{
 "paper": "kammer-2026-foveal-feedback",
 "reviewed_on": "2026-09-12",
 "model": "supplied:runs/kammer-2026-foveal-feedback/edge-review.answer.json",
 "note": "Relations for claims that had none, asked of the claim set as it stands. Validated by edges.edges_from_raw, so the direction rules and the corpus vocabulary hold exactly as they do for edge-inference.",
 "edges": [
  {
   "source": "parametric-effects-not-global-fluctuations",
   "target": "ips-candidate-driver-foveal-feedback",
   "relation": "supports",
   "why": "The paper offers claim 13 (span results-078, Figure 4-figure supplement 1) as the reason its parametric-modulation effects are not global brain fluctuations or signal-to-noise, which is the confound that would otherwise undercut claim 10's reading of IPS as a candidate mediator."
  }
 ],
 "dropped_already_connected": []
}