{
  "schema": "pw365.observed.2",
  "observedAt": "2026-10-19T11:51:51.289874+00:00",
  "frozenAt": "2026-10-19T11:47:25.251291+00:00",
  "kind": "Actual screenshot-only model responses, not simulated observations",
  "readers": 4,
  "cases": 16,
  "overall": {
    "responses": 32,
    "correct": 32,
    "confident_errors": 0,
    "evidence_supported": 32,
    "confidence_values": [
      100
    ]
  },
  "by_view": {
    "table": {
      "responses": 16,
      "correct": 16,
      "confident_errors": 0,
      "evidence_supported": 16,
      "confidence_values": [
        100
      ]
    },
    "aligned": {
      "responses": 16,
      "correct": 16,
      "confident_errors": 0,
      "evidence_supported": 16,
      "confidence_values": [
        100
      ]
    }
  },
  "by_reader": {
    "R1": {
      "responses": 8,
      "correct": 8,
      "confident_errors": 0,
      "evidence_supported": 8,
      "confidence_values": [
        100
      ]
    },
    "R2": {
      "responses": 8,
      "correct": 8,
      "confident_errors": 0,
      "evidence_supported": 8,
      "confidence_values": [
        100
      ]
    },
    "R3": {
      "responses": 8,
      "correct": 8,
      "confident_errors": 0,
      "evidence_supported": 8,
      "confidence_values": [
        100
      ]
    },
    "R4": {
      "responses": 8,
      "correct": 8,
      "confident_errors": 0,
      "evidence_supported": 8,
      "confidence_values": [
        100
      ]
    }
  },
  "paired_cases": [
    {
      "case_id": "K01",
      "family": "touching",
      "expected": "Yes",
      "table": {
        "reader_id": "R1",
        "position": 1,
        "case_id": "K01",
        "view": "table",
        "expected": "Yes",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K01",
          "decision": "Yes",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Preparations and consent are confirmed. Alder's bookings meet at 09:00, Cedar's at 10:00, and Birch's at 09:30, with no overlap."
        },
        "evidence_supported": true
      },
      "aligned": {
        "reader_id": "R3",
        "position": 8,
        "case_id": "K01",
        "view": "aligned",
        "expected": "Yes",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K01",
          "decision": "Yes",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Preparations are complete and consent is approved. Alder's bookings meet at 09:00, Birch's at 09:30, and Cedar's at 10:00. The record explicitly permits an interval to end exactly when another starts."
        },
        "evidence_supported": true
      }
    },
    {
      "case_id": "K02",
      "family": "parallel_crews",
      "expected": "Yes",
      "table": {
        "reader_id": "R3",
        "position": 7,
        "case_id": "K02",
        "view": "table",
        "expected": "Yes",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K02",
          "decision": "Yes",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Preparations are complete and consent is approved. Fern's first booking ends at 09:00 before its next starts at 09:10. Hazel's bookings are 08:00–09:00 and 10:00–11:00; Elm's are 08:00–09:00 and 10:10–11:10. No same-crew intervals overlap."
        },
        "evidence_supported": true
      },
      "aligned": {
        "reader_id": "R1",
        "position": 2,
        "case_id": "K02",
        "view": "aligned",
        "expected": "Yes",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K02",
          "decision": "Yes",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Preparations and consent are confirmed. Elm has bookings at 08:00–09:00 and 10:10–11:10; Fern at 08:00–09:00 and 09:10–10:00; Hazel at 08:00–09:00 and 10:00–11:00. None overlap within a crew."
        },
        "evidence_supported": true
      }
    },
    {
      "case_id": "K03",
      "family": "nested",
      "expected": "No",
      "table": {
        "reader_id": "R3",
        "position": 6,
        "case_id": "K03",
        "view": "table",
        "expected": "No",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K03",
          "decision": "No",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Crew Ivy's B2 runs from 09:00 to 09:30 within its B4 booking of 08:00 to 10:30. Ivy therefore has two bookings in progress simultaneously."
        },
        "evidence_supported": true
      },
      "aligned": {
        "reader_id": "R1",
        "position": 3,
        "case_id": "K03",
        "view": "aligned",
        "expected": "No",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K03",
          "decision": "No",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Crew Ivy's B4 runs 08:00–10:30 while its B2 runs 09:00–09:30, creating a same-crew overlap."
        },
        "evidence_supported": true
      }
    },
    {
      "case_id": "K04",
      "family": "five_minute",
      "expected": "No",
      "table": {
        "reader_id": "R1",
        "position": 4,
        "case_id": "K04",
        "view": "table",
        "expected": "No",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K04",
          "decision": "No",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Crew Maple's B2 runs 08:00–09:05 and B4 runs 09:00–10:00, overlapping from 09:00–09:05."
        },
        "evidence_supported": true
      },
      "aligned": {
        "reader_id": "R3",
        "position": 5,
        "case_id": "K04",
        "view": "aligned",
        "expected": "No",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K04",
          "decision": "No",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Crew Maple's B2 runs until 09:05 while its B4 starts at 09:00, creating an overlap from 09:00 to 09:05."
        },
        "evidence_supported": true
      }
    },
    {
      "case_id": "K05",
      "family": "touching_chain",
      "expected": "Yes",
      "table": {
        "reader_id": "R1",
        "position": 5,
        "case_id": "K05",
        "view": "table",
        "expected": "Yes",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K05",
          "decision": "Yes",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Preparations and consent are confirmed. Rowan's three bookings meet at 09:00 and 10:00; Spruce's meet at 09:50; Willow has one booking. There is no same-crew overlap."
        },
        "evidence_supported": true
      },
      "aligned": {
        "reader_id": "R3",
        "position": 4,
        "case_id": "K05",
        "view": "aligned",
        "expected": "Yes",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K05",
          "decision": "Yes",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Preparations are complete and consent is approved. Rowan's three bookings meet at 09:00 and 10:00, and Spruce's two meet at 09:50. Exact boundary handovers are permitted. Willow has only one booking."
        },
        "evidence_supported": true
      }
    },
    {
      "case_id": "K06",
      "family": "overlap_chain",
      "expected": "No",
      "table": {
        "reader_id": "R3",
        "position": 3,
        "case_id": "K06",
        "view": "table",
        "expected": "No",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K06",
          "decision": "No",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Crew Yew's B3 and B5 overlap from 09:00 to 09:20. Its B5 and B2 also overlap from 10:10 to 10:20. The plan therefore fails the crew condition."
        },
        "evidence_supported": true
      },
      "aligned": {
        "reader_id": "R1",
        "position": 6,
        "case_id": "K06",
        "view": "aligned",
        "expected": "No",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K06",
          "decision": "No",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Crew Yew's B3 and B5 overlap from 09:00–09:20; its B5 and B2 also overlap from 10:10–10:20."
        },
        "evidence_supported": true
      }
    },
    {
      "case_id": "K07",
      "family": "small_gaps",
      "expected": "Yes",
      "table": {
        "reader_id": "R3",
        "position": 2,
        "case_id": "K07",
        "view": "table",
        "expected": "Yes",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K07",
          "decision": "Yes",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Preparations are complete and consent is approved. Holly's bookings end at 08:55 and resume at 09:00; Lime's end at 09:50 and resume at 09:55; Laurel's end at 09:35 and resume at 09:40. No crew has overlapping bookings."
        },
        "evidence_supported": true
      },
      "aligned": {
        "reader_id": "R1",
        "position": 7,
        "case_id": "K07",
        "view": "aligned",
        "expected": "Yes",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K07",
          "decision": "Yes",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Preparations and consent are confirmed. Holly's bookings have a gap from 08:55–09:00, Laurel's from 09:35–09:40, and Lime's from 09:50–09:55. No crew has overlapping bookings."
        },
        "evidence_supported": true
      }
    },
    {
      "case_id": "K08",
      "family": "two_crews",
      "expected": "No",
      "table": {
        "reader_id": "R1",
        "position": 8,
        "case_id": "K08",
        "view": "table",
        "expected": "No",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K08",
          "decision": "No",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Crew Poplar's B2 and B3 overlap from 09:10–09:30, and Crew Sallow's B4 and B5 overlap from 09:50–10:10."
        },
        "evidence_supported": true
      },
      "aligned": {
        "reader_id": "R3",
        "position": 1,
        "case_id": "K08",
        "view": "aligned",
        "expected": "No",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K08",
          "decision": "No",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Crew Poplar's B2 and B3 overlap from 09:10 to 09:30. Crew Sallow's B4 and B5 also overlap from 09:50 to 10:10. These violate the one-booking-per-crew condition."
        },
        "evidence_supported": true
      }
    },
    {
      "case_id": "K09",
      "family": "touching",
      "expected": "Yes",
      "table": {
        "reader_id": "R2",
        "position": 1,
        "case_id": "K09",
        "view": "table",
        "expected": "Yes",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K09",
          "decision": "Yes",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Preparations are complete and consent is approved. Alder's bookings meet at 10:15, Birch's at 09:15 and Cedar's at 09:45. The rules explicitly permit these boundaries without overlap."
        },
        "evidence_supported": true
      },
      "aligned": {
        "reader_id": "R4",
        "position": 8,
        "case_id": "K09",
        "view": "aligned",
        "expected": "Yes",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K09",
          "decision": "Yes",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Preparations are complete and consent approved. Alder's bookings meet at 10:15, Birch's at 09:15, and Cedar's at 09:45. These exact endpoint handovers are allowed and create no overlap."
        },
        "evidence_supported": true
      }
    },
    {
      "case_id": "K10",
      "family": "parallel_crews",
      "expected": "Yes",
      "table": {
        "reader_id": "R4",
        "position": 7,
        "case_id": "K10",
        "view": "table",
        "expected": "Yes",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K10",
          "decision": "Yes",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Preparations are complete and consent approved. Fern's earlier booking ends at 09:15 and its next starts at 10:25; Hazel's end and restart are 09:15 and 09:25; Elm's are 09:15 and 10:15. No same-crew intervals overlap."
        },
        "evidence_supported": true
      },
      "aligned": {
        "reader_id": "R2",
        "position": 2,
        "case_id": "K10",
        "view": "aligned",
        "expected": "Yes",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K10",
          "decision": "Yes",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Preparations are complete and consent is approved. Elm's first booking ends at 09:15 and its next starts at 10:15; Fern's ends at 09:15 and its next starts at 10:25; Hazel's ends at 09:15 and its next starts at 09:25. No crew has overlapping bookings."
        },
        "evidence_supported": true
      }
    },
    {
      "case_id": "K11",
      "family": "nested",
      "expected": "No",
      "table": {
        "reader_id": "R4",
        "position": 6,
        "case_id": "K11",
        "view": "table",
        "expected": "No",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K11",
          "decision": "No",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Crew Juniper's B1 runs from 08:15 to 10:45, encompassing its B6 booking from 09:15 to 09:45. Juniper therefore has two bookings in progress during that interval."
        },
        "evidence_supported": true
      },
      "aligned": {
        "reader_id": "R2",
        "position": 3,
        "case_id": "K11",
        "view": "aligned",
        "expected": "No",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K11",
          "decision": "No",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Crew Juniper is assigned B1 from 08:15–10:45 and B6 from 09:15–09:45. Both bookings are in progress from 09:15–09:45, violating the one-booking-per-crew condition."
        },
        "evidence_supported": true
      }
    },
    {
      "case_id": "K12",
      "family": "five_minute",
      "expected": "No",
      "table": {
        "reader_id": "R2",
        "position": 4,
        "case_id": "K12",
        "view": "table",
        "expected": "No",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K12",
          "decision": "No",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Crew Oak is assigned B4 from 08:15–09:20 and B1 from 09:15–10:15. These overlap from 09:15–09:20."
        },
        "evidence_supported": true
      },
      "aligned": {
        "reader_id": "R4",
        "position": 5,
        "case_id": "K12",
        "view": "aligned",
        "expected": "No",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K12",
          "decision": "No",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Crew Oak's B4 runs from 08:15 to 09:20, while its B1 starts at 09:15. They overlap from 09:15 to 09:20, violating the one-booking-at-a-time condition."
        },
        "evidence_supported": true
      }
    },
    {
      "case_id": "K13",
      "family": "touching_chain",
      "expected": "Yes",
      "table": {
        "reader_id": "R2",
        "position": 5,
        "case_id": "K13",
        "view": "table",
        "expected": "Yes",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K13",
          "decision": "Yes",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Preparations are complete and consent is approved. Spruce's three bookings meet at 09:15 and 10:15; Willow's two bookings meet at 10:05; Rowan has one booking. No same-crew bookings overlap."
        },
        "evidence_supported": true
      },
      "aligned": {
        "reader_id": "R4",
        "position": 4,
        "case_id": "K13",
        "view": "aligned",
        "expected": "Yes",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K13",
          "decision": "Yes",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Preparations are complete and consent approved. Spruce's bookings meet exactly at 09:15 and 10:15; Willow's meet exactly at 10:05. Rowan has one booking. Exact endpoint handovers are explicitly allowed."
        },
        "evidence_supported": true
      }
    },
    {
      "case_id": "K14",
      "family": "overlap_chain",
      "expected": "No",
      "table": {
        "reader_id": "R4",
        "position": 3,
        "case_id": "K14",
        "view": "table",
        "expected": "No",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K14",
          "decision": "No",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Crew Ash's B5 overlaps B1 from 09:15 to 09:35, and B1 overlaps B2 from 10:25 to 10:35. The plan therefore violates the one-booking-at-a-time condition."
        },
        "evidence_supported": true
      },
      "aligned": {
        "reader_id": "R2",
        "position": 6,
        "case_id": "K14",
        "view": "aligned",
        "expected": "No",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K14",
          "decision": "No",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Crew Ash's B5 and B1 overlap from 09:15–09:35. Its B1 and B2 also overlap from 10:25–10:35. These violate the one-booking-per-crew condition."
        },
        "evidence_supported": true
      }
    },
    {
      "case_id": "K15",
      "family": "small_gaps",
      "expected": "Yes",
      "table": {
        "reader_id": "R4",
        "position": 2,
        "case_id": "K15",
        "view": "table",
        "expected": "Yes",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K15",
          "decision": "Yes",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Preparations are complete and consent approved. Lime's bookings end at 09:50 and resume at 09:55; Laurel's end at 09:10 and resume at 09:15; Holly's end at 10:05 and resume at 10:10. No crew has overlapping bookings."
        },
        "evidence_supported": true
      },
      "aligned": {
        "reader_id": "R2",
        "position": 7,
        "case_id": "K15",
        "view": "aligned",
        "expected": "Yes",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K15",
          "decision": "Yes",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Preparations are complete and consent is approved. Holly has a gap from 10:05–10:10, Laurel from 09:10–09:15 and Lime from 09:50–09:55 between their bookings. No crew has overlapping bookings."
        },
        "evidence_supported": true
      }
    },
    {
      "case_id": "K16",
      "family": "two_crews",
      "expected": "No",
      "table": {
        "reader_id": "R2",
        "position": 8,
        "case_id": "K16",
        "view": "table",
        "expected": "No",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K16",
          "decision": "No",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Crew Sallow's B1 and B3 overlap from 09:25–09:45. Crew Walnut's B2 and B4 overlap from 10:05–10:25. Either overlap prevents authorisation exactly as listed."
        },
        "evidence_supported": true
      },
      "aligned": {
        "reader_id": "R4",
        "position": 1,
        "case_id": "K16",
        "view": "aligned",
        "expected": "No",
        "correct": true,
        "confident_error": false,
        "response": {
          "case_id": "K16",
          "decision": "No",
          "confidence": 100,
          "needs_checking": "Nothing further under the stated rules",
          "evidence": "Crew Sallow's B1 and B3 overlap from 09:25 to 09:45. Crew Walnut's B2 and B4 also overlap from 10:05 to 10:25. These conflicts violate the one-booking-at-a-time condition."
        },
        "evidence_supported": true
      }
    }
  ],
  "rows": [
    {
      "reader_id": "R1",
      "position": 1,
      "case_id": "K01",
      "view": "table",
      "expected": "Yes",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K01",
        "decision": "Yes",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Preparations and consent are confirmed. Alder's bookings meet at 09:00, Cedar's at 10:00, and Birch's at 09:30, with no overlap."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R1",
      "position": 2,
      "case_id": "K02",
      "view": "aligned",
      "expected": "Yes",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K02",
        "decision": "Yes",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Preparations and consent are confirmed. Elm has bookings at 08:00–09:00 and 10:10–11:10; Fern at 08:00–09:00 and 09:10–10:00; Hazel at 08:00–09:00 and 10:00–11:00. None overlap within a crew."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R1",
      "position": 3,
      "case_id": "K03",
      "view": "aligned",
      "expected": "No",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K03",
        "decision": "No",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Crew Ivy's B4 runs 08:00–10:30 while its B2 runs 09:00–09:30, creating a same-crew overlap."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R1",
      "position": 4,
      "case_id": "K04",
      "view": "table",
      "expected": "No",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K04",
        "decision": "No",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Crew Maple's B2 runs 08:00–09:05 and B4 runs 09:00–10:00, overlapping from 09:00–09:05."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R1",
      "position": 5,
      "case_id": "K05",
      "view": "table",
      "expected": "Yes",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K05",
        "decision": "Yes",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Preparations and consent are confirmed. Rowan's three bookings meet at 09:00 and 10:00; Spruce's meet at 09:50; Willow has one booking. There is no same-crew overlap."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R1",
      "position": 6,
      "case_id": "K06",
      "view": "aligned",
      "expected": "No",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K06",
        "decision": "No",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Crew Yew's B3 and B5 overlap from 09:00–09:20; its B5 and B2 also overlap from 10:10–10:20."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R1",
      "position": 7,
      "case_id": "K07",
      "view": "aligned",
      "expected": "Yes",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K07",
        "decision": "Yes",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Preparations and consent are confirmed. Holly's bookings have a gap from 08:55–09:00, Laurel's from 09:35–09:40, and Lime's from 09:50–09:55. No crew has overlapping bookings."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R1",
      "position": 8,
      "case_id": "K08",
      "view": "table",
      "expected": "No",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K08",
        "decision": "No",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Crew Poplar's B2 and B3 overlap from 09:10–09:30, and Crew Sallow's B4 and B5 overlap from 09:50–10:10."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R2",
      "position": 1,
      "case_id": "K09",
      "view": "table",
      "expected": "Yes",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K09",
        "decision": "Yes",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Preparations are complete and consent is approved. Alder's bookings meet at 10:15, Birch's at 09:15 and Cedar's at 09:45. The rules explicitly permit these boundaries without overlap."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R2",
      "position": 2,
      "case_id": "K10",
      "view": "aligned",
      "expected": "Yes",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K10",
        "decision": "Yes",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Preparations are complete and consent is approved. Elm's first booking ends at 09:15 and its next starts at 10:15; Fern's ends at 09:15 and its next starts at 10:25; Hazel's ends at 09:15 and its next starts at 09:25. No crew has overlapping bookings."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R2",
      "position": 3,
      "case_id": "K11",
      "view": "aligned",
      "expected": "No",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K11",
        "decision": "No",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Crew Juniper is assigned B1 from 08:15–10:45 and B6 from 09:15–09:45. Both bookings are in progress from 09:15–09:45, violating the one-booking-per-crew condition."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R2",
      "position": 4,
      "case_id": "K12",
      "view": "table",
      "expected": "No",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K12",
        "decision": "No",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Crew Oak is assigned B4 from 08:15–09:20 and B1 from 09:15–10:15. These overlap from 09:15–09:20."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R2",
      "position": 5,
      "case_id": "K13",
      "view": "table",
      "expected": "Yes",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K13",
        "decision": "Yes",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Preparations are complete and consent is approved. Spruce's three bookings meet at 09:15 and 10:15; Willow's two bookings meet at 10:05; Rowan has one booking. No same-crew bookings overlap."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R2",
      "position": 6,
      "case_id": "K14",
      "view": "aligned",
      "expected": "No",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K14",
        "decision": "No",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Crew Ash's B5 and B1 overlap from 09:15–09:35. Its B1 and B2 also overlap from 10:25–10:35. These violate the one-booking-per-crew condition."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R2",
      "position": 7,
      "case_id": "K15",
      "view": "aligned",
      "expected": "Yes",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K15",
        "decision": "Yes",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Preparations are complete and consent is approved. Holly has a gap from 10:05–10:10, Laurel from 09:10–09:15 and Lime from 09:50–09:55 between their bookings. No crew has overlapping bookings."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R2",
      "position": 8,
      "case_id": "K16",
      "view": "table",
      "expected": "No",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K16",
        "decision": "No",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Crew Sallow's B1 and B3 overlap from 09:25–09:45. Crew Walnut's B2 and B4 overlap from 10:05–10:25. Either overlap prevents authorisation exactly as listed."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R3",
      "position": 1,
      "case_id": "K08",
      "view": "aligned",
      "expected": "No",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K08",
        "decision": "No",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Crew Poplar's B2 and B3 overlap from 09:10 to 09:30. Crew Sallow's B4 and B5 also overlap from 09:50 to 10:10. These violate the one-booking-per-crew condition."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R3",
      "position": 2,
      "case_id": "K07",
      "view": "table",
      "expected": "Yes",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K07",
        "decision": "Yes",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Preparations are complete and consent is approved. Holly's bookings end at 08:55 and resume at 09:00; Lime's end at 09:50 and resume at 09:55; Laurel's end at 09:35 and resume at 09:40. No crew has overlapping bookings."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R3",
      "position": 3,
      "case_id": "K06",
      "view": "table",
      "expected": "No",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K06",
        "decision": "No",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Crew Yew's B3 and B5 overlap from 09:00 to 09:20. Its B5 and B2 also overlap from 10:10 to 10:20. The plan therefore fails the crew condition."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R3",
      "position": 4,
      "case_id": "K05",
      "view": "aligned",
      "expected": "Yes",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K05",
        "decision": "Yes",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Preparations are complete and consent is approved. Rowan's three bookings meet at 09:00 and 10:00, and Spruce's two meet at 09:50. Exact boundary handovers are permitted. Willow has only one booking."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R3",
      "position": 5,
      "case_id": "K04",
      "view": "aligned",
      "expected": "No",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K04",
        "decision": "No",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Crew Maple's B2 runs until 09:05 while its B4 starts at 09:00, creating an overlap from 09:00 to 09:05."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R3",
      "position": 6,
      "case_id": "K03",
      "view": "table",
      "expected": "No",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K03",
        "decision": "No",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Crew Ivy's B2 runs from 09:00 to 09:30 within its B4 booking of 08:00 to 10:30. Ivy therefore has two bookings in progress simultaneously."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R3",
      "position": 7,
      "case_id": "K02",
      "view": "table",
      "expected": "Yes",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K02",
        "decision": "Yes",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Preparations are complete and consent is approved. Fern's first booking ends at 09:00 before its next starts at 09:10. Hazel's bookings are 08:00–09:00 and 10:00–11:00; Elm's are 08:00–09:00 and 10:10–11:10. No same-crew intervals overlap."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R3",
      "position": 8,
      "case_id": "K01",
      "view": "aligned",
      "expected": "Yes",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K01",
        "decision": "Yes",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Preparations are complete and consent is approved. Alder's bookings meet at 09:00, Birch's at 09:30, and Cedar's at 10:00. The record explicitly permits an interval to end exactly when another starts."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R4",
      "position": 1,
      "case_id": "K16",
      "view": "aligned",
      "expected": "No",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K16",
        "decision": "No",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Crew Sallow's B1 and B3 overlap from 09:25 to 09:45. Crew Walnut's B2 and B4 also overlap from 10:05 to 10:25. These conflicts violate the one-booking-at-a-time condition."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R4",
      "position": 2,
      "case_id": "K15",
      "view": "table",
      "expected": "Yes",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K15",
        "decision": "Yes",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Preparations are complete and consent approved. Lime's bookings end at 09:50 and resume at 09:55; Laurel's end at 09:10 and resume at 09:15; Holly's end at 10:05 and resume at 10:10. No crew has overlapping bookings."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R4",
      "position": 3,
      "case_id": "K14",
      "view": "table",
      "expected": "No",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K14",
        "decision": "No",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Crew Ash's B5 overlaps B1 from 09:15 to 09:35, and B1 overlaps B2 from 10:25 to 10:35. The plan therefore violates the one-booking-at-a-time condition."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R4",
      "position": 4,
      "case_id": "K13",
      "view": "aligned",
      "expected": "Yes",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K13",
        "decision": "Yes",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Preparations are complete and consent approved. Spruce's bookings meet exactly at 09:15 and 10:15; Willow's meet exactly at 10:05. Rowan has one booking. Exact endpoint handovers are explicitly allowed."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R4",
      "position": 5,
      "case_id": "K12",
      "view": "aligned",
      "expected": "No",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K12",
        "decision": "No",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Crew Oak's B4 runs from 08:15 to 09:20, while its B1 starts at 09:15. They overlap from 09:15 to 09:20, violating the one-booking-at-a-time condition."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R4",
      "position": 6,
      "case_id": "K11",
      "view": "table",
      "expected": "No",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K11",
        "decision": "No",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Crew Juniper's B1 runs from 08:15 to 10:45, encompassing its B6 booking from 09:15 to 09:45. Juniper therefore has two bookings in progress during that interval."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R4",
      "position": 7,
      "case_id": "K10",
      "view": "table",
      "expected": "Yes",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K10",
        "decision": "Yes",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Preparations are complete and consent approved. Fern's earlier booking ends at 09:15 and its next starts at 10:25; Hazel's end and restart are 09:15 and 09:25; Elm's are 09:15 and 10:15. No same-crew intervals overlap."
      },
      "evidence_supported": true
    },
    {
      "reader_id": "R4",
      "position": 8,
      "case_id": "K09",
      "view": "aligned",
      "expected": "Yes",
      "correct": true,
      "confident_error": false,
      "response": {
        "case_id": "K09",
        "decision": "Yes",
        "confidence": 100,
        "needs_checking": "Nothing further under the stated rules",
        "evidence": "Preparations are complete and consent approved. Alder's bookings meet at 10:15, Birch's at 09:15, and Cedar's at 09:45. These exact endpoint handovers are allowed and create no overlap."
      },
      "evidence_supported": true
    }
  ],
  "response_sha256": {
    "responses/R1.json": "f1b516d46680440a6fb1d09e01950c9b2537fc2b99b038e15c873d6c4943a69e",
    "responses/R2.json": "9d65cf4fc3197fe4d4143eebe8143c15e68bb8783fc56a8765d20b1ed5a434ef",
    "responses/R3.json": "ffe9f4cc1c31718752d40d6259839e04364cd91fab59113c6114d408ed54c97a",
    "responses/R4.json": "e112e7e399ca455efb29573c905ac6400255bb3277d71342bf8c76ee203d6932"
  },
  "freeze_verified": 74,
  "interpretation": "No observed accuracy advantage: both forms reached a ceiling. This pilot does not establish equivalence, a speed advantage, a cognitive mechanism, or human benefit."
}