Remember the Answer, or Remember the Procedure?

When an independent reviewer reuses working memory, does a stale stored conclusion poison later decisions, does a deterministic revalidation guard remove the stale part, and does a reusable procedure help more than a stored answer?

Builds on Round 48 by storing working memory between reviews and testing it in two domains: 20 contract items and 16 quality release record items (aircraft inspections, food lot tests, calibration certificates, bridge load tests). Structurally, a quote check alone leaves 9 of 36 stale conclusions intact and adding a coverage check withholds all 36 while keeping all 36 correct notes. Live, Witness D is correct on 26 of 36 with no memory, 27 with a correct, stale, or guarded note, and 30 with a procedure card, and never adopts a stale conclusion. The near miss FAIL survives every stored answer and moves only with the procedure card (5 of 9). A stored extract first commitment lock blocks the one wrong revision a right brief causes. Live note following on 24 exhibits is zero in all three note styles.

• F49.1 (deterministic, E49-A): A quote check alone leaves 9 of 36 stale conclusions intact; adding a coverage check withholds all 36 and leaves all 36 correct notes intact.
• F49.2 (live, E49-B and E49-D): Witness D never adopts a stale stored conclusion (0 of 36); accuracy is 26 of 36 with no memory and 27 of 36 with a correct, stale, or guarded note.
• F49.3 (live): Witness D answers FAIL on 9 of 9 near miss items under every answer memory arm, including the correct note that stores HOLD with its reason.
• F49.4 (live): A procedure card raises Witness D to 30 of 36, with near miss items 5 of 9, unsafe verdicts 4 against 9, and false holds 2 against 1, in both domains.
• F49.5 (live): Witness C scores 28, 27, and 29 of 36 under no memory, stale, and guarded notes, with 0 stale follows.
• F49.6 (live): A wrong brief changes 0 of 36 provisional verdicts; a right brief changes 1, wrongly, and the lock blocks it.
• F49.7 (live, E49-C): Live note following is 0 of 24 across three note styles; five of ten CERTIFY exhibits rewritten to HOLD cite verified condition precedent text.

• E49-A Structural diagnostic for memory revalidation and the commitment lock Complete
• E49-B Contract working memory and stored commitment study Complete
• E49-C Public exhibit note styles on 24 new filings Complete
• E49-D Quality release record working memory study Complete

• H49.1 A stale stored note lowers Witness D accuracy relative to no memory, and Witness D follows the stale conclusion on at least 9 of 36 items.
• H49.2 A correct stored note is at least as accurate as no memory for Witness D.
• H49.3 The guarded note is followed less often than the stale note, and guarded accuracy is within 2 items of no memory in each domain.
• H49.4 A procedure card is at least as accurate as no memory with no more unsafe verdicts.
• H49.5 The directions in H49.1 and H49.3 hold in both domains.
• H49.6 Under a stale note, Witness C gives the stale wrong verdict on more items than under no memory.
• H49.7 Under a wrong brief, the locked final is at least as accurate as the unlocked revision; under a right brief, the locked final is at least as accurate as the provisional.
• H49.8 At least one note style yields nonzero live note following on HOLD exhibits, and the gate routes every followed release.

Research diagnostic and live benchmark only. Recommends that any reused memory pass both a quote check and a coverage check against the current record, and that reusable procedure memory be preferred over stored conclusions. No production system, release gate, or Noetis behavior changed.

• Is the shared near miss FAIL of Witnesses C and D a joint error that removing or swapping a role exposes?
• Does stored memory with the structure of a ledger and an automated outcome anchor where a plain stored conclusion does not?

Builds on Round 48, Round 47, Round 46. Continued in Round 54, Round 53, Round 52, Round 51, Round 50.