurement: in the same individual, over time, through a perturbation with a decline arm and a recovery arm, jointly record (a) the felt signal and (b) an INDEPENDENT viability endpoint — one not derived from the felt signal. This note argues that four adjacent paradigms — intensive longitudinal self-report (ESM/EMA), animal physiology/behaviour, appraisal/coping psychology, and viability-indicator studies — are each, for reasons internal to how they are built, not architected to provide that joint measurement. — A documented counterexample search — To test the gap by trying to break it, five web-literature searches were run (queries recorded verbatim in the note), each aimed at finding one study meeting all four conditions. No such study surfaced. What surfaced were strong near-misses, each dropping a DIFFERENT one of the four conditions: chronic-stress/sucrose-preference reversal in rodents (drops the independent viability endpoint); digital-phenotyping relapse prediction in humans (endpoint is symptom-defined relapse, not independent); breeding-condition perturbation with behavioural flexibility and telomere dynamics in seabirds (no felt signal); immune/proteomic markers of residual impairment in remission (no dense felt signal, no decline+recovery arm). The recurring failure in the human and animal affect literatures is the independent viability endpoint: the outcome is operationalised as a symptom-derived quantity. This sharpens the claim from "absence" to a JOINT-MEASUREMENT GAP in which each near-miss drops a different condition. — What this is NOT — It does not claim no such study exists anywhere: the search is five queries of bounded web reconnaissance, not a systematic database audit, and three of the four paradigm rows are non-exhaustive. Only the ESM/EMA row is an exhaustive audit; the others are a reconnaissance, a concept mapping and a single probe. A single counterexample — one dataset satisfying all four conditions — would falsify the empirical half and would be welcomed. The gap's existence does not depend on any theory; its salience does (it matters conditional on caring whether felt states track viability). — v1.1 (2026-07-26): a second probe, on the organism route — v1.1 adds §3b and corrects nothing in v1.0's text. Where v1.0 surveyed paradigms that carry a felt signal, §3b probes the disjoint half of the landscape: systems in which condition (3) is absent by construction (microbial populations and communities), so that only conditions (1), (2) and (4) are in play. Three candidates were examined — yeast collapse under a deteriorating gradient, 48 nutrient-cycling microcosms, and a synthetic three-species gut community in minibioreactors (Rios Garza et al., Nat Commun 2025). The last has a genuine decline-and-recovery perturbation; per-vessel counts were obtained by retrieving the repository directly rather than from the text. The probe corrects a diagnosis that was too coarse. The recurring blocker is not that a recovery arm is missing — in microbial systems the recovery arm is present and well sampled. What does not exist is a densely sampled DECLINE arm and a recovery arm in the same experiment. The designs split into (a) a slow push, densely sampled, with no release, and (b) a short 12–24 h shock with a densely sampled recovery and a release; in (b) the decline arm is short by design, not through any sampling failure. That split is principled rather than accidental, and both obvious remedies are closed. A slower push lets adaptation track the gradient, so the measured threshold reflects evolutionary rescue rather than the ecological tipping point — and that adaptation can move a community's tipping point in either direction is itself an experimental result (Blake et al., Nat Ecol Evol 2024, over 4,000 generations). Sampling faster fails quantitatively: a continuous culture cannot change composition faster than dilution allows, so its residence time is the resolution limit. Obtaining ~50 INDEPENDENT decline-arm observations requires a decline spanning ~50 residence times, which at steady state is ~70 generations. In a chemostat, dynamical resolution and evolutionary quiescence are not independent knobs; they are the same knob, pulled in opposite directions. v1.1 is careful about what 70 generations does and does not establish, and this CORRECTS the addendum previously carried in this record's notes field, which wrote that ~70 generations is the regime in which adaptation is "known to shift" the tipping point. Blake et al. ran for over 4,000 generations and licenses no claim at 70. What can be said at 70 comes from the chemostat-adaptation literature instead: adaptive mutations that later sweep are present within roughly 13-26 generations of founding and detectable at 1% or more by about 100 (Gresham et al., PLoS Genet 2008, in yeast), and reach 50% frequency in less than a few hundred generations (Gresham and Hong, FEMS Microbiol Rev 2014). Seventy generations therefore falls after adaptive lineages exist and are expanding, and before their sweep is routinely detected. The argument holds on the narrower ground that the confound does not require fixation: a rising lineage already changes the population whose threshold is being measured. No threshold shift is claimed to have been demonstrated at 70 generations. These figures are from yeast and single-species chemostat work applied to a three-species bacterial community, so they carry an order of magnitude, not a transferred measurement. The conclusion is unchanged; its warrant is different. The ~50-per-arm figure in this note is therefore not a neutral specification: it is readily met in systems that do not evolve on the timescale of their own relaxation, and close to unreachable in those that do. On the organism side the route is blocked by that constraint rather than by a gap in anyone's attention. §3b is a bounded probe of three named paradigms, not an audit of the microbial literature; a dataset combining a slow push, dense sampling and a release would falsify its central sentence and would be welcome. — The design that closes it — A study satisfying the four conditions would attach an independent viability endpoint (behavioural flexibility, physiological reserve, or an external functional outcome) to dense within-person affect through an engineered decline and recovery. The analysis that would then adjudicate path-dependence (hysteresis) versus a smooth gradient is pre-registered and frozen (10.5281/zenodo.21366132), and remains un-run precisely because no accessible dataset meets the four conditions. — Independence and disclosure — The gap was found by probing whether Valence Order Theory's central claim could be tested; it outlives the theory. Prepared with generative-AI assistance under the author's direction; the author is responsible for the content. Not peer reviewed. The load-bearing step that remains — systematic database searches with screening, and addressing the named near-misses by title — is not done.