Skip to main content
Formal Specification

Walbiri (Landmark Memory Game) — Formal Specification

Australia · guessing-hidden · Confidence: low

← Back to game page

Implementation Suitability

Suitability: adaptation-requiredFidelity: discretizedGap-fills: 2
# Walbiri — AI-optimized specification ## General idea A pass-the-baton memory chain where failure hands the sequence to the next player rather than ending the round — so the group collectively works through the same fixed sequence, and the deciding factor is who breaks first, not who recites fastest. ## Formal model - **Abstraction taken:** the landmark arrangement becomes a fixed ordered sequence of N labelled items; recall becomes a per-item success-probability check that compounds down the sequence. - **Player/agent count:** 4-10, acting in sequential turns within a shared round. - **Turn model:** sequential — one player recites as far as they can, then the next continues from the point of failure. ## Formal rules R1. At round start, a fixed sequence of N labelled landmark-objects is generated and shown to all players for a study period. R2. The active player performs `recall(item)` for the next position in the sequence, succeeding with `P(success) = f(memory_skill, position, study_time)`. [gap-fill] no explicit memory-model formula is given; added because a deterministic perfect-recall game has no content. R3. On success, the active player may continue with `recall()` for the next position; on failure, play passes to the next player, who continues from the same (failed) position. R4. The round ends when either a player recalls every item in the sequence correctly (round win, that player arranges the next round) or all players in turn have failed at the same position without anyone completing it (round restarts with a re-shuffled sequence). ## Edge cases - **Tie at the end condition:** not applicable — the round resolves to exactly one completing player, or a restart; no comparative score. - **Player with no legal move:** not applicable — recall() is always available at the active player's turn. - **Repetition / cycles:** possible if no player ever completes the sequence; explicitly resolved by R4's restart-with-reshuffle, which is the source's own documented mechanic. - **Simultaneous claims:** not applicable — strictly sequential single-player attempts. - **Mid-round elimination and turn-order repair:** not applicable — no elimination; every player gets a turn at the current failure point. - **Resource exhaustion:** not applicable. - **Termination guarantee:** guaranteed per round-cycle by R4; the overall game (multiple rounds) is bounded by an agreed number of rounds or time limit, [gap-fill] since the source describes indefinite continued play. ## State representation Persistent: sequence length N, study time, memory-skill parameters per player. Per-round: the generated sequence, current recall position, whose turn is active. Per-turn ephemeral: this player's recall attempt outcome. Information set: fully observable once studied — this is a memory-accuracy game, not a hidden-information game; the 'hidden' element is purely the player's own imperfect recall, not concealed game state. ## Action space - `recall(item)` — precondition: actor is the active player. ## Playability notes - **Single-player-vs-AI:** natural as a personal recall-length challenge against a fixed or growing sequence. - **Group games (n>4):** scales directly — more players simply means more turns before a restart or completion. - **Physical-skill model:** not applicable; the memory-skill-and-position-dependent recall function (R2) is a cognitive-proxy analogue to this framework's physical dexterity models. - **What is lost:** the pedagogical link to real topographic knowledge — the source's point that this trained genuine navigational recall — has no analogue in a spec using arbitrary generated sequences instead of real, meaningful landmarks. ## Implementation hints Model recall difficulty as increasing with sequence position (R2) — real serial-recall tasks get harder deeper into a sequence, and this is the natural source of the game's escalating tension as players fail progressively further in.

Download as JSON

Download the full game corpus entry for this game as structured JSON.

Download JSON
Cite this page

Walbiri (Landmark Memory Game) (Australia). (n.d.). Games of Countries. https://gamesofcountries.com/games/au-australia/walbiri-memory

@misc{gamesofcountries-au-australia-walbiri-memory,
  title = {Walbiri (Landmark Memory Game) (Australia)},
  year = {n.d.},
  publisher = {Games of Countries},
  url = {https://gamesofcountries.com/games/au-australia/walbiri-memory}
}