Assessing believability

Julian Togelius, Georgios N. Yannakakis, Sergey Karakovskiy, Noor Shaker

    Research output: Chapter in Book/Report/Conference proceedingChapter

    Abstract

    We discuss what it means for a non-player character (NPC) to be believable or human-like, and how we can accurately assess believability. We argue that participatory observation, where the human assessing believability takes part in the game, is prone to distortion effects. For many games, a fairer (or at least complementary) assessment might be made by an external observer that does not participate in the game, through comparing and ranking the performance of human and non-human agents playing a game. This assessment philosophy was embodied in the Turing Test track of the recent Mario AI Championship, where non-expert bystanders evaluated the human-likeness of several agents and humans playing a version of Super Mario Bros. We analyze the results of this competition. Finally, we discuss the possibilities for forming models of believability and of maximizing believability through adjusting game content rather than NPC control logic.

    Original languageEnglish (US)
    Title of host publicationBelievable Bots: Can Computers Play Like People?
    PublisherSpringer-Verlag Berlin Heidelberg
    Pages215-230
    Number of pages16
    Volume9783642323232
    ISBN (Print)9783642323232, 3642323227, 9783642323225
    DOIs
    StatePublished - Oct 1 2012

    ASJC Scopus subject areas

    • Computer Science(all)

    Cite this

    Togelius, J., Yannakakis, G. N., Karakovskiy, S., & Shaker, N. (2012). Assessing believability. In Believable Bots: Can Computers Play Like People? (Vol. 9783642323232, pp. 215-230). Springer-Verlag Berlin Heidelberg. https://doi.org/10.1007/978-3-642-32323-2_9

    Assessing believability. / Togelius, Julian; Yannakakis, Georgios N.; Karakovskiy, Sergey; Shaker, Noor.

    Believable Bots: Can Computers Play Like People?. Vol. 9783642323232 Springer-Verlag Berlin Heidelberg, 2012. p. 215-230.

    Research output: Chapter in Book/Report/Conference proceedingChapter

    Togelius, J, Yannakakis, GN, Karakovskiy, S & Shaker, N 2012, Assessing believability. in Believable Bots: Can Computers Play Like People?. vol. 9783642323232, Springer-Verlag Berlin Heidelberg, pp. 215-230. https://doi.org/10.1007/978-3-642-32323-2_9
    Togelius J, Yannakakis GN, Karakovskiy S, Shaker N. Assessing believability. In Believable Bots: Can Computers Play Like People?. Vol. 9783642323232. Springer-Verlag Berlin Heidelberg. 2012. p. 215-230 https://doi.org/10.1007/978-3-642-32323-2_9
    Togelius, Julian ; Yannakakis, Georgios N. ; Karakovskiy, Sergey ; Shaker, Noor. / Assessing believability. Believable Bots: Can Computers Play Like People?. Vol. 9783642323232 Springer-Verlag Berlin Heidelberg, 2012. pp. 215-230
    @inbook{314d79244d864ef2bf2b22a9c11e97bf,
    title = "Assessing believability",
    abstract = "We discuss what it means for a non-player character (NPC) to be believable or human-like, and how we can accurately assess believability. We argue that participatory observation, where the human assessing believability takes part in the game, is prone to distortion effects. For many games, a fairer (or at least complementary) assessment might be made by an external observer that does not participate in the game, through comparing and ranking the performance of human and non-human agents playing a game. This assessment philosophy was embodied in the Turing Test track of the recent Mario AI Championship, where non-expert bystanders evaluated the human-likeness of several agents and humans playing a version of Super Mario Bros. We analyze the results of this competition. Finally, we discuss the possibilities for forming models of believability and of maximizing believability through adjusting game content rather than NPC control logic.",
    author = "Julian Togelius and Yannakakis, {Georgios N.} and Sergey Karakovskiy and Noor Shaker",
    year = "2012",
    month = "10",
    day = "1",
    doi = "10.1007/978-3-642-32323-2_9",
    language = "English (US)",
    isbn = "9783642323232",
    volume = "9783642323232",
    pages = "215--230",
    booktitle = "Believable Bots: Can Computers Play Like People?",
    publisher = "Springer-Verlag Berlin Heidelberg",

    }

    TY - CHAP

    T1 - Assessing believability

    AU - Togelius, Julian

    AU - Yannakakis, Georgios N.

    AU - Karakovskiy, Sergey

    AU - Shaker, Noor

    PY - 2012/10/1

    Y1 - 2012/10/1

    N2 - We discuss what it means for a non-player character (NPC) to be believable or human-like, and how we can accurately assess believability. We argue that participatory observation, where the human assessing believability takes part in the game, is prone to distortion effects. For many games, a fairer (or at least complementary) assessment might be made by an external observer that does not participate in the game, through comparing and ranking the performance of human and non-human agents playing a game. This assessment philosophy was embodied in the Turing Test track of the recent Mario AI Championship, where non-expert bystanders evaluated the human-likeness of several agents and humans playing a version of Super Mario Bros. We analyze the results of this competition. Finally, we discuss the possibilities for forming models of believability and of maximizing believability through adjusting game content rather than NPC control logic.

    AB - We discuss what it means for a non-player character (NPC) to be believable or human-like, and how we can accurately assess believability. We argue that participatory observation, where the human assessing believability takes part in the game, is prone to distortion effects. For many games, a fairer (or at least complementary) assessment might be made by an external observer that does not participate in the game, through comparing and ranking the performance of human and non-human agents playing a game. This assessment philosophy was embodied in the Turing Test track of the recent Mario AI Championship, where non-expert bystanders evaluated the human-likeness of several agents and humans playing a version of Super Mario Bros. We analyze the results of this competition. Finally, we discuss the possibilities for forming models of believability and of maximizing believability through adjusting game content rather than NPC control logic.

    UR - http://www.scopus.com/inward/record.url?scp=84949175640&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84949175640&partnerID=8YFLogxK

    U2 - 10.1007/978-3-642-32323-2_9

    DO - 10.1007/978-3-642-32323-2_9

    M3 - Chapter

    AN - SCOPUS:84949175640

    SN - 9783642323232

    SN - 3642323227

    SN - 9783642323225

    VL - 9783642323232

    SP - 215

    EP - 230

    BT - Believable Bots: Can Computers Play Like People?

    PB - Springer-Verlag Berlin Heidelberg

    ER -