Simulation trajectories and analysis code for corrective-information timing in coupled rumor-spreading networks
Description
This dataset supports a model-conditional study of corrective-information timing in coupled online and interpersonal rumor-spreading networks. It contains 930 previously audited scientific trajectories across the reference, return-route, propagation-scale, and response-duration experiments, plus 240 additional trajectories from an online-rewiring sensitivity experiment. Four technical controls and retained unsuccessful starts are identified separately and are not counted as scientific replications. The archive provides raw simulation outputs, queue and build manifests, analysis scripts, derived tables, figure-source outputs, and exploratory predictor diagnostics, including unfavorable transfer results. It excludes third-party row-level Community Notes data and the proprietary AnyLogic application. The included materials permit analysis replay from retained trajectories; an independent AnyLogic simulation rerun and fresh dependency installation have not been verified. The previously published Mendeley Data Version 1 record (doi: 10.17632/535w3ghw8m.1) covers earlier related materials but not this complete evidence set.
Files
Steps to reproduce
1. Download CNSNS_Data_20260925_v1_release.zip and verify its SHA-256: 8bc52b36b0532edd3adb8ebd9fa639ddacf85f15bc9ab36c0dc9d65c98a4d510. Extract the archive and compare every project file to MANIFEST.json. 2. Use Python 3.12 with the package versions in project/requirements.txt; change into project/. 3. Run experiment/audit_complete.py, information_sciences_extension/audit_progress.py, information_sciences_extension/range_stage/audit.py, information_sciences_extension/duration_stage/audit.py, post_INS_20260924/topology_stage/pro897_analyze.py, preactivation_audit.py and headroom_audit.py. 4. Re-run the analysis scripts in experiment/, information_sciences_extension/, and post_INS_20260924/stage_diagnostic/ (including scale_adjusted_sensitivity and topology_transfer), then compare result tables with the archived files. 5. Run post_INS_20260924/topology_stage/plot_topology.py to regenerate the topology figure. Analysis from retained trajectories was checked in a separate extraction using existing matching dependencies. A fresh dependency installation and third-party AnyLogic simulation rerun were not verified; the proprietary AnyLogic runtime is not included. External Community Notes row-level data are available from their original OSF source, not this archive.