The question we are testing
Does moving the same hollow copper sphere between two fixed positions produce a repeatable resonance difference that survives controls for temperature, mounting, timing, and other measured conditions?
The registered LOC-1 design
LOC-1 is a controlled, same-laboratory comparison of a test sphere at two fixed positions, with a reference artifact and environmental monitoring. The optical readout tracks mechanical resonance; the registration defines the primary outcome, sampling plan, operational variables, and analysis rules. The OSF record controls if this overview and the registration differ.
The archived technical whitepaper describes the proposed instrumentation architecture, environmental controls, differential metrology, and bounded-null framework. It is background and technical context; the OSF registration is the source of truth for the preregistered study plan.
Controls before conclusions
- Thermal characterization: deliberately vary temperature during calibration to measure each artifact's frequency response.
- Mount repeatability: lift and replace the sphere at one position to quantify handling-induced scatter.
- Shared reference: track common clock and environmental drift while retaining independent checks for differential effects.
- Blinded labels: an independent position keeper holds the placement key; the analyst receives coded data.
- Sham moves and rotations: distinguish a positional signal from handling, orientation, and mounting effects.
- Environmental logs: preserve temperature, humidity, pressure, magnetic-field, and vibration records alongside measurement data.
Matched objects are not assumed to be identical, and no enclosure makes every extrinsic variable perfectly equal. Calibration and an uncertainty budget are essential. These controls do not change the registered plan unless a formal amendment is made and documented.
Proposed amendment: four-object acquisition New · Oct 9, 2026
LOC-1 as written compares a moved test sphere with one fixed reference. A proposed amendment, published for review on October 9, 2026, would capture four objects over the same observation window with a shared clock, so that drift, handling effects and reference-specific artifacts can be separated from a position contrast instead of inferred from sequential readings.
| Object | Proposed role | Diagnostic value |
|---|---|---|
| S1 | Stationary reference | Tracks shared drift |
| S2 | Test sphere moved between A and B | Primary position contrast |
| S3 | Second reference / comparator | Tests reference-specific artifacts |
| S4 | Sham-move control | Tests handling and settling effects |
- Concurrent channels, not device counts. Four objects need four genuinely simultaneous readout paths for each observable captured in parallel. A scanning instrument that visits four points in turn is not equivalent.
- Three layers of synchronization. A common frequency reference with logged lock status, shared or aligned sample clocks, and a hardware trigger with verified channel delays.
- Rotating roles. Scheduled calibration blocks swap object, channel and position assignments so each bias can be separated.
- Not six replications. Four objects yield six pairwise comparisons, but pairs sharing an object are correlated, so analysis stays at the trial or block level.
Read the full proposed protocol, the timing and false-positive error budget, and why Earth’s motion alone does not settle the question. For equipment categories see the hardware guide, and tell us what is missing in the community design review.
Official registration and finalized specifications
The sampling plan, variable definitions, primary outcome, exclusions, covariates, uncertainty treatment, multiplicity rules, stopping criteria, and analysis specifications have been finalized and published in the official OSF preregistration. The permanent record—not earlier working summaries or proposed amendments—governs the registered plan. The technical whitepaper is archived at Zenodo DOI 10.5281/zenodo.23583932. The preregistration is a plan, not a report of measurements or results.
How outcomes will be interpreted
Null at the tested sensitivity
A preregistered equivalence criterion can support a bounded null result under the tested conditions, not universal disproof of every interpretation of the source idea.
Candidate signal for follow-up
A signal must meet the registered decision criteria, survive controls and uncertainty checks, and show consistency across independent blocks. This would trigger investigation and replication, not a claim of new physics.
Inconclusive
If neither equivalence nor the signal criteria are met, or controls fail, report the study as inconclusive. Larger error bars must not be disguised as evidence of no effect.
Release and replication
The planned release includes raw measurements, environmental logs, metadata, analysis scripts, deviations, and a replication package. Results are to be published within 60 days of unblinding, regardless of outcome.
Review the research charter for the stage gates, browse the synchronized-measurement research pages, or offer replication expertise.