{
  "handle": "StoppingRule",
  "gloss": "Stop integrating stressor handling into a design once it survives a fresh round of unseen simulated stressors, or once further integration costs more than it buys; tested by experiment, not proven from a probability distribution",
  "mechanism": "To make a design robust without predicting the future, imagine many possible {{exogenous_stress}} events, including absurd ones, and change the design so it survives each, integrating the changes into one structure. The question is when to stop. A statistical stopping rule assumes a distribution and stops when estimates are accurate enough; here no probabilities are assigned, so that rule cannot apply. Instead, after integrating one round, test the design against a new round of unseen simulated stressors. Complex systems tend to settle into a small number of stable states that each survive many different inputs, so a design that survives enough stressors starts surviving ones it has never met: it has generalised, or reached criticality. When most of a fresh round is survived, stop. Cost is the second dimension: integration stops earlier if adding more costs more than it protects, and building partly integrated functions, as a {{fractional_meta_model}} does, lowers that cost so the generalisation point can actually be reached.",
  "invariants": [
    "No probability is assigned to stressors; the rule is checked by experiment",
    "Stopping is signalled by survival of stressors not used in the design",
    "Integration cost is weighed alongside generalisation"
  ],
  "preconditions": [
    "Stressors can be simulated and the design's survival of each can be judged"
  ],
  "failure_modes": [
    "Stopping after the stressors used in design are survived, without testing unseen ones",
    "Ignoring integration cost so the process never reaches a usable design",
    "Re-introducing probability estimates and missing possible but unestimable stressors"
  ],
  "_meta": {
    "path": [
      "Mind",
      "Strategy"
    ],
    "ring": 2,
    "tier": 2,
    "source": "Barry O'Reilly, Residuality Theory; Stuart Kauffman (criticality, attractors); Adam Timlett, On the Origin of Risk (2025), Technical Appendix I p.139-147"
  },
  "sema_id": "sema:StoppingRule#mh:SHA-256:9a1d4616af79fa4b654e36ec26704887ff8f8ac246c1bf71be5207948aa863fc",
  "sema_ref": "StoppingRule#9a1d",
  "sema_stub": "9a1d",
  "dependencies": {
    "references": {
      "exogenous_stress": "sema:ExogenousStress#mh:SHA-256:6a33d65f1361a95ad72b8a8d59c90bf253e08f804e2a79532977d5113987b672",
      "fractional_meta_model": "sema:FractionalMetaModel#mh:SHA-256:b25a96009ba8797d2c4a8673cebfe38ebb0326d237a3669a2a2d7f56aeb5e2c2"
    }
  },
  "sema_layer": "Mind",
  "sema_category": "Strategy",
  "used_by": {}
}