Abstract
TWe develop a laboratory–only test of hidden–sector emission based on a deterministic, tunable threshold into the mixed in–medium eigenmode. The key signature is an on/off contrast obtained by bracketing the condition β nA′ (ω) > 1 with two otherwise identical runs (“Off” just below; “On” just above), so that smooth detector effects cancel by design. We keep the method simulation–light: the incremental threshold yield Nth is built from a transparent scaling |εeff (ω)|2× (Frank–Tamm spectrum) × S(ω) A(ω, θ), while the selected background Nbrem is reused from a vetted missing–momentum baseline (e.g. LDMX). The on/off contrast C and expected significance Z follow in closed form in the Gaussian regime, with profile– likelihood CLs and Asimov construction used for exclusions. Dataset–native falsifiers are built in: a step only at the predicted crossing, thickness scaling (linear → plateau), material swaps that move the threshold, an angular check (cos θ ≃ 1/(β ℜ[nA′ ])), band–limited excess, null samples, and a calorimetric null. Formation–length and damping effects are handled analytically near turn–on. We position the method within the current/prospective landscape (NA64, BaBar/Belle II, LDMX) and provide a reuse–first adapter to interface directly to published projections. Complete derivations (effective mixing, dispersion, threshold, statistics) are collected in Apps. A–C. Cosmology and model–building implications are deferred to a separate companion paper. We release minimal scripts for Nth, a medium–model card for εr (ω), and a template Combine workspace to reproduce the results.



![Author ORCID: We display the ORCID iD icon alongside authors names on our website to acknowledge that the ORCiD has been authenticated when entered by the user. To view the users ORCiD record click the icon. [opens in a new tab]](https://www.cambridge.org/engage/assets/public/coe/logo/orcid.png)