Toward Purrfect Advice: Static Versus Adaptive Advice Systems for Narrative Puzzle Games
Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment · 08.11.2026 · 22 (1) · 97-107
DOI: 10.1609/aiide.v22i1.43568 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment, online erschienen am 08.11.2026. Verfasst von Naia Kamawar-MacLeod und 1 weiteren Autor:innen.
Abstract (Originalsprache)
Balancing difficulty is an important concern in video game design, as players have a variety of experience and skill levels. While adjusting difficulty by modifying game elements (e.g., enemy damage rates or hit points) can work well in some genres, doing so in narrative puzzle games can risk disrupting the game's design and narrative. Difficulty in puzzle games can also be adjusted by providing hints or advice, but doing so in puzzles with multiple solutions is challenging because the player's intended solution is unknown. We examine the effects of two ways to generate advice in a novel narrative puzzle game: one that provides steps toward a static target solution, and one that dynamically directs the player toward the solution nearest the puzzle's current state. In a user study with forty participants, we found that the adaptive advice system caused players to take significantly fewer actions to solve the puzzles, on average. We found no significant effect on player enjoyment across the two systems. A thematic analysis revealed that players differ in terms of how much support they want from an advice system, highlighting that such systems should allow players to customize this parameter.
Zitieren
APA
Kamawar-MacLeod, N., & Thue, D. (2026). Toward Purrfect Advice: Static Versus Adaptive Advice Systems for Narrative Puzzle Games. Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment, 22(1), 97-107. https://doi.org/10.1609/aiide.v22i1.43568
Vancouver
Kamawar-MacLeod N, Thue D. Toward Purrfect Advice: Static Versus Adaptive Advice Systems for Narrative Puzzle Games. Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment. 2026;22(1):97-107. doi:10.1609/aiide.v22i1.43568
BibTeX
@article{KamawarMacLeod2026,
author = {Kamawar-MacLeod, Naia and Thue, David},
title = {Toward Purrfect Advice: Static Versus Adaptive Advice Systems for Narrative Puzzle Games},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment},
year = {2026},
volume = {22},
number = {1},
pages = {97--107},
doi = {10.1609/aiide.v22i1.43568}
}
EUVIMED