Game Studies

The “Moral Disengagement in Violent Videogames” Model

Abstract

How do violent videogames, as entertainment products, communicate violence in the context of warfare and in other settings? Also, why do users enjoy virtual violence? The present article introduces the Moral Disengagement in Violent Videogames model to tackle these important questions. The model resulted from empirical insights gained in (Media) Psychology and Communication Science. These disciplinary research findings are reviewed in the present article to substantiate the four core propositions of the model. First, the model refutes the view that users enjoy virtual violence primarily because they are constantly aware that "this game is not real." Instead, the model follows experimental findings suggesting that videogame users are inclined to automatically feel present in virtual environments and (despite better knowledge) may intuitively feel like actually enacting violence against social beings. Second, if users feel as if virtual characters have a mind of their own, they may also assign a moral status to them. Third, this finding implies that enacted unjustified transgressions against virtual characters may trigger discomfort in users. Fourth, most violent videogames are designed for entertainment purposes. Therefore, the games might frequently embed cues that effectively frame violence enacted against seemingly social beings as "okay." Therefore, users may become morally disengaged and enjoy the violence, rather than experiencing feelings of subjective discomfort, such as guilt. The article concludes with a brief discussion of the boundaries of the proposed model and alternative explanations.

Keywords: violent videogame, violence, virtual violence, moral disengagement, morality, enjoyment, guilt, social perception, justification, entertainment

How do violent videogames communicate warfare and violence, and how do users, in turn, experience virtual violence? The present article introduces the "moral disengagement in violent videogames model" (in short "MoDViG model") to tackle this important question. The MoDViG model consists of four propositions that were derived based on empirical findings from communication-scientific and media-psychological studies (Hartmann, 2011, 2012; Hartmann & Vorderer, 2010; Hartmann, Toz, & Brandon, 2010; Hartmann et al., 2014; Klimmt et al., 2006). The purpose of this article is to introduce the MoDViG model and its media-psychological underpinnings to the game studies community. The hope is that this cross-disciplinary introduction is appreciated and inspires further debate because, of course, crossing disciplinary boundaries is a challenging task. While I tried to reflect upon some of the existing game studies literature, admittedly, there is considerably more to be acknowledged. Accordingly, my hope is that introducing the media-psychological MoDViG model might spark further cross-disciplinary debate and pollination.

Research leading to the model was originally motivated by the question of why users enjoy virtual violence, rather than feeling guilty about it. From a media-psychological perspective, the answer to this question seemed challenging. On the one hand, if asked in interviews, most gamers stressed that they do not feel bad when enacting virtual violence because they distinguish reality from fiction and know that "this is not truly happening" (Klimmt et al., 2006). This distinction seemed to provide a plausible explanation. However, on the other hand, empirical studies from different contexts (e.g., on human-computer interaction, virtual reality, artificial social agents) increasingly suggested that, despite better knowledge, users might also automatically feel like they are present in a videogame, are actually acting in the virtual environment and are encountering actual social beings (e.g., Blankendaal et al., 2015; Nass & Moon, 2000; Slater et al., 2006). Furthermore, many players reported feeling uneasy regarding conducting certain virtual transgressions, such as shooting innocent children or committing rape, despite their fictional status (Whitty, Young, & Goodings, 2011). Accordingly, "knowing that this is not truly happening" does not provide a conclusive answer to the question of why virtual violence is enjoyable. The MoDViG model therefore proposes a different explanation.

At its core, the MoDViG model builds on Bandura's (1991) social-cognitive theory of moral thought and action. More specifically, the model adapts the idea that individuals perceive transgressions as justified (or as not morally significant) if they morally disengage. Bandura argues that moral disengagement techniques are frequently embedded in warfare and military training (see for an overview Moore, 2015). Similarly, according to the MoDViG model, users may experience violence in videogames primarily in a state of moral disengagement in which they perceive their intentionally violent conduct against seemingly social beings on the screen as appropriate and legitimate. Virtual violence may be enjoyable, not primarily because users were continuously aware that "what is happening is not real" but because videogames effectively morally disengage users in seemingly real violent conduct. In the MoDViG model, this central argument is explicated in four propositions that will be further discussed and substantiated in the remainder of this article:

1. Despite knowing otherwise (e.g., if asked in ex-post interviews), players automatically or intuitively perceive videogame characters as social beings who have "a mind of their own" while playing.
2. In general, the moral socialization of individuals involves the idea that social beings are deemed worthy of proper moral treatment. Accordingly, while playing, users may tend to automatically perceive videogame characters as beings worthy of proper moral treatment.
3. Improper treatment of videogame characters can violate player's norms and trigger empathetic distress and feelings of guilt. Guilt effectively diminishes enjoyment.
4. Videogames are entertainment products — and they are designed in such a way as to make violence enjoyable. Violent videogames frequently embed moral disengagement cues that effectively frame the violence as justifiable. Accordingly, players tend to enjoy videogame violence and related warfare scenarios (rather than feeling guilty or empathetic distress) because they are morally disengaged while playing.

Proposition 1: Videogame Characters Are Intuitively Perceived as Social Beings

All of us are familiar with laughing about good events happening to characters or crying about bad events happening to characters we only perceive on a screen. Why do we respond in such ways? Literature in the humanities and the related game studies community provides a number of approaches that tackled this question, including the widely debated paradox of fiction (Mason, 2014) or gamer's dilemma (Luck, 2009). An important unifying theoretical account is proposed by Frome (2006). This researcher proposes that we derive emotions, e.g., if exposed to videogames, based on at least two mental systems that appraise perceived events in different ways. The researcher vividly explains how both systems differ by referring to how users might respond to a scary alien in a film or videogame (p. 15): "Some of the viewer's mental systems respond to the alien as if it is a real danger, raising her heart rate and increasing her adrenaline. Other mental systems, such as the ones responsible for running and hiding, recognize that the alien is fictional and thus not a real danger that she must avoid." Frome's explication of how users emotionally respond to mediated representations might hint at an important psychological explanation of why video games are best described as half-real (Juul, 2005), and feature a porous, rather than completely sealed, magic circle (Castronova, 2005). Indeed, scholars working in media psychology and neighbouring fields that investigated how human-beings respond to audio-visual mediated representations of objects or social figures consistently show that they respond "almost as if" the displayed objects or figures are real (see for an overview Hartmann & Vorderer, 2010; Hartmann, 2011). For example, based on a range of experiments, the widely influential media equation theory (Reeves & Nass, 1996; Nass & Moon, 2000) boldly suggests that people tend to treat computers and other media as if they were either real people or real places. Reeves and Nass assume that the reason is that the human brain has not evolved into fully understanding 21st century technology. Other scholars support at least a weaker version of the media equation by showing that human users tend to automatically perceive even simple animations as something social (rather than objects). For example, Mar and Macrae (2006, p. 110) observe that we "routinely view quite abstract nonliving representations as if they were intentional agents." Similarly, Heberlein and Adolphs (2004, p. 7490) find that "anthropomorphizing [...] occurs when we attribute social meanings to stimuli that are not social, such as computers or clouds, presumably based on cues that signal the presence of agency or emotion. That we do so universally and automatically is a hallmark of human cognition."

Research on parasocial interaction (Hartmann & Goldhoorn, 2012) and social presence (e.g., Bailenson, Blascovich, Beall, & Loomis, 2003) also suggests that users can temporarily feel as if the mediated, and occasionally even completely computer-generated, characters they encounter are actually social, even if they are not existing social beings. Contemporary artificial characters, such as robots or digital characters, displayed in videogames or virtual reality environments often feature subtle cues that can provoke automatic social responses (Holtgraves et al., 2007; Morrison & Ziemke, 2005; Shapiro, Peña & Hancock, 2006; Simion et al., 2013), including eye-gazing, natural vocal tones, biological motion, displays of natural facial activity and emotions, and displays of intelligence. Consequently, users may intuitively feel as if they have encountered an actually living social being that has "a mind of its own" when encountering these technology-generated characters (see for an explanation of this process Arico et al., 2011)[1]. Other lines of research concur with this assumption, including those on users' general illusions of non-mediation (Lombard & Ditton, 1995) and on what users intuitively believe to be true or false ("magic window" in the perceived realism research, Busselle & Greenberg, 2000; Hall, 2003).

In this context, I recently attempted to distinguish between what users simultaneously feel and know concerning mediated objects or people. The core argument of my dual process model of media reality (Hartmann, 2011, 2012) is that media illusions may typically imply that users simultaneously know that "this is not true," but "feel as if it is true." Knowing and feeling diverge and suggest different conclusions about what is actually true. The dual-process model of media reality combines the existing insights on analytic vs. involved modes of experiencing media (Cupchick, 2002; Vorderer, 1993) with the psychological dual-process logic that human-beings can process information either in a highly automatic, bottom-up, sensory, and associative manner (System I) or in a more elaborate, effortful, top-down, logical manner (System II). From this perspective, accurate sensory input (e.g., realistically displayed behaviour and facial features of virtual characters) may give rise to the intuitive feeling that the social character would really exist — right here, right now (System I process). If accurately animated, users may perceive something social (and alive), rather than an inanimate object, and might even feel compelled to perceive that character as an actual existing entity. At the same time, top-down knowledge and plausibility checks may embed this intuitive feeling into a "knowing state" that this character does not truly exist, that it is not truly alive, and that his or her behaviour is not truly happening right and here and now (System II process). Accordingly, from a dual-process perspective, videogame characters are simultaneously perceived as existing — or at least living — creatures (namely, within System I), and as virtual representations (namely, within System II)[2].

Certainly, further empirical research is needed to fully understand how much users "knowing that this is not real" and their apparent "feeling that this is real" relate to each other in their encounters with, for instance, virtual beings, animated agents, and robots. However, given the efforts most media users, particularly in an entertainment context and including videogame users, spend not to be reminded of what is real (e.g., by shutting off the lights or mobile phones in a cinema; also see Leino, 2015, for an illuminating discussion of this motivation in the context of game love), it seems safe to say that often the videogame user's feeling that "this character is truly alive and present" may dominate their overall experience. In contrast, users' knowledge that "these are just pixels on a screen" may often temporarily recede into the background of their minds. Accordingly, although users' responses may be less intense under virtual conditions compared to real ones, they seem sufficiently substantial to matter.

Proposition 2: To Feel that "This Is Alive" Implies Treating It with Respect

The moral socialization of most people emphasizes the idea that social beings are deemed worthy of proper moral treatment. Accordingly, if videogame users intuitively see something social instead of pixels and even feel as if encountered videogame characters are alive or really exist, they should also tend to automatically perceive them as agents worthy of proper moral treatment, i.e., while playing, videogame users may be inclined to assign a moral status (Warren, 1997) to encountered videogame characters.

As Pizarro, Detweiler-Bedell, and Bloom (2006, p. 82) observed, "Nobody feels guilty about kicking a rock for the simple pleasure of doing so, but doing the same thing to a child is universally forbidden. What's the difference?" Olthof, Rieffe, Terwogt, Lalay-Cederburg, Reijntjes, and Hagenaar (2008, p. 234) argue that an entity is perceived to possess a moral status if it "belongs to the category of entities to which moral subjects have moral obligations" (see also "moral circle," Singer, 1981). Accordingly, if people assign a moral status to objects, these objects become moral patients. Olthof et al. (2008) examined the characteristics that determine why others assign a moral status to an entity. These researchers assumed that the assignment of a moral status depends on whether others believe the entity is able to suffer, able to think, and - as a most basic mental capacity - able to perceive. These researchers' study shows that younger individuals rely on an entities' perceived ability to suffer when assigning a moral status, whereas the impression that the other can think becomes a more important criterion for older individuals.

Most likely, the most coercive evidence for proposition 2 of the MoDViG model stems from research by Gray, Waytz, and colleagues (e.g., Gray, Young, & Waytz, 2012; Gray, Gray, & Wegner, 2007; see for an overview Epley & Waytz, 2010). These authors show that the intuitive perception that another entity has a mind provides the essence of moral judgement. Mind perception, in turn, rests on two factors: People may intuitively feel an entity seems capable of perceiving things ("experience," i.e., emotion and sensations, such as pain, pleasure, hunger, and desire) and is able to intentionally do things ("agency," i.e., self-control and action, including planning, controlled thoughts, etc.). Gray et al. (2007) show that people are the more inclined to protect another entity from destruction the more they intuitively feel that the entity is able to experience things.

Once videogame players (despite better knowledge) intuitively feel as if encountered videogame characters are alive and may even existent, they will also automatically perceive them to have a mind (Arico et al., 2011). Accordingly, users will be inclined — while playing — to assign a moral status to videogame characters and to intuitively perceive them as individuals worthy of proper moral treatment (Epley & Waytz, 2009). In the context of videogame violence, this should imply that they feel sympathetic towards virtual victims of unjustified violence. This notion corresponds to an argument by Pizarro (2006, p. 92) that "given the right conditions, one can even feel sympathy for inanimate objects."

Accordingly, observing that virtual characters became victims of unnecessary or unjustified violent conduct may trigger sympathetic distress in users ("suffering with" as a form of empathy; Morrison & Ziemke, 2005). For example, in a related study examining users' affective responses to robots, Rosenthal-von der Pütten, Krämer, Hoffmann, Sobieraj, and Eimler (2013) observed increased physiological, and fewer positive and more negative feelings, and increased empathetic concern among participants that watched a video in which a robot was seemingly tortured compared to a normal video portraying the robot. Other studies provide more direct evidence for the assumption that users routinely feel sympathetic towards virtual characters. For example, in a replication of the famous Milgram experiment by Slater et al. (2006), participants reported that they felt like they were inflicting pain on a virtual character, despite simultaneously knowing that this was impossible. In another study by Bouchard et al. (2013), participants were empathetic towards a virtual character expressing pain in a virtual environment, particularly if they felt a sense of co-presence with the character. In summary, these findings are congruent with the assumption that users are inclined to intuitively feel as if encountered videogame characters are alive and that they, as a consequence, tend to perceive them as moral entities (see also Young, 2013).

Proposition 3: Improper Treatment of Videogame Characters Violates Norms and Triggers Guilt

If videogame players intuitively perceive videogame characters as social beings worthy of proper moral treatment, seemingly harming them for no good reason should be perceived as a norm violation and, accordingly, result in at least mild forms of self-condemnation or guilt. Bandura (1991) proposed a general theoretical model of people's moral thoughts and actions. In his model, moral transgressions or norm violations (e.g., hurting others without good reason) are regulated based on two sources, external sanctions and internalized self-sanctions, like, for example, self-contempt in the form of guilt. Bandura argues that the fear of anticipated guilt normally prevents people from engaging in moral transgressions.

Guilt is not an uncommon emotion among videogamers, as the opening of Sicart's (2009) widely received book on the ethics of computer games already suggests ("I am not quite sure how it happened, but I felt guilty," p. 1; also see Jørgensen, 2016). Media-psychological studies also provide evidence that users can feel guilty, particularly if committing seemingly unjustified harm against other videogame characters. For example, videogamers have been shown to shy away from taking moral decisions in a game that would violate their moral convictions, if not urged by the game to do otherwise. It may be argued that users do so to prevent experiencing guilt or discomfort. In an experimental study by Joeckel, Bowman, and Dogruel (2012; also see Dogruel et al., 2013), participants played a modified computer game in which they were confronted with decisions (among two choices) that either violated or upheld aspects of their morality. The study showed that if the game scenario did not appeal to participants' most important or salient moral dispositions (addressed in the study as "moral intuitions"), participants opted for both options with equal likelihood. However, if the game scenario appealed to participants' most important or salient moral dispositions, participants opted more often for the choice that was in line with their moral conviction. Joeckel et al. (2012, p. 460) "interpret this as evidence for a gut or game explanation of decision making in videogames. When users encounter virtual scenarios that prime their moral sensitivities, they rely on their moral intuitions; otherwise, they make satisficing decisions not as an indication of moral corruption but merely as a continuation of the virtual experience." In another study by Weaver and Lewis (2012), participants reported their moral dispositions before they played the first full act of the violent videogame, Fallout 3 (Bethesda Game Studios, 2008). The gameplay was recorded and content-analysed for the moral decisions that participants made while playing. The scholars observed that "the majority of players made moral decisions and behaved towards the nonplayer game characters they encountered as if these were actual interpersonal interactions" (p. 1, see also Schulzke, 2009, for a discussion of the moral choices in Fallout 3).

Often, fear of anticipated guilt may prevent people in everyday life settings and users in videogame settings from engaging in moral transgressions. However, this mechanism may not be flawless. People may sometimes be forced to engage in moral transgressions, they may act on impulse, they may not anticipate feeling guilty, or they may simply not be aware of potential moral conflicts when conducting a behaviour. Accordingly, individuals may sometimes perform actions they feel guilty for later. If people engage in moral transgressions that violate a personally relevant norm, and if they become aware of it, they experience self-condemnation in the form of guilt (Kugler & Jones, 1992). According to Bandura, anticipated guilt prevents immoral behaviour, but actually feeling guilty about one's own behaviour functions as an ex-post correction to discourage future norm violations.

If users intuitively perceive videogame characters as moral entities, they should also feel guilty when hurting these characters without a good reason. Some experimental evidence suggests that this is indeed the case. Lin (2011) observed that participants that played a videogame in which they were fighting against monster opponents (presumably an opponent less worthy of proper moral treatment than human opponents) experienced less guilt and shame, and more enjoyment, than participants that were fighting against human opponents. Another study by Lin (2010) found that, on average, users felt guiltier if playing a morally unjustified character than a morally justified character. Likewise, Hartmann and Vorderer (2010) observed that users fighting against opponents that were guilty of a minor transgression reported feeling more guilt, shame, nervousness, and irritability than players fighting against opponents that committed a condemnable act.

In another study reported by Hartmann and Vorderer (2010), participants had to play a modified version of Operation Flashpoint (Bohemia Interactive, 2001), a first-person shooter game (FPS). All participants watched a cinematic introduction scene in which a torture camp was portrayed that was approached by UN soldiers to restore justice in the name of humanity. Participants were randomly assigned to either play a paramilitary soldier defending the torture camp (unjustified condition), or a UN soldier attacking the camp in the name of humanity (justified condition). In the study, participants that played the paramilitary soldier and shot UN soldiers felt more guilt than participants that played the UN soldier and shot paramilitary forces (for a replication of this finding see Grizzard, Tamborini, Lewis, Wang, & Prabhu, 2014).

Hartmann, Toz, and Brandon (2010) re-analysed this study and found that particularly more empathetic participants felt more guilt if they performed unjustified virtual violence. In another experiment by the same authors, participants had to shoot virtual characters whose social background was either known (e.g., name and job of targets, family status, and individual preferences) or unknown to them. Participants — particularly more empathic ones — reported feeling more guilt if they shot characters whose social background they knew. Similarly, in a study by Schmierbach and Limperos (2013, p. 538) players "reported correspondingly high levels of guilt when their actions were less morally defensible." In summary, these studies suggest that (particularly more empathetic) users tend to feel guilty if they conduct unjustified virtual violence.

In addition to this experimental empirical evidence, anecdotal evidence also suggests that videogame users may feel guilty if engaging in unjustified violence against game characters who are seemingly alive. In the past, even heavy gamers seemed irritated, if not outright morally disgusted, about videogame sequences that urged them to engage in unjustified virtual violence against seemingly social characters. One such example is the interactive torture sequence, called "By the Book," that is featured in the best-selling videogame, Grand Theft Auto V (Rockstar North, 2013). Searching the Internet for player's opinions about this sequence reveals that many users expressed their discomfort with this sequence and that more than a few (albeit not every user) felt intuitively guilty or morally disgusted by it. The "No Russian" mission in Call of Duty: Modern Warfare II (Infinity Ward, 2009), in which users need to commit (along with other computer-navigated characters) a massacre among civilians at an airport, provides another example of seemingly unjustified videogame violence that caused irritation among the gaming community. In summary, these findings provide evidence for the idea that the improper treatment of videogame characters can violate users' norms and make them feel guilty. This effect should particularly apply if innocent video game characters are harmed without any justification or rationalization. However, most (albeit, of course, not all) videogames seek to provide entertainment. Accordingly, the fourth proposition of the MoDViG model is that these videogames frame violence (against seemingly social characters) as morally justified actions.

Proposition 4: For the Sake of Enjoyment - Moral Disengagement in Violent Videogames

According to Bandura's (1991, p. 45) social-cognitive theory of moral thought and action, "self-reactive influences [e.g., self-contempt in the form of guilt] do not operate unless they are activated, and there are many processes by which self-sanctions can be disengaged from inhumane conduct." Bandura argues that particularly in the absence of strong countervailing standards, individuals' moral judgements of behaviour may be susceptible to external influences. For example, most individuals may lack the strong endorsement of non-violent ideals that may be necessary to generally (or absolutely) condemn physical violence. Instead, most people's moral judgements of physical violence may dynamically adapt to the characteristics of the situation or scenario that contextualize a violent act. Under certain circumstances, physical violence may seem legitimate or justified. Accordingly, in certain conditions, physical violence may not appear to be immoral. Bandura (1991, 2002) argues that certain cognitive strategies that may be effectively cued in a situation allow individuals to morally disengage from condemning physical violence and to perceive it as morally acceptable behaviour. "What is culpable can be made righteous through cognitive reconstrual" (1991, p. 72). Bandura (1991, 2002) identifies eight factors that may be cued in certain situations or by scenarios and that effectively trigger cognitive moral disengagement:

1. Moral justification implies that an otherwise reproachable violent act is interpreted as serving some higher purpose (e.g., killing for a greater good or religious purpose).
2. Euphemistic labelling means that an otherwise reproachable violent act is labelled in a way that makes it seem less severe (e.g., killing of civilians as "collateral damage").
3. Advantageous comparison makes an otherwise reproachable violent act seem more righteous by comparing it to an even more heinous act ostensibly committed by the victim (e.g., death sentence of murderers).
4. Attribution of blame implies that victims or circumstances are blamed for necessitating the act (e.g., "that's what they asked for").
5. Displacement of responsibility means that the responsibility for an otherwise reproachable violent act is placed on someone else, often an authority figure (e.g., "I just followed my orders").
6. Diffusion of responsibility implies placing responsibility for an otherwise reproachable violent act on the shoulders of many others (e.g., "I wasn't alone, others did the same").
7. Disregard or distortion of consequences entails that potentially distressful consequences of an otherwise reproachable violent act are concealed or displayed in distorted ways (e.g., if long-distance weapons, such as drones, prevent the perception of the suffering of victims).
8. Dehumanization implies that human victims are portrayed as non-human (and either as animals, e.g., rats, vermin, or non-living beings), which allows the perception of an otherwise reproachable violent act to be viewed as morally legitimate. Dehumanization also occurs if victims are portrayed as interchangeable numbers or targets, rather than individual personalities.

Bandura (1991, 2002) argues that these eight moral disengagement factors may be a primary reason why actors in real-life contexts, such as warfare scenarios, do not feel guilty about violent conduct they would otherwise condemn if occurring under different circumstances.

According to the fourth proposition of the MoDViG model, these eight moral disengagement factors are also frequently embedded in violent videogames. Thus, these factors effectively allow users to perceive violent conduct against seemingly alive characters as a morally legitimate action. Indeed, both controlled experiments and in-depth interviews with players revealed that the presence of moral disengagement factors in violent videogames effectively reduces feelings of guilt among players (Hartmann, 2012; Hartmann & Vorderer, 2010; Klimmt et al., 2006). For example, the justification given for violent acts (Hartmann et al., 2010; Hartmann & Vorderer, 2010; Lin, 2010), the dehumanization of opponents (Gollwitzer & Melzer, 2012; Lin, 2011), and portrayal of consequences of violent actions (Hartmann & Vorderer, 2010) effectively frame violence in videogames against seemingly existing social beings as an "okay action." These findings suggest that moral disengagement factors allow users to enjoy virtual violence, rather than feeling troubled about it (Hartmann, 2011, 2012).

Many (successful commercial) videogames are designed to entertain their users and to provide fun and enjoyment. Accordingly, it may not be surprising that both qualitative and quantitative content analyses of violent videogames also revealed that moral disengagement factors are frequently embedded in violent videogames. Based on an extensive literature review of existing content-analyses of violent videogames, Hartmann et al. (2014, pp. 315-316) conclude that "the findings of previous studies suggest that violence in videogames is often portrayed as justified because opponents are displayed as gruesome (advantageous comparison), players have to fight for a higher good (e.g., to save the world), violence is rewarded, and victims rarely include civilians, females, or children. Quite often, violence also seems to be embedded in humorous or euphemistic contexts. Furthermore, consequences are often portrayed in distorted ways in violent videogames, either because victims are barely visible (e.g., long-distance fights) or because realistic and upsetting forms of the suffering of victims are not displayed (e.g., ongoing physical and psychological pain)."

Supporting this hypothesis, a content-analysis of 189 popular military-themed FPS games (Breuer, Festl, & Quandt, 2012) found that most games (64%) featured a WWII scenario — a war with "clear 'moral frontlines'" (p. 228) that provides a clear moral justification for violence. Another group of analysed games (roughly 7%) featured the war on terror, and presented "military operations [...] such as acts of self-defence of the United States or the Western world" (p. 229), evoking advantageous comparisons and attributions of blame. Furthermore, the authors found that "enemies are typically Russian nationalists, terrorist groups from the Middle East, South American rebels or drug dealing organizations" (p. 229), while unpleasant details, "such as suffering or civilian casualties" were usually not portrayed (p. 230). The latter finding suggests that many of the analysed games portrayed the consequences of military warfare in a distorted way.

Further evidence for the prevalence of moral disengagement factors in violent videogames is provided by a quantitative content-analysis of moral disengagement factors embedded in the narratives and gameplay of 14 top-ranked FPS games (Hartmann et al., 2014). The study suggests that moral disengagement factors are frequently embedded in the narratives and gameplay of FPS games. According to the study, the distortion of consequences, dehumanization, and moral justification are the three most commonly provided moral disengagement factors in FPS games.

In summary, these findings suggest that moral disengagement factors are embedded in the narratives, scenarios, and gameplay of many violent videogames, which effectively free users from feeling guilty if committing violent acts against seemingly alive videogame characters. Videogames are entertainment products, but guilt may disturb the enjoyment received from playing. Accordingly, moral disengagement factors that effectively diminish guilt may be considered a crucial design element of videogames as entertainment products. Furthermore, this notion, together with the finding that moral disengagement factors are indeed frequently embedded in violent videogames, may suggest that these factors characterize how violent videogames typically communicate (or display) violence: namely, as a justified and clean action with negligible consequences against a largely anonymous enemy (whose atrocities seem ill-motivated). Interestingly, this framing of violence seems to be quite similar to the one identified in war movies (Pötzsch, 2014).

Exceptions to the Rule and Alternative Accounts

The MoDViG model aims to explain how videogamers may experience virtual violence. The four propositions of the MoDViG model receive, to a large extent, support by the empirical literature. However, as with any theoretical model, important exceptions to the rule exist. Although it may be true for many best-selling commercial games, not all violent videogames aim to be only fun and communicate violence as justified and clean actions of little consequence (Jørgensen, 2016). Important exceptions have already been noted, such as the shooting of innocent civilians in the "No Russian" mission of Call of Duty: Modern Warfare II (Infinity Ward, 2009) and the torturing of a character in the "By the Book" mission featured in Grand Theft Auto V (Rockstar North, 2013). Both scenarios are different from many other violent videogame scenarios, particularly because they portray the disturbing consequences of physical violence[3]. Similar sequences are featured in other best-selling violent videogames, as well. For example, the FPS game, Spec Ops: The Line (Yager Development, 2012), features a "white phosphorus" scenario in which the user (together with computer-navigated characters) shoots mortar grenades loaded with white phosphorus on the opponents. The gruesome consequences of this attack are portrayed in full detail, both in interactive and cinematic sequences. Another frequently mentioned example of potentially irritating virtual violence is provided by the fictional FPS game, BioShock (2k Boston et al., 2007), in which players can choose to either rescue or "harvest" (i.e., kill) drug-addicted girls called "little sisters" to gain more energy (from the drug they contain). The debate surrounding the videogame, Hatred (Destructive Creations, 2015), in which users play a psychopath that sets out to kill anybody - including fleeing civilians - for no convincing reason, may be considered another example of how users' experience may shift if moral disengagement factors are largely absent in a videogame.

Additionally, an increasing number of alternative or independent videogame productions, designed to inspire reflection rather than only amusement, provide exceptions to the rule that videogames communicate violence as a justified and clean action of little consequence. For example, the simple but intriguing game, September 12th (Frasca, 2003), portrays cartoon-like terrorists and civilians that can be killed by what seems to be a drone strike; other characters mourn their death (and potentially turn into terrorists). In This War of Mine (11 Bit Studios, 2014) users experience the struggle of survival among civilians affected by a war. The game features relatively little physical violence, but many moral dilemmas portraying the consequences of moral transgressions, such as stealing bread from others that need it just as desperately as oneself. In summary, these sequences and videogames provide examples in which moral disengagement factors are largely absent. However, these examples may only represent exceptions to the rule that most commercially successful violent games contextualize violence within a set of moral disengagement factors.

Nevertheless, an important boundary of the MoDViG model should be noted in this context. The MoDViG model builds on the assumption that violent games are designed for entertainment and that hedonic enjoyment is the primary if not the only user experience game designers seek to attire. This notion, of course, can be challenged. For example, as Pitts (2012) recalls, the designers of Spec Ops: The Line (Yager Development) "wanted to create a game that would challenge the player, and bother them a little. They didn't want it to be necessarily fun. They wanted to say something." Whenever self-enacted virtual violence starts to bother users, they may be inclined to engage in higher-order cognitive processing to reflect upon it. This reflection, in turn, may resemble a sense-making process that eventually produces new insights about violence (Schulzke, 2009). The users' reflection marks an important distinction made in entertainment research between hedonic and eudemonic entertainment (Oliver, Bartsch & Hartmann, 2014; see also Jorgensen, 2014). Hedonic entertainment builds on care-free enjoyment, on "simply having a good time." Eudemonic entertainment, in contrast, builds on personal growth, and on meaningful experiences that users appreciate but do not simply enjoy. As an important boundary, the MoDViG model explains the violent videogame experience as a hedonic experience, not as a eudemonic experience.

The MoDViG model argues that moral disengagement factors explain why users enjoy the killing of seemingly alive videogame characters instead of feeling guilty. A similar account is the sanctioned equivalence approach by Young and Whitty (e.g., Young, 2013; Young & Whitty, 2011; Whitty et al., 2011). These researchers argue that users find symbolic behaviour acceptable if it is (at least occasionally) sanctioned in real life (like many forms of physical violence, cannibalism for survive, torture of one to prevent the killing of many) or if it is sanctioned within the game (e.g., killing children to gain energy). However, users may deem symbolic behaviour inappropriate and may feel morally disgusted concerning it if there is no justification for the behaviour in real life, e.g., as is the case for rape. Whitty, Young, and Goodings (2011, p. 268) refer to this type of behaviour as "symbolic taboo activities."

Other important alternative explanations for the observation that players do not feel guilty when engaging in videogame violence can be proposed. For example, maybe with greater literacy, videogame users develop a more automatically triggered understanding that "this is not real" and that they become less inclined to perceive videogame characters as social (see "cognitive orientation," Chalmers, 2016). Users might also develop a tunnel vision in the heat of the gameplay, and, in turn, start perceiving video game characters as object-like targets rather than social beings (e.g., Breuer, Scharkow, & Quandt, 2014). Alternatively, maybe users enjoy doing something that feels wrong while knowing that the behaviour does not really matter and won't be sanctioned. Finally, videogames' magic circle, which bends and reshapes existing norms, might be less porous than the MoDViG presumes.

Conclusions

This paper introduced the MoDViG model, which resulted from research in Media Psychology and Communication Science. Based on four propositions that link to empirical literature, the model aims to explain how videogame users experience virtual violence. In essence, the model argues that violent videogames morally disengage users by framing violence against seemingly alive characters as an "okay action." This disengagement, in turn, allows players to enjoy the virtual violence, rather than feeling troubled by it.

As a concluding step, it may be worthwhile to recall the general scope and boundaries of the model. The MoDViG model focuses on users' hedonic entertainment experience but not on users' more eudemonic and reflective entertainment experiences (e.g., as in Jørgensen, 2016). Furthermore, the MoDViG model seeks to contribute to a better understanding of how violent videogames communicate violence and how players experience it. However, assumptions regarding potential long-term effects of violent videogames, e.g., on players' worldviews or their subjective beliefs about warfare (e.g., Festl, Scharkow, & Quandt, 2013), are beyond the scope of the MoDViG model. Additionally, the MoDViG model examines virtual violence from the distinct disciplinary tradition of (Media) Psychology and Communication Science. The present article reviewed related disciplinary research and seeks to introduce the model to the broader community of game studies scholars that have already provided a rich understanding of videogame violence and of players' experiences of virtual warfare (e.g., Sicart, 2009; Mortensen, Linderoth & Brown, 2015). In my opinion, it would be useful to more closely link these elaborated disciplinary views on virtual violence in the future. The present introduction of the MoDViG model may be a small but important step in that direction.

Endnotes

[1] Psychological research revealed that simple cues suffice in determining whether or not we perceive an object as social (and even as possessing conscious mental states; see Arico et al., 2011; Simion et al., 2013; Johnson, 2003; Mar & Macrae, 2006 for overviews). According to Arico et al.'s (2011) AGENCY model, an object must possess eyes (or sensory organs), display distinctive motions (self-propelled motion), and contingent interactive behaviour. Video game characters do that. Even simple animated geometrical figures, as in the famous experiments by Heider and Simmel (1944), provide a sense of agency. Most videogame characters portray more complex self-propelled motion and appear to be in control of their own movements. Johnson (2003, p. 551) shows that "the presence of a face, or the propensity to interact contingently" is important in triggering the perception that the other may be capable of experiencing. Videogame characters usually respond contingently to their environment, and often feature a face, or at least some form of sensory organs that may effectively cue the intuitive impression that they may experience things.

[2] Accordingly, media equation theory may only be partly correct if it boldly argues that users treat videogame characters as real social actors. This may only be true to the (admittedly important) extent that users mindlessly treat virtual characters without accessing their knowledge that these characters are not real (System II). However, to the extent users stay aware of the artificiality of videogame characters, they may treat them differently — and respond to them differently (e.g., with less intense emotions) — than existing social beings. The notion that mediated representations are processed in two systems (also see Frome, 2006) may provide an explanation as to why many studies, including research on the media equation theory, find that participants' responses to virtual characters are structurally equivalent to the responses they show to existing beings in real life (e.g., people are more scared if encountering a scary character than a friendly character), while the general intensity of their responses (e.g., the intensity of the fear response) is usually much lower if encountering virtual compared to real characters. Furthermore, the magnitude of observed differences between different social encounters seems to be smaller in virtual, compared to real, environments (e.g., the magnitude of the fear response difference if encountering a scary vs. friendly character). For example, Blankendaal et al. (2015) observed substantial stress responses both among participants confronted with an aggressive outburst of either a virtual agent or a real human-being (structural equivalence), but the stress response among participants encountering the aggressive human-being was much higher than the stress response of participants encountering the aggressive virtual agent. Furthermore, the difference in the observed stress response before and after the aggressive outburst was substantially larger among participants in the real encounter compared to participants in the virtual encounter. From the perspective of my dual process model of media reality, this finding may be interpreted as showing that System I processes give rise to structurally similar feelings under real or virtual conditions, however System II processes ("knowing that this is not real") may dampen users' responses under virtual conditions.

[3] It is probably more than a coincidence that these two missions (unlike most other missions of both games) are heavily debated among gamers.

References

Arico, A., B. Fiala, R.F. Goldberg, & S. Nichols (2011). The folk psychology of consciousness. Mind & Language, 26(3), 327-352.

Bailenson, J.N., Blascovich, J., Beall, A.C., & Loomis, L.M. (2003). Interpersonal distance in immersive virtual environments. Personality and Social Psychology Bulletin, 29(7), 819-833.

Bandura, A. (1991). Social cognitive theory of moral thought and action. In W. M. Kurtines & J. L. Gewirtz (Eds.), Handbook of moral behavior and development (pp. 45-103). Hillsdale, NJ: Erlbaum

Bandura, A. (2002). Selective moral disengagement in the exercise of moral agency. Journal of Moral Education, 31, 101-119. doi:10.1080/0305724022014322

Blankendaal, R.A.M., Bosse, T., Gerritsen, C., Jong, T. de & Man, J. de (2015). Are Aggressive Agents as Scary as Aggressive Humans? Proceedings of the Fourteenth International Conference on Autonomous Agents and Multi-Agent Systems, AAMAS'15. ACM Press.

Bouchard, S., Bernier, F., Boivin, E., Dumoulin, S., Laforest, M., Guitard, T., Robillard, G., Monthuy-Blanc, J., & Renaud, P. (2013). Empathy Toward Virtual Humans Depicting a Known or Unknown Person Expressing Pain. Cyberpsychology, Behavior, and Social Networking, 16(1), 61-71.

Breuer, J., Festl, R., & Quandt, T. (2012). Digital war: An empirical analysis of narrative elements in military first-person shooters. Journal of Gaming & Virtual Worlds, 4(3), 215-237.

Breuer, J., Kowert, R., Festl, R., & Quandt, T. (2015). Sexist games = sexist gamers? A longitudinal study on the relationship between video game use and sexist attitudes. Cyberpsychology, Behavior, and Social Networking, 18(4), 1-6.

Breuer, J., Scharkow, M., & Quandt, T. (2014). The effect of interactivity and frequency of use on the perception and evaluation of violence in digital games. Journal of Media Psychology, 26, 176-188.

Busselle, R. W. & Greenberg, B. S. (2000). The nature of television realism judgments: a reevaluation of their conceptualization and measurement. Mass Communication and Society, 3, 249-258.

Castronova, E. (2005). Synthetic worlds: the business and culture of online games. Chicago: University of Chicago Press.

Chalmers, D. (2016). The virtual and the real. Retrieved online from consc.net/papers/virtual.pdf

Cupchik, G. C. (2002). The evolution of psychical distance as an aesthetic concept. Culture and Psychology, 8(2), 155-187.

Dogruel, L., Joeckel, S., & Bowman, N. D. (2013): Elderly People and Morality in Virtual Worlds. A cross-cultural analysis of elderly people's morality in interactive media. New Media & Society, 15(2), 276-293.

Epley, N., & Waytz, A. (2010). Mind Perception. In S.T. Fiske, D.T. Gilbert, & G. Lindsay (Eds.), The Handbook of Social Psychology (5th ed., pp. 498-541). New York: Wiley.

Festl, R., Scharkow, M., & Quandt, T. (2013). Militaristic attitudes and the use of digital games. Games and Culture, 8, 392-407.

Frome, J. (2006). Representation, reality, and emotions across media. Film Studies, 8, 12-25.

Gollwitzer, M., & Melzer, A. (2012). Macbeth and the joystick: Evidence for moral cleansing after playing a violent video game. Journal of Experimental Social Psychology, 48, 1356-1360.

Gray, H. M., Gray, K. & Wegner, D. M. (2007). Dimensions of mind perception, Science, 315, 619.

Gray, K., Young, L., Waytz, A. (2012). Mind perception is the essence of morality. Psychological Inquiry, 23, 101-124.

Grizzard, N., Tamborini, R., Lewis, R.J., Wang, L., & Prabhu, S. (2014). Being Bad in a video game Can Make Us Morally Sensitive. Cyberpsychology, Behavior, and Social Networking, 17(8), 499-504.

Hall, A. (2003). Reading Realism: Audiences' Evaluations of the Reality of Media Texts. Journal of Communication, 53(4), 624-641.

Hartmann, T. (2011). Players' experiential and rational processing of virtual violence. In S. Malliet & K. Poels (Eds.), Vice City Virtue. Moral Issues in Digital Game Play (pp. 135-150). Leuven: Acco.

Hartmann, T. (2012). Moral disengagement during exposure to media violence: Would it feel right to shoot an innocent civilian in a video game? In R. Tamborini (Ed.), Media and the Moral Mind (pp. 109-131). New York, NY: Routledge.

Hartmann, T., & Goldhoorn, C. (2011). Horton and Wohl revisited: Exploring viewers' experience of parasocial interaction. Journal of Communication, 61, 1104-1121.

Hartmann, T., & Vorderer, P. (2010). It's okay to shoot a character: Moral disengagement in violent video games. Journal of Communication, 60, 94-119.

Hartmann, T., Krakowiak, M., & Tsay-Vogel, M. (2014). How violent video games communicate violence: A literature review and content analysis of moral disengagement factors. Communication Monographs, 81(3), 310-332.

Hartmann, T., Toz, E., & Brandon, M. (2010). Just a game? Unjustified virtual violence produces guilt in empathetic players. Media Psychology, 14, 339-363.

Heberlein, S., & Adolphs, R. (2004). Impaired spontaneous anthropomorphizing despite intact perception and social knowledge. Proceedings of the National Academy of Sciences USA, 101(19), 7487-7491.

Heider F. & Simmel, M. (1944). An experimental study of apparent behavior. American Journal of Psychology, 57, 243-249.

Holtgraves, T. M., Ross, S. J., Weywadt, C.R., & Han, T.L. (2007). Perceiving artificial social agents. Computers in Human Behavior, 23, 2163-2174.

Joeckel, S., Bowman, N. D., & Dogruel, L. (2012). Gut or game? The influence of moral intuitions on decision in video games. Media Psychology, 15(4), 460-485.

Johnson, S.C. (2003). Detecting agents. Philosophical Transactions of the Royal Society, 358, 549- 559.

Jørgensen, K. (2014). Devil's plaything: On the boundary between playful and serious. Proceedings of Nordic DiGRA 2014 Conference.

Jørgensen, K. (2016). The positive discomfort of Spec Ops: The Line. Game Studies, 16(2).

Juul, J. (2005). Half-real. Video games between real rules and fictional worlds. Cambridge: MIT Press.

Klimmt, C., Schmid, H., Nosper, A., Hartmann, T., & Vorderer, P. (2006). How players manage moral concerns to make video game violence enjoyable. Communications: The European Journal of Communication Research, 31, 309-328.

Kugler, K., & Jones, W. H. (1992). On conceptualizing and assessing guilt. Journal of Personality and Social Psychology, 62(2), 318-327.

Leino, O.T. (2015). "I know your type, you are a player": Suspended fulfilment in Fallout: New Vegas. In J. Enevold & E. MacCallum-Stewart (eds.). Game Love: Essays on Play and Affection (pp. 218-242). Jefferson: McFarland.

Lin, S. F. (2010). Gender differences and the effect of contextual features on game enjoyment and responses. Cyberpsychology, Behavior, and Social Networking, 13, 533-537.

Lin, S. F. (2011). Effect of opponent type on moral emotions and responses to video game play. Cyberpsychology, Behavior, and Social Networking, 14, 695-698.

Lombard, M. & Ditton, T. (1997). At the Heart of It All: The Concept of Presence. Journal of Computer-Mediated Communication, 3(2). http://www.ascusc.org/jcmc/vol3/issue2/lombard.html.

Luck, M. (2009). The gamer's dilemma: An analysis of the arguments for the moral distinction between virtual murder and virtual paedophilia. Ethics and Information Technology, 11, 31-36.

Mar, R. A., & Macrae, C. N. (2006). Triggering the intentional stance. In G. Bock, & J. Goode (Eds.), Empathy and Fairness (pp. 110-119). Chichester, U.K.: Wiley.

Mason, D. (2014). Video games, theater, and the paradox of fiction. The Journal of Popular Culture, 47, 1109-1121.

Moore, C. (2015). Moral disengagement. Current Opinion in Psychology, 6, 199-204.

Morgan, M. & Shanahan, J. (2010). The state of cultivation. Journal of Broadcasting & Electronic Media, 54, 337-355.

Morrison, I., & Ziemke, T. (2005). Empathy with computer game characters: A cognitive neuroscience perspective. Proceedings of Joint Symposium on Virtual Social Agents (pp. 73-79). Hatfield: AISB.

Mortensen, T.E., Linderoth, J., & Brown, A.M.L. (Eds.). (2015). The dark side of gameplay. Controversial issues in playful environments. London: Routledge.

Nass, C., & Moon, Y. (2000). Machines and mindlessness: Social responses to computers. Journal of Social Issues, 56(1), 81-103.

Oliver, M.B., Bartsch, A., & Hartmann, T. (2014). Negative emotions and the meaningful sides of media entertainment. In Parrott, W.G. (Ed.). The positive side of negative emotions (pp. 224 - 246).New York: Guilford Press.

Olthof, T., et al. (2008). The assignment of moral status: Age-related differences in the use of three mental capacity criteria. British Journal of Developmental Psychology, 26(2), 233-247.

Pitts, R. (27^th August 2012). Don't be a hero. The full story behind Spec Ops: The Line. Retrieved online from http://www.polygon.com.

Pizarro, D. A., Detweiler-Bedell, B., & Bloom, P. (2006). The creativity of everyday moral reasoning: Empathy, disgust and moral persuasion. In J. Kaufman, & J. Baer (Eds.), Creativity and reason in cognitive development (pp. 81-98). Cambridge, U.K.: Cambridge University Press.

Potter, W.J. (2014). A critical analysis of cultivation theory. Journal of Communication, 64, 1015-1036.

Pötzsch, H. (2014). Ubiquitous absence: Character engagement in the contemporary war film. Nordicom Review, 34(1), 125-144.

Reeves, B., & Nass, C. (1996). The media equation: How people treat computers, television, and new media like real people and places. Cambridge University Press.

Rosenthal-von der Pütten, A. M., Krämer, N. C., Hoffmann, L., Sobieraj, S., & Eimler, S. C. (2013). An experimental study on emotional reactions towards a robot. International Journal of Social Robotics, 5(1), 17-34.

Schmierbach, M., & Limperos, A.M. (2013).Virtual justice: Testing disposition theory in the context of a story-driven video game. Journal of Broadcasting & Electronic Media, 57(4).

Schulzke, M. (2009). Moral decision making in Fallout. Game Studies, 9(2).

Shapiro, M. A., Peña, J., & Hancock, J. T. (2006). Realism, imagination, and narrative video games. In P. Vorderer & J. Bryant (Eds.), Playing computer games: Motives, responses, and consequences (pp. 275-289). Mahwah, NJ: Erlbaum.

Sicart, M. (2009). The ethics of computer games. Cambridge: MIT Press.

Simion, F., Bardi, L., Mascalzoni, E., & Regolin, L. (2013). From motion cues to social perception: Innate predispositions. In M. D. Rutherford & V.A. Kuhlmeier (Eds.)., Social Perception: Detection and Interpretation of Animacy, Agency, and Intention (pp. 37-60).

Singer, P. (1981). The expanding circle: Ethics and sociobiology. New York: Farrar, Straus & Giroux.

Slater, M., Antley, A., Davison, A., Swapp, D., Guger, C., et al. (2006). A virtual reprise of the Stanley Milgram obedience experiments. PLoS ONE, 1(1), e39.

Vorderer, P. (1993). Audience involvement and program loyalty. Poetics, 22, 89-98.

Warren, M. A. (1997). Moral status: Obligations to persons and other living things. Oxford: Clarendon Press.

Weaver, A.J., & Lewis, N. (2012). Mirrored morality: an exploration of moral choice in video games. Cyberpsychology, Behavior, & Social Networking, 15, 610-614.

Whitty, M.T., Young, G., & Goodings, L. (2011). What I won't do in pixels: Examining the limits of taboo violation in MMORPGs. Computers in Human Behavior, 27(1), 268-276.

Young, G. (2013). Ethics in the virtual world. The morality and psychology of gaming. Durham: Acumen.

Young, G., & Whitty, M. T. (2011). Should gamespace be a taboo-free zone? Moral and psychological implications for single-player video games. Theory & Psychology, 21(6), 802-820.

Ludography

11 Bit Studios. (2014). This War of Mine. Deep Silver.

2K Boston et al. (2007). BioShock. 2K Games.

Bethesda Game Studios. (2008). Fallout 3. Bethesda Softworks.

Bohemia Interactive. (2001). Operation Flashpoint. Codemasters.

Destructive Creations. (2015). Hatred. Destructive Creations.

Frasca, Gonzalo. (2003). September 12^th: A toy world. Newsgaming.com.

Infinity Ward. (2009). Call of Duty: Modern Warfare II. Activision.

Rockstar North. (2013). Grand Theft Auto V. Rockstar Games.

Yager Development. (2012). Spec Ops: The Line. 2k Games.