MUSEUM OF ONE / ENCOUNTER DIFFERENTIAL
DEMONSTRATION STUDY 1.0.0 / 21 OCTOBER 2026

START HERE
This is a constructed demonstration, not participant research. Four illustrative
participant codes each have clear (1x) and magnifying (2x) glass conditions:
eight completed trials. One scripted rehearsal is explicitly excluded. Nothing
here establishes gaze, preference, aesthetic quality or empirical validity.
The application itself is not included. No private research notebook was read.

1. Open accessible-study-report.html in a browser. It works offline and includes
   graph tables, source labels, formulas, exclusions and the protocol summary.
2. Inspect demonstration-protocol.json and regions-of-interest.json. The locked
   hash records the demo design; it is not an external preregistration. A 60-second
   trial and four constructed people are examples, not sample-size advice.
3. Open graph-*.csv for the derived accessible views, or evidence/tables/*.csv
   for original tidy projections. Full evidence/records/*.json is authoritative.
4. Verify every package file: python3 verify-demonstration.py
   Then check the original export and selected calculations:
   python3 evidence/verify-and-recompute.py
5. Regenerate graph data independently: python3 recompute-graph-tables.py
   This compares the recalculated values with every packaged graph CSV and JSON.
6. Read museum-pilot-draft-0.1.0.html as a separate proposed study. It is a draft
   with unresolved host/consent/retention details, not approval or registration.

MEASURES
Work share = selected milliseconds / active milliseconds * 100. Unassigned time
stays in the denominator. Cohort values average within each participant/condition
first, then across participants; these records contain one trial per condition.
Paired difference = magnifying share minus clear share, in percentage points.
Transitions count consecutive selections of different works; they are not gaze
paths. ROI dwell integrates the pointer centre, includes polygon boundaries and
is not the lens footprint. See evidence/data-dictionary.txt for full semantics.
All zeros shown are recorded/calculated; blank values mean unavailable.

SOURCE AND REUSE
Public documentation: https://www.museumofone.art/research-encounter-differential/
Source evidence ZIP SHA-256 is retained in graph-data.json.
The original export files and checksums are unmodified in evidence/.
Source image records are retained in the protocol and acquisition manifest.
Metropolitan Museum of Art Open Access images are the source collection. See
SOURCE-AND-REUSE.txt for individual source URLs and limits on other material.
No general software/data licence or DOI is asserted by this documentation kit.
Contact research@museumofone.art for tool/repository access and reuse terms.
