Playing with Character: A Framework of Characterization in Video Games

Abstract

Characterization refers to the process of attributing character traits to narrative entities called ‘characters'. While there is a long tradition of characterization theory in literary studies, the topic has not been examined extensively in game research. Based on insights from literary, film, and game studies, this article creates a theoretical model of how ‘character,' or character traits, can be attributed in video games, and offers a methodological vocabulary for further character(ization) research. First, this paper synthesizes the tradition of characterization research in literary studies. Second, it identifies three participants in video game characterization (developers, actors, and players) and introduces the concept of ergodic characterization to describe those instances in which players produce nontrivial characterization efforts. Finally, the framework itself is presented through application to various game titles, and several answers to methodological problems within game characterization analysis are suggested.

Keywords

characterization ergodic video game character narratology game analysis

While there is no shortage of scholarship on game characters, the topic of characterization in games (cf. Schröter, 2013; Vella, 2015; Willumsen, 2018) has not received elaborate attention. Characterization is a term from literary studies and narratology broadly referring to the authorial attribution of character information or ‘traits' to a narrative agent. Several models of characterization have been proposed in literary theory (see below) but these have only rarely been applied in game studies, and when they have, the broad scope of textual cues that may produce characterization in games has not always been taken into account. The current paper, therefore, proposes a comprehensive framework of characterization in games, furthering our understanding of the multimodal ways through which games present character traits.

Video game characters (VGCs), used here as an umbrella term for both playable characters (PCs) and nonplayable characters (NPCs), have been the object of much academic study. They are usually considered as the combination of a (functional) avatar and a (fictional) character pole (e.g., Aldred, 2014/2016; Caracciolo, 2015; Klevjer, 2006; Schröter, 2013; Tronstad, 2008; Vella, 2015; Willumsen, 2018): the avatar refers to the player's in-game embodiment, while the character pole describes the VGC's mimetic dimension as a fictional entity in a ludonarrative storyworld. This paper is primarily concerned with the character pole, although it is recognized in the Discussion that players may choose to ignore the VGC's existence as a fictional character in favor of purely avatarial engagement (Bayliss, 2007; Tronstad, 2008).

We will first survey the tradition of character and characterization research in literary studies in order to contextualize the discussion on game characterization. Subsequently, we identify the participants operative within video game characterization, and introduce the concept of ergodic characterization. Finally, the framework itself is presented, building on insights from literature, film, and game studies, followed by a discussion on methodological difficulties related to studying characterization in games. At the outset, it is worth mentioning that we are aware that the reliance on theories from neighboring fields has been considered as “theoretical imperialism” (Aarseth, 1997, p. 16) or interdisciplinary “colonization” (e.g., Eskelinen, 2001; Simons, 2007), but we must admit that most research on characterization stems from adjacent disciplines, and that much of the relevant literature on game characters/characterization similarly depends on literary theories (e.g., Lankoski et al., 2003/2010; Vella, 2015; Willumsen, 2018). As stressed by transmedial narratology, we should “acknowledge both similarities and differences” (Thon, 2017, p. 288) in how storytelling (here, characterization) operates across media. Furthermore, Lankoski (2010) specifically argued that “[a] multidisciplinary approach is needed” (p. 15) to fully grasp VGCs, so our recourse to literature and film seems legitimate.

Character(ization) in Literary Studies

Research on fictional characters dates back to Aristotle, who used the term prattōn (literally: a person who acts, an agent) to refer to a/the ‘character' as narrative agent in the text (Poetics, 1448a1), and ēthos (literally: a custom, a character[istic]) for ‘character' as a prattōn's individual personality (Poetics, 1449b38–1450a5; Aristotle, 1995). Modern character theory has, among other topics, focused on the ontological dimensions of fictional characters (e.g., Heidbrink, 2010; Phelan, 1989), affective engagement with/cognitive responses to characters (e.g., Eder et al., 2010; Margolin, 2010; Schneider, 2001; Smith, 1995), structuralist/narratological character typologies (e.g., Forster, 1927/1969; Greimas, 1966; Propp, 1928/1968; Vogler, 2007), or transmedia characters (e.g., Bertetti, 2014; Richardson, 2010; Thon, 2019).

Important for our purposes is that characters are seen as constellations of different character traits, combined into a coherent “bundle of differential elements” (Frow, 2014, p. 24) that composes the prattōn's ēthos. Chatman (1978) defined a trait as a “relatively stable or abiding personal quality” that may “emerge earlier or later in the course of a story, or […] disappear and be replaced by another” (p. 126). Traits transform basic character functions such as ‘the helper' or ‘villain' (called “actants” by Greimas, 1966) into specific mimetic individuals (“acteurs”; ibid.). Traits are static or dynamic, depending on whether they remain the same throughout the narrative or change as an indication of character development (Eder et al., 2010; Margolin, 1986).

Characterization, then, refers to the ascription of ēthos (or character traits) to a certain prattōn. Scholars agree on a distinction between direct or explicit characterization on the one hand, and indirect or implicit characterization on the other (e.g., Bal, 1997/1999; De Temmerman and van Emde Boas, 2018; Eder et al., 2010; Herman & Vervaeck, 2005; Rimmon-Kenan, 1983/2004), depending on whether the character information is attributed directly (e.g., by the narrator, the character itself, or another character) or implied within other textual elements (e.g., the character's appearance or setting). A narrator's statement that ‘John is bad' directly assesses John's (alleged) badness, whereas showing John doing bad things or surrounding himself with bad people indirectly presents him as bad. One of the most recent models of characterization was created by De Temmerman and van Emde Boas (2018), which is subdivided into (1) name-giving (characterization through the character's proper name), (2) direct, and (3) indirect characterization. Indirect characterization, they argue, further consists of (3.1.) metaphorical (i.e., comparisons or contrasts with, for example, other characters) and (3.2.) metonymical characterization. The latter refers to traits inferred from “an aspect related to [the character] by contiguity” (De Temmerman and van Emde Boas, 2018, p. 20), and comprises seven characterization techniques: emotions, membership of a specific group, action, speech, focalization, appearance, and setting. These terms will prove crucial for our framework of characterization in games.

Who Characterizes in Games? Ergodic and Non-Ergodic Characterization

First, we must inquire into the entities that characterize in games. We identify three participants in the process of video game characterization. The first is the developer, a term used here as a form of studio authorship (Hadas, 2020) to describe everyone involved in the process of creating the game and its VGCs, including directors, writers, art designers, programmers, composers, and more. They give VGCs specific traits, and/or program a system through which players may customize PCs themselves.

The second participant is the (voice) actor. As technology has evolved, the possibilities for VGCs to be modeled after a specific actor have increased (Eder and Thon, 2012; Schröter, 2013), facilitating VGC appearances that convincingly look like the actors who portrayed them (‘motion capture'). Such is the case with, for example, Kassandra in Assassin's Creed Odyssey (Ubisoft Quebec, 2018), played by Melissanthi Mahut. As these actors lend their appearance and/or voice to the VGC, their involvement has consequences for characterization (see below).

Finally, players can actively participate in the characterization process (e.g., Bayliss, 2007; Fizek, 2012; Lankoski and Björk, 2008/2010a; Schröter, 2013; Vella, 2015). Players are often empowered to, for example, make meaningful choices, customize their PCs’ appearance, or role-play their PC as a specific character type. We refer to this form of characterization as ergodic characterization, after Aarseth's (1997) “ergodic literature” for any kind of literature where “nontrivial effort is required to allow the reader to traverse the text” (p. 1). Aarseth (1997) uses “nontrivial” to describe any effort beyond “(for example) eye movement and periodic or arbitrary turning of pages” (p. 2). Consequently, ergodic characterization in games can be defined as any form of characterization where nontrivial effort (e.g., customization, choice-making) is required to allow the player to characterize their PC.

The major peculiarity of ergodic characterization (see the Discussion section) is that different players (or the same player over different play sessions) can actualize different PCs. Vella (2015) discusses how players of The Last of Us (Naughty Dog, 2013) may focus on Joel's “stealth and spatial awareness,” or “make a point of eliminating every hostile individual encountered” (p. 414). This leads to different characterizations of the character Joel, depending on the player's ergodic input. Borrowing a term from literary studies to describe possible different actualizations of narrative actions, we may describe different player-actualized PCs as “variant[s]” (Ryan, 1985, p. 747) of one another. The term ‘variant' is also used in research on transmedia characters (e.g., Richardson, 2010) to describe versions of the ‘same' characters in different texts (e.g., Sherlock Holmes, Batman, Lara Croft) and is therefore an apt name for different actualizations of PCs.

The above leads to a basic triadic structure of video game characterization as a potential interplay between game designer, (voice) actor, and player. Only those cases in which players perform nontrivial characterization efforts can be considered ergodic. Not all three participants are necessarily involved in every game: silent characters do not involve voice actors, and players are usually unable to ergodically impact the characterization of NPCs. In the case of NPCs without voice acting, the developer is the only characterizing entity that remains.

Framework of Characterization in Games

While several scholars have described singular characterization cues (see below), few have attempted to integrate different game studies insights into a specific model of video game characterization in the vein of those constructed by literary scholars. Fernández-Vara (2011, cited in Vella, 2015, p. 380) listed “name, image, animation, speech, [and] backstory” as “identity markers” in games, although Vella (2015) pointed out that research could go further in “taxonomizing the various avenues by which characterization might occur” (p. 380). Schröter (2013) investigated how film techniques can convey character in games, and distinguished the character's name, film-stylistic devices (e.g., mise-en-scène, camera, editing, music, sound design), appearance, movement, and the HUD as cues that may convey character information. The most elaborate scholarship on this topic so far is the model by Vella (2015), who adapts Margolin's (1986) influential character theory and distinguishes between static mimetic elements (i.e., information that is relatively fixed), dynamic mimetic elements (i.e., those determined by actions) and formal patterns, referring to similarities or contrasts between characters in the text. Vella's static elements comprise (1) represented elements (name, physical appearance, costume(s), voice, animations), (2) contextual elements (possessions, environment, role), and (3) ludic/mechanical elements (capabilities and limitations, passion, goal, attributes, development). Within the dynamic elements, Vella distinguishes between “player actions”, performed through player input and “perceived by the player as being her own as much as they are the character's” (p. 405), and “character actions” (p. 401), which players are unable to influence, and may lead to “character autonomy” (Willumsen, 2018) if the character operates independently from the player. Yet, we can expand Vella's list, for example, with Schröter's filmic cues, or other cues discussed in later scholarship (e.g., haptic feedback; Willumsen and Jaćević, 2018). Additionally, the distinction between direct and indirect characterization does not feature in these models.

In order to combine the above literary scholarship with game studies, the present paper constructs a new characterization framework suited for VGCs. Building on the aforementioned scholarship, our framework distinguishes between storyworld cues and medium cues of characterization: the former concern elements within the ludonarrative storyworld, whereas the latter describe characterization through the medium's depiction of the narrative agent(s). Our framework is inspired by the literary models explained above, insights from film studies, and the aforementioned game character research. To account for the specific nature of the video game medium, several of the above cues are conceptualized differently, and others are added to account for the medium's specific semiotic operations. The particular digression to film studies is legitimized by the affinity between films and games, apparent through games’ use of film aesthetics (Egenfeldt-Nielsen et al., 2008/2016; Thon, 2016) and representational languages (Eder and Thon, 2012; Schröter, 2013). The complete framework is represented in Figure 1.

Storyworld Cues

We use ‘storyworld cues' to describe narrative elements within the storyworld that give specific character information. We conceptualize the narrative storyworld as transmedial (Herman, 2009; Thon, 2016; 2017), with the potential for mediation through different media in different ways. Inspired first and foremost by De Temmerman and van Emde Boas (2018), our list of storyworld cues consists of name-giving, action, appearance, speech, emotions, membership of a specific group, setting, and objects.

Name-giving. Barthes (1970/1975) considered the proper name as a “cohesion device” (Heidbrink, 2010, p. 86) for the structuralist constellation of character traits. Names may immediately reveal social information such as gender, heritage, class, and more (Frow, 2014). For example, Barthes (1970/1975) describes the name ‘Sarrasine' in Balzac's novel of the same name (1830) as having a connotation of “femininity, which will be obvious to any French-speaking person, since that language automatically takes the final ‘e’ as a specifically feminine linguistic property” (p. 17).

Game researchers have also discussed the characterization potential of the given name (Fizek, 2012; Lankoski and Björk, 2007/2010b; Meretzky, 2001; Schröter, 2013; Vella, 2015). For example, the name ‘Kratos’ in God of War (Santa Monica Studio, 2005) already reveals him as powerful through its original meaning (Greek kratos, meaning ‘strength'). In games, PC names can be determined by the developer (e.g., Nathan Drake, Mario), created ergodically by players during character customization (cf. Hagström, 2008), or even be absent to foster “monadic” (Klimmt et al., 2009) identification between player and PC.

Action. According to Margolin (1986), there is universal agreement that actions “are one of the main sources” (p. 208) for characterization. Smith (2010) even writes that characters who lack meaningful action (e.g., unnamed characters in the background of a busy street) “[dissolve] into an aspect of setting” (p. 235). ‘Action' comprises anything from one-time deeds to habitual activities (Eder et al., 2010; Rimmon-Kenan, 1983/2004) or large actions that change the course of the plot (Eder et al., 2010), and may even include actions that were deliberately not executed (e.g., not helping someone in need) or merely contemplated (Rimmon-Kenan, 1983/2004). Crucial to action are choices: Aristotle already described the moral decision between doing something good or not (proairesis; Poetics, 1454a18; Aristotle, 1995) as an important indicator of character in narrative. Screenwriting theory similarly stresses that “[t]rue character is revealed in the choices a human being makes under pressure” (McKee, 1998/2014, p. 101, emphasis removed).

Game scholars have treated action in much the same way (e.g., Fizek, 2012; Lankoski et al., 2003/2010; Lankoski and Björk, 2007/2010b; Vella, 2015), as have game design handbooks (e.g., Sheldon, 2004). Galloway (2006), for instance, described games as an “action-based medium” (p. 3, original emphasis). We have already discussed how Vella (2015) distinguishes between player and character actions, which refer not only to what the character actually does, but also to the ludic system of possible actions (cf. Neitzel, 2014) as a set of capabilities at the player's/PC's disposal (Vella, 2015). Action also includes the VGC's movement, which may serve characterization (Schröter, 2013): in Smite (Titan Forge Games, 2014-), Mercury's fast movement (and additional quick speech) is one element of his characterization as a funny speedster, while Thanatos’ low movement speed (in addition to his ability to instantly execute enemies below a certain Health level) contributes to his representation as a silent, sudden killer. Choices are equally crucial in games: in many games, choices result in character development/growth (Jørgensen, 2010) or actualize variant versions of the narrative (Schröter and Thon, 2014). Finally, since players perform an input in games, characterization through action is also determined by player skill: Macías Villalobos (2020) wrote that Kratos “will not be the same when guided by a novice as when guided by an expert player” (p. 76).

For NPCs, the functions they fulfill in the game also fall under action (Schröter, 2013). Bartle (2003) listed eight possible functions of NPCs: “buy, sell, and make stuff,” “provide services,” “guard places,” “get killed for loot,” “dispense quests,” “supply background information,” “do stuff for players,” and “make the place look busy” (p. 382). Depending on which function(s) they perform, NPCs are characterized differently from those who perform other functions.

Appearance. What characters look like is often assumed to hint at character traits. The pseudo-science of physiognomy (e.g., the work by J. K. Lavater in the eighteenth century) aimed to infer personality traits from external features (Eder et al., 2010; Rimmon-Kenan, 1983/2004). While the validity of such work has been “completely discredited” in modern times, “the metonymic relation between external appearance and character-traits has remained a powerful resource in the hand of many writers” (Rimmon-Kenan, 1983/2004, p. 31). Here, all external aspects of the character are included, both those “beyond the character's control” (e.g., facial features) and “those which at least partly depend on [them]” (e.g., clothes) (Rimmon-Kenan, 1983/2004, pp. 65–66).

Appearance is a powerful character indicator in games as well (Grimes, 2003; Isbister, 2006; Meretzky, 2001; Schröter, 2013; Vella, 2015). Like Rimmon-Kenan (1983/2004), Vella (2015) distinguishes between physical appearance and costume(s), but both can be seen as subcategories of one larger category. A VGC's appearance can be determined by the developer (e.g., Mario) or open to player customization (Schröter and Thon, 2014), and it may change throughout the game or stay the same (Kromand, 2007). Players are often in control of the PC's appearance: in The Elder Scrolls V: Skyrim (Bethesda Game Studios, 2011), they are able to create their PC out of nine different races and 47 different customizable categories (e.g., Complexion, Cheek Color, Chin Width), allowing for myriads of possible PCs. If they so choose, players can have their character resemble their own appearance (Boellstorff, 2008; Vella, 2015). Customization can be purely cosmetical, or linked to ludic traits and statistics (Klastrup and Tosca, 2009). Finally, appearance may also serve a social function (Klastrup and Tosca, 2009): Smite's diamond skins, which only become available after reaching a PC's highest mastery level, color the PC into diamond hues and indicate the player's skill, perseverance, and investment in the game.

Speech. Both the “manner” (how something is said) and “matter” (what is said; p. 210) of a speech utterance is important for characterization (Margolin, 1986). Speech style (e.g., intonation, register, grammatical errors) may for example be “indicative of origin, dwelling place, social class, or profession” (Rimmon-Kenan, 1983/2004, p. 64).

In Mass Effect (BioWare, 2007), the Hanar species refuses to talk in the first person, which they deem egotistical. Instead, Hanar refer to themselves as “this one,” conveying their belief system through speech style. Schröter and Thon (2014) give the example of Spec Ops: The Line (Yager Development, 2012), where the deterioration of protagonist Martin Walker's mind is conveyed through his evolution from military speech (e.g., “Hostile down”) to increasingly unhinged, unrestrained language (e.g., “Kill everything that fucking moves”). Games often let players ergodically select their PC speech options (e.g., Mass Effect), or allow them to type themselves (e.g., RuneScape (Jagex, 2001-)). As in film or drama, speech in games is often defined by the actor's voice and delivery (Fizek, 2012; Vella, 2015).

Emotions. De Temmerman and van Emde Boas (2018) distinguish between a character's “permanent characteristics” and “emotions, temporary feelings more easily influenced than ēthos” (p. 22). How a character reacts emotionally to narrative situations may point to certain mental or psychological traits. Evidently, the absence of emotions in a given situation is just as telling as the presence of them.

In game studies, scholars initially doubted whether VGCs were capable of having emotions. Early research often considered PCs as mere cursors for the player's actions—a view called “cursor theory” by Klevjer (2006). Oft-cited is Aarseth's (2004) quote “when I play, I don’t even see [Lara Croft's] body, but see through and past it” (p. 48), which strips Lara from her character dimension and relegates her exclusively to avatar functionality. Similarly, Simons (2007) wrote that characters “are functional and not emotionally and psychologically characterized entities as their counterparts in narratives.” Later work attributed more depth to VGCs: as Caracciolo (2015) points out, (contemporary) VGCs very often possess their own “more or less fleshed-out identity, complete with past experiences, personality traits, and motivations” (p. 238). Scholars (e.g., Jørgensen, 2010; Schröter and Thon, 2014; Willumsen, 2018) started to consider the complexity of VGCs, indicating that VGCs could be discussed similarly to characters found in “traditional” media (Caracciolo, 2015; Willumsen and Jaćević, 2018). Characterization through emotions for example happens in The Last of Us Part II (Naughty Dog, 2020) where Ellie's emotional reaction when Joel is killed, as well as her subsequent revenge quest, characterizes Ellie as someone who deeply cared for Joel.

Membership of a specific group. De Temmerman and van Emde Boas (2018) distinguish between “macro-social” groups (e.g., the society one lives in), “micro-social” groups (e.g., friends, heritage, affiliations), and “educated-intellectual peer groups” (p. 22), referring to levels of education, as different types of groups to which characters may belong, and whose membership in turn characterizes them.

VGCs are equally characterized by the groups they belong to (Eder and Thon, 2012; Schröter and Thon, 2014; Vella, 2015): in The Elder Scrolls V: Skyrim, joining the Stormcloaks or Imperials has characterization consequences. We may also refer to character classes or guilds (Schröter and Thon, 2014) as characterizing: the former often change the PC's statistical qualities, while the latter potentially associate PCs with certain guild ideologies and/or specific clothing or insignias.

Setting. Setting refers to the characters’ surroundings as a source for character information. De Temmerman and van Emde Boas (2018) give the example of the Trojan prince Paris in Homer's Iliad, often shown in the women's quarters while the Trojan War rages outside the city walls, indicating “his predilection for female company and his problematic ēthos as a warrior” (p. 21).

Setting also characterizes in games (Eder and Thon, 2012; Vella, 2015): Aerith's lush and flowerful garden in Final Fantasy VII Remake (Square Enix, 2020) hints at her innocence and tranquility, especially when compared to the messier Sector 7 slums. In Minecraft (Mojang Studios, 2010) or No Man's Sky (Hello Games, 2016), players are given command of their own environments, and can build any structure allowed by the game system. The structures they create may influence characterization, as those who for example actualize their Minecraft PC as a king may characterize them as such by constructing castles.

Objects. One type of cue that we may add to De Temmerman and van Emde Boas' (2018) list is objects. Hamon (1972) writes that objects may serve as “a concretization of certain of [a character's] psychological, moral or physical qualities” (p. 109, my translation). De Jong (2018) cites Ajax’ “tower-like shield” (p. 34) in the Iliad (7.219; Homer, 1999) as an indicator of “his status as a warrior on whom the others can rely and who will fight to the utmost to protect his men and ships.”

Objects operate as characterization cues in games as well (Fizek, 2012; Vella, 2015), and players often ergodically choose which objects the character possesses. If players of Assassin's Creed Valhalla (Ubisoft Montreal, 2020) equip Eivor with mythical/legendary weapons (e.g., Mjölnir, Excalibur) instead of more historical weaponry, their Eivor variant is more strongly mythologized than is otherwise the case. Like appearance, objects can offer statistical/strategic advantages that change the character, or be tied to the PC's progression or development. In role-playing games, players often enter a game world with simple clothing and no items (cf. Klastrup and Tosca, 2009), but dispose of a large array of powerful equipment by the end.

Medium Cues

While the aforementioned cues are applicable to storyworlds regardless of the medium, the following focus more strictly on how the game medium portrays the storyworld through representation and simulation (Frasca, 2003). As mentioned above, there are similarities between games and other media (Thon, 2016), which are accounted for by reference to insights from literary studies and film studies. Our list consists of focalization, acting, film arts, audio, statistics, controls, haptics and interface.

Focalization. Focalization broadly refers to the perspective from which a story is told (Genette, 1972). Characterization through focalization occurs when the character's “habits of reasoning and of formulating thoughts in the mind” (Eder et al., 2010, p. 32) indicate traits. In Margaret Atwood's (2005/2018) The Penelopiad, the image Penelope mentally constructs of her cousin Helen of Troy as the person who ruined her life (chapter 11) characterizes Penelope herself as someone who harbors negative feelings towards Helen.

Thon (2016) discussed various ways in which games allow players access to the subjectivity of their PCs, including blurred vision as the PC's physical/mental health decreases, hallucinations, or interior monologues. Characterization through focalization can take place when players have access to their PC's thoughts (Arjoranta, 2013). When the protagonist of Assassin's Creed Odyssey meets their father Nikolaos, players have the option to select “I’ve come to take my revenge” (killing Nikolaos), or “I’ve come to get answers” (sparing him, while remaining angry). This mental choice (presented in a close-up shot, symbolizing the decision-making process) reflects the protagonist's “habits of reasoning” (Eder et al., 2010, p. 32): it is a binary mode of thinking that does not leave room to shake hands, forgive, or reconcile. No matter what the protagonist chooses, they will be characterized as vengeful, although the choice to kill Nikolaos actualizes a patricidal protagonist acting on their vengeful feelings.

Acting. It is often said that characters are ‘brought to life' by actors. McKee (1998/2004) stresses that screenwriters “should leave room for the actor” (pp. 383–384). Through acting, the character's appearance and voice is connected to that of the actor, who additionally shapes the character's actions through gestures, expressions, posture, and more (Dyer, 1998/2004; Michaels, 1998).

A related aspect serving characterization is the actor's stardom and persona (Dyer, 1998/2004). In Troy (Petersen, 2004), Achilles is played by Brad Pitt, who matches the Greek hero in star status: arguably, Troy's Achilles would be a different character if played by an unknown actor. In games, we may point to Far Cry 6's (Ubisoft Toronto, 2021) villain Antón Castillo, played by Giancarlo Esposito, who is known for his quiet yet terrifying television villains (e.g., Gus in Breaking Bad; Gilligan et al., 2008–2013), which fits the series’ approach to antagonists:

The most important part of crafting a Far Cry villain isn’t the fact that they’re evil or shocking. It's that they’re someone that could sit across from you at the dinner table and charm you into thinking that everything they believe, no matter how twisted or horrifying, could be absolutely true. (Bouali in Alvelo, 2021)

As such, the weight of Esposito's former roles at least partially informs the character of Castillo, a character similar to the villains he has portrayed previously.

Film arts. The broad umbrella term ‘film arts’ here includes all filmic devices of representation, including camera positioning and movement, editing, mise-en-scène, lighting, and more (cf. Schröter, 2013). For example, Quentin Tarantino's famous trunk shots—in which the camera is placed in the perspective of (someone in) a car trunk, looking upwards to the characters in the shot—visually imbues the characters within the shot with authority, as viewers literally look up to them from the perspective of a powerless (trunk-bound) character (Mascelli, 1965/1998).

Such cinematic characterization techniques also appear in games (Schröter, 2013). The third DLC for Immortals Fenyx Rising (Ubisoft Quebec, 2020) adopts an isometric perspective to depict its new PC, Ash. The perspective represents the viewpoint of the NPCs Athena and Fenyx, both of whom are gods, and look down at Ash from Olympus. Therefore, this camera position reveals Ash as looked after by the gods, and supports the construction of her heroic identity. Most other cinematic techniques in games work similar to those of film (e.g., lighting), but one that deserves special attention is the camera. The camera in games can be fixed or free, or even disappear in the case of virtual reality games (Arcagni and D’Aloia, 2021). While fixed cameras may serve characterization (as illustrated above), the characterization possibilities of a free, player-controlled camera are less apparent (cf. Schröter, 2013). In theory, players could move the camera in spectacular ways (e.g., 360° around the PC) to highlight spectacular events, but such camera movements would most likely impede gameplay.

Audio. Music and sound design have been discussed as important characterization cues in film (Eder et al., 2010). For example, the dark, authoritative “Imperial March” used in The Empire Strikes Back (Kershner, 1980) to accompany Darth Vader “signifies “villain” from its first appearance” (Kassabian, 2000, p. 57).

Similarly, Summers (2016) discussed how Street Fighter II (Capcom, 1991) uses stereotypical musical themes to “characterize the fighters beyond the limited visual aspect in a perceptually salient and cognitively impactful way” (p. 62): Spanish characters are accompanied with “flamenco rhythms,” while the Chinese Chun-Li receives “a wooden-timbred parallel fourth motif in the Orientalist tradition” (ibid.). While “morally questionable” (Summers, 2016, p. 62), these scores do immediately indicate a social background for these PCs.

Statistics. VGCs are “a set of numerical variables” (Vella, 2015, p. 9), and how this “character data set” (Fitzpatrick et al., 2005) is statistically configured reveals character traits (cf. Bizzocchi, 2007). In Smite, Mercury has the game's highest basic Attack Speed (1.48 at level 20), characterizing him as fast. Statistics can also hint at flaws or weaknesses (called “negative traits” by Berman, 1988/1997, p. 52): in The Outer Worlds (Obsidian Entertainment, 2019), the PC receives the permanent flaw ‘cynophobia' (Perception −2, Temperament −1) if they die too many times at the hands of ‘canids'. Dynamic statistic evolutions can also be important: in Apotheon (Alientrap, 2015), the PC Nikandreos evolves from a Greek hero to a god, and grows increasingly powerful throughout his deification. Apotheon does not include scaling (Paprocki, 2021), meaning that enemies remain on the same power level while the PC evolves. Nikandreos’ character evolution is thereby marked by the contrast between the dynamism of his statistics versus the stasis of those of other characters.

Controls. Malliet (2007) stressed the importance of studying game controls, which he described as “the mental and physical efforts […] required of a player in order to successfully and efficiently interact with the game program.” Lankoski et al. (2010) further pointed out that controls are highly important in the characterization of fighting game PCs. For example, in Mortal Kombat X (NetherRealm Studios, 2015), the character Kenshi may be classified as a “zoner,” or a PC who works best when positioned at a distance and keeping the enemy stuck in a specific zone (cf. Telesinski, 2016), based on several of his attacks and combo controls (e.g., Demon Assault, Tele-Flurry).

Haptics. Haptic feedback (i.e., controller rumble) has also been identified as having the potential for characterization (Lankoski, 2011; Willumsen and Jaćević, 2018). For example, one scene in Marvel's Avengers (Crystal Dynamics, 2020) shows protagonist Kamala Khan captured and interrogated. During the scene, the DualShock 4 controller vibrates in the player's hands, indicating Kamala's fear and distress: as Kamala trembles under the pressure of being kidnapped, so do our hands literally shake during the scene.

Interface. Finally, the game's user interface may serve characterization. Bizzocchi (2007) considers the visual lay-out of the cursors in Ceremony of Innocence (Real World Media, 1997) as indicative of the character traits of the in-game characters (cf. Bizzocchi and Tanenbaum, 2011). Similarly, the fireball-like aesthetics of the quick-time events of God of War III (Santa Monica Studio, 2010) indicate that every act performed by Kratos is an act of fiery brutalism, and that the PC is by extension capable of executing such acts.

Discussion

The previous taxonomy identified different characterization cues in games. These cues should of course be interpreted in relation to their context: Vella (2013) is right to point out that “[Nathan] Drake and Mario both jump, but while Drake's leap from one precarious ledge to another is the desperate act of a man doing what he has to do to survive, Mario's bound across platforms is an expression of sheer, kinaesthetic joy” (p. 5).

It is important to acknowledge that cues may overlap with one another. Objects may, for example, be tied to a character's appearance. Likewise, audio and interface can be used diegetically through integration into the storyworld (Iacovides et al., 2015; Kassabian, 2000). Similarly, not all of the aforementioned cues necessarily operate at the same time. Acting does not play a role in the case of silent 8-bit characters, and haptic feedback is not present in games played with a keyboard and mouse. However, a certain degree of abstraction is required for a heuristic model of this kind (Thon, 2016), and we, therefore, propose the above framework not as rigid or inflexible, but rather as a systematic and comprehensive critical vocabulary with which characterization in games can be discussed. Certain cues will not be used with certain characters (or will be less appropriate in certain genres) but may still be productive of characterization in other contexts.

As described above, previous scholarship distinguished between direct and indirect characterization, depending on whether character information is given explicitly or implicitly. This dichotomy operates similarly in games: when Assassin's Creed Odyssey's Herodotos describes Kassandra as a “mighty mercenary and traveler,” he is directly characterizing her as “mighty.” However, we argue that these could be considered more flexible modalities depending on how they are used. A character's name, for example, may convey traits directly in the case of ‘telling names' (Eder et al., 2010), but can equally hint at traits more indirectly (e.g., in the case of a character of nobility with multiple surnames and a title).

The above model incorporated De Temmerman and van Emde Boas’ (2018) metonymical characterization techniques as storyworld cues (with the exception of focalization, which pertains to the medium's construction of the narrative). The cues can be used metaphorically as well if they draw comparisons or contrasts between a character and someone/something else: in Assassin's Creed Odyssey, the politician Kleon proclaims he will “make Athens great again”—a clear reference to the ‘Make America Great Again’ slogan of the 2016 U.S. election, metaphorically imbuing the game's Kleon with Trumpian personality traits. Likewise, musical leitmotifs (Eder et al., 2010) can produce metaphorical characterization: in Assassin's Creed Origins (Ubisoft Montreal, 2017), an Egyptian-themed version of the franchise leitmotif ‘Ezio's Family' plays when Aya and Bayek create the Brotherhood of Assassins/The Hidden Ones. The theme opens a comparison with other assassins in the series, and inducts Aya and Bayek into this larger group. An ergodic variant of metaphorical characterization occurs when players role-play their PC as a specific persona or type (e.g., wizard, superhero).

Character Variants

As mentioned above, players may ignore the character aspects of their PC, and purely consider the PC as an avatar, or the prosthetic (Klevjer, 2006) in-game extension of their real-life selves. Similarly to how they could “refuse” (Juul, 2005, p. 139) the narrative dimension of a game, players may skip cutscenes with character information, select choices at random, or play the game purely for ludic interests such as competition (Caracciolo, 2015). This brings up an important methodological issue for game character researchers (cf. Carr, 2019): since games are ergodic (Aarseth, 1997), how can we accurately describe character X as characterized as Y in game Z if different players may actualize different characterizations of X within the same game system?

It is, for example, important to be aware of how the text system is configured. Lankoski et al. (2003/2010) wrote that the characterization of Garrett in Thief II: The Metal Age (Looking Glass Studios, 2000) depends on the difficulty level chosen by the player: since the highest difficulty forbids the player/Garrett to kill enemies, his actions will differ from those on the normal or easy difficulty. It thus seems advisable to advocate for reflexivity and transparency in game character research: since video games are an ergodic and, by extension, personal medium (i.e., players actualize the game system differently, to varying extents), Daneels et al. (2022) have pleaded that game scholars should be aware of their own impact on the analyzed text, and provide insight into their influence while reporting their analysis results. Such transparency may be achieved by describing relevant boundaries placed on the game (e.g., software version, platform, in-game choices) or through the online publication of recorded gameplay scenes in the form of a “game corpus” (Daneels et al., 2022). Such corpora offer readers insight into researchers’ personal experiences with the investigated game and its characters.

Researchers may additionally play the game repeatedly and consciously adopt different strategies of play (e.g., Aarseth, 2003), organize experiments with various researchers or players (Daneels et al., 2022) to compare character variants, or investigate “meta-ludic texts” (Masso, 2009) such as gameplay videos or game journalism articles to gain awareness of latent play content. However, such repeated play or multi-researcher experiments are highly time-consuming (Bizzocchi and Tanenbaum, 2011) and therefore not feasible in every research project. In those cases, the above plea for transparency in game research gains special importance.

Future Research

Future research could empirically apply the theoretical model to specific (sets of) VGCs. The framework is currently used in the first author's research on the characterization of Greco-Roman gods and heroes in video games. Further applications of the model on different game samples could, for example, provide insight into consistencies and deviations in VGC characterization within or across game genres, historical evolutions of VGC characterizations (e.g., Gilbert et al., 2021; Lynch et al., 2016), or the prevalence of certain characterization cues in specific types of games (e.g., genres, developers, production contexts). As games and game technologies evolve, it is also important to keep testing and updating the framework in follow-up research.

Additionally, future research could investigate in more detail why and how players characterize their PC variants in specific ways. Such analyses will illuminate the ergodic, narrative, cognitive, emotional, and/or strategic ways in which players treat PCs. In order to assess empirically which decisions shape players’ individual characterization processes, the authors are currently experimenting with a questionnaire based on the above framework, including questions such as What did you make the character say, and why? referring to speech, What did you name the character, and why? referring to name-giving, and so on. When applied to specific samples of game audiences, such questionnaire-based analyses will further our understanding of how players engage with PCs.

Conclusion

This paper has forwarded a theoretical framework to examine and study characterization in video games, based on insights from literary, film, and game studies. The framework offers a comprehensive list and methodological vocabulary of characterization cues in games that may be used to study the characterization of both specific characters or broader samples of game characters across genres, time, or production contexts. Finally, in line with similar calls in game studies, this article has advocated for researcher transparency in character research.

Footnotes

Acknowledgments

The authors would like to thank the anonymous reviewer, Marco Caracciolo, and Koen De Temmerman for their constructive thoughts and most valuable feedback on an earlier version of this manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Fonds Wetenschappelijk Onderzoek.

ORCID iD

Alexander Vandewalle

Author Biographies

Alexander Vandewalle is a Joint PhD Researcher at the University of Antwerp (Department of Communication Studies) and Ghent University (Department of Literary Studies) in Belgium, where he studies the characterization of Greco-Roman gods and heroes in video games. He holds an MA in Linguistics & Literature: Latin & Greek from Ghent University (2017) and an MSc in Film Studies and Visual Culture from the University of Antwerp (2018). He has previously published and presented on game analysis methodology, characterization in video games, Homeric intertextuality in Assassin's Creed Odyssey, epigraphy in classical antiquity games and pedagogical applications of historical video games. He is also the creator of Paizomen (), a work-in-progress database of video games set in classical antiquity.

Steven Malliet is an Associate Professor at the Department of Communication Studies, University of Antwerp, Belgium, and a Senior Lecturer/Researcher at the LUCA School of Arts in Genk, Belgium. He holds a PhD in Communication Sciences from KU Leuven, and his research focuses on digital game effects, player experiences, game design, and game realism.

Kristoffel Demoen is a Professor at the Department of Literary Studies, Ghent University, where he teaches or has taught Ancient Greek Literature, Translation Theory, and Classical Mythology. His approach to literary texts is consistently informed by the theoretical frameworks of rhetoric, intertextuality, and narratology.

References

Aarseth

(1997). Cybertext. Perspectives on ergodic literature. John Hopkins University Press.

Aarseth

(2003). Playing research: Methodological approaches to game analysis. In Melbourne DAC – The 5th International Digital Arts and Culture conference (pp. 1–7).

Aarseth

(2004). Genre trouble: Narrativism and the art of simulation. In Wardrip-Fruin

Harrigan

(Eds.), First person: New media as story, performance, and game (pp. 45–55). MIT Press.

Aldred

(2016). Characters. In Wolf

M. J. P.

Perron

(Eds.), The Routledge companion to video game studies (pp. 355–363). Routledge. (Original work published in 2014).

Alientrap. (2015). Apotheon [Video game]. Alientrap.

Alvelo

(2021). Far Cry 6’s villain DLC makes me want to catch up on the series . HP Critical. https://www.hpcritical.com/all-post/2021/6/13/far-cry-6s-villain-dlc-makes-me-want-to-catch-up-to-the-series

Arcagni

D’Aloia

(2021). VR storytelling: Potentials and limitations of virtual reality narratives. Cinergi – Il Cinema e le Altre Arti, 10(19), 1–7. https://doi.org/10.6092/issn.2280-9481/13412

Aristotle. (1995). Poetics (S. Halliwell, Trans.). Loeb Classical Library.

Arjoranta

(2013, Oct 31-Nov 1). Meaning effects in video games. Focalization, granularity and mode of narration in games [Paper presentation]. First International Conference in Games and Literary Theory, Valetta, Malta). https://jonne.arjoranta.fi/2013/meaning-effects-in-video-games-talk/

10.

Atwood

(2018). The Penelopiad. Canongate. (Original work published in 2005)

11.

Bal

(1999). Narratology. Introduction to the theory of narrative (2nd ed.). University of Toronto Press. (Original work published in 1997).

12.

Barthes

(1975). S/Z (R. Miller, Trans.). Jonathan Cape. (Original work published in 1970)

13.

Bartle

(2003). Designing virtual worlds. Creative Commons Licence.

14.

Bayliss

(2007). Beings-in-the-gameworld: Characters, avatars and players. In Proceedings of the 4th Australasian Conference on Interactive entertainment: IE2007 (pp. 1–6).

15.

Berman

R. A.

(1997). Fade in: The screenwriting process (2nd ed.). Michael Wiese Productions. (Original work published in 1988).

16.

Bertetti

(2014). Toward a typology of transmedia characters. International Journal of Communication, 8, 2344–2361. https://ijoc.org/index.php/ijoc/article/view/2597/1201

17.

Bethesda Game Studios. (2011). The Elder Scrolls V: Skyrim [Video game]. Bethesda Softworks.

18.

BioWare. (2007). Mass Effect [Video game]. Microsoft Game Studios.

19.

Bizzocchi

(2007). Games and narrative: An analytical framework. Loading… The Journal of the Canadian Games Studies Association, 1(1), 1–10. https://journals.sfu.ca/loading/index.php/loading/article/view/1

20.

Bizzocchi

Tanenbaum

T. J.

(2011). Well read: Applying close reading techniques to gameplay experiences. In Davidson

(Ed.), Well played 3.0. Video games, value and meaning (pp. 289–315). ETC Press.

21.

Boellstorff

(2008). Coming of age in Second Life: An anthropologist explores the virtually human. Princeton University Press.

22.

Capcom. (1991). Street Fighter II [Video game]. Capcom.

23.

Caracciolo

(2015). Playing Home: Videogame experiences between narrative and ludic interests. Narrative, 23(3), 231–251. https://doi.org/10.1353/nar.2015.0022

24.

Carr

(2019). Methodology, representation, and games. Games and Culture, 14(7–8), 707–723. https://doi.org/10.1177/1555412017728641

25.

Chatman

(1978). Story and discourse. Narrative structure in fiction and film. Cornell University Press.

26.

Crystal Dynamics. (2020). Marvel’s Avengers [Video game]. Square Enix.

27.

Daneels

Denoo

Vandewalle

Dupont

Malliet

(2022). The Digital Game Analysis Protocol (DiGAP): Introducing a guide for reflexive and transparent game analyses. Game Studies, 22(2). https://gamestudies.org/2202/articles/gap_daneels_denoo_vandewalle_dupont_malliet

28.

De Jong

(2018). Homer. In De Temmerman, K., & van Emde Boas, E. (Eds.), Characterization in ancient Greek literature (pp. 27–45). Brill.

29.

De Temmerman

van Emde Boas

(2018). Character and characterization in ancient Greek literature: An introduction. In De Temmerman, K., & van Emde Boas, E. (Eds.), Characterization in ancient Greek literature (pp. 1–23). Brill.

30.

Dyer

(2004). Stars. BFI Publishing. (Original work published in 1998).

31.

Eder

Jannidis

Schneider

(2010). Characters in fictional worlds. An introduction. In Eder

Jannidis

Schneider

(Eds.), Characters in fictional worlds. Understanding imaginary beings in literature, film, and other media (pp. 3–64). De Gruyter.

32.

Eder

Thon

J.-N.

(2012). Digitale Figuren in Kinofilm und Computerspiel. In Segeberg

(Ed.), Film im Zeitalter neuer Medien II: Digitalität und Kino (pp. 139–182). Fink.

33.

Egenfeldt-Nielsen

Smith

J. H.

Tosca

S. P.

(2016). Understanding video games (3rd ed.). Routledge. (Original work published in 2008).

34.

Eskelinen

(2001). The gaming situation. Game Studies, 1(1). https://www.gamestudies.org/0101/eskelinen/

35.

Fernández-Vara

(2011, May 25). The paradox between player control and player character in adventure games [Paper presentation]. Annual Meeting of the International Communication Association 2011, Boston (MA), USA.

36.

Fitzpatrick

Walsh

Nitsche

(2005, Oct 14–15). Character data sets and parameterized morality [Paper presentation]. Aesthetics of Play Conference, Bergen, Norway.

37.

Fizek

(2012). Pivoting the player. A methodological toolkit for player character research in offline role-playing games. [Doctoral dissertation, Bangor University].

38.

Forster

E. M.

(1969). Aspects of the novel. Edward Arnold (Publishers) Ltd. (Original work published in 1927).

39.

Frasca

(2003). Simulation versus narrative: Introduction to ludology. In Wolf

M. J. P.

Perron

(Eds.), The video game theory reader (pp. 221–236). Routledge.

40.

Frow

(2014). Character & person. Oxford University Press.

41.

Galloway

A. R.

(2006). Gaming. Essays on algorithmic culture. University of Minnesota Press.

42.

Genette

(1972). Figures III. Seuil.

43.

Gilbert

Lynch

Burridge

Archipley

(2021). Formidability of male video game characters over 45 years. Information, Communication & Society, 26(1), 1–17. https://doi.org/10.1080/1369118X.2021.2013921

44.

Gilligan

., Johnson

., & MacLaren

. (Executive Producers). (2008–2013). Breaking Bad . [TV series]. Sony Pictures Television.

45.

Greimas

A. J.

(1966). Sémantique structurale. Recherche de méthode. Larousse.

46.

Grimes

S. M.

(2003). “You shoot like a girl!” The female protagonist in action-adventure video games. In DiGRA ‘03 – Proceedings of the 2003 DiGRA International Conference: Level up, 2, 1–15.

47.

Hadas

(2020). Authorship as promotional discourse in the screen industries. Selling genius. Routledge.

48.

Hagström

(2008). Playing with names: Gaming and naming in World of Warcraft. In Corneliussen

Rettberg

J. W.

(Eds.), Digital culture, play and identity: A World of Warcraft reader (pp. 265–285). MIT Press.

49.

Hamon

(1972). Pour un statut sémiologique du personnage. Littérature, 6, 88–110. https://doi.org/10.3406/litt.1972.1957

50.

Heidbrink

(2010). Fictional characters in literary and media studies. A survey of the research. In Eder

Jannidis

Schneider

(Eds.), Characters in fictional worlds. Understanding imaginary beings in literature, film, and other media (pp. 67–110). De Gruyter.

51.

Hello Games. (2016). No Man’s Sky [Video game]. Hello Games.

52.

Herman

(2009). Basic elements of narrative. Wiley-Blackwell.

53.

Herman

Vervaeck

(2005). Handbook of narrative analysis. University of Nebraska Press.

54.

Homer. (1999). Iliad. Books 1–12 (A. T. Murray, Trans.). Loeb Classical Library. (Original work published in 1924)

55.

Iacovides

Cox

Kennedy

Cairns

Jennett

(2015). Removing the HUD: The impact of non-diegetic game elements and expertise on player involvement. In CHI Play ‘15: Proceedings of the 2015 Annual Symposium on Computer-Human Interaction in Play (pp. 13–22).

56.

Isbister

(2006). Better game characters by design. A psychological approach. Morgan Kaufmann.

57.

Jagex. (2001-). RuneScape [Video Game]. Jagex.

58.

Jørgensen

(2010). Game characters as narrative devices. A comparative analysis of Dragon Age: Origins and Mass Effect 2. Eludamos. Journal for Computer Game Studies, 4(2), 315–331. https://doi.org/10.7557/23.6051

59.

Juul

(2005). Half-real. Video games between real and fictional worlds. MIT Press.

60.

Kassabian

(2000). Hearing film. Tracking identifications in contemporary Hollywood music. Routledge.

61.

Kershner

(Director) . (1980). The Empire Strikes Back. Lucasfilm Ltd.

62.

Klastrup

Tosca

S. P.

(2009). “Because it just looks cool!” Fashion as character performance: The case of WoW. Journal of Virtual Worlds Research, 1(3), 4–17. https://jvwr-ojs-utexas.tdl.org/jvwr/article/view/305

63.

Klevjer

(2006). What is the avatar? Fiction and embodiment in avatar-based singleplayer computer games. [Doctoral dissertation, University of Bergen].

64.

Klimmt

Hefner

Vorderer

(2009). The video game experience as “true” identification: A theory of enjoyable alterations of players’ self-perception. Communication Theory, 19(1), 351–373. https://doi.org/10.1111/j.1468-2885.2009.01347.x

65.

Kromand

(2007). Avatar categorization. In Situated play: Proceedings of the 2007 DiGRA conference (pp. 400–406). University of Tokyo Press.

66.

Lankoski

(2010). Character-driven game design. A design approach and its foundations in character engagement. WS Bookwell Ltd.

67.

Lankoski

(2011). Player character engagement in computer games. Games and Culture, 6(4), 291–311. https://doi.org/10.1177/1555412010391088

68.

Lankoski

Björk

(2010a). Character-driven game design: Characters, conflict, and gameplay. In Lankoski

(Ed.), Character-driven game design. A design approach and its foundations in character engagement (pp. 163–180). WS Bookwell Ltd. (Original work published in 2008).

69.

Lankoski

Björk

(2010b). Gameplay design patterns for believable non-player characters. In Lankoski

(Ed.), Character-driven game design. A design approach and its foundations in character engagement (pp. 117–130). WS Bookwell Ltd. (Original work published in 2007).

70.

Lankoski

Heliö

Ekman

(2010). Character in computer games: Toward understanding interpretation and design. In Lankoski

(Ed.), Character-driven game design. A design approach and its foundations in character engagement (pp. 77–90). WS Bookwell Ltd. (Original work published in 2003).

71.

Looking Glass Studios. (2000). Thief II: The Metal Age. Eidos Interactive.

72.

Lynch

Tompkins

J. E.

van Driel

I. I.

Fritz

(2016). Sexy, strong, and secondary: A content analysis of female characters in video games across 31 years. Journal of Communication, 66(4), 564–584. https://doi.org/10.1111/jcom.12237

73.

Macías Villalobos

(2020). The videogame hero in his labyrinth. The hero in electronic gaming. In López Gregoris

Macías Villalobos

(Eds.), The hero reloaded: The reinvention of the classical hero in contemporary mass media (pp. 73–102). John Benjamins.

74.

Malliet

(2007). Adapting the principles of ludology to the method of video game content analysis. Game Studies, 7(1). https://gamestudies.org/7010701/articles/malliet

75.

Margolin

(1986). The doer and the deed: Action as a basis for characterization in narrative. Poetics Today, 7(2), 205–225. https://doi.org/10.2307/1772759

76.

Margolin

(2010). From predicates to people like us. Kinds of readerly engagement with literary characters. In Eder

Jannidis

Schneider

(Eds.), Characters in fictional worlds. Understanding imaginary beings in literature, film, and other media (pp. 400–415). De Gruyter.

77.

Mascelli

J. V.

(1998). The five C’s of cinematography. Motion picture filming techniques. Silman-James Press. (Original work published in 1965).

78.

Masso

I. C.

(2009). Developing a methodology for corpus-based computer game studies. Journal of Gaming and Virtual Worlds, 1(2), 145–169.

79.

McKee

(2014). Story. Substance, structure, style, and the principles of screenwriting. Methuen. (Original work published in 1998).

80.

Meretzky

(2001). Building character: An analysis of character creation . Game Developer. https://www.gamedeveloper.com/design/building-character-an-analysis-of-character-creation

81.

Michaels

(1998). The phantom of the cinema. Character in modern film. State University of New York Press.

82.

Mojang Studios. (2010). Minecraft [Video game]. Mojang Studios.

83.

Naughty Dog. (2013). The Last of Us [Video game]. Sony Computer Entertainment.

84.

Naughty Dog. (2020). The Last of Us Part II [Video game]. Sony Interactive Entertainment.

85.

Neitzel

(2014). Narrativity of computer games. In Hühn

Meister

J. C.

Pier

Schmid

(Eds.), Handbook of narratology (2nd ed., pp. 608–622). De Gruyter.

86.

NetherRealm Studios. (2015). Mortal Kombat X [Video game]. Warner Bros. Interactive Entertainment.

87.

Obsidian Entertainment. (2019). The Outer Worlds [Video game]. Private Division.

88.

Paprocki

(2021). By the Power of Zeus: Apotheon, Divine Power-Ups and their Classical Inspirations . Paizomen. https://paizomen.com/2021/10/19/by-the-power-of-zeus-apotheon-divine-power-ups-and-their-classical-inspirations-by-maciej-paprocki/

89.

Petersen

(Director) . (2004). Troy . Warner Bros. Pictures.

90.

Phelan

(1989). Reading people, reading plots. Character, progression, and the interpretation of narrative. University of Chicago Press.

91.

Propp

V. J.

(1968). Morphology of the folktale. University of Texas Press. (Original work published in 1928).

92.

Real World Multimedia. (1997). Ceremony of Innocence [Video game]. Real World Multimedia.

93.

Richardson

(2010). Transtextual characters. In Eder

Jannidis

Schneider

(Eds.), Characters in fictional worlds. Understanding imaginary beings in literature, film, and other media (pp. 527–541). De Gruyter.

94.

Rimmon-Kenan

(2004). Narrative fiction. Contemporary poetics (2nd ed.). Routledge. (Original work published in 1983).

95.

Ryan

M.-L.

(1985). The modal structure of narrative universes. Poetics Today, 6(4), 717–755. https://doi.org/10.2307/1771963

96.

Santa Monica Studio. (2005). God of War [Video game]. Sony Computer Entertainment.

97.

Santa Monica Studio. (2010). God of War III [Video game]. Sony Computer Entertainment.

98.

Schneider

(2001). Towards a cognitive theory of literary character: The dynamics of mental-model construction. Style, 35(4), 607–640. https://www.jstor.org/stable/10.5325/style.35.4.607

99.

Schröter

(2013). ‘Don’t show it, play it!’ Filmische und nicht-filmische Figurenkonzeption im Computerspiel. Rabbit Eye. Zeitschrift für Filmforschung, 5, 22–39. https://www.rabbiteye.de/2013/5/schroeter_figurenkonzeption.pdf

100.

Schröter

Thon

J.-N.

(2014). Video game characters. Theory and analysis. Diegesis, 3(1), 40–77.

101.

Sheldon

(2004). Character development and storytelling for games. Thomson Course Technology.

102.

Simons

(2007). Narrative, games, and theory. Game Studies, 7(1). https://gamestudies.org/0701/articles/simons

103.

Smith

(1995). Engaging characters: Fiction, emotion, and the cinema. Clarendon Press.

104.

Smith

(2010). Engaging characters: Further reflections. In Eder

Jannidis

Schneider

(Eds.), Characters in fictional worlds. Understanding imaginary beings in literature, film, and other media (pp. 232–258). De Gruyter.

105.

Square Enix. (2020). Final Fantasy VII Remake [Video game]. Square Enix.

106.

Summers

(2016). Understanding video game music. Cambridge University Press.

107.

Telesinski

(2016). Kenshi | System. Mortal Kombat X Guide . GamePressure. https://guides.gamepressure.com/mortalkombatx/guide.asp?ID=30623

108.

Thon

J.-N.

(2016). Transmedial narratology and contemporary media culture. University of Nebraska Press.

109.

Thon

J.-N.

(2017). Transmedial narratology revisited: On the intersubjective construction of storyworlds and the problem of representational correspondence in films, comics, and video games. Narrative, 25(3), 286–320. https://doi.org/10.1353/nar.2017.0016

110.

Thon

J.-N.

(2019). Transmedia characters: Theory and analysis. Frontiers of Narrative Studies, 5(2), 176–199. https://doi.org/10.1515/fns-2019-0012

111.

Titan Forge Games. (2014-). Smite [Video game]. Hi-Rez Studios.

112.

Tronstad

(2008). Character identification in World of Warcraft: The relationship between capacity and appearance. In Corneliussen

Rettberg

J.W.

(Eds.), Digital culture, play and identity: A World of Warcraft reader (pp. 249–264). MIT Press.

113.

Ubisoft Montreal. (2017). Assassin’s Creed Origins [Video game]. Ubisoft.

114.

Ubisoft Montreal. (2020). Assassin’s Creed Valhalla [Video game]. Ubisoft.

115.

Ubisoft Quebec. (2018). Assassin’s Creed Odyssey [Video game]. Ubisoft.

116.

Ubisoft Quebec. (2020). Immortals Fenyx Rising [Video game]. Ubisoft.

117.

Ubisoft Toronto. (2021). Far Cry 6 [Video game]. Ubisoft.

118.

Vella

(2013). “It’s A-Me/Mario!”: Playing as a ludic character. In Foundations of Digital Games Conference Proceedings, 8, 31–38.

119.

Vella

(2015). The ludic subject and the ludic self: Analyzing the ‘I-in-the-gameworld’. [Doctoral dissertation, IT University of Copenhagen].

120.

Vogler

(2007). The writer’s journey. Mythic structure for writers (3rd ed.). Michael Wiese Productions. (Original work published in 1998).

121.

Willumsen

E. C.

(2018). Is my avatar MY avatar? Character autonomy and automated avatar actions in digital games. In DiGRA ‘18 – Proceedings of the 2018 DiGRA International Conference: The Game is the Message (pp. 1–17).

122.

Willumsen

E. C.

Jaćević

(2018). Avatar-kinaesthetics as characterisation statements in Horizon: Zero Dawn. In DiGRA ‘18 – Abstract Proceedings of the 2018 DiGRA International Conference: The Game is the Message (pp. 1–4).

123.

Yager Development. (2012). Spec Ops: The Line [Video Game]. 2K Games.