Research Design: Community falsification matrix and signal-detection protocol modules

[ProhibitedTV]

Parent roadmap: #3
Related: #12, #13, #14, #22, #23

Objective

Capture the useful parts of the community's proposed falsification matrix as LaserLab analysis modules and protocol templates. The goal is not to endorse every theoretical claim; the goal is to turn community hypotheses into testable, conservative, well-labeled protocol options.

This issue should preserve the scientific intent of the community discussion while filtering out unsafe, speculative, or non-software-operational content.

Community-source themes to encode safely

The shared community notes propose moving beyond informal reports by building tests that can separate:

• wall-anchored structure vs eye/retina/cortical artifacts;
• external spatial coordinate claims vs internally generated percepts;
• signal sensitivity vs liberal response bias;
• laser-specific effects vs generic visual-noise pareidolia;
• intersubjective agreement vs individual report variability.

LaserLab should treat these as hypothesis tests, not proof claims.

Protocol module candidates

1. Gaze stability / microsaccade decoupling

Question: Does the reported symbol/structure stay anchored to wall coordinates or move with gaze?

Implementation support:

• Optional eye-tracking import fields: gaze x/y, timestamp, confidence.
• Low-cost fixed-fixation metadata: wall marker coordinates, fixation target notes.
• Report overlay: candidate ROI vs gaze position over time.

Acceptance:

• Reports can show whether candidate coordinates remain stable relative to the projection surface.
• UI labels this as a visual-stability analysis, not proof of externality.

2. Motion parallax / Z-axis mapping

Question: Does a reported symbol behave like it occupies a wall plane, intermediate plane, or subject-relative plane?

Implementation support:

• Optional head/camera position metadata.
• Low-cost reference-object metadata: string, glass mark, tripod, grid, or wall marker.
• Report fields for parallax observations and uncertainty.

Acceptance:

• LaserLab can store and report parallax protocol context without overclaiming depth reconstruction.

3. Focus / accommodation context

Question: Does perceived sharpness track the wall, a foreground target, or neither?

Implementation support:

• Metadata for foreground object distance, wall distance, focus instructions, camera focus state.
• Optional structured subjective annotations: sharp_on_wall, sharp_on_foreground, ambiguous, not_collected.

Acceptance:

• Reports separate observer focus reports from image/video detector evidence.

4. Binocular consistency

Question: Are reports stable between left/right eye viewing, or do they behave like monocular/entoptic phenomena?

Implementation support:

• Session metadata for left-eye/right-eye/alternating-eye observations.
• Optional coordinate transcriptions per eye.
• Report comparison of reported coordinates/symbols by viewing condition.

Acceptance:

• The app can store binocular protocol data without treating it as detector evidence.

5. Polarization / entoptic exclusion

Question: Does perceived structure rotate or change with head tilt or polarization changes?

Implementation support:

• Metadata for polarizer presence, angle, head-tilt condition, LCD/polarizer test notes.
• Warnings that these are exploratory controls and require safe optical handling.

Acceptance:

• Reports can compare candidate observations across polarizer/head-tilt conditions.

6. Carrier interaction / Moiré / occlusion tests

Question: Does a physical mesh, grating, grid overlay, or occluder interact with the perceived structure?

Implementation support:

• Capture metadata for mesh/grating/transparent grid overlays.
• ROI fields for grid coordinates.
• Computer-vision support for grid registration if visible in footage.

Acceptance:

• LaserLab can compare symbol/structure reports against visible grid coordinates and physical overlay conditions.

7. Signal Detection Theory / Gaussian noise controls

Question: Is a participant more sensitive to a real weak signal, or simply more likely to report seeing structure?

Implementation support:

• Trial schema: stimulus type, embedded-signal truth, participant response, confidence, timestamp.
• SDT outputs: hit rate, false alarm rate, miss rate, correct rejection rate, d-prime, criterion/bias estimate.
• Separate visual-noise controls from laser-wall projection runs.

Acceptance:

• Reports distinguish sensitivity (d') from response bias/criterion.
• No run can claim improved detection without false-alarm controls.

8. PRNG/noise-loop recognition

Question: Can participants detect repeated structure in controlled visual noise better than chance?

Implementation support:

• Metadata for generated stimuli, seed, loop interval, and ground-truth hidden pattern.
• Blind randomized trial support.
• Report chance-baseline performance and confidence intervals.

Acceptance:

• Results are framed as psychophysics task outcomes, not evidence of metaphysical content.

9. Temporal modulation / bandwidth tests

Question: Can participants detect changes in a modulated visual stimulus beyond ordinary conscious report thresholds?

Implementation support:

• Store modulation frequency, waveform, encoded symbol/message metadata, and hardware/controller notes.
• Require safe laser modulation warnings and avoid eye-level exposure.
• Prefer non-laser display fixtures for software tests unless safety reviewed externally.

Acceptance:

• Reports compare participant responses against hidden ground truth and chance baselines.
• No documentation includes unsafe laser construction or exposure instructions.

10. Intersubjective synchronization / grid coordinate transcription

Question: Do independent observers report the same symbol at the same wall coordinate without communication?

Implementation support:

• Grid overlay coordinate system for wall projection footage.
• Pseudonymous participant IDs.
• Independent transcription records with timestamps.
• Agreement statistics: exact match, coordinate-distance tolerance, symbol-class similarity, chance baseline.

Acceptance:

• Reports can compare intersubjective agreement without revealing identities or overclaiming consensus.

11. Wavelength / surface / shielding metadata

Question: Do reports or detector outputs change with wavelength, surface, optical path, or environment?

Implementation support:

• Extend Research Design: Wall-projection protocol metadata, safety, and subjective-report context #23 metadata fields for wavelength, surface, lighting, optical component, and shielding/environment notes.
• Treat Faraday/mu-metal/shielding claims as environment metadata only unless a formal validated setup exists.

Acceptance:

• Reports avoid claims about EM/neutrino/simulation mechanisms unless the run only states the measured variable and observed outcome.

12. Symbol transcription statistics

Question: Do reported glyph distributions look structured beyond random response generation?

Implementation support:

• Store symbol transcription sequences.
• Compute simple descriptive stats: symbol counts, entropy, repeat rate, transition matrix, Zipf-like rank-frequency fit with caveats.
• Add warnings that Zipf-like behavior alone does not prove language or external information.

Acceptance:

• Reports can present transcription stats with null models and caveats.

Data model tasks

• Add a protocol_module field to runs or sessions.
• Add subjective_trial records separate from detector_result records.
• Add optional grid_coordinate, eye_condition, focus_condition, head_position_condition, stimulus_truth, response, and confidence fields.
• Add generated-stimulus provenance: seed, frame count, frequency, embedded-signal truth, and checksum.
• Add explicit privacy/redaction warnings for community-submitted subjective reports.

UI tasks

• Add an expandable Falsification Modules section under Study Context / Capture Protocol.
• Keep the default UI simple: Wall projection, Matched controls, Grid transcription, Subjective report attached.
• Put advanced modules behind an Experimental psychophysics modules section.
• Show clear badges for footage evidence, subjective report, psychophysics trial, and speculative mechanism.

Report tasks

• Include a Falsification matrix section showing which alternative explanations were tested and which were not.
• Clearly separate:
  1. camera/image detector evidence;
  2. subjective reports;
  3. psychophysics trial outcomes;
  4. speculative interpretation notes.
• Never phrase one successful module as proof of external origin.
• Include not tested rows for major confounds, rather than silently omitting them.

Safety / scope boundaries

• Do not include DMT extraction, procurement, dosing, or administration instructions.
• Do not provide instructions for unsafe laser construction or direct viewing.
• Treat participant work as requiring appropriate legal, ethical, and safety review outside the app.
• Treat speculative mechanisms such as simulation, neutrino, microtubule, EM pickup, or metaphysical language as report-context labels only, not as validated scientific conclusions.

Acceptance criteria

• LaserLab supports at least one concrete falsification module beyond generic laser/control comparison, preferably grid-coordinate transcription or SDT/noise-control trials.
• Subjective reports and psychophysics responses remain separate from image/video detector scoring.
• Reports include a falsification matrix with tested/not-tested confounds.
• UI stays simple for normal wall-projection footage analysis while allowing advanced community-science modules.
• Documentation preserves scientific skepticism and does not embed unsafe or illegal operational instructions.

Notes

The community is generating useful falsification ideas. LaserLab should not absorb them as beliefs; it should turn them into structured, auditable, falsifiable protocols.