Building on the separation of table truth vs. character truth, what concrete procedures can you give players (e.g., calling for an ‘audit scene,’ invoking a ‘no-Carcosa recap,’ or trading SAN for a clean record check) so they can actively manage when memories and documents are allowed to warp in a King in Yellow campaign, and how does that player-facing control change the design of sanity pressure, clue obfuscation, and safety tools compared to a purely GM-driven distortion model?

Answer

Procedures are short, named moves players can call that only alter character truth while table truth stays logged. Below are four core tools and their design effects.

1. Core rule

• Table log: short, shared recap of what actually happened (from 7f8b342b-…). Never altered.
• Character layer: memories, files, labels that can shift.
• Every move states: affects character truth only.

2. Player moves

A) “Audit scene”

• When: after a weird or high-stakes scene.
• Call: any player says “Audit scene.”
• Procedure:
  • 3–5 min in-fiction cross-check (compare notes, talk to witnesses, pull CCTV, reread files).
  • GM must answer 1–2 factual questions from the table log, not from warped memories.
  • PCs get 1–2 anchors (facts that can’t later be rewritten without explicit consent).
• Costs (pick one): small SAN/stress, time pressure, or a public Carcosa-flag in the fiction.

B) “No-Carcosa recap”

• When: start or end of session.
• Call: any player.
• Procedure:
  • Group states a 3–5 bullet recap strictly from table truth.
  • GM may not introduce Carcosa distortions during this recap.
  • Result: a clean reference card; later distortions must be framed as "what your PC now remembers" against this.
• Use: re-orient when reality play gets dense; reinforce that players are not being gaslit.

C) “Clean record check”

• When: a PC reviews a specific document/record.
• Call: name the doc and say “Clean check.”
• Procedure:
  • GM shows the original version as first presented at table.
  • Then shows the current in-fiction version (with edits, stamps, Yellow Signs, new codes).
  • Group explicitly notes the delta.
• Cost trade: gain mechanical benefit (e.g., bonus to research / advantage on next investigative roll) but pay SAN/stress or tick a public Carcosa-status box (from 68425436-…). PCs see more clearly, but are now on someone’s watchlist.

D) “Revision consent”

• When: GM wants a retroactive memory/file change.
• Call: GM proposes; any player can instead say “Revision with consent.”
• Procedure:
  • GM states: what changes for PCs, and what stays in the table log.
  • Group can veto specific elements (e.g., core backstory, trauma lines) or narrow the scope.
  • Change is marked as character-only in notes.

3. How this changes design

Sanity pressure

• Less: surprise GM-driven reality edits.
• More: voluntary trade-offs:
  • Use audits/clean checks to stay grounded but pay SAN or reputational Carcosa marks.
  • Skip them to stay “safe” socially but accept fuzzier truth and higher narrative risk.
• SAN can model the strain of seeing inconsistencies, not just of exposure to symbols.

Clue obfuscation

• Obfuscation is opt-in and legible: players can always spend a move to see what changed.
• GM designs clues with two faces (16ea78b0-…):
  • Stable core fact (lives in table log).
  • Mutable interpretation (lives in character layer, can slide Carcosa-ward).
• Distortion becomes a resource: GM offers tempting benefits for not calling an audit or for accepting revision fallout.

Safety tools

• “No-Carcosa recap” and the stable log keep players oriented.
• “Revision consent” makes memory erosion negotiated, not imposed.
• Audit and clean checks are natural check-in points: you can pause, ask OOC if everyone is okay with the current level of drift, then proceed.

Compared to a GM-only distortion model, these moves shift horror from "the GM might rewrite anything" to "we choose when to look too closely, and that choice hurts." That keeps ambiguity sharp while preserving player clarity and consent.