Expressive AI: Games and Artificial Intelligence
Open accessAbstract
In recent years, as dramatic increases in graphic sophistication began yielding diminishing returns, the technical focus in game design has been turning towards Artificial Intelligence (AI). While game AI might be considered a “purely technical” phenomenon not of interest to game designers and theorists, this paper argues that AI-based art and entertainment constitutes a new interdisciplinary agenda linking games studies, design practice, and technical research. I call this new interdisciplinary agenda expressive AI.
AI snapshot
Written from the paper’s full text.
TL;DR
This paper argues that AI-based art and entertainment constitutes a new interdisciplinary agenda, "expressive AI," linking games studies, design practice, and technical AI research. It proposes game AI as a "white box" language of behavior for analyzing and designing games, illustrated through Pac-Man's ghost AI and the author's Façade project.
Key findings
- Game AI is defined not by specific techniques (pathfinding, neural nets, finite-state machines, etc.) but by a concern with behavior the player can read as produced by an intelligence with its own desires — related to Dennett's intentional stance and Newell's knowledge level.
- Game AI is distinguished from game physics: physics is the "dead," purely mechanical part of a game, while AI is the part supporting a psychological read. The paper argues the enemies in *Space Invaders* are "really part of the physics of the game, not AI," since they move mechanically and appear to shoot randomly.
- Analyzing Pac-Man shows the ghosts' behavior "defines the primary action of the game"; without them "there is no challenge in clearing a maze of pellets, and hence no game." Designer Toru Iwatani states the ghost algorithm was the most difficult part of designing the game.
- Multiple conflicting player accounts of the ghosts' behavior (chasing, ambushing, wandering, randomness) are read as evidence of the AI design's success: simple algorithms yield behavior rich enough to support multiple interpretations and give the ghosts an "inner life." None of these accounts mention the attack waves Iwatani describes.
- Two stated goals of game AI as research: it focuses on audience experience (traditional AI measures like time-to-task and error counts are "insufficient"; the AI must perform "with style"), and it supports human authorship (designers need architectural frameworks for particular visions, not "generic" reusable AI).
Methods
- Conceptual/theoretical essay with illustrative case analysis, not an empirical study.
- Analysis of Pac-Man's ghost AI draws on an interview with designer Toru Iwatani plus several gamer/player discussions of ghost behavior, comparing these accounts against each other.
- Design discussion draws on the author's own work with Andrew Stern on *Façade*, a first-person interactive drama, as a concrete design space defined by an AI architecture.
Limitations
- The paper states that Iwatani's interview description "is far from a complete description of the ghost AI" and that the author "has been unable to find a detailed description," so the Pac-Man analysis rests on incomplete and partly conflicting sources; Iwatani's account is "not detailed enough to resolve the dispute" among player readings.
- Evident from the text: the argument is programmatic and illustrative rather than empirically tested — no formal evaluation, metrics, or systematic study of player experience is reported, and the Façade discussion is presented as the author's own design exploration rather than as evidence.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
0 of 13 cited sources checked · 0 supported · 0 flagged
- Findingown data
There is little agreement on what the ghosts' behavior actually is in gamer discussions.
“In fact, I have been unable to find a detailed description, but rather many gamer discussions hypothesizing about the ghost AI.It is interesting that in these discussions there is agreement that that ghost AI is central to the experience of the game, but little agreement on what the ghosts' behavior actually is.”
Quote in the paperNot a substantial claim - Interpretationreasoning
Much of Pac-Man's gameplay, and thus much of the player experience, is determined by the fine details of the ghosts' behavior.
“Thus we can see that much of Pac-Man's gameplay, and thus much of the player experience, is determined by the fine details of the ghosts' behavior.”
Quote in the paperSubstantial claim - Interpretationreasoning
The ghosts in Pac-Man define the primary action of the game.
“The ghosts in fact define the primary action of the game.”
Quote in the paperSubstantial claim - Interpretationreasoning
Game AI supports human authorship.
“Game AI supports human authorship.”
Quote in the paperNot a substantial claim - Interpretationreasoning
For game design languages to be able to talk about behavior requires an engagement with game AI.
“Thus for game design languages to be able to talk about behavior requires an engagement with game AI.”
Quote in the paperNot a substantial claim
- Classified as probably not a substantial claim. 38
- May combine two claims. 9
- The quote only partly states this claim. 1
Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).
Findingnone givenIwatani's description in his interview is far from a complete description of the ghost AI.
“Iwatani's description in this interview, however, is far from a complete description of the ghost AI.”
Quote in the paperNot a substantial claimMay combine two claims.
FindingcitationThe turn towards game AI is manifest in increasing interest at game design venues, at AI research venues, and in game marketing that touts AI sophistication.
“This change is manifest both in increasing interest in game AI at game design venues (e.g. the AI roundtable at the Game Developers Conferences), at AI research venues (e.g. the series of AAAI Symposia on AI and Interactive Entertainment [5,15,7]), and in game marketing that touts the sophistication of the AI rather than the graphic resolution or polygon count.”
Quote in the paperNot a substantial claim0 of 3 sources checked- cites [5]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites [15]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites [7]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Probably not a substantial claim: may not pass S4 (based).
FindingreasoningThe technical focus in game design has been turning towards Artificial Intelligence (AI).
“In recent years, as dramatic increases in graphic sophistication began yielding diminishing returns, the technical focus in game design has been turning towards Artificial Intelligence (AI).”
Quote in the paperNot a substantial claimMay combine two claims.
PremisecitationGame designers have called for a language of design, noting that designers currently lack a unified vocabulary for describing existing games and thinking through new games.
“For several years, game designers have called for a language of design [11,2,12], noting that designers currently lack a unified vocabulary for describing the design of existing games and thinking through the design of new games.”
Quote in the paperNot a substantial claim0 of 3 sources checked- cites [11]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites [2]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites [12]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Probably not a substantial claim: may not pass S4 (based).
Premisenone givenAcademic game studies practitioners generally seal off implementation details, treating implementation as a black box that supports an autonomous discourse about games.
“Like the design patterns and rules of game designers, academic game studies practitioners generally seal off implementation details, treating the implementation as a black box that supports an autonomous discourse about games.”
Quote in the paperNot a substantial claimClassified as probably not a substantial claim. May combine two claims.
PremisecitationAcademic game studies is developing methods for analyzing games at an abstract content level above the implementation, drawing from humanistic disciplines.
“Academic game studies is similarly developing various methods for analyzing games at an abstract content level above the implementation.”
- Also stated in: § AI AND DESIGN Language for Design and Analysis · plain
“Such analyses tend to draw methodologies and terminologies from various humanistic disciplines, producing a broad range of work that, for example, looks at games in terms of their use of space [10,24], as semiotic sign systems [13], as a narrative form [24,26], or in terms of the temporal relationships between actions and events [6].”
Quote in the paperSubstantial claim0 of 4 sources checked- cites [10]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites [24]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites [13]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites [6]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Also stated in: § AI AND DESIGN Language for Design and Analysis · plain
Probably not a substantial claim: may not pass S2 (checkable), S4 (based).
Interpretationnone givenTo use game AI as a language for analysis requires opening up the black box of the implementation and considering the relationship between the generation of intelligent behavior, the player experience, and the cultural context of the game.
“To use game AI as a language for analysis requires opening up the black box of the implementation and considering the relationship between the generation of intelligent behavior, the player experience, and the cultural context of the game.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable).
InterpretationreasoningThe behavior of the ghosts in Pac-Man is critical to understanding the game.
“The behavior of the ghosts is critical to understanding the game.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable), S4 (based). May combine two claims.
InterpretationreasoningThis white box language is the language of behavior, the language of game AI.
“This is the language of behavior, the language of game AI.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable), S4 (based).
Interpretationnone givenIn the context of game analysis and design, expressive AI provides a language for talking about
“In the context of game analysis and design, expressive AI provides a language for talking about”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable), S4 (based). The quote only partly states this claim.
InterpretationreasoningAI-based art and entertainment constitutes a new interdisciplinary agenda linking games studies, design practice, and technical research (expressive AI).
“While game AI might be considered a”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S3 (load-bearing).
Premisenone givenDesign patterns and design rules are usually articulated at a level of abstraction above implementation details, dealing purely with relationships between visual elements, game rules, and player interaction.
“Both design patterns, which name and describe design elements, and the closely related notion of design rules, which offer imperative advice in specific design situations, are usually articulated at a level of abstraction above the implementation details of the game, dealing purely with relationships between visual elements, game rules, and player interaction.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable).
InterpretationreasoningThe multiple readings of the ghosts' behavior speak to the success of the AI design.
“The fact that there are these multiple readings of the ghosts' behavior in fact speaks to the success of the AI design.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S1 (asserted), S2 (checkable).
PremisereasoningWhat links game AI practices is a concern with
“What links these practices together under the single term”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable), S4 (based).
Interpretationnone givenIn the context of AI research, expressive AI opens up a new technical research agenda.
“In the context of AI research, expressive AI opens up a new technical research agenda.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable), S4 (based). May combine two claims.
Interpretationnone givenGame AI is not just about hyperrational problem solving, but also about creating a sense of aliveness, the sense that there is an entity living within the computer that has its own life independently of the player and cares about how the player's actions impact this life.
“But game AI is not just about hyperrational problem solving, but also about creating a sense of aliveness, the sense that there is an entity living within the computer that has its own life independently of the player and cares about how the player's actions impact this life.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable).
InterpretationreasoningGame AI lies at the intersection of player perception (the player is able to read part of the game behavior as alive) and the game code that supports this perception.
“Game AI lies at the intersection of player perception (the player is able to read part of the game behavior as alive) and the game code that supports this perception.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable).
InterpretationreasoningThere is a place for a
“However, this paper argues that there is a place for a game language that opens up the box and considers the implementation, a”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable). May combine two claims.
InterpretationreasoningThe relatively simple algorithms of the ghost AI yield a richness and unpredictability of behavior that supports multiple interpretations, giving the ghosts an
“The relatively simple algorithms of the ghost AI yields a richness and unpredictability of behavior that supports multiple interpretations, giving the ghosts an”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable).
InterpretationreasoningFor game designers, the game AI strongly influences the design space.
“For game designers, the game AI strongly influences the design space.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable).
InterpretationreasoningThe gameplay can't be understood without a
“The gameplay can't be understood without a”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable). May combine two claims.
InterpretationreasoningOpening up the game AI is essential not only for analysis, but as part of the language of design.
“Thus, opening up the game AI is essential not only for analysis, but as part of the language of design.”
Quote in the paperNot a substantial claimClassified as probably not a substantial claim.
InterpretationreasoningGame AI focuses on audience experience.
“Game AI focuses on audience experience.”
Quote in the paperSubstantial claimProbably not a substantial claim: may not pass S2 (checkable). May combine two claims.
InterpretationreasoningThe game AI must not only accomplish tasks within the game world, but do it with style.
“The game AI must not only accomplish tasks within the game world, but do it with style.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing). May combine two claims.
InterpretationreasoningFor AI researchers, game AI offers new high-level research goals, thus opening up new technical challenges.
“And for AI researchers, game AI offers new high-level research goals, thus opening up new technical challenges.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable), S4 (based).
Interpretationnone givenGame design can have a real impact on the technical practice of AI, opening up a new research agenda that asks novel technical research questions.
“While the technical practice of AI has an influence on the design and analysis of games, conversely, game design can have a real impact on the technical practice of AI, opening up a new research agenda that asks novel technical research questions.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable).
InterpretationreasoningThe AI's fundamental goal is to create an experience for the player.
“The AI's fundamental goal is to create an experience for the player.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable).
InterpretationreasoningThe behavioral design space is structured by the architecture and algorithms of the game AI.
“The behavioral design space is structured by the architecture and algorithms of the game AI.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable), S4 (based).
Interpretationnone givenMuch of the question of interactive story revolves around whether and how a game can provide moment-by-moment interaction that produces an immediate sense of agency while helping to structure long-term interaction into a dramatically pleasing arc without diminishing the sense of agency.
“Much of the question of interactive story revolves around whether and how a game can provide moment-by-moment interaction that produces an immediate sense of agency while helping to structure long-term interaction into a dramatically pleasing arc (e.g.complication, crisis, climax and resolution) without diminishing the sense of agency.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable).
InterpretationreasoningTo really understand Pac-Man, to understand the player experience and the player interpretations supported by this experience, would require a detailed understanding of the ghost AI.
“So to really understand Pac-Man, to understand the player experience and the player interpretations supported by this experience, would require a detailed understanding of the ghost AI.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable).
InterpretationreasoningA full understanding of a game requires understanding the behavior of the
“A full understanding of a game requires understanding the behavior of the”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).
Interpretationnone givenFor game analysis, game AI is the language of intentional behavior.
“For game analysis, game AI is the language of intentional behavior.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable).
InterpretationreasoningThe AI architecture must support game artists in expressing the behavioral potential needed for a specific design.
“Because game AI is a fundamental part of the player experience, the AI architecture must support game artists in expressing the behavioral potential needed for a specific design.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable).
InterpretationreasoningGame AI is more than an implementation issue only of interest to game designers.
“Game AI is more than an implementation issue only of interest to game designers.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable).
InterpretationreasoningSince game AI is a deep part of the design, novel AI architectures and approaches will need to be continuously developed within the context of specific designs and particular creative visions.
“Since game AI is a deep part of the design, novel AI architectures and approaches will need to be continuously developed within the context of specific designs and particular creative visions.”
Quote in the paperNot a substantial claimClassified as probably not a substantial claim.
InterpretationcitationGame AI is an instance of expressive AI, AI research that takes AI-based art and entertainment as a first class problem.
“Game AI is an instance of expressive AI, AI research that takes AI-based art and entertainment as a first class problem [18,19,20].”
Quote in the paperSubstantial claim0 of 3 sources checked- cites [18]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites [19]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- cites [20]
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
Probably not a substantial claim: may not pass S2 (checkable).
InterpretationreasoningThe measures of effectiveness used in traditional AI research, such as time to accomplish a task, number of errors, and so forth, are insufficient for guiding work in game AI.
“The measures of effectiveness used in traditional AI research, such as time to accomplish a task, number of errors, and so forth, are insufficient for guiding work in game AI.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable), S4 (based).
Interpretationnone givenThese questions of interactive story are primarily questions of behavior, and can thus be most profitably explored within concrete design spaces defined by specific AI approaches, rather than through abstract argumentation.
“These are primarily questions of behavior, and can thus be most profitably explored within concrete design spaces defined by specific AI approaches, rather than through abstract argumentation.”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing), S4 (based).
Interpretationnone givenGame designers need architectural frameworks within which they can express their particular visions, rather than
“Rather than”
Quote in the paperNot a substantial claimProbably not a substantial claim: may not pass S2 (checkable).
Interpretationnone givenAI, as a language of behavior, operates at the interface between everyday language for describing intentional behavior and technical implementations that provide a specific, narrow interpretation of this everyday terminology.
“AI, as a language of behavior, operates at the interface between everyday language for describing intentional behavior (e.g.”
Quote in the paperNot a substantial claim
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Citation 26 has no reference entry
MinorThe text cites reference 26 for work on games as narrative, but the provided reference list ends at 25.
“as a narrative form [24,26]”
“25. Ryan, M. Beyond Myth and Metaphor: The Case of Narrative In Digital Media. Games Studies: The International Journal of Computer Games Research.”
Dramatic-beats citation points to a different work
MinorThe discussion cites [17] for dramatic beats, but reference 17 is a paper on semiotic analysis; the listed screenwriting book is reference 16.
“the idea of dramatic beats [17]”
“17. Mateas, M. Expressive AI: A Semiotic Analysis of Machinic Affordances. Proceedings of the 3 rd Conference on Computational Semiotics and New Media, University of Teesside, UK. 2003.”
Newell citation points to Ryan
MinorThe text cites [25] for Newell's knowledge level, but reference 25 is by Ryan. Newell's The Knowledge Level is reference 24.
“Newell's notion of a knowledge level [25]”
“25. Ryan, M. Beyond Myth and Metaphor: The Case of Narrative In Digital Media. Games Studies: The International Journal of Computer Games Research.”
Pac-Man ghost quotation cites an AI symposium
MinorThe quoted description of individual Pac-Man ghosts is attributed to [14], but reference 14 is an AI symposium volume; the Pac-Man source appears under reference 13.
“Pokey is the orange ghost and is slow and moves in a random manner.[14]”
“14. Laird, J., and van Lent, M. (Eds.). Papers from the AAAI Spring Symposium on Artificial Intelligence and Interactive Entertainment. Menlo Park, CA: AAAI Press. 2001.”
Citation for the authors' architecture points to Murray
MinorThe text attributes beat goals and conversational handlers in the authors' architecture to [23], but reference 23 is Murray's Hamlet on the Holodeck, not a work by the authors.
“defines beat goals and conversational handlers [23]”
“23. Murray, J. Hamlet on the Holodeck. Cambridge, MA: MIT Press, 1998.”
Iwatani interview cites a screenwriting book
MinorThe cited reference [16] is McKee's screenwriting book, not the Iwatani interview. The interview is listed within reference 15.
“In an interview, Toru Iwatani, Pac-Man's designer, clearly describes the motivation for adding ghosts to Pac-Man [16]:”
“16. McKee, R. Story: Substance, Structure, Style, and the Principles of Screenwriting. New York, NY: HarperCollins, 1997.”
AI symposium citation points to an interview book
MinorThe text cites [15] as part of a series of AAAI symposia, but reference 15 is a book of programmer interviews. The 2001 symposium is listed as reference 14.
“the series of AAAI Symposia on AI and Interactive Entertainment [5,15,7]”
“15. Lammers, Susan (ed.). Programmers at Work: Interviews with 19 Programmers Who Shaped the Computer Industry. Tempus Books, 1989.”
Cites
(1987) · Medical Entomology and Zoology · Cited by 4,269 · see citing works
Hamlet on the holodeck: the future of narrative in cyberspace
(1997) · Choice Reviews Online · doi:10.5860/choice.35-1342 · Cited by 2,780 · see citing works
Newell, A. (1982) · Artificial Intelligence · doi:10.1016/0004-3702(82)90012-1 · Cited by 2,589 · cited 1× · see citing works
“This is related to Dennett’s notion of the intentional stance [4] or Newell’s notion of a knowledge level [25].”
BackgroundCybertext: Perspectives on Ergodic Literature
Aarseth, E. (1997) · Portuguese National Funding Agency for Science, Research and Technology (RCAAP Project by FCT) · Cited by 1,088 · see citing works
Story: Substance, Structure, Style, and the Principles of Screenwriting
McKee, R. (1997) · Methuen eBooks · Cited by 810 · see citing works
First Person: New Media As Story, Performance, And Game
Wardrip-Fruin, N., & Harrigan, P. (2004) · Choice Reviews Online · doi:10.5860/choice.42-0713 · Cited by 620 · see citing works
Facade: An Experiment in Building a Fully-Realized Interactive Drama
Mateas, M., & Stern, A. (2003) · Cited by 409 · see citing works
Self-citationMurray, J. H. (1997) · Cited by 371 · see citing works
A behavior language for story-based believable agents
Mateas, M., & Stern, A. (2002) · IEEE Intelligent Systems · doi:10.1109/mis.2002.1024751 · Cited by 236 · cited 1× · see citing works
“The details of how our architecture operationalizes the idea of dramatic beats [17], defines beat goals and conversational handlers [23], and supports natural language understanding [21], reveal interactive story opportunities as well as non-obvious constraints that force us to iterate either on the architecture or on…”
BackgroundSelf-citationBeyond Myth and Metaphor* -The Case of Narrative in Digital Media
Ryan, M. (2001) · Scholarly Commons (Embry–Riddle Aeronautical University) · Cited by 223 · see citing works
Videogames of the oppressed :videogames as a means for critical thinking and debate
Frasca, G. (2001) · Cited by 207 · see citing works
Interactive drama, art and artificial intelligence
Mateas, M., Bates, J., & Carbonell, J. (2002) · Cited by 190 · see citing works
Self-citationEskelinen, M. (2001) · Digital Creativity · doi:10.1076/digc.12.3.175.3232 · Cited by 149 · cited 2× · see citing works
“[1, 6, 8]), richly interactive stories that provide strong player agency are in principle impossible, and thus a source of much wasted effort in the games community.”
BackgroundThe Case For Game Design Patterns
Kreimeier, B. (2002) · Cited by 109 · see citing works
Expressive AI: A Hybrid Art and Science Practice
Mateas, M. (2001) · Leonardo · doi:10.1162/002409401750184717 · Cited by 102 · cited 1× · see citing works
“Game AI is an instance of expressive AI, AI research that takes AI-based art and entertainment as a first class problem [18, 19, 20].”
BackgroundSelf-citationPerspectives of Computer Game Philology.
Kücklich, J. (2003) · Cited by 43 · see citing works
Leonardo: Journal of the International Society for the Arts, Sciences and Technology
Dorin, A., & McCormack, J. (2001) · Cited by 34 · see citing works
Programmers at work : interviews with 19 programmers who shaped the computer industry
Lammers, S. M. (2006) · Medical Entomology and Zoology · Cited by 16 · see citing works
Flagged first, then by citations
1 of the 19 references couldn’t be matched to a known paper.
Research works printed but not linked: 5
- Dobson, W. (Ed.). Papers from the 2000 AAAI Spring Symposium on Artificial Intelligence and Interactive Entertainment. Menlo Park, CA: AAAI Press. 2000.
- Forbus, K. and El-Nasr, M. (Eds.). Papers from the 2002 AAAI Spring Symposium on Artificial Intelligence and Interactive Entertainment. Menlo Park, CA: AAAI Press. 2002.
- Laird, J., and van Lent, M. (Eds.). Papers from the AAAI Spring Symposium on Artificial Intelligence and Interactive Entertainment. Menlo Park, CA: AAAI Press. 2001.
- Mateas, M. Expressive AI: A Semiotic Analysis of Machinic Affordances. Proceedings of the 3 rd Conference on Computational Semiotics and New Media, University of Teesside, UK. 2003.
- Mateas, M. & Stern, A. Integrating Plot, Character and Natural Language Processing in the Interactive Drama Facade. In Göbel, S., Braun, N., Spierling, ,U., Dechau, J., and Diener, H. (Eds.) Proceedings of Technologies in Interactive Digital Storytelling and Entertainment. ZDGV, Darmstadt, Germany, 2003.
Cited by
Sorted by citations
Yannakakis, G. N. (2012) · Proceedings of the 9th conference on Computing Frontiers · doi:10.1145/2212908.2212954 · Cited by 179 · see citing works
“…game conference at Harvard [33] and 12 years after Laird’s and van Lent’s seminal article [26] that, in part, established the foundations of game artificial intelligence (AI) and inspired early work in the field [34, 22, 25, 14, 3, 30, 58] the game AI term needs to be revisited and restructured.”
Procedural literacy: educating the new media practitioner
Mateas, M. (2005) · On the Horizon The International Journal of Learning Futures · doi:10.1108/10748120510608133 · Cited by 103 · see citing works
“The fundamentally procedural nature of games can be seen by looking at the two sources of activity within a game: game AI and game physics (Mateas 2003).”
BackgroundSelf-citationLudoliteracy: Defining, Understanding, and Supporting Games Education
Zagal, J. P. (2024) · Figshare · doi:10.1184/r1/6686828 · Cited by 97 · see citing works
Build It to Understand It: Ludology Meets Narratology in Game Design Space
Mateas, M., & Stern, A. (2005) · doi:10.26503/dl.v2005i1.172 · Cited by 79 · see citing works
“Thus the game AI itself becomes a design resource, not a mere “implementation detail”; the AI architecture provides the language for thinking about game behavior [19].”
BackgroundSelf-citationPlaying With Non-Humans: Digital Games as Techno-Cultural Form
Giddings, S. (2005) · doi:10.26503/dl.v2005i1.219 · Cited by 57 · see citing works
A framework for games literacy and understanding games
Zagal, J. P. (2008) · Conference on Future Play · doi:10.1145/1496984.1496991 · Cited by 41 · see citing works
“In order to really understand Pac-Man, to understand the player experience and the player interpretations supported by the experience, requires a detailed understanding of the AI of the ghosts [25].”
BackgroundInternational Journal of Role-playing 1: Full Issue
Drachen, A., Hitchens, M., Montola, M., et al. (2008) · International Journal of Role-Playing · doi:10.33063/ijrp.vi1.176 · Cited by 36 · see citing works
Video game pathfinding and improvements to discrete search on grid-based maps
Anguelov, B. (2011) · UpSpace Institutional Repository (University of Pretoria) · Cited by 24 · see citing works
AI-Based Game Design : Enabling New Playable Experiences
Eladhari, M. P., Sullivan, A., Smith, G., & McCoy, J. (2011) · Cited by 23 · see citing works
Personalizing the Player Experience in MMORPGs
Tychsen, A., Tosca, S. P., & Brolund, T. (2006) · Lecture notes in computer science · doi:10.1007/11944577_26 · Cited by 20 · see citing works
“Work such as [9],[12],[16] point towards game formats featuring advanced userresponsiveness.”
BackgroundAntBot: Ant Colonies for Video Games
Recio, G., Martín, E., Estébanez, C., & Sáez, Y. (2012) · IEEE Transactions on Computational Intelligence and AI in Games · doi:10.1109/tciaig.2012.2212194 · Cited by 17 · see citing works
“The behavior of the enemies of Pac-Man in the original game was developed by Toru Iwatani using a mixture of patterns, randomness, and tracking behaviors for each ghost [34].”
Fizek, S. (2018) · Digital Culture & Society · doi:10.14361/dcs-2018-0112 · Cited by 16 · see citing works
Player talk—the functions of communication in multplayer role-playing games
Drachen, A., & Smith, J. H. (2008) · Computers in entertainment · doi:10.1145/1461999.1462008 · Cited by 15 · see citing works
“Empirical data on the communicative activity of players interacting via multiplayer games in any medium is, with a few exceptions [Trappl 1997; Hutchison 2003; Mateas 2003; Eladhari and Eindley 2004; Castronova 2005; Aylett et al. 2006], mostly absent in the academic literature, although massively multiplayer games ha…”
BackgroundZagal, J. P., & Tomuro, N. (2010) · doi:10.1145/1930488.1930492 · Cited by 13 · see citing works
“If AI is such a central component of gameplay, why doesn’t it feature prominently? One possible explanation is that we haven’t yet developed a language (and corresponding vocabulary) for talking about behavior in games [21].”
BackgroundProcedural literacy: educating the new media practitioner
Mateas, M. (2008) · Cited by 13 · see citing works
Self-citationBehavioral Modeling Based on Probabilistic Finite Automata: An Empirical Study
Tîrnȃucȃ, C., Montaña, J. L., Ontañón, S., et al. (2016) · Sensors · doi:10.3390/s16070958 · Cited by 11 · see citing works
“For example, the artificial intelligence for the two virtual characters in the game Façade [1] took more than five person-years to develop.”
BackgroundGhost Worlds – Time and Consequence in MMORPGs
Tychsen, A., & Hitchens, M. (2006) · Lecture notes in computer science · doi:10.1007/11944577_30 · Cited by 11 · see citing works
“[2],[16],[21]), it has become possible to integrate advanced (and possibly meaningful) NPC-interaction in computer games.”
BackgroundKeeping the Story Straight: A Comparison of Commitment Strategies for a Social Deduction Game
Eger, M., & Martens, C. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment · doi:10.1609/aiide.v14i1.13015 · Cited by 10 · see citing works
“Mateas (2003) has thus argued that for the development of believable agents for games intentionality plays a key role. Eger et al. (2017) describe how agents that use and communicate intentions achieve a better performance in a collaborative game when compared to agents that simply follow a set protocol.”
BackgroundModeling of Emotional Effects on Decision-making by Game Agents
Spraragen, M., & Madni, A. M. (2014) · Procedia Computer Science · doi:10.1016/j.procs.2014.03.088 · Cited by 9 · see citing works
Procedural Content Generation of Custom Tower Defense Game Using Genetic Algorithms
Kraner, V., Fister, I., & Brezočnik, L. (2021) · Lecture notes in networks and systems · doi:10.1007/978-3-030-75275-0_54 · Cited by 8 · see citing works
The Mind Module—Using an Affectand Personality Computational Modelas a Game-Play Element
Eladhari, M. P. (2014) · IEEE Transactions on Affective Computing · doi:10.1109/taffc.2014.2309123 · Cited by 8 · see citing works
“The central question explored in this work is how representations of affective processes can be used in VGWs as game-play elements, such that players are able to create mental models of the game-play derived from these affective processes.”
BackgroundA Model-Based Approach to Optimizing Ms. Pac-Man Game Strategies in Real Time
Foderaro, G., Swingler, A., & Ferrari, S. (2016) · IEEE Transactions on Computational Intelligence and AI in Games · doi:10.1109/tciaig.2016.2523508 · Cited by 8 · see citing works
“Pac-Man game and profit function based on guidelines and resources describing the behaviors of the characters, such as [49].”
MethodRoles and Worlds in the Hybrid RPG Game of Oblivion
Champion, E. (2008) · International Journal of Role-Playing · doi:10.33063/ijrp.vi1.186 · Cited by 7 · see citing works
“So the sensory and spatially explorative aspects of the game are powerful and rich, but Oblivion seems to struggle with the second type of world, the social one, it lacks expressive agents (Mateas 2003).”
BackgroundSilva, A., Rocha, F., Santos, A. L. M., et al. (2011) · doi:10.1109/sbgames.2011.35 · Cited by 6 · see citing works
“One of these tasks is the pathfinding of an agent over a game map, which has already achieved a better performance on GPU, but is still limited.”
BackgroundMethodExploring the Potential of A.I.-Driven Opponents in Video Games
Quadir, A. M., & Khder, M. A. (2022) · ASU International Conference in Emerging Technologies for Sustainability and Intelligent Systems (ICETSIS) · doi:10.1109/icetsis55481.2022.9888909 · Cited by 6 · see citing works
“…was able to defeat the reigning World Chess Champion Gary Kasparov in 1997 (as seen in fig 1), which was a milestone in both the history of AI, as well as the history of chess [14]. cause the player to immediately lose if they touched Pac-Man, each seemingly with its own unique personality [15].”
BackgroundA Methodology for Requirements Analysis of AI Architecture Authoring Tools
Grow, A., Gaudl, S. E., Gomes, P. F., Mateas, M., & Wardrip–Fruin, N. (2014) · The University of Bath Online Publications Store (The University of Bath) · Cited by 6 · see citing works
Self-citationMazes, monsters and multicursality. Mastering Pac-Man 1980–2016
Newman, J. (2016) · Cogent Arts and Humanities · doi:10.1080/23311983.2016.1190439 · Cited by 5 · see citing works
“Many of these, as (Mateas, 2003) and Jones (2008) have noted, seek to move beyond the manifest to reveal the game’s latent meanings by making allegorical connections between Pac-Man ’s insatiable hunger and constant movement and Marxist readings of consumption practices and market expansion (as in…”
BackgroundInfluentialThe Achievement of Animals: An Ethology of AI in Video Games
Giddings, S. (2020) · Neue Perspektiven der Medienästhetik · doi:10.1007/978-3-658-30721-9_6 · Cited by 5 · see citing works
A Historical Review of Immersive Storytelling Technologies
Bienvenido, H. P., Barinaga, B., & Mora-Fernández, J. (2021) · Advances in business information systems and analytics book series · doi:10.4018/978-1-7998-6985-6.ch027 · Cited by 4 · see citing works
Narrative Review of Game AI 2000 Onwards and Future Research Directions
Gera, R., & Chadha, P. (2021) · Advances in logistics, operations, and management science book series · doi:10.4018/978-1-7998-8497-2.ch013 · Cited by 3 · see citing works
What Design Choices are Effective in Inducing Fear and Tension in First-Person PC Horror Games?
Liao, K.-L., Huang, M., & Yang, R. (2022) · International Conference on Human System Interaction (HSI) · doi:10.1109/hsi55341.2022.9869445 · Cited by 3 · see citing works
“These two kinds of movement AI are first implemented in the classic arcade game “Pac-Man” [5].”
BackgroundInfluentialIntegrating Players’ Perspectives in AI-Based Games: Case Studies of Player-AI Interaction Design
Villareale, J., Maram, S. S., El‐Nasr, M. S., et al. (2023) · International Conference on the Foundations of Digital Games (FDG) · doi:10.1145/3582437.3582451 · Cited by 3 · see citing works
“For example, Mateas developed the notion of expressive AI to explore AI as a design material [25, 26].”
BackgroundButt, R., & Johansson, S. J. (2009) · International Conference on Foundations of Digital Games (FDG) · doi:10.1145/1536513.1536558 · Cited by 3 · see citing works
“a promising research field for Ai. Game Ai is now helping game developers to make game logics for characters that can behave more believable which in turn makes them much more fun to play against [ 15 ].”
BackgroundNtina, C. C., Ma, X., & Deng, Z. (2013) · Computers in entertainment · doi:10.1145/2582186.2633437 · Cited by 2 · see citing works
Performance Comparison of AI Algorithms Anytime Algorithms
Butt, R. (2008) · Cited by 2 · see citing works
Supporting learning about games
Bruckman, A. S., & Zagal, J. P. (2008) · Cited by 2 · see citing works
Game-Playing Agents and Non-player Characters
Merrick, K. E. (2016) · Springer eBooks · doi:10.1007/978-3-319-33459-2_3 · Cited by 1 · see citing works
Modeling emotional effects on decision-making by agents in game-based simulations
Zyda, M., & Spraragen, M. (2013) · Cited by 1 · see citing works
Expressividade procedural: narrativas e jogos digitais
Ferreira, D. P. (2016) · doi:10.11606/t.27.2016.tde-22092016-141634 · Cited by 1 · see citing works
WRITING FOR EACH OTHER: DYNAMIC QUEST GENERATION USING IN SESSION PLAYER BEHAVIORS IN MMORPG
Mendonca, S. C. (2020) · doi:10.15368/theses.2020.54 · Cited by 1 · see citing works
Fizek, S. (2018) · Digital Culture & Society · doi:10.14361/dcs-2018-040111
“This diverse set of techniques and practices united under one capacious term may refer to “pathfinding, neuralnetworks, models of emotion and social situations, finite-state machines, rule systems, decision-tree learning”, amongst many others (Mateas 2003).”
BackgroundClaure, H., Candon, K., Clark, O., & Vázquez, M. (2024) · ACM/IEEE International Conference on Human-Robot Interaction (HRI) · doi:10.1145/3610978.3640687
“This game extends the well-known single-player Space Invaders game, which has a rich history in psychology [19] and HRI [7, 8, 18], to a multiplayer setting, which can be either collaborative or competitive.”
BackgroundSel, F. (2024) · Advances in human and social aspects of technology book series · doi:10.4018/979-8-3693-7235-7.ch005
Exploring extensions of neurotransmitter-based emotion models
Pinheiro, J. R., Fernandes, P. D. T., Baffa, A. C. E., & Feijó, B. (2025) · Journal on Interactive Systems · doi:10.5753/jis.2025.5642 · Cited by 1 · see citing works
Entre tradition narrative et innovations technologiques
Hagemann, S. (2024) · Sens public · doi:10.7202/1118959ar
HeRoN: a mediated RL-LLM framework for adaptive NPC behavior in interactive environments
Cimino, G., Deufemia, V., & Selice, A. (2026) · Neural Computing and Applications · doi:10.1007/s00521-026-12275-w
“Early games such as Tennis for Two and Pac-Man relied on minimal AI logic [20, 21], while the 1990 s introduced more strategic behavior in titles like Civilization and StarCraft [22].”
- Authors
- Mateas, M.
- Affiliations
- Georgia Institute of Technology (US)
- Corresponding author
- Mateas, Michael
- ORCID
- Mateas, Michael
- Venue
- DiGRA Digital Library
- Published
- 2003-01-01
- Volume
- 2
- ISSN
- 2342-9666
- OpenAlex
- W2175156397
- Cited by
- 49
- References
- 19
- Citations by year
- 2026: 1 · 2025: 1 · 2024: 4 · 2023: 1 · 2022: 2 · 2021: 3 · 2020: 2 · 2019: 1 · 2018: 4 · 2017: 1 · 2016: 5 · 2014: 3 · 2013: 2 · 2012: 2
- FWCI
- 1.22
- Type
- conference paper
- Language
- en
- Open access
- gold
- doi.org
- Locations
- Digital Games Research Association Conference
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Artificial Intelligence in Games
- Social Sciences › Sociology and Political Science › Digital Games and Media
- Social Sciences › Arts and Humanities › Visual Arts and Performing Arts › Art, Technology, and Culture
- Keywords
- game AI, game design, game studies, entertainment, interdisciplinary research
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-19.
Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-19.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review