A
n
e
x
p
e
r
i
m
e
n
t
o
n
p
l
a
y
e
r
s
'
m
o
r
a
l
i
n
t
u
i
t
i
o
n
s
,
r
e
a
s
o
n
i
n
g
,
a
n
d
m
o
r
a
l
r
e
a
s
o
n
i
n
g
i
n
a
v
i
d
e
o
g
a
m
e
c
o
n
t
e
x
t
Acknowledgements
At this time, I wish to thank my supervisor Sindy Sumter for her support and guidance throughout this process and for helping me take on a project much more voluminous (and fun!) than I envisioned at the start.
In addition, I would like to express my appreciation to the students and staff of the University of Amsterdam who participated, or in some other way helped, with this experiment.
Lastly, I would like to thank my family and friends for their nonstop support and encouragement throughout this master’s programme.
Martijn Samuels
1
Abstract
Current research on the subject of video games and morality shows that, in contrast to the concept of the magic circle, gamers are influenced by their moral foundations. This study examines participants’ (N=100) innate moral foundations as influencing video game play decisions and reasoning. In an experimental design, participants were presented with a video game in which they were confronted with several moral and strategic dilemmas. Their decisions and reasonings for those dilemmas were recorded. The players’ reasonings were coded as either moral, strategic, or enjoyment reasoning, and, in turn, moral reasons were classified using Moral Foundations Theory into five different types. Furthermore, the social context and the time participants had to provide their reasons, after or during the game, were manipulated to study their possible influences. Results indicated a person’s moral foundations might influence the type of moral reasons they provide for the dilemmas. For example, a higher score on the moral foundation Fairness-Reciprocity was positively related to the use of moral reasoning that reflected Fairness/Reciprocity. Further analyzing one of the dilemmas, results indicated that a person’s moral foundations might also guide in-game decisions. However, no strong effects of the social context nor the timing of reasoning on the type of (moral) reasoning or in-game decisions were found. Considerations and vocal points for future research are discussed, stressing an importance on taking into account gamers’ personal perspectives on their in-game decisions and the contextual meanings they entail.
2
Being Batman: An experiment on players’ moral intuitions, reasoning, and
moral reasoning in a video game context
Much of the earliest research regarding video game players took a social scientific approach with a focus on concerns that surrounded the question on how various types of players were affected by violence in video games (Ferguson, 2015; Hartmann, Krakowiak, & Tsay-Vogel, 2014; Lin, 2011; Sherry, 2001). Although this focus on violence has been criticized by many (Consalvo, 2005; Goldstein, 2005), it probably helped shape the paradigm that gamers are seen as being negatively affected by the content within games. Even though the effects of video games on their players is an interesting field of study, some have argued this focus might lead to a negation of the question how players constantly make choices about their in-game behaviour, as well as consider and frame their in-game actions relative to other daily situations and contexts (e.g., Consalvo, 2005).
A defining feature of video games is their interactivity, which demands activity and volition and enables players to make choices; choices that alter the experience of a video game. Studying the outcome of these choices, e.g. being violent or immoral, is insufficient when trying to understand how these decisions came to be, what their justifications are and what (context specific) meanings these hold for their respective players. For example, when looking at how players manage moral concerns arising from violence in video games, gamers report that committing moral transgressions in video games is not related to actual-life
experiences. For gamers, virtual violence is seen as an experience distinct from reality (Klimmt, Schmid, Nosper, Hartmann, & Vorderer, 2006). This justification for virtual
violence as being distinct from reality stems from the argument, later discussed in more detail, that when players enter a game world their identities shift from their every day, “ordinary” selves, into their ludic selves and adopt a so-called ‘lusory attitude’; a disposition one enters into when in the bounded space of a game (Calleja, 2015). This lusory attitude, however,
3 might not be devout of a sense of morality, as research findings suggest that a person’s moral intuitions might influence the level of enjoyment of video games (Joeckel, Bowman, & Dogruel, 2013) and the decisions they make (Krcmar & Cingel, 2016).
Studying participants in the age range 18 till 31 years, the current study expands on findings suggesting that an individual’s sense of morality plays a non-redundant part in the video game experience and that it might be indicative of how one experiences and plays certain games (Joeckel, Bowman, & Dogruel, 2012; Krcmar & Cingel, 2016). Specifically, this study contributes to emerging research on morality and gaming in three ways. First, it focuses on the effects of a person’s actual-life moral orientation on their observed in-game decisions and how these decisions might be influenced by the social context in which they were made and by the method participants had to provide justifications for those decisions. Second, it looks at how these decisions were guided by morality, strategy and/or enjoyment; this is compelling as it helps to address the debate on the meanings of video game play in terms of understanding how players approach dilemmas presented to them in video games. Lastly, this study will analyze moral justifications players present for in-game decisions, which will contribute to understanding these justifications and how they might be guided by a player’s moral intuition. This is important as it helps scholars focus on broader issues of morality in video games and stresses them to attend to video game players by seeking out their views and incorporating their viewpoints and justifications for in-game decisions into constructions of knowledge and methodology.
Theoretical Background
Morality, Strategy, and Enjoyment within the Magic CircleWhat does it mean to enter the system of a game? A prevailing concept in discussions on video games in contemporary game studies is the term ‘magic circle’, originally coined by Johan Huizinga (1950). For Huizinga, the magic circle is a space created for playing, a
4 material or conceptual temporary world dedicated to the act, apart from the actual (Stenros, 2014). However, the concept of the magic circle as it is understood in game studies today comes from the book ‘Rules of Play’ by Katie Salen and Eric Zimmerman (2004). They describe the magic circle as shorthand for “the idea of a special place in time and space created by a game” (Salen & Zimmerman, 2004, p.95). It is argued that the circle is both finite and infinite, while the magic is in the transformation of meaning within the circle. In their formulation, the boundaries of play are fuzzy and permeable, but the borders of games are more formal (Stenros, 2014). This fits with the thought that video games, for many,
represent a kind of mentally walled-off space where our actions have no consequences (Heron & Belford, 2014). The idea that a video game creates a space, or game world, separate from the actual, or actual world would, theoretically, lead to gamers ignoring the rules of actual life in favor of the rules of the game. In other words, if video games exist within a magic circle, then morality or moral logic of the outside world, would be substandard to strategy and game logic. When we would consider morality through this lens it can be argued that the player of a video game is explicitly divesting himself of the need to consider the moral consequences of his actions because the normal rules regarding what is and is not appropriate do not pertain (Heron & Belford, 2014).
However, some scholars argue that the magic circle does not exist and that actual world rules do apply when playing video games (Consalvo, 2009) and that it is hardly possible for the game-space to block out the complexity of social and personal relations (Calleja, 2015). Consalvo (2009) argues that actual world rules do apply, but these rules exist in addition to, or in competition with, other rules. Moreover, these rules stand in relation to multiple contexts, including cultures, social groups, legal situations, and homes. For her, the boundaries of the magic circle are less strict and the concept of the magic circle emphasizes form at the cost of function, without attention to the context of actual gameplay. Certainly one
5 of the challenges in video game research is the extent to which we can attribute a player making a certain decision to be a consequence of the game’s design or to the player’s background and gameplay context. Any attempt to create a clear demarcation between the game experience and the experience of the world (supposedly) external to it, will find it difficult to explain how the players’ personal and social histories can be excluded from the game activity (Calleja, 2015). Surely it is possible a game’s architecture and design inhabits a certain ethical system, but those ethical constructions are not what is considered here. As Babij (2013, p.159) explains a video game might impose an ethical value system on a player, but this does not subsequently mean players will view or consider their actions as moral decisions.
Considering the concept of the magic circle and the critiques on it, this study will look if and how in-game decisions and justifications for those decisions are influenced by a
players’ moral intuitions. Previous research on the subject has found that gamers do approach in-game decisions sometimes as moral (Krcmar & Cingel, 2016), experience enjoyment or guilt when engaging in moral transgressions (Hartmann, Toz, & Brandon, 2010; Lin, 2011), and moral considerations play a role in the appreciation of certain media (Raney, 2011; Tamborini, 2011). These recent approaches to morality have suggested that moral considerations might be the result of intuitive reactions rather than rational processing (Graham, Haidt, & Nosek, 2009; Haidt, Graham, & Hersh, 2008). This focus on moral intuitions serves to explain the seemingly quick gut reactions players experience when confronted with morally charged game scenarios. The aforementioned intuitive nature of morality is explained by Moral Foundations Theory [MFT] (Haidt et al., 2008), which explicates morality as an innate set of moral foundations that guides our decisions and reactions, and which will be used in this study to measure moral salience and classify reasoning.
6 But besides moral and strategic decision-making in video games another possible type of reasoning needs to be discussed, namely acts that are solely for the purpose of pleasure or enjoyment. Bareither (2017) discusses, as he calls them, ‘playful transgressions’ in video games, which are acts of virtual violence with the aim of experiencing the pleasure of transgression. Bareither (2017) tries to explain how it is possible for some gamers to find pleasure in virtual violent acts without any moral or strategic justification, even though transgressive acts are not the regular means by which game developers intended pleasure to be had. The behavior of playful transgressions can be seen as intentionally provoking
incongruences by doing something unexpected that breaks established social patterns, and this incongruence is experienced as funny and therefore pleasurable (Bareither, 2017). Bareither (2017) argues that when we look closely to these playful transgressions, we see that the pleasure of transgression does not emerge from simply feeling good about unjust violence. In fact, the pleasure seems to emerge exactly because feeling good about unjust violence creates a contrast to how one should actually feel about it.
When relating this to the concept of the magic circle it becomes clear that the pleasure of transgression is not bound to a space separated from everyday life or solely within the realm of the magic circle, but it can only take place at the intersection of the emotional space of everyday life and the space created through playing video games. The magic circle then serves as a kind of inoculation against considering our relationship between our characters and their game world and as an extension between ourselves and our characters (Heron & Belford, 2014). So besides moral and strategic justifications for decisions within the magic circle, another type of justification, namely purely for the sake of pleasure and enjoyment, needs to be considered when studying video game in-game reasoning. This third type of reasoning is, throughout this study, referred to as enjoyment reasoning.
7 This study will look at how players explain their decisions in a video game that
presents them with both strategic and moral dilemmas and will classify their reasoning as either strategic, moral, or focused on enjoyment; for the sake of determining which is most prevalent.
Social Context
The social context in which a game is played, when talking about in-game decisions, might play an influential role. Consider for example playing against your boss or against your love-interest, would your goals for the game be the same as when playing against a friend or your younger sibling? As Juul (2010, p.126-127) explains, the social context in which a game is played, influences the choices within a game; he distinguishes three different
considerations. The first consideration for evaluating an in-game action or decision is the game as goal orientation; this is simply wanting to win the game. The second consideration is the game as experience; in this case, wanting the game to be fun and pleasurable is guiding game decisions. A player might, for example, balance the game so nobody falls behind and the outcome of the game is less certain, thus more interesting. The third consideration that Juul (2010) discusses is the game as a social event, in this case, the player has a desire for management of the social situation. The player knows the choices he makes or the outcome of the game might make other players happy or sad and that this may influence his social
standing or the social dynamic of the group. Salen and Zimmerman (2004) distinguish social play interactions as internally or externally derived. With internally they mean social
interaction within the magic circle, within the game, as a product of the formal system of a game. Social interaction is derived externally when social roles are brought into the game from outside the magic circle. Pre-existing friendships and rivalries that affect in-game strategic choices, for example, are externally derived elements of social play (Salen & Zimmerman, 2004, p.462).
8 One of the oft-touted benefits of video games is that they permit multiple ways of engagement and this allows players with different styles to find their own paths into the game (Hung, 2011). However, when players have to interact with other people, this creates potential problems, especially when players are playing against one another and need to be on a
perceived and mutually agreed on level playing field. In other words, while players can have different orientations to a given game, these differences need to be sorted out if they are to play together in a shared game space; this sorting constitutes the meaning the game will have to this particular group of players (Hung, 2011).
In the case of this study, some players were asked to play a single player game together with someone else and to make the in-game decisions together; the consequence being a need to interact with the other player and a certain alignment of goals. Most of the participants playing together in this study knew each other and were friends, which suggests that the social interaction during gameplay was influenced externally. The social context might have an influence on the (shared) goals of the players and thus on the outcome of choices. Also, the context might have an influence on the type of reasoning players employ since playing together with a friend might be more enjoyable than playing alone and in that sense, the players might reflect more reasoning based on enjoyment. The focus on the social context has resulted in the following research question (RQ1): does the social context influence the type of reasoning players employ when confronted with strategic and moral dilemmas?
System 1 and System 2
When studying in-game decisions and reasoning, researchers can ask players to explain their reasoning during or after the gameplay. However, as theorized by Krcmar and Cingel (2016), asking players during the game, using a think-aloud protocol, to provide their justifications, might influence the decision-making process and thus the reasoning. Research
9 in thinking and reasoning has proposed that there are two distinct cognitive systems
underlying reasoning (Evans, 2003). In his book ‘Thinking, Fast and Slow’, Daniel Kahneman (2011) explains the difference between System 1 and System 2 thinking, sometimes described as implicit and explicit thinking. System 1 is considered to be fast, instinctive, and emotional, System 2 is associated with slow, effortful, logical, and conscious thinking. The assumption regarding the differences between System 1 and 2 and their effects on reasoning, is that System 2 thinking is both volitional and responsive to verbal instructions, and System 1 thinking is not (Evans, 2003). Hence, the influence of System 1 can only be suppressed indirectly by asking people to make a strong effort to reason deductively. This means that one of the features of System 2 seems to be the ability to override or inhibit default responses emanating from System 1 (Stanovich, 1999).
The question which System video game players employ when playing games that present them with strategic and moral dilemmas is an interesting one. Difficult and complex dilemmas might trigger System 2 thinking, however, the fast nature of video games might not allow for this slower and more conscious thinking process. Asking players to supply their reasoning during the game requires more of them, in terms of actions and thinking, than asking them after gameplay; meaning this could spark more conscious thinking related to System 2 and a suppression of System 1 thinking. The effect on the type of reasoning might be that System 2 thinking results in more strategic reasoning in contrast to moral or
enjoyment reasoning. System 1, in addition, might be more closely related to moral intuitions as it is associated with fast and emotional gut reactions, resulting in more moral reasoning.
Looking at the difference between players that supplied their reasoning during or after the gameplay is not the same as looking at the difference between System 1 and 2, but they are related because reasoning during the game demands more of players and might spark more conscious thinking. So besides the question if the social context influences the type of
10 reasoning (RQ1), this study will also try to answer if the type of reasoning is influenced by the timing of reasoning (RQ2), during or after the game.
After coding the type of reasons as either moral, strategic or based on enjoyment, the reasons established as moral will be further classified, using the MFT classification, as relating to one of five moral foundations for the purpose of studying their relation with the participants’ moral intuitions. This will answer the question (RQ3) if moral foundation salience can predict the type of moral reasoning. Additionally, the possible influences of the social context (RQ4) and the timing of reasoning (RQ5) on the type of moral reasoning will be considered.
Current Study
In summary, this study will look at individual moral intuitions and their relation to the type of moral reasoning. Furthermore, the social context (i.e. Solo vs Duo) and the way participants had to provide their reasons (i.e. In-Game vs Post-Game) were manipulated to study their possible influences.
In addition to relating an individual’s moral intuitions to the type of moral reasoning and to the main research questions, this study will also have a closer look at the outcome of one specific dilemma presented to the participants. This dilemma, named Harm-Care choice, is of specific interest because it presents the participants with a clear choice to either hurt a non-player-character or to ease his pain. The influence of individual moral intuitions on the outcome of the choice and the type of (moral) reasoning provided for this choice will be tested and discussed. Also, the possible influences of the different conditions will be taken into account. This focus on a Harm-Care choice resulted in the following research questions: does moral foundation salience predict the outcome of a Harm-Care choice (RQ6), the type of reasoning for a Harm-Care choice (RQ7), and the type of moral reasoning for a Harm-Care choice (RQ8)?
11
Method
Participants
Participants (N = 100) were recruited from a recruitment pool website of the University of Amsterdam for communication science students. These participants received compensation for their participation in the form of research credits. Requirements for participation were mentioned on the website and included that participants should be older than 18 years, that participants should be able to understand enough English to play the game in English, and that participants should not have played the Batman Telltale game before.
Participants themselves could choose to sign-up alone or as a couple, however, in some cases participants that signed-up alone were placed together with someone else to play as a couple. The average age of the sample was 20.61 (SD = 2.45), ranging from 18 to 31 years and the sample included 35 different nationalities, with 27 Dutch, 18 German, 5
Bulgarian, and 5 Italian nationals as being the most represented nationalities. The majority of the sample was female (77%) and included mostly non-gamers, with 58% of participants reporting they never play video games.
Participants were placed in one of four conditions, namely Game-Solo, Post-Game-Duo, In-Game-Solo, and In-Game-Duo. The participant distribution per condition can be found in Table 1. As mentioned above, participants could decide for themselves to sign-up alone or as a couple, thus deciding themselves if they were placed in the Solo or Duo
condition. The first 56 sessions in the lab were placed in the Post-Game condition, after which the remaining participants were placed in the In-Game condition. Independent-Samples T-Tests were conducted to determine if the conditions differed significantly on measures like age, gender, and moral foundation salience, however, none significant differences were found.
12 Table 1.
Participant Distribution (N=100)
Stimuli
Participants were asked to play the second episode (‘Children of Arkham’) of the video game ‘Batman: The Telltale Series’, which is an episodic point-and-click graphic adventure game developed by Telltale Games and published in 2016. The game offers a branching narrative, giving the player choices, mainly in dialogue responses, that influence a situation and affect later events in the game. For example, the story presents the player with a choice to approach certain situations as Batman or Bruce Wayne, influencing situations and how the non-player characters in the game react. Participants were handed a character sheet before playing, which included a picture and a small description of the main characters whom the participants would have known, had they played the first episode (See Appendix 1 – Character Sheet). The gameplay was completely in English with English subtitles.
This game was chosen because the narrative of the game presents the player with several difficult moral and strategic choices that could possibly produce different moral reasonings. The player is presented with several choices throughout the game, of which six
Average age M (SD)
Female Male Same sex /
Mixed sex duo’s
In-Game Solo 21.63 (2.45) 12 4 n.a.
Post-Game Solo 21.13 (3.14) 11 5 n.a.
In-Game Duo 20.75 (2.44) 19 9 13/1
13 were selected as primary choices. The player is, for example, asked to choose to either help a non-player-character or stop helping or is given the option to betray an ally1.
Study Procedure
The study took place in a lab space, which consisted of two separate rooms divided by a wall with a window; in one room the participants played the game, in the other the
researcher observed the participants. In the gaming room, a large screen was present with, before it, a wireless keyboard and mouse. The lighting in the gaming room was dimmed when the gameplay started. There was also a laptop present so participants in the Duo condition could complete the surveys individually.
Participants were explained, in English or Dutch, the study consisted of a two-part survey and them playing a video game. Participants first read and agreed to the informed consent form, presented to them digitally on a computer, after which they filled in the first part of the survey. Participants were then explained more about the specifics of the game; that they would be able to influence the storyline, how their in-game choices had consequences, and, more practically, that they had to use the mouse and the keys E, Q, SHIFT, and the arrow keys. Next, participants were asked to read the character sheet, since they would be playing episode 2 of the game and they needed the character sheet to understand the storyline.
Participants in the Duo condition were told they had to make the decisions in the game together. Some duo’s controlled the mouse and keyboard together, dividing the keys, some switched during the game sometimes, and with some duo’s there was only one player controlling the game.
1 For a more extensive discussion and analysis of the choices presented to the players, see Appendix 2 – Choices presented in the game
14 Participants in the In-Game condition were told that the researcher would, at times, pause the game and switch a sign hanging behind the window in the researcher’s room. The two-sided sign depicted a thinking cloud and on the other side a silence sign. The participants were explained that when they saw the thinking cloud they were supposed to explain, out-loud in the microphone, their reasons for their choices. It was explained that, if possible, they had to explain their reasons before making the choices. In the In-Game condition, the door between the researcher’s space and the game room was not closed during the first scene of the game, this was to allow the researcher to comment on how participants were doing and to check if they fully understood their task. Participants in the In-Game-Solo condition wore a headset, participants in the In-Game-Duo condition were placed before a standing desk-microphone. Participants were told that the researcher in the other room would not be able to hear them reason when the door was closed.
When the game started, screen-capture and recording software was turned on, recording the voices of participants and everything visible on the screen of the participants. The duration of the gameplay was approximately an hour but was influenced by the decisions the participants made in the game and on how well they did. For example, dying in the game prolonged the duration. The Post-Game condition took slightly longer because participants in this condition had to explain their reasons after rather than during the game. After the
gameplay was finished participants were requested to fill in the second part of the survey on the computer, or, in the duo condition, on the computer and the laptop. After filling out the rest of the survey participants were told they were done and were rewarded in the form of research credits. When the data collection was complete a debriefing email was sent to all participants explaining the real purpose of the study.
15 Measures
Moral Foundations. The Moral Foundations Questionaire [MFQ] (Graham et al., 2009; Haidt et al., 2008) is a two-part 32-item measure designed to determine the importance to individuals of five moral intuitions. These intuitions include Harm/Care,
Fairness/Reciprocity, In-Group/Loyalty, Authority/Respect, and Purity/Sanctity and all consisted of 6 items. The first part of the questionnaire consisted of sixteen items (three for
each intuition) with the stem: “When you decide whether something is right or wrong, to what
extent are the following considerations relevant to your thinking?”. Participants responded on
a 7-point Likert scale ranging from 1 (=extremely irrelevant) to 7 (=extremely relevant). The second part consisted of sixteen statement items (three for each intuition) and participants had to rate the extent with which they agreed with the statements on a 7-point Likert scale ranging from 1 (=strongly agree) to 7 (=strongly disagree). Each part contained one screening item
(“Whether or not someone is good at math” and “It is better to do good than to do bad”) that
did not pertain to moral relevance and served to remove people not responding seriously, however, no participants had to be deleted. The internal consistency (Cronbach’s alpha) for
each moral intuition scale (Harm/Care (α = .494), Fairness/Reciprocity (α = .538),
In-group/Loyalty (α = .589), Authority/Respect (α = .622), and Purity/Sanctity (α = .632)) was not particularly high, but these scores are congruent with other studies using MFQ (Dogruel, Joeckel, & Bowman, 2012; Zhang, Kong, & Li, 2017) and might be explained by the divers national backgrounds of the participants. No items were deleted and for each moral
foundation the intuition items were averaged to form a composite salience score for each
participant: Harm/Care (M = 5.68, SD = 0.59), Fairness/Reciprocity (M = 5.71, SD = 0.59),
In-group/Loyalty (M = 4.43, SD = 0.80), Authority/Respect (M = 4.22, SD = 0.81), and
16 Harm-Care choice. One of the choices has been analyzed as being a Harm-Care choice, this was the first important choice of the game and regarded the non-player-character Carmine Falcone. Participants knew from the character sheet he was a powerful mob boss and that he was hospitalized after an encounter with Batman. Bruce visits Falcone to question him about the murder of his parents; he suspects Falcone ordered this murder. Falcone expresses he cannot talk through the physical pain he is experiencing and asks Bruce to give him some morphine. The participants are offered the following choice: ease his pain or hurt him. This is a choice to either harm a person or to supply some care to him; this is the reason this choice, and its justifications, are further analyzed to see what influences this Harm-Care choice. The outcome of the choice was recorded for each participant with 0 (=hurt him) or 1 (=ease his pain).
Player Reasoning: type of reasoning and type of moral reasoning. All recordings of participants in the In-Game condition were partly transcribed; only the parts concerning relevant choices were transcribed into an Excel document. In some cases, the reasons participants provided during the In-Game condition were inaudible on the recording and sometimes participants failed to provide a reason, these cases were treated as missing. Next, the written and transcribed reasons of all conditions were imported in Atlas.ti 8 software for coding purposes. When explaining a decision sometimes participants offered multiple reasons, each of these reasons were treated separately. All reasons were coded for two different factors, i.e. 1) the type of reasoning, and 2) the moral foundation.
The first coding determined if the reason was found to be focussing on strategy, morality, or enjoyment. When a reason was coded ‘Strategic’ it means that the reasoning expressed the participants’ will to improve the position of Bruce/Batman or aid their progress in the game. Strategic reasoning thus expressed a will to achieve in-game goals, for example,
17 “To get more information and details.” or “We decided that it would only be a farce, and only chose the option in order to get Hill to talk.”.
When a reason was coded ‘Moral’ it means that the reasoning expressed reflected morality in some sense. Moral reasoning thus included reasons for helping others or statements about what the player found wrong or right actions in that specific situation, for example, “I don’t want him to feel pain anymore, somehow help him. He seemed defenceless and helpless, so why hurt him?.” or “It is wrong to expect from someone to still help you (in this case financially) and support you, even though you are not willing to do the same.”.
Finally, when a reason was coded with ‘Enjoyment’ it means that the reasoning expressed a certain preference for the most enjoyable option for the player. Enjoyment
reasoning thus included reasons for choosing the option that the player would enjoy more, for example, “It makes the story kind of more thrilling.” or “Batman is much more entertaining to play as than Bruce Wayne.”.
The second coding only concerned reasons that were coded as ‘Moral’ during the first coding round. For each moral reason, it was identified as referencing one out of the five moral foundations. Reasons were coded as ‘Harm’ if they involved things like protecting or
physically hurting others, for example, “I did not want to kill him.” or “We chose to save Harvey because he was defenceless. Selina would've been able to protect herself.”.
Reasons were coded as ‘Fairness’ if they involved things like reciprocity, property ownership or consequences for actions, for example, “He was guilty and had it coming.” or “Hij zou batman niet meer steunen dus batman hoeft hem ook niet meer te steunen.”.
Reasons were coded as ‘Loyalty’ if they involved things concerning commitment to characters considered friends or members of one’s in-group, for example, “Because I had already made prior commitments with Harvey.” or “I told her this information because Vicki is my main ally in the press.”.
18 Reasons were coded as ‘Respect’ if they involved things concerning respect or
authority, for example, “since she was a good reporter.“ or “I thought that he was not only unsympathetic, but also the way he asked me for the medicine was impolite and without respect.”.
Reasons were coded as ‘Purity’ if they involved things concerning about whether or not something was considered dirty or impure in some way, for example, “I did not want to kiss her because I felt it was inappropriate to kiss her after killing several people.” or “When you do business with someone, feelings should not be involved.”.
Results
Type of Reasoning: Moral, Strategic, and Enjoyment
In total 660 reasons that were provided for the 6 primary choices have been analyzed, of which 55.52% were coded as moral reasons (see Table 2); suggesting strategic and
19 Table 2.
Descriptives of Type of Reasoning (N=100)
Variables M SD Min Max Sum
Moral reasons 4.04 1.51 1 8 364 Strategic reasons 2.61 1.11 1 5 235 Enjoyment reasons 0.57 0.64 0 3 51 % Moral reasons of total reasons 55.52 16.59 16.67 87.50 - % Strategic reasons of total reasons 36.55 15.22 12.50 71.43 -
& Enjoyment reasons
of total reasons 7.93 9.15 0 42.86 -
Influence of conditions on the type of reasoning. To test for RQ1 and RQ2, the possible influences of the conditions on the type of reasoning, a two-way multivariate analysis of variance (MANOVA) was conducted. The type of reasoning was not significantly
predicted by social context, Wilk's Λ = 0.963, F(3, 53) = 0.963, p = .572, partial η2 = .04, nor by the timing of reasoning (In-Game versus Post-Game), Wilk's Λ = 0.996, F (3, 53) =
0.996, p = .402, partial η2 = .05). In addition, no significant interaction effect between the two conditions was found on the type of reasoning (F (3, 53) = 1.233, p = .307; Wilk's Λ = 0.935, partial η2 = .07). However, the effect of the timing of reasoning on the number of moral reasons approached significance, F(1, 59) = 2.838, p = .098. Participants in the In-Game condition, M = 4.43, SD = 1.59, provided more moral reasons than participants in the Post-Game condition, M = 3.78, SD = 1.27.
20 Type of Moral Reasoning: Harm, Fairness, Loyalty, Respect, and Purity
Moral foundation salience and type of moral reasoning. To test the relationships between a participant’s individual moral foundation salience and the type of moral reasoning they employed (RQ3), five multiple linear regressions were conducted (see Table 3).
Fairness-Reciprocity. The multiple linear regression to predict fairness-reciprocity reasoning based on moral foundation salience was significant, F(5, 84) = 2.308, p = .051, with an R2 of .121. Moral foundations fairness/reciprocity (β = .19, p = .088) and authority/respect (β = .41, p = .010) were both significant predictors of the number of moral reasonings that reflected fairness-reciprocity. A higher score on fairness/reciprocity or authority/respect salience predicted more reasoning that reflected fairness-reciprocity.
In-Group-Loyalty. In-group-loyalty reasoning (R2 = .111, F(5, 84) = 2.095, p = .074)
found one significant predictor, namely fairness/reciprocity salience (β = .19, p = .004). A higher score on fairness/reciprocity salience predicted less reasoning that reflected in-group-loyalty.
Authority-Respect. Finally, authority-respect reasoning (R2 = .067, F(5, 84) = 1.198, p
= .317) was predicted by authority/respect salience (β = .32, p = .045). A higher score on authority/respect salience predicted an increase in reasoning coded as authority-respect.
Table 3.
Multiple Linear Regression Models Estimating Effects of Moral Foundation Salience on Type of Moral Reasoning (n=90)
Moral reasoning
Harm-Care Fairness-Reciprocity In-Group-Loyalty Authority-Respect Purity-Sanctity
Variables B SE b* B SE b* B SE b* B SE b* B SE b* (Constant) 0.01 1.31 0.02 0.95 3.26 1.46 -0.20 0.91 0.74 0.66 Harm/Care 0.23 0.18 0.15 -0.12 0.13 -0.10 0.39 0.20 0.22 0.15 0.13 0.14 -0.12 0.09 -0.16 Fairness/Reciprocity -0.04 0.18 -0.03 0.22 0.13 0.19† -0.58 0.20 -0.33** -0.01 0.12 -0.01 0.05 0.09 0.06 In-Group/Loyalty 0.01 0.16 0.01 -0.18 0.12 -0.21 -0.03 0.18 -0.02 -0.13 0.11 -0.17 -0.06 0.08 -0.10 Authority/Respect -0.07 0.17 -0.06 0.34 0.13 0.41** 0.01 0.20 0.01 0.25 0.12 0.32* 0.05 0.09 0.10 Purity/Sanctity 0.09 0.16 0.09 -0.12 0.11 -0.17 -0.15 0.17 -0.14 -0.16 0.11 -0.24 -0.01 0.08 -0.02 R2 / Adjusted R2 .029 / -.029 .121† / .068 .111† / .058 .067 / .011 .034 / -.034 †p < .10, *p < .05, **p < .01, and ***p < .001.
Influence of conditions on type of moral reasoning. To test RQ4 and RQ5, the possible influences of the conditions on the type of moral reasoning, a two-way MANOVA was conducted. This test showed that there were no statistically significant differences in type of moral reasoning based on the social context, Wilk's Λ = 0.915, F (5, 51) = 0.944, p = .461, partial η2 = .09 or on the In-Game versus Post-Game condition, Wilk's Λ = 0.934, F (5, 51) = 0.721, p = .611, partial η2 = .07. In addition, there was no significant interaction effect between social context and time of reasoning, Wilk's Λ = 0.903, F (5, 51) = 1.097, p = .373, partial η2 = .10. However, social context did have a marginally significant effect on purity-sanctity reasons, F(1, 59) = 3.161, p = .081. Participants in the Solo condition (M = 0.39, SD = 0.50) provided more purity-sanctity reasons than participants in the Duo condition (M = 0.16, SD = 0.37).
Harm-Care Choice
In total 121 reasons for the Harm-Care choice were provided, of which 80 were coded as moral reasons (see Table 4); 66.12% of all reasons were moral reasons. To test the possible influences of the social context and the In-Game versus Post-Game conditions on the type of (moral) reasoning for the Harm-Care choice and the outcome of the choice, three two-way MANOVA’s were conducted, however, no significant main or interaction effects of the conditions were found.
23 Table 4.
Descriptives of Type of (Moral) Reasoning for Harm-Care choice (N=100)
Variables M SD Min Max
Moral reasons 0.83 0.59 0 3 Harm-Care 0.51 0.58 0 2 Fairness-Reciprocity 0.25 0.49 0 2 In-Group-Loyalty 0 0 0 0 Authority-Respect 0.07 0.26 0 1 Purity-Sanctity 0 0 0 0 Strategic reasons 0.40 0.49 0 1 Enjoyment reasons 0.03 0.18 0 1
Moral foundation salience on the outcome of Harm-Care choice. To answer RQ6, if moral foundation salience predicts the outcome of a Harm-Care choice, a logistic regression was performed to ascertain the effects of moral foundation salience and gender (0 = male and 1 = female) on the likelihood that participants chose to ease the pain of Falcone (0 = hurt him and 1 = ease his pain)(see Table 5). The logistic regression model was statistically significant, χ2(6) = 23.298, p = .001. The model explained 29.1% (Nagelkerke R2) of the variance in the outcome of the Harm-Care choice and correctly classified 73% of cases. Males were 0.08 times more likely to choose to ease the pain of Falcone than females. An increasing score on harm/care salience was associated with an increased likelihood of choosing to ease his pain, with an odd ratio of 3.21, in contrast, an increasing purity/sanctity salience score was associated with an increased likelihood (OR = 0.39) of choosing to hurt Falcone.
24 Table 5.
Binomial Logistic Regression Model Estimating Effects of Moral Foundations Salience and Gender on the outcome of Harm-Care choice (N=100)
Variables B SE p-Value OR 95% C.I.
Harm/Care 1.17 0.48 .015 3.21 1.25 – 8.19 Fairness/Reciprocity 0.43 0.48 .373 1.53 0.60 – 3.91 In-Group/Loyalty -0.37 0.36 .304 0.69 0.34 – 1.40 Authority/Respect 0.61 0.42 .152 1.84 0.80 – 4.21 Purity/Sanctity -0.95 0.44 .032 0.39 0.16 – 0.92 Gender -2.48 0.96 .010 0.08 0.01 – 0.55 (Constant) -3.22 3.18 .311 Model χ2 = 23.298, df = 6, p = .001
Type of reasoning for Harm-Care choice. To test RQ7, which looks at the relationships between participants’ individual moral foundation salience, gender and the outcome of the Harm-Care choice and the type of reasoning they employed for the Harm-Care choice, three multiple linear regressions were conducted. Each regression included moral foundation salience, gender, and the outcome of the Harm-Care choice as independent variables (see Table 6).
Moral reasons for choice. The multiple linear regression to predict moral reasoning
for the Harm-Care choice was significant, F(7, 88) = 4.151, p < .001, with an R2 of .248. The moral foundations harm/care (β = .27, p = .033), in-group/loyalty (β = .24, p = .052) and authority/respect (β = .25, p = .071) were positively related to the number of moral
reasonings. A higher score on each of these moral foundations predicted more reasoning that reflected morality for the HarmCare choice. A higher score on purity/sanctity salience (β = -.44, p = .003), however, predicted a lower number of moral reasoning for the Harm-Care choice. The outcome of the Harm-Care choice also predicted the number of moral reasoning
25 for this choice (β = -.42, p < .001), participants that chose to hurt Falcone significantly
provided more moral reasons than those who chose to ease the pain of Falcone.
Strategic reasons for choice. The regression model to predict strategic reasoning was also found significant (R2 = .172, F(7,88) = 2.605, p = .017). No significant effects from any of the moral foundations were found, however, the outcome of the Harm-Care choice
predicted the number of strategic reasoning for this choice (β = .36, p = .001); participants that chose to ease the pain of Falcone provided more strategic reasons for the Harm-Care choice than those who chose to hurt him.
Enjoyment reasons for choice. The regression to predict enjoyment reasoning for the Harm-Care choice (R2 = .152, F(7, 88) = 2.255, p = .037) found two significant predictors. A higher score on fairness/reciprocity (β = .07, p = .037) predicts a higher number of enjoyment reasons provided for the Harm-Care choice. In addition, those who chose to hurt him provided significantly more enjoyment reasons than those who chose to ease his pain (β = -.26, p = .020).
26 Table 6.
Multiple Linear Regression Models Estimating Moral Foundation Salience, Gender and Choice Outcome on Type of Reasoning for Harm-Care choice (n=96)
Type of reasoning for Harm-Care choice
Moral reasons Strategic reasons Enjoyment reasons
Variables B SE b* B SE b* B SE b* (Constant) -0.61 0.75 0.61 0.65 -0.03 0.23 Harm/Care 0.23 0.11 0.23* -0.08 0.09 -0.09 -0.02 0.03 -0.08 Fairness/Reciprocity 0.09 0.10 0.09 0.02 0.09 0.02 0.02 0.03 0.07 In-Group/Loyalty 0.19 0.09 0.24† -0.12 0.08 -0.18 0.06 0.03 0.28* Authority/Respect 0.19 0.10 0.25† 0.08 0.09 0.12 -0.04 0.03 -0.19 Purity/Sanctity -0.28 0.09 -0.44** 0.03 0.08 0.06 0.02 0.03 0.12 Gender -0.07 0.14 -0.05 -0.10 0.12 -0.09 -0.06 0.04 -0.16 Choice Outcome -0.53 0.13 -0.42*** 0.38 0.11 0.36** -0.10 0.04 -0.26* R2 / Adjusted R2 .248*** / .188 .172* / .106 .152* / .085 Note. †p < .10, *p < .05, **p < .01, and ***p < .001. Gender: 0 = male and 1 = female, Choice outcome: 0 = hurt him and 1 = ease his pain
27 Type of moral reasoning for Harm-Care choice. For the Harm-Care choice, no reasoning that reflected in-group-loyalty or purity-sanctity were recorded. To test RQ8 and the relationships between a participant’s individual moral foundation salience, gender, and the outcome of the Harm-Care choice and the type of moral reasoning they employed for the Harm-Care choice, three multiple linear regressions were conducted (see Table 7). For each of these three models, the independent variables were moral foundation salience, gender, and the outcome of the Harm-Care choice.
Harm-Care reasons for choice. The multiple linear regression to predict harm-care reasoning (R2 = .115, F(7,88) = 1626, p = .138) found two significant predictors. The score on the moral foundation in-group/loyalty (β = .25, p = .067) was found to be a positive
significant predictor of the number of harm-care reasoning; a higher score on in-group/loyalty salience predicted more reasoning that reflected harm-care for the Harm-Care choice. Also, gender (β = -.19, p = .077) was found to be a significant predictor; males provided more moral reasons for the Harm-Care choice that were categorized as harm-care than females.
Fairness-Reciprocity reasons for choice. The regression to predict
fairness-reciprocity reasoning (R2 = .338, F(7,88) = 6421, p < .001) found four significant predictors. A higher score on the moral foundation authority/respect (β = .24, p = .067) predicted more reasons classified as fairness-reciprocity for the Harm-Care choice, in contrast, a higher score on the moral foundation purity/sanctity (β = -.24, p = .078) predicted less fairness-reciprocity reasoning. There were also differences between genders (β = .21, p = .030), with females reflecting more fairness-reciprocity reasoning, and on the outcome of the choice (β = -.49, p < .001), with those who chose to hurt Falcone providing more fairness-reciprocity reasoning than those who chose to ease his pain.
Authority-Respect reasons for choice. The multiple linear regression to predict authority-respect reasoning (R2 = .081, F(7,88) = 1.114, p = .362) found two significant
28 predictors. A higher score on harm/care salience (β = .20, p = .086) predicts a higher number of authority/respect reasons provided for the Harm-Care choice. In addition, a higher score on purity/sanctity (β = -.37, p = .022) predicted a lower number of authority/respect reasons.
Table 7.
Multiple Linear Regression Models Estimating Moral Foundation Salience, Gender and Choice Outcome on Type of Moral Reasoning for Harm-Care choice (n=96)
Moral reasoning for Harm-Care choice
Harm-Care Fairness-Reciprocity Authority-Respect
Variables B SE b* B SE b* B SE b* (Constant) -1.47 0.79 0.67 0.54 -0.27 0.36 Harm/Care 0.16 0.12 0.15 -0.02 0.08 -0.02 0.09 0.05 0.20† Fairness/Reciprocity 0.14 0.11 0.14 -0.01 0.07 -0.01 -0.03 0.05 -0.08 In-Group/Loyalty 0.19 0.10 0.25† -0.05 0.07 -0.08 0.04 0.05 0.13 Authority/Respect 0.01 0.11 0.01 0.14 0.08 0.24† 0.06 0.05 0.20 Purity/Sanctity -0.09 0.10 -0.14 -0.12 0.07 -0.24† -0.11 0.05 -0.37* Gender -0.26 0.15 -0.19† 0.22 0.10 0.21* -0.02 0.07 -0.04 Choice outcome -0.02 0.14 -0.02 -0.48 0.10 -0.49*** -0.04 0.06 -0.07 R2 / Adjusted R2 .115 / .044 .338*** / .285 .081 / .008
Note. †p < .10, *p < .05, **p < .01, and ***p < .001. Gender: 0 = male and 1 = female, Choice outcome: 0 = hurt him and 1 = ease his pain
29
Discussion and Conclusion
This study tried to understand the relations between a person’s innate moral
foundations and the decisions, reasoning, and moral reasoning they produce in a video game context. Similar to other studies on the subject (Krcmar & Cingel, 2016; Weaver & Lewis, 2012), results indicate that players use moral reasoning in a video game context in addition to strategic reasoning. Of the 660 reasons that were analyzed in this study, 55.52% were moral reasons; which is similar to Krcmar and Cingel’s (2016) findings that showed 54% moral reasoning compared to 46% strategic reasoning. However, they did not code separately for enjoyment reasoning which explains their higher percentage of strategic reasoning, and further suggests that taking enjoyment reasoning into account is mostly at the expense of strategic reasoning.
The results showing moral reasoning to be prevalent in a video game context suggest that the boundaries of the magic circle do not constitute a strict demarcation between the game and the actual-world and that in-game reasoning is influenced, in part, by a person’s moral foundations; which is in contrast to the concept of the magic circle still prevalent in video game studies (Consalvo, 2009; Stenros, 2014). Also, a number of reasonings reflecting enjoyment were found, suggesting not only strategy and morality can guide decisions, but identifying the choice that seems the most fun or pleasurable can steer the decision-making process too. These results are interesting because they can be used to argue that a player of a video game is not divesting himself of the need to consider the moral consequences of his actions, even though there are no actual-world consequences, and only use strategic viewpoints, but take, in part, morality and enjoyment into consideration when making in-game decisions.
30 The Influence of the Conditions
Even though some effects that approached significance, or were moderately significant, were found for the two manipulated conditions (see RQs 1, 2, 4, and 5), the overall impact of social context and timing of reasoning on the type of reasoning and type of moral reasoning was limited. However, to conclude that the social context has no influence at all on in-game decisions and reasoning would be too strong of a statement. The current study looked at differences between players playing alone or with a friend for a single player game, however, a multiplayer game might constitute a totally different social context with different, stronger effects on (moral) reasoning. In addition, not all relationships between participants in the Duo condition were the same; most were friends, but some were strangers to each other. Since there is research (Ravaja et al., 2006) suggesting playing games against a friend elicits greater engagement and physiological arousal compared to playing against a stranger, the same might be true when playing with a friend is compared to playing with a stranger. This could mean the type or strength of the relationship between players in the Duo condition might have had an influence on their engagement with the game, possibly resulting in an effect on their reasoning. More research studying the variations of video game experiences between particular social situations is needed to further understand the various influences the social context could have.
With regard to the lack of differences between In-Game and Post-Game reasoning, it might be important to reflect on the limitations of the method. As discussed in the theoretical background, comparing the method of collecting reasonings from players, before or after gameplay, to the two different systems of thinking is not an exact one-on-one comparison. Even though this was the first study to focus on In-Game versus Post-Game reasoning and found no very strong main effects, further inquiry into System 1 and System 2 thinking (Evans, 2003; Kahneman, 2011) during video games, and, specifically, which one gamers
31 employ in what situation, is needed in order to make more interesting statements about in-game decisions and reasoning.
Moral Foundations and Moral Reasoning
One of the main aims was to determine the possible relations between moral foundations and the type of moral reasoning gamers employ. Research question 3, which asked if an individual’s moral foundations influence the type of moral reasoning they employed, emanated in some compelling results. The results indicate, again in congruence with previous research (Krcmar & Cingel, 2016) and MFT (Haidt et al., 2008), that the level of importance an individual allocates to specific moral foundations was a predictor for the number of reasons used that reflected that moral foundation. More accurately, the
fairness/reciprocity and authority/respect foundations were positive predictors of the use of reasons referencing that respective foundation. This suggests that, at least for some, players of video games may use the same kind of moral reasoning processes in the game and actual world and place importance to the same kind of moral considerations. However, the salience of the in-group/loyalty, purity/sanctity, and harm/care foundations were not found to be related to the use of reasoning reflecting those respective foundations for all of the combined dilemmas presented to the players. The results also indicated that a higher score on one moral foundation can make reasonings which reflect other types of moral reasoning less likely; for example, a higher score on fairness/reciprocity salience predicted less reasoning that reflected in-group-loyalty. This might be in line with research suggesting the existence of fundamental differences between people’s personalities on moral foundations (Medjedovic & Petrovic, 2016), for example, concerning politics liberal and conservatives differ on their sets of moral foundations (Graham et al., 2009). The effects of a moral foundation on reasoning reflecting another moral foundation found in this study might in that sense be effects of a person’s complete set of foundations. More research looking at the correlations and influences between
32 the moral foundations could explain how they relate to each other better and how this
influences their effects on moral reasoning in video games. This could be done by, for example, categorizing participants into groups with similar complete sets of moral foundations and study their possibly different video game experiences.
Harm-Care Choice
Besides looking at the combined dilemmas, this study looked at a specific dilemma to get more insight into how participants approached this choice and what influenced this. The analyzed dilemma can be considered exemplary of a Harm-Care choice, with the option being to hurt a non-player-character or to ease his pain. The question if moral foundation salience predicts the outcome of a Harm-Care choice (RQ6) can be answered positively. The results show that for this specific dilemma, as one would expect, the harm/care salience predicted the outcome of the choice; a higher score predicted easing the pain of the non-player-character Falcone instead of hurting him. This is in line with other studies finding results that indicate video game players can act socially towards non-player-characters (Hartmann et al., 2010). In contrast to harm/care salience, purity/sanctity salience predicted hurting Falcone and made it less likely that the reasoning provided for the Harm-Care choice reflected morality.
The results for studying the Harm-Care choice (pertaining to RQs 6, 7, and 8) show that those who choose to hurt Falcone were more likely to be female and provided more moral and enjoyment reasoning than those who chose to ease his pain; also their moral reasoning reflected more fairness-reciprocity. This shows that hurting Falcone was justified morally or based on enjoyment, instead of strategically, and that the moral justifications for hurting him were mostly based on fairness-reciprocity. Since there is research suggesting that unjustified virtual violence produces guilt (Hartmann et al., 2010), providing a moral reason to hurt a non-player-character might be a tactic of gamers to reduce the effects of guilt. The fact that hurting Falcone was related to moral reasoning that reflected fairness-reciprocity adds to that,
33 because when arguing hurting him was fair there is no need for experiencing guilt. Easing the pain of Falcone was significantly related to strategic reasoning which suggests that not hurting Falcone was, for some, not a moral consideration but a strategic one. The fact that hurting Falcone was related to providing more enjoyment reasoning fits with the idea (Bareither, 2017) that acts of virtual violence can be done with the aim of experiencing the pleasure of transgression. Hurting Falcone for the player’s enjoyment can be seen as intentionally provoking incongruences by doing something that breaks normal patterns and this incongruence is experienced as enjoying and adding to the enjoyment of the video game experience.
In contrast to findings on all dilemmas combined, the results for this specific Harm-Care dilemma do not indicate that a higher score on a moral foundation can make moral reasoning reflecting that respective foundation more likely. For example, even though
harm/care salience predicted the outcome of the choice, it did not predict reasoning reflecting harm-care. The consequence of this finding is that, in order to find effects of a moral
foundation on reasoning reflecting that foundation, multiple different dilemmas need to be studied. This allows for different approaches by players to different dilemmas and results in various reasonings, which in turn can be linked to moral foundations. Focussing on one dilemma gives more insight into the decision-making process of players, but more dilemmas are needed to be able to link moral foundations to moral reasons.
Limitations
An interesting topic, not addressed in this study, is the relation between the player and the game character. There is a debate on identification (see, for example, Christoph, Dorothée, & Peter, 2009) that tries to answer whether a player inhabits a character, simply controlling it, or if a player merges his personality with the character being controlled. This merging is different from inhabitation because elements of the game’s protagonist then do influence the
34 player; the player adopts (parts of) the perceived identity of the target character. Considering this study, taking on the role of Batman, a well-known cultural iconic character (Brooker, 2005), might have influenced players to feel, reason, and act more, for example, heroically than as they would otherwise. Future research on video game reasoning should take into account the perspective of the reasoning to determine if the media-character being controlled influences decisions or reasoning.
Even though this study provides new insights into the (moral) decision-making process of gamers, one of the limitations is the fact that only one video game was used, making generalization troublesome. This game was chosen due to its narrative-branching based structure which allowed gamers to choose between dialogue options when presented with several dilemmas. The options were limited, which made studying the outcome and consequential reasoning pretty straightforward. However, to get a better understanding of reasoning and moral processing in video games, different types of video games need to be studied and compared. Especially free roaming games, giving players higher degrees of freedom of movement, actions, and decision-making within their game world, could be of specific interest.
The sample of this study also included some limitations, consisting, for example, mostly out of non-gamers. Previous research found that participants from different nations score differently on the Moral Foundation Questionnaire (Dogruel et al., 2012), which relates to the fact that different cultures have differing ideas about morality. In this way, the various nationalities of participants in this study’s sample could be considered a weakness.
Conclusion
To conclude, this study looked at the influence of an individual’s moral intuitions on their moral reasoning in a video game context. Results indicate that for some moral
35 study also looked at possible influences on reasoning of the social context in which the game was played and of the timing of measuring reasoning, however, not many and no strong influences were found. This experiment makes a humble contribution to the emerging field of research on the intersection of morality and gaming and stresses video game scholars to not only address (im)moral decision outcomes but also to relate them to the players’ justifications and reasoning.
36
Bibliography
Babij, K. (2013). The Good, the Bad and the Neutral: Problems with the Ethical Constructions of Video and Computer Games. In M. Wysocki (Ed.), CTRL-ALT-PLAY: Essays on Control in Video Gaming (pp. 158–168). Jefferson, NC and London: McFarland & Company.
Bareither, C. (2017). “That was so mean :D” – Playful virtual violence and the pleasure of transgressing intersecting emotional spaces. Emotion, Space and Society, 25, 111–118. https://doi.org/10.1016/j.emospa.2016.12.005
Brooker, W. (2005). Batman unmasked: analyzing a cultural icon. New York: Continuum. Calleja, G. (2015). Ludic identities and the magic circle. In V. Frissen, S. Lammes, M. De
Lange, J. De Mul, & J. Raessens (Eds.), Playful Identities: The Ludification of Digital Media Cultures (pp. 211–224). Amsterdam: Amsterdam University Press.
Christoph, K., Dorothée, H., & Peter, V. (2009). The Video Game Experience as “True” Identification: A Theory of Enjoyable Alterations of Players’ Self-Perception. Communication Theory, 19(4), 351–373. https://doi.org/10.1111/j.1468-2885.2009.01347.x
Consalvo, M. (2005). Rule Sets, Cheating, and Magic Circles: Studying Games and Ethics. International Review of Information Ethics, 4(July), 7–12.
https://doi.org/10.1087/20110109
Consalvo, M. (2009). There is no magic circle. Games and Culture, 4(4), 408–417. https://doi.org/10.1177/1555412009343575
Dogruel, L., Joeckel, S., & Bowman, N. D. (2012). 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. https://doi.org/10.1177/1461444812451571
37 Cognitive Sciences, 7(10), 454–459. https://doi.org/10.1016/j.tics.2003.08.012
Ferguson, C. J. (2015). Does media violence predict societal violence? It depends on what you look at and when. Journal of Communication, 65(1), E1–E22.
https://doi.org/10.1111/jcom.12129
Goldstein, J. (2005). Violent video games. In J. Raessens & J. Goldstein (Eds.), Handbook of computer game studies (pp. 341–357). Cambridge, MA: MIT Press.
Graham, J., Haidt, J., & Nosek, B. A. (2009). Liberals and conservatives rely on different sets of moral foundations. Journal of Personality and Social Psychology, 96(5), 1029–1046. https://doi.org/10.1037/a0015141
Haidt, J., Graham, J., & Hersh, M. A. (2008). The Moral Foundations Questionnaire. Retrieved February 2, 2018, from http://www.moralfoundations.org/questionnaires Hartmann, T., Krakowiak, K. 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. https://doi.org/10.1080/03637751.2014.922206
Hartmann, T., Toz, E., & Brandon, M. (2010). Just a Game? Unjustified Virtual Violence Produces Guilt in Empathetic Players. Media Psychology, 13(4), 339–363.
https://doi.org/10.1080/15213269.2010.524912
Heron, M. J., & Belford, P. H. (2014). Do You Feel Like a Hero Yet? Externalised Morality in Video Games. Journal of Games Criticism, 1(2), 1–22. Retrieved from
http://gamescriticism.org/articles/heronbelford-1-2
Huizinga, J. (1950). Homo ludens. Proeve eener bepaling van het spel-element der cultuur. In L. Brummel et al. (Ed.), Johan Huizinga, Verzamelde werken V. Cultuurgeschiedenis III (pp. 26–246). Haarlem: H.D. Tjeenk Willink & Zoon. Retrieved from
38 Hung, A. C. Y. (2011). The work of play: meaning-making in videogames. New York, NY:
Peter Lang Publishing.
Joeckel, S., Bowman, N. D., & Dogruel, L. (2012). Gut or Game? The Influence of Moral Intuitions on Decisions in Video Games. Media Psychology, 15(4), 460–485.
https://doi.org/10.1080/15213269.2012.727218
Joeckel, S., Bowman, N. D., & Dogruel, L. (2013). The Influence of Adolescents’ Moral Salience on Actions and Entertainment Experience in Interactive Media. Journal of Children and Media, 7(4), 480–506. https://doi.org/10.1080/17482798.2013.781513 Juul, J. (2010). A Casual Revolution: Reinventing Video Games and their Players.
Cambridge, MA; London: MIT Press.
Kahneman, D. (2011). Thinking fast and slow (1st ed.). New York: Farrar, Straus and Giroux. Klimmt, C., Hefner, D., Vorderer, P., Roth, C., & Blake, C. (2010). Identification with video
game characters as automatic shift of self-perceptions. Media Psychology, 13(4), 323– 338. https://doi.org/10.1080/15213269.2010.524911
Klimmt, C., Schmid, H., Nosper, A., Hartmann, T., & Vorderer, P. (2006). How players manage moral concerns to make video game violence enjoyable. Communications, 31(3), 309–328. https://doi.org/10.1515/COMMUN.2006.020
Krcmar, M., & Cingel, D. P. (2016). Moral Foundations Theory and Moral Reasoning in Video Game Play: Using Real-Life Morality in a Game Context. Journal of
Broadcasting & Electronic Media, 60(1), 87–103. https://doi.org/10.1080/08838151.2015.1127246
Lin, S.-F. (2011). Effect of opponent type on moral emotions and responses to video game play. Cyberpsychology, Behavior and Social Networking, 14(11), 695–8.
https://doi.org/10.1089/cyber.2010.0523
