Preregistered Confirmatory Replication of a QRT-Consistent Coherence Result at Stated Parameters on a Second Superconducting Device (v5)

Amber Anson

Published Jul 5, 2026• 0 citations

Abstract

A preregistered Ramsey test (v4.1, job d94hi4lgc6cc73ff4cc0, ibm_fez) of Quantum Resonance Theory's coherence-enhancement prediction (Anson 2024, DOI https://doi.org/10.5281/zenodo.21263717) returned G_A = +2.87% [+2.35%, +3.46%] at the theory's stated parameters (ωr = 10 MHz, R0 = 10⁻³ rad) — QRT-consistent, treated per its registration as an artifact pending escalation. A first escalation arm (v4.2, job d94j866vtlqs73fucnvg, ibm_marrakesh) excluded the amplitude-scaling mechanism class (10× arm: G_scale = +1.07% [−0.18%, +2.24%], DOES_NOT_SCALE) but, by its own declared scope, did not adjudicate the paper's point claim: its 1× replication index was preregistered as descriptive, underpowered (4-point grid), and carries no inferential weight. This study is the confirmatory adjudication of the point claim itself: the v4.1 protocol at full power (12-point delay grid, 6,000 shots, 72 circuits) on a second device (ibm_fez refused in code), with the primary index defined at the paper's stated amplitude and frequency against symmetric neighbor controls (9 and 11 MHz). Secondary arms (10× amplitude, per-pulse phase randomization, undriven reference) ride in the same packed session as declared descriptive vectors. The result will be published regardless of outcome. If the point claim does not replicate here, the v4.1 result is attributed to a device-specific or statistical artifact and the program's adjudication of QRT's coherence prediction closes; if it replicates, the effect survives to independent-party verification.

DOI: 10.17605/osf.io/jw9d6

Related papers

Preregistered Amplitude-Scaling and Multi-Vector Adjudication of a QRT-Consistent Coherence Result on Superconducting Hardware (v4.2)

Amber Anson

A preregistered Ramsey test (v4.1, job d94hi4lgc6cc73ff4cc0, ibm_fez, July 2026) of Quantum Resonance Theory's coherence-enhancement prediction (Anson 2024, DOI https://doi.org/10.5281/zenodo.21263717) returned a QRT-consistent result: G_A = +2.87%, 95% CI [+2.35%, +3.46%], at the theory-designated ωr = 10 MHz with R0 = 10⁻³ rad. Per that registration's own terms, the result is treated as an artifact until it survives escalation. This study executes the highest-information single-session adjudication available within remaining hardware quota, exploiting two facts: (1) a genuine drive-coupled effect must scale with drive amplitude, while calibration-floor and noise-spectrum artifacts at milliradian angles do not — so a 10× amplitude arm converts a marginal effect into a decisively resolvable one; (2) the packed six-qubit layout carries six conditions per circuit, so amplitude scaling, second-device replication at original amplitude, neighbor peak structure, drive-phase randomization, and an undriven reference all execute in the same 24 circuits. The primary decision rests on the amplitude arm alone; all other vectors are declared secondary and descriptive. The backend is required in code to differ from ibm_fez (the v4.1 device). The result will be published regardless of outcome.

Jul 4, 20260 citations

Preregistered Test of Quantum Resonance Theory's Coherence-Enhancement Prediction via Ramsey Interferometry: Corrected Implementation (v4.1)

Amber Anson

Quantum Resonance Theory (QRT; Anson 2024, DOI https://doi.org/10.5281/zenodo.21263717) predicts that superconducting-qubit coherence increases by 5–10% under a periodic physical drive at ωr = 10 MHz with amplitude R0 = 10⁻³ rad. A prior execution of this design (v4, job d940k9lgc6cc73feh290, July 2026) was found post-hoc to contain two implementation flaws — condition-to-qubit position confounding from incomplete rotation, and a decay-only fit model unable to represent Ramsey fringes from residual per-qubit detuning (~50–85 kHz observed) — and its verdict was withdrawn as improperly derived rather than claimed. This registration corrects the implementation with the hypothesis, parameters, frequencies, and drive amplitude unchanged: (1) full six-fold condition-to-qubit rotation, so every condition visits every physical qubit and position effects cancel in same-qubit contrasts; (2) a model-free primary index — the coherence-envelope area, the integral of |P(0) − 0.5| over the delay grid — which is monotone in coherence time, immune to fringe phase and detuning, and requires no curve fitting; (3) a same-qubit paired difference contrast as the primary comparison, which the v4 dataset demonstrated is the only clean comparison on this hardware. The drive is delivered as native fractional RX pulses (physical microwave drive) with a preregistered physicality assertion. Six conditions (four off-resonance frequencies, the theory-designated 10 MHz, and an undriven descriptive reference) run in parallel across six qubits per circuit. The result will be published regardless of outcome.

Jul 4, 20260 citations

Preregistered Ramsey Test of Quantum Resonance Theory's 10 MHz Coherence Prediction on Superconducting Quantum Hardware

Amber Anson

Quantum Resonance Theory (QRT; Anson 2024, DOI https://doi.org/10.5281/zenodo.21263717) predicts that resonance extends superconducting-qubit coherence: the source paper states that T2 should rise by 5-10% at wr = 10 MHz and identifies Ramsey interferometry as the appropriate quantum-computing test. The paper also gives a phase-amplitude form R(t) = R0 cos(wr t), with R0 approximately 10^-3 rad, and a Hamiltonian perturbation form H(t) = H0 + V0 cos(wr t), with V0 approximately 10^-3 eV. Two prior preregistered Bell-state tests were informative but implementation-limited. Test #1 used virtual RZ phase rotations and, at the theory-designated 10 MHz point, sampled the sine drive at zero crossings. Test #2 used physical fractional RX pulses and compared frequency structure across driven arms, but its stroboscopic schedule was still a discrete approximation near the Nyquist boundary rather than the continuous perturbation described in the source paper. This study is therefore a new, non-rescue preregistration focused on the paper's explicit Ramsey/T2 prediction rather than on the prior Bell-fidelity operationalization. This study tests whether a 10 MHz physically delivered periodic drive improves Ramsey coherence relative to matched off-resonance and no-drive controls. The physical-drive implementation follows the Test #2 correction in kind: the confirmatory hardware path uses native fractional RX pulses with use_fractional_gates=True, and it aborts if driven circuits transpile without native rx operations. The execution script will include a waveform audit before hardware submission: no confirmatory data collection begins unless the actual discrete schedule avoids zero-crossing lock-in, documents its effective frequency content, and samples the 10 MHz waveform with enough phase coverage to make the intended stimulus identifiable. The result will be reported publicly regardless of outcome.

Jul 3, 20260 citations

Preregistered Test of Quantum Resonance Theory's Bell-State Fidelity Prediction on Superconducting Quantum Hardware

Amber Anson

Quantum Resonance Theory (QRT; Anson 2024, DOI https://doi.org/10.5281/zenodo.21263717) predicts that a weak periodic resonance drive, Rθ = R0·sin(ωr·t) with R0 = 10⁻³ rad and ωr = 10 MHz, applied to an entangled pair during a hold period, increases Bell-state fidelity by approximately 5% relative to an undriven baseline (paper's stated figures: 0.95 → 0.9975). Standard quantum mechanics predicts no fidelity gain from this intervention: a milliradian-amplitude phase drive is approximately three orders of magnitude below the pulse amplitudes required for dynamical decoupling and constitutes a negligible coherent perturbation. The two theories therefore diverge on the outcome of a single inexpensive experiment. This study implements the QRT drive as stroboscopic phase modulation on IBM superconducting hardware in an interleaved A/B design with a frequency sweep, and tests the prediction against a preregistered decision rule. The result will be published regardless of outcome.

Jul 3, 20260 citations

Preregistered Test of Quantum Resonance Theory Under a Physically Delivered Periodic Drive on Superconducting Quantum Hardware

Amber Anson

Quantum Resonance Theory (QRT; Anson 2024, DOI https://doi.org/10.5281/zenodo.21263717) predicts a resonance enhancement of Bell-state fidelity under a periodic drive with amplitude R0 = 10⁻³ rad at ωr = 10 MHz. A prior preregistered test (Test #1, this author, July 2026) falsified the prediction as delivered by phase modulation, which superconducting hardware implements as virtual (non-physical) frame rotations. This study tests the physically delivered form of the drive, corresponding to the paper's Hamiltonian H = H0 + V0·cos(ωr·t): the drive is implemented as native fractional RX pulses on IBM Heron-class hardware — real microwave energy entering the device, modulated at the drive frequency — with a preregistered physicality assertion that aborts the run if the transpiler decomposes the drive into virtual rotations. Because physical pulses can degrade fidelity uniformly, the design compares fidelity ACROSS drive frequencies at identical amplitude schedules rather than driven-versus-idle: QRT's resonance mechanism predicts fidelity at 10 MHz elevated above off-resonance frequencies; standard quantum mechanics predicts a flat profile. This is a new prediction under a new registration, per the rules of Test #1's registration; it does not reopen or rescue Test #1's result. The result will be published regardless of outcome.

Jul 3, 20260 citations

Preregistered Balanced Packed Ramsey Test of Quantum Resonance Theory's 10 MHz Coherence Prediction on Superconducting Quantum Hardware

Amber Anson

Quantum Resonance Theory (https://doi.org/10.5281/zenodo.21263717) predicts that superconducting-qubit coherence should increase by 5-10% at wr = 10 MHz, with Ramsey interferometry identified as the appropriate quantum-computing test. This registration tests that same substantive claim using a revised execution design. The prior Ramsey preregistration used a serial circuit layout: one condition and one Ramsey delay per circuit. IBM runtime estimates made that layout difficult to execute within the available public allocation. This new registration uses a balanced packed layout: all six drive conditions are measured in parallel within each Ramsey-delay circuit, and the condition-to-qubit assignment is rotated across two planned repetitions to reduce fixed-qubit-position confounding. The hypothesis, waveform audit, physical fractional-RX drive implementation, T2* endpoint, fit model, control aggregate, and decision rule remain aligned with the prior Ramsey registration. The execution methodology is different enough that this is a new preregistration, not an amendment. The result will be reported publicly regardless of outcome.

Jul 3, 20260 citations