prediction_id,hypothesis_id,rival_hypothesis_ids,conditions,observable,expected_if_focal,expected_if_rivals,falsifier,indeterminate_result,boundary_conditions,measurement_ids,negative_control_ids,analysis_ids,uncertainty
P1,H1,H2,"Random assignment, successful manipulation, calibrated measurements, and masked validation","Ordering of synthetic intermediate M and outcome Y","Intermediate M changes before outcome Y in condition X","H2 predicts no condition-specific intermediate change and a shift shared with the reference quantity","Outcome Y changes without the prespecified intermediate change after successful manipulation and valid measurement","Manipulation or measurement check fails, or uncertainty is too large to distinguish the patterns","Synthetic eligible units over one prespecified interval",M-INTERMEDIATE;M-OUTCOME,NC1,A1,"Magnitude is not assumed; the planned interval and measurement resolution determine whether patterns are distinguishable"
P2,H2,H1,"Matched synthetic acquisition and calibration workflow","Shift in the synthetic negative-control reference quantity","Reference quantity and outcome Y shift together by acquisition period","H1 predicts outcome Y changes without a corresponding reference shift","Reference quantity is stable across independently calibrated periods while outcome Y changes reproducibly by assignment","Reference measurement fails or acquisition periods do not overlap assigned conditions","Synthetic acquisition batches covered by the declared calibration process",M-NEGATIVE-CONTROL;M-OUTCOME,NC1,A2,"A control shift can have causes other than the stated measurement-drift rival"