Paper T14F: Whole-VROS Pair Formation, Separation, and Pair-Created Relational Corridors Reduced Pair Audit and Constructive Tests of Topological Linking and Protected Phase Winding
This paper evaluates a three-stage reduced model of whole-VROS pair formation, separation, and pair-created relational carriers. In the first stage, a deterministic census of one hundred thousand primary models and forty thousand independently seeded holdout models tests binding, interior stability, mixed response, advantage over a matched factorized control, partner specificity, and null controls. About fifteen percent of both model banks pass the complete static pair gate. Ranked subsets of the supported models also show strong resistance to coefficient perturbations and broad accessibility under the declared gradient-relaxation rule. The second stage follows the supported pairs through increasing separation. The direct strain interaction produces finite-range cross-response, but every tested primary and holdout pair approaches factorization as the interaction decays. Correlated phase and axis states may remain as historical records, but the factorized post-interaction equations provide no continuing live response between the separated structures. Direct strain therefore does not supply the proposed long-distance carrier. The third stage evaluates four explicitly constructed carrier families using assigned formation, retention, propagation, addressing, energy, covariance, and destruction laws. Topological linking and protected phase winding pass the frozen constructive carrier gates in reproducible primary and holdout subsets. Closure-class memory retains a finite static subset but fails the selected protocol threshold, while persistent corridor deformation fails the complete static conjunction. These results establish the internal consistency of a reduced architecture combining stable local pair formation, direct-strain factorization, and protected carrier classes. They do not derive a native whole-VROS linking or winding invariant, a physical carrier propagation law, detector outcomes, Bell correlations, no-signalling, or quantum entanglement.
This paper evaluates a three-stage reduced model of whole-VROS pair formation, separation, and pair-created relational carriers. In the first stage, a deterministic census of one hundred thousand primary models and forty thousand independently seeded holdout models tests binding, interior stability, mixed response, advantage over a matched factorized control, partner specificity, and null controls. About fifteen percent of both model banks pass the complete static pair gate. Ranked subsets of the supported models also show strong resistance to coefficient perturbations and broad accessibility under the declared gradient-relaxation rule. The second stage follows the supported pairs through increasing separation. The direct strain interaction produces finite-range cross-response, but every tested primary and holdout pair approaches factorization as the interaction decays. Correlated phase and axis states may remain as historical records, but the factorized post-interaction equations provide no continuing live response between the separated structures. Direct strain therefore does not supply the proposed long-distance carrier. The third stage evaluates four explicitly constructed carrier families using assigned formation, retention, propagation, addressing, energy, covariance, and destruction laws. Topological linking and protected phase winding pass the frozen constructive carrier gates in reproducible primary and holdout subsets. Closure-class memory retains a finite static subset but fails the selected protocol threshold, while persistent corridor deformation fails the complete static conjunction. These results establish the internal consistency of a reduced architecture combining stable local pair formation, direct-strain factorization, and protected carrier classes. They do not derive a native whole-VROS linking or winding invariant, a physical carrier propagation law, detector outcomes, Bell correlations, no-signalling, or quantum entanglement.
Paper T14D investigates how a supplied Bell-correlated relational rule can be converted into two separately finalized detector records. The calculation uses the previously established Toner–Bacon one-bit communication protocol together with an angle and residual schedule passed from Paper T14C. Two shared random directions determine Alice’s ideal response and a one-bit message that is used in Bob’s response. Separate noisy first-passage detector processes then convert the ideal responses into records with variable finalization times, rare sign changes, finite-horizon handling, record hashing, and later trial matching. The record generator does not consult the target Bell correlation, a joint probability table, or a globally sampled outcome pair. An independent-seed hardening run generated and retained six million three hundred thousand primary records. The ideal responses produced a Bell value near the expected maximum, while the final records showed the small attenuation predicted by the measured pointer flip rates. The exact attenuation calculation reproduced the final record result to numerical precision. Communication analysis showed that the message carries one-half bit of information about a choice between orthogonal Alice settings when the shared directions are known. Removing T14C frame and state provenance did not change the generated outcomes, while removing the scheduled residual altered Bob’s response in the residual audit. Message removal and cross-trial message permutation damaged the Bell correlation, detector asymmetry produced measurable marginal bias, and no primary records were discarded. The finalized outcome marginals were consistent with the tested no-signalling equalities, although the complete protocol contains explicit communication from Alice to Bob. The result supports a communication-assisted first-passage record-encoding architecture. It does not derive the one-bit update from microscopic Holosphere dynamics, establish Bell locality, or provide a relativistically local model of spacelike measurement
Paper T14A connects finite partner selection, transported surface information, angular readout,detector response, and environmental coherence within one reduced Holosphere architecture.The source is an occupied contact organization; its three lanes are observables of a collectivemode. A controlled twelve-slot pairing problem has a unique antipodal minimum, with an explicitrobustness bound. The same source amplitudes then enter the selected pair and the local phaserecords directly. An explicitly assigned two-alternative reporting map produces a general angularlaw containing source imbalance and relative phase; its balanced, opposite-sign limit gives thefamiliar Bell correlation.The conservative response and backreaction of Paper T14 extend to four joint channels. Formatched response and a uniform fixed-total readiness preparation, first-release volumes reproducethe normalized channel inventories, with an exact no-record fraction. Environmental branchoverlap suppresses the interference component while leaving a population correlation intact. Thiscorrects a scalar visibility approximation and separates physical loss from selection of coincidentrecords. Exhaustive matching and deterministic amplitude, polytope-volume, and conservationcalculations verify the stated results without generated detection events. The remaining physicalinterface is the realization of the reporting map and joint outcome selection in an occupiedapparatus with separated readout sites.
Paper 3 of 11 — A General Theory of Answerability The eleven-paper TUS OS mathematical research programme Abstract This paper develops an exact finite local-to-global analysis of the six-dimensional Hamming-cube foundation Atlas underlying the TUS OS construction. Every admissible completed foundation carries the same exact local qubit prepare–measure representation, while the complete repeated-common-effect Atlas has exact minimum complex Hilbert dimension four. The paper classifies the full ambient family of completed four-state foundations in the six-dimensional Hamming cube. Every foundation is uniquely an affine coset associated with a perfect matching of the six coordinates. There are exactly 240 unordered foundations and 5,760 ordered codebooks. Every foundation induces the same exact 27-class tetrahedral-SIC qubit operational quotient, with carrier-fibre profile 1⁸, 2¹², 4⁶, 8¹ and the corresponding Bloch-radius structure of eight pure states, twelve states of radius √(2/3), six states of radius 1/√3, and one maximally mixed state. Each ambient chart reduces losslessly to a signed perfect matching. The chart-coloured union of two matchings obeys an exact component-parity collision law, yielding the complete seven-value image spectrum over all 28,680 ambient foundation pairs. Exactly 10,800 pairs jointly reconstruct the full 64-state carrier, giving an ambient mathematical chart-cover number of two. The frozen TUS selector defines a canonical subatlas of 36 distinct foundations and 64 anchor-indexed chart instances across nine matching classes. Its selected-foundation reconstruction graph is bipartite and eight-regular with 144 edges, while selector multiplicity yields exactly 400 reconstructing pairs among the 2,016 canonical chart-instance pairs. Only this canonical subatlas inherits the native LANDING and GUIDANCE structure. At the sequential level, all 384 active map/chart descent tests fail: no active LANDING map descends to a state-only transformation on a fixed 27-class local chart quotient. Under the complete principal future language, every canonical chart refines to all 64 deterministic predictive states, with maximum separating depth two. The subordinate GUIDANCE witness is also classified: its endpoint map has image size 34, eight two-cycle attractors and maximum transient depth five, while the additional GUIDANCE-selected chart raises principal path reconstruction from 138/192 contexts to 169/192. The paper then establishes the exact global static common-effect Hilbert dimension. Although every individual foundation admits an exact qubit representation, no unrestricted qutrit state/common-effect family can reproduce all 4,096 Atlas probabilities while simultaneously satisfying all 240 overlapping foundation POVM identities. Together with an explicit chart-independent dimension-four construction, this proves d_q^atlas = 4 for the complete static Atlas. The attaining construction uses two declared qubit sectors; unrestricted uniqueness or rigidity of minimum representations is not claimed. The result therefore exhibits a precise local-to-global separation: exact qubit prepare–measure form exists on every individual foundation, but an exact chart-independent repeated-common-effect representation of the complete Atlas requires complex Hilbert dimension four, while the complete deterministic future language preserves all 64 predictive identities. Local operational state, global relational state and deterministic predictive state are consequently distinct mathematical structures. The paper explicitly keeps these distinct from the still-unresolved whole-position process state. The scope boundary is explicit. The paper does not establish a physical quantum carrier, operational access to all 240 ambient foundations, LANDING as a quantum channel, GUIDANCE as a quantum observable, endpoint channel or instrument, contextuality, nonlocality, quantum advantage, unrestricted rigidity of minimum representations, or whole-position process synthesis. Its conclusions are exact at the static Atlas, canonical-chart and deterministic predictive scopes defined in the manuscript. Supporting General Theory Programme This paper is Paper 3 in the TUS OS quantum-operational and Answerability research programme. It extends the results of Papers 1–2 from the canonical local geometry and exact qubit prepare–measure behaviour to the complete finite relational Atlas. It classifies the full family of admissible foundations, proves that every local chart carries the same exact 27-class qubit operational quotient, identifies the canonical TUS subatlas, and shows that the complete native future language refines those local quotients back to all 64 deterministic predictive states. Its principal global result is that the complete common-effect Atlas has exact minimum quantum dimension d=4, while remaining distinct from the larger 64-state predictive structure and from later process-level dynamics. The paper also classifies the subordinate GUIDANCE witness and its finite reconstruction contribution without promoting it into a channel or instrument. The wider programme then develops complete-future and process closure in Paper 4, the Anchored Analog Quantum System framework in Paper 5, contract-relative resource and operational-observability theory in Papers 6–7, generic Answerability theory in Papers 8–9, complete TUS OS formal and executable realisation in Paper 10, and culminates in Paper 11 with A General Theory of Answerability. Paper 1 — Canonical Tetrahedral Qubit-State Geometry in the TUS OS TriadXOR Completion, Receipt Verification and the Qubit-Geometric Semantic WitnessEstablishes the canonical tetrahedral geometry, unique XOR completion, exact qubit-SIC representation and centred-face trine geometry. Paper 2 — The TUS OS Operational CalculusClassical Construction of Exact Qubit Prepare–Measure Behaviour from a Receipt-Custodied Engine ArchitectureEstablishes exact full-carrier qubit prepare–measure behaviour from the deterministic classical engine under the declared Coordinate-Incidence Operational Harness. For each lawful foundation, the 64 six-bit carriers map to exactly 27 anchor-relative qubit preparations under one foundation-defined tetrahedral-SIC Harness, while the separately declared face Harness gives the exact trine table. Paper 3 — From Finite Relational Structure to Qubit Operational FormAmbient Foundations, Canonical Subatlas, Exact Global Quantum Dimension, GUIDANCE Witness, Sequential Closure and Global Representation in TUS OSExtends the local prepare–measure construction to the complete ambient family of 240 completed foundations and the canonical TUS OS subatlas. Every completed foundation has the same 27-class qubit operational quotient; pair relations between charts provide exact carrier-reconstruction structure; the complete repeated-common-effect static Atlas has exact minimum complex Hilbert dimension 4; and the frozen complete principal future language has deterministic predictive-state minimum 64. The dimension-4 result is a static representation theorem and does not by itself supply common quantum process dynamics. Paper 4 — Technical Volume I — Formal Foundations and the Complete-Future CorpusNative Contract, Statistical Discrimination and the Invariant Probability ModuleFixes the formal native domain, complete principal terminal future-experiment language and one-common-word experiment class; establishes complete statistical separation of distinct carrier pairs, exact parity-resolved discrimination results and complement reachability; and proves that the rational invariant probability module has dimension 33. Paper 4 — Technical Volume II — Orthogonality, Process Closure and Native Triadic SynthesisEstablishes universal complement-support orthogonality and its propagation under exact common-CPTP dynamics, yielding the certified same-parity support graph and the bound 32 ≤ d_CPTP^min ≤ 64, with cross-parity forcing and the exact state-only common-CPTP minimum remaining unresolved. It further establishes recovery-complete LANDING and GUIDANCE results, exact 64-dimensional selected-line instrument/readout minima at their declared contracts, an exact classical–quantum hybrid construction, and the native structural/predictive quotient correspondence. Paper 5 — TUS OS as an Anchored Analog Quantum SystemNative Witnesses, Predictive Synthesis, and Access-Contract Resource SeparationDefines Analog Quantum Views and Anchored Analog Quantum Systems, proves that TUS OS is a proper Anchored Analog Quantum System on its 64-input abstract canonical corpus, elevates the inherited structural/predictive correspondence into the Predictive Synthesis Identity Theorem, organises the inherited exact quantum-operational results through a typed access-contract resource law, and composes the recognition, predictive, resource and downstream-conservation results into the TUS OS Anchored Analog Quantum System Theorem. Paper 6 — Contract-Relative Quantum Realisations of Classical ProcessesPredictive Compression, Vanishing Worst-Case Causal-State Transparency, and Finite-Dimensional BreakdownExtends the programme beyond the TUS-specific discovery corpus by formalising the contract-relative achievable-resource object (S,C,𝔐,ε) ⟼ Ach(S,C;𝔐,ε), and establishes three exact contract-relative regimes: strict finite separation, unbounded finite separation and finite-capacity breakdown. These are instantiated respectively by the TUS OS 4→64 separation, the periodically modulated decay 2→N exact prediction-to-transparency separation, and dual-Poisson finite-capacity breakdown for every positive uniform worst-case transparency threshold. Paper 7 — Operational Observability and Contract-Induced Resource Transitions in Quantum RealisabilityStructural Classification, Family Scaling, and Finite-Resource BreakdownDevelops a structural classification
This record freezes the analytical and computer-assisted successor protocol to the HLV Golden Gauge Forge finite-family study. Previous work established an exact golden 6→3 cut-and-project carrier fingerprint, a nested bounded-locality exhaustion, maximum vertex degree 12, maximum edge–face incidence 10, volume-independent positive HLV coefficient bounds, and a strictly positive physical-sector gap for every fixed finite connected complex satisfying the declared hypotheses. A preregistered exact Z3 finite-family study successfully constructed the controlled carrier sequence but did not support the frozen positive volume-uniform-gap candidate. This negative result motivates a locality-based analytical programme rather than further unrestricted enlargement of finite matrices. The primary system is a fixed-lattice Z3 gauge Hamiltonian on a declared class of finite oriented two-complexes with uniformly bounded vertex degree, face incidence, interaction range, and positive electric and magnetic coefficients. The protocol freezes a sequential theorem programme consisting of: an explicit volume-uniform gap for the electric product-state backbone; derivation of system-size-independent locality and interaction constants; a strong-coupling perturbative stability theorem with a volume-independent spectral gap; exponential clustering of spatially separated gauge-invariant observables; a van-Hove/Følner theorem for the canonical HLV origin-connected exhaustion; and, only if these gates are closed, construction of a fixed-cutoff thermodynamic-limit state. The computer-assisted component freezes an eight-radius geometry audit, graph-boundary layers of thickness one to three, radius-two rooted patch-frequency diagnostics, exact local norm and interaction-overlap calculations, a global exact-diagonalization cap at cycle rank 10, independent numerical verification, and complete deterministic replay. HLV carrier specificity is tested against matched controls consisting of uniform weights on the same complex, 50 exact histogram-preserving type-stratified weight permutations, 12 deterministic internal-window phason shifts, and three rational Fibonacci-slope periodic approximants. The protocol explicitly forbids replacing a local perturbation estimate by the global magnetic operator norm, which grows with system size and cannot establish a bulk spectral gap. It also keeps the thermodynamic limit, representation-cutoff removal, and continuum lattice-spacing limit strictly separate. A positive result under this protocol would establish, at most, a fixed-Z3, fixed-lattice strong-coupling theorem for a class of aperiodic bounded-locality gauge complexes and a separate finite test of whether the HLV geometric assignment produces effects not reproduced by matched controls. It would not establish full SU(3) dynamics, representation-cutoff removal, a continuum Yang–Mills theory, or the Yang–Mills existence and mass-gap problem.
We report the confirmatory stage of a finite, preregistered test of phase-resolved response ina two-channel quantum 6j recoupling model. The upstream construction fixes a regular localq-deformation coordinate, an explicit finite carrier atlas, and analytic recoupling observables.A disjoint calibration stage selected the integer q-curvature mode m = 5 against a matchednoninteger control frequency 9.5. Before any confirmatory family was evaluated, the protocolfroze eight previously unused external-label families, all 106 ordered carriers belonging tothem, a 1101-point deformation grid, a cubic baseline projection, an anchored complex-columnprimary observable, a signed-angular-speed replication observable, exact decision thresholds,and a one-run stop rule.The primary endpoint was positive in all eight families. The median family energy advantage was 0.12716387712459315, compared with the preregistered threshold 0.02; the exact onesided family sign-test value was 1/256 = 0.00390625. The primary family phases were stronglyconcentrated, with R1 = 0.9967284980027259 and R2 = 0.9869553189520441. The frozendecision rule therefore assigned the protocol outcome A_UNIVERSAL_PHASE_LOCK. The replication observable also gave eight positive families, median advantage 0.09240553845217536,and R1 = R2 = 1. It did not participate in outcome assignment. Numerical audits found nohard failure, nonfinite value, residual degeneracy, undefined phase, or anchor-sign failure; thelargest reported orthogonality and evaluator discrepancies were of order 10−15.The controlled conclusion is deliberately narrow: on the frozen finite atlas, at the frozenmode/control pair and under the declared observation protocol, the primary residual phasesexhibit preregistered directional concentration together with positive predictor-over-controlenergy advantage. The result is local in the deformation coordinate and finite in labels. It isnot a theorem for all SU(2)k data, not a pentagon, braiding, or modular-coherence statement,not global gluing, not a large-label or continuum limit, and not evidence of physical timeoscillations, dynamics, mass, matter, or gravity.
Dynamic Quasi-Closure and Relational Organization of Quantum Systems Residual Certificates, Geometric Holonomies, Environmental Correlations, and Structural Transfer Conditions Robert Kupski · BKT–74–QREC · Edition 2.6 Abstract A mathematical and physical account of organizational retention in finite-dimensional quantum systems is developed within the Law of One Mechanism (LOM) research programme. Organization is specified by a distinguished family of states, while its retention depends on the dynamical generator, preparation, interaction with the environment, and observation interval. State recurrence, endpoint retention, quasi-closure of the entire trajectory, family invariance, and attraction are distinguished. Error certificates based on the residual of the equation of motion, their geometric refinements, invariance conditions, and bounds for quantum channels are derived. An exactly solvable two-coupling model relates protection of a selected sector to destructive interference between transition amplitudes. At resonance, for coupling frequencies of 0.30 and 0.40 MHz, leakage after 0.25 µs is 0 for the dark state, 0.2304 after complete dephasing of the preparation before evolution, and 0.4608 after a relative-phase shift of π. All three preparations have the same initial populations. In the ideal resonant model, keeping leakage at or below the one-percent threshold throughout evolution allows a maximum absolute phase deviation from the dark preparation of approximately 0.2087119452 rad. In an open three-qubit model, with the same Hamiltonian and preparation and local relaxation and coherence times T₁ = 80 µs and T₂ = 40 µs, replacing independent dephasing with common dephasing increases the time to one-percent total organizational loss from 0.483439 to 0.804027 µs, or by 66.31%. For these two realizations, the lower bound on the worst-case absolute error of a deterministic threshold-time prediction based solely on their shared local data is 0.1602938517 µs. An independent measurement of normalized relative-coherence decay identifies the correlation parameter within the model family and determines the coefficient of cubic leakage onset into the third qubit without fitting its subsequent trajectory. This transfer assumes an unchanged environmental generator across protocols and specified conditions of Markovian dynamics. The account is extended to the quantum geometric tensor, dark-state transport, and interferometric readout. A full coupling-phase cycle produces an adiabatic geometric phase of −0.72π and an ideal interference probability of 0.1812880051. Exact evolution determines finite-duration corrections. In the open model, return fidelity is bounded by excitation survival: for one initial excitation, no pumping, equal local relaxation times, and an excitation-preserving Hamiltonian, the upper bound is exp(−T/T₁), where T denotes the cycle duration. Geometric records and environmental memory are distinguished, and a conditional two-cycle order test is formulated. LOM postulates that fields and particles emerge from a more fundamental relational and informational organization and constitute its effective realizations. The article gives selected elements of this hypothesis an operational form through states, couplings, generators, and measurement relations. The results are conditional theorems and calculations for specified models; they are neither new experimental data nor evidence establishing the existence of a metafield. Additional constitutive content of the programme may take the form of independent constraints linking admissible generators and protocols. The study establishes a common account of retention, perturbation tolerances, environmental correlations, geometric transport, and dynamical identifiability. Introduction Persistence of organization in a quantum system is a property of the relations between the state, a distinguished class of configurations, and the dynamics. Two different, even orthogonal states may belong to the same organizational class, whereas recurrence to the initial state may follow an earlier departure from that class. Retention at one instant, control of the distance of an entire trajectory from a state family, and attraction to that family therefore describe distinct physical properties. Their quantitative separation identifies the contributions of autonomous evolution, preparation, perturbations, and measurement interventions to observed organizational retention. The study develops the LOM–GTSFC–USC–GTCW programme. The Law of One Mechanism (LOM; Polish: Prawo Jednego Mechanizmu, PJM) provides the overarching framework; GTSFC denotes the Global Theory of Shared Fundamental Couplings, USC the Universal Structural Code, and GTCW the Gravitational–Thermodynamic Cyclicity of Multiverses. Within this framework, the metafield is a hypothetical structure of admissible relations, couplings, and tolerance bands. USC describes their admissibility rules, while GTSFC organizes couplings and their sectoral realizations. The hypothesis of a common organizational principle permits different effective Hamiltonians and environmental channels. Structural information concerns the organization of relations that sustain a selected class of states; preservation of a bit record and state entropy do not exhaust this concept. A possible LOM principle more fundamental than the informational layer itself remains a separate source hypothesis. The article continues the BKT–74 line of research on memory, retention, and stabilization. BKT–74A distinguishes organization from instantaneous geometry; BKT–74B separates preparation and return memory from the stabilizing response; and BKT–74C relates retention and first exit to the system–environment coupling operator. BKT–74–QREC develops these distinctions through explicit state families, leakage operators, trajectory certificates, and relations between protocols. The protection analysis concerns a specified preparation and generator. The existence of an invariant subsector and the formation of such organization from general initial conditions are separate questions. Experimental context is provided by Dong and co-workers’ studies of regular trajectories in a variational description of an interacting 24-qubit system and their search and stabilization through hybrid feedback control (Dong et al., 2026; DOI: 10.1038/s41567-026-03431-z). Exact quantum evolution, dynamics on a variational manifold, and a loop comprising measurement, optimization, and re-preparation represent different levels of description. The two-coupling model considered here is a self-contained analytical construction. It provides an explicit example of interference-based protection, and its parameters and results refer to that model, without identifying it with the qubit-ladder Hamiltonian or the data of the cited experiment. Dynamic quasi-closure bounds the distance of a trajectory from a distinguished family of states at every instant of the interval considered. The residual of the equation of motion measures the mismatch between an approximate trajectory and a specified generator, yielding certificates for state and observable errors. Removing the component associated with the global phase gives an angular certificate in ray space. For organization described by a subspace, the operator K = QHP/ℏ, where P projects onto that subspace, Q = I − P, I is the identity operator, H is the Hamiltonian, and ℏ is the reduced Planck constant, identifies direct leakage directions. Full organizational retention also depends on the action of the generator within the distinguished sector. This approach connects protection properties to the dynamics and separates instantaneous cancellation of leakage, invariance, and attraction. In the resonant two-coupling model, the dark-state condition is g₁a₁ + g₂a₂ = 0. The quantities g₁ and g₂ are angular coupling frequencies, related to ordinary frequencies by gⱼ = 2πfⱼ, while a₁ and a₂ are preparation amplitudes. For f₁ = 0.30 MHz and f₂ = 0.40 MHz, the dark state is |D⟩ = 0.8|1⟩ − 0.6|2⟩, where |1⟩ and |2⟩ are orthonormal states spanning the distinguished sector. Comparing the dark preparation, its completely dephased mixture, and a preparation with reversed relative phase isolates the role of coherence at unchanged populations. Phase-error and coupling-direction tolerances quantify the robustness of this protection. Certificates based on the integrated residual norm provide sufficient conditions; loss of information about phase cancellations can limit their sharpness. The extension to open systems uses Gorini–Kossakowski–Lindblad–Sudarshan (GKLS) generators with specified jump operators. In the three-qubit model considered, identical local relaxation and coherence times T₁ and T₂ do not specify dephasing correlations and therefore do not uniquely determine retention time. The parameter c interpolates between independent noise, c = 0, and common noise, c = 1. At T₁ = 80 µs and T₂ = 40 µs, the difference between the times to one-percent total loss yields a lower error bound for predictions based on local data that do not distinguish these realizations. Total loss includes both leakage into the third qubit and relaxation to the state with no excitation. This non-identifiability concerns the scope of the calibration data, while the predictions of a fully specified model remain unambiguous. An additional measurement of relative coherence provides information about environmental correlations. With the couplings switched off, the decay rate of population-normalized coherence determines c within the family considered, provided that the pure-dephasing rate is positive. The calibration result then determines the coefficient A₃ in the expansion P₀₀₁(t) = A₃t³ + O(t⁴), where P₀₀₁ denotes the population of the state with the third qubit excited, and t is the time since dark-state preparation and activation of the couplings. The relation holds for th