A laboratory for explicit assumptions.

World simulates choices under declared behavioural models. It is not empirically calibrated and does not predict real people’s actions with validated accuracy.

People are synthetic agents. A chosen alternative is a model decision; the realised monetary outcome is the subsequent recorded change in wealth. Different parameter settings need not produce different choices.

Scientific foundations

Expected utility is a reference framework. Prospect theory was developed by Kahneman and Tversky (1979); cumulative prospect theory by Tversky and Kahneman (1992). Samuelson and Zeckhauser (1988) studied status-quo bias. World’s additive status-quo premium is an explicit modelling assumption, not a parameter estimated by those papers.

Reproducibility and records

Saved specifications, seeds, policy versions, provenance and individual traces remain available in Results → Details. Repeated runs describe simulation variation, not confidence in predictions about human populations. Friendly names are local to the browser; saved canonical agent identifiers are authoritative.

System status

Advanced and compatibility workspaces

Advanced reuses the existing version-2 builder and its validation: distributions, correlated sampling, subgroups, sensitivity sweeps, model comparisons, definitions, background jobs and detailed exports. These controls are not yet incorporated into the guided People editor.

Template compatibility retains the version-1 template laboratory. Its original URL and API identifiers are preserved; it is not redirected into a different scientific contract.