Between the Game System and the Fictional World

Introduction

Contrary to the prophecies of Janet Murray (1997), games have not developed into a medium of interactive film. Over 30 years after their commercialization, computer games have established themselves as a media format with its own aesthetic conventions and styles. The “language” of computer games has not, however, been developed in a vacuum, but rather it builds on the conventions from other formats such as film and human–computer interaction (HCI). Modern computer games are typically set in virtual environments partly inspired by the audiovisual conventions of film fiction in their presentation of characters, locations, and events. At the same time, they are pieces of software that need well-developed systems that allow the player to interact with them in a consistent manner and that provide information necessary for coherent gameplay. These conventions are combined into a communicative system unique to games. This is a system that puts emphasis on usability and functionality to the detriment of a coherent fictional world, something that signals a major break from the conventions of traditional fictional media.

Today’s game user interfaces (UI) vary considerably, from the minimal and transparent head-up displays (HUD) of first-person shooters, to the solid overlays and frequent pop-up windows superimposed onto the screen in real-time strategy games (see illustration below). Also physical interfaces vary from traditional input devices such as controllers, mouse and keyboard, to the “natural” input (Nash, 2011) of systems such as Microsoft’s Kinect and Sony’s Play station Move. For all kinds of interfaces, the trend is to minimize them as much as possible by integrating parts of them into the game world (Andrews, 2010; Fagerholt & Lorentzon, 2009, p.5). While general principles of aesthetics and tastefulness motivate this design approach, it has also been argued that any perceived mediation lessens the sense of involvement (Fagerholt & Lorentzon, 2009, p. 3). Today there seem to be two fields of thought about this practice: those that argue that any sign of mediation, or visibility of the system behind, ruins the sense of involvement in the game world (Wilson, 2006); and those that argue that such features do no harm since they are needed and have become a conventional form of communication in games (Breda, 2008; Weise, 2008).

This article discusses how players experience the game UI and how it affects the interpretation of the game world. First, I will discuss and delimit how the game interface is understood in this article, before going on to present developers’ ideals for game interface design. The main portion of the article will discuss how players interpret the UI with respect to the game world. As we will see, players rarely consider system features intrusive to the game world, but accept them as conventions established as part of modern computer game aesthetics. From this perspective, the UI and the game world are not as delimited as one may think, but operate together in a coherent manner.

The article relies on two sources of data: Group interviews with two American game developers and interviews with and observations of Norwegian computer game players between the age of 18–40. The two developers, Harmonix Music Systems and Turbine, Inc., were selected because they develop games that represent very different genres, and on the basis of availability during my stay as a visiting scholar in the United States. Both developers underwent focus group interviews; the interview with Turbine was held with four members of the UI design team, and the Harmonix interview took place with two members of the UI design and testing teams.

The recruitment method for player respondents was self-selection and the snowball method. Five players underwent a focus group interview, while another 17 interviews were carried out with single individuals. The gender ratio was 4 females and 13 males. The focus group were asked to discuss a range of questions about the game interface and the game world on the basis of screenshots from games from four different genres, while the individual interviews were based on video captures of the participants’ own gameplay and a comparison with other genres based on screenshots. In both cases, the four games chosen were the real-time strategy game Command & Conquer 3: Tiberium Wars; the first-person shooter Crysi; the hack and slash role-playing classic Diablo 2; and the dollhouse simulator The Sims 2 (see illustration below).

OPEN IN VIEWER

Image 1: Crysis (Crytek, 2007). Image 2: The Sims 2 (Maxis, 2004). Image 3: Diablo 2 (Blizzard, 2000). Image 4: Command & Conquer 3: Tiberium Wars (EA, 2007). All screenshots captured by the author.

The games were selected on the basis of popularity and variation in gameplay and player perspective. Games that position the player with different perspectives with respect to the game world were chosen, and while three games use traditional overlay interfaces, Crysis uses a minimalist interface where most elements are explained as parts of the game world. The Sims 2, however, was chosen as an interesting hybrid of genres, and in order to attract female players to the study. In order to compare games with the same physical interface, I decided to use games from one platform only, and the PC platform was selected due to limitations connected to technological equipment.

The Game Interface

What is the interface of a computer game? According to game developer Brent Fox, the game interface “is the part of the game that allows the user to interact with the game. [It] is the connection between the user and the game” (2005, p. xv). Jesse Schell describes the interface more specifically as “the infinitely thin membrane that separates (…) player and (…) game”, and that provides the player with access to the game world (Schell, 2008, p. 222). He puts emphasis on the idea that game UIs consist of two features, namely the physical and the virtual (Schell, 2008, p. 226). The physical interface consists of the hardware, such as controllers and display, while the virtual interface is the auditory and visual software features that allow the player to interact with the game system. Common for the two is that they allow player input and provide game output, and that they map computer code to specific actions in the game (2008, p. 224–227). Taking this as point of departure, we may see a game’s interface as graphical, auditory, and tactile features related to hardware and software that allow the player to interact with the system beyond. Taking a step back to the field of HCI where the term interface is most commonly used, we can understand the UI as “the part of the system that you see, hear and feel” (Lauesen, 2005, p. 4). Lauesen thus takes a broader perspective than Fox and Schell by not limiting the definition to system control features, but including all aspects of the system that are available to the user. This perspective is in concert with the arguments proposed by designer Marcus Andrews (2010) and Fagerholt and Lorentzon’s study of the UI in first-person shooters (2009). They include system information that is integral to the game environment, as well as borderline features such as blood splatter rendered on screen to signal relative health level (Killzone 2), and silhouettes that outline the relative location of companions (Left4Dead).

Although I will make reference to the physical interface where relevant, I will in this article largely exclude the physical interface and focus the argument to what Schell calls the virtual interface. By doing so, I am acknowledging the broader perspective that the interface may include any features of the represented game environment. This means that in my understanding of the term, the game UI is not limited to system features such as overlay menus and HUD, but may also be implemented into the game world as color schemes that signal the importance of specific objects, or animations that provide information about the current state of objects or characters.

This means that the game interface is not a static or absolute entity. Being integrated into the game world, many interface features balance on the border between the universe of the game and the layer that provides information to the player (Jørgensen, 2007). According to Alexander Galloway, interfaces have commonly been compared to a doorway or window into another dimension, “a gateway that opens up and allows passage to some place beyond” (Galloway, 2008, p. 936). He argues that this metaphor is limited. While doors may be closed or open, the interface is better compared to the threshold, as it is a tool for communication between two spaces, challenging the distinction between what counts as the artwork’s edge and centre (Galloway, 2008, p. 938). The interface is a relation effect that creates coherence between otherwise uncombinable domains (Galloway, 2008, p. 941). It should therefore be seen not so much as a feature that is similar for all games, but more in terms of a collection of techniques used for communicating necessary information and allowing the player to interact with the mediated environment.

Looking at computer game interfaces more closely, we see that these techniques are realized in different ways. The common denominator is that they all position themselves in a dimension somewhere between the screen and the interior of the game world. Often, however, an interface feature may be part of the game world while communicating to the player on the outside, or vice versa (Jørgensen, 2007, 2010). These features may seem to be part of the game world even though they clearly communicate usability information to the player, or may appear to be external to the game world although they are responded to by inhabitants internal to that world. Since there is ambiguity as to whether such features belong to the game world or the interface layer, they must be interpreted as borderline features placed at varying levels between the two. From this perspective, there is a tight relationship between the game interface and the game world, and as we will see, this relationship affects how the players understand both.

It may be tempting to associate these elements with “diegetic” and “non-diegetic” features which describe the relative “interiority” and “exteriority” of features with respect to story worlds. However, I am trying to avoid this association in this article, as these terms do not grasp the functional and interactive aspects of the game interface and the game world (Collins, 2008, p. 180; Jørgensen, 2010, p. 81). Being developed for narrative media, diegetic and non-diegetic are also confusing when used in connection with games since they focus on the idea of a narrated story world instead of a playable game world (Jørgensen, 2010, p. 87; Klevjer, 2007, p. 46). Game interfaces approach the relationship between game world and system information in a different manner, where the goal is to provide necessary usability information.

We may identify different ways of integrating the interface into the game world that are typically used in computer games (Jørgensen, 2010, pp. 92–93). The examples below are not absolute, and it is likely that there may be other ways of integrating the interface into the game world. While integrated into the game world, these features must also be seen as part of the interface since they have clear usability functions in that they provide relevant gameplay information to the player.

An interface feature may be iconic or “invisible” in the sense that it corresponds to features that occur naturally in the physical world, at the same time as it provides gameplay information to the player: the specific armor that an avatar wears in World of Warcraft signals relative power, but it is not highlighted or otherwise enhanced. Objects that appear to be natural to the game world can also be specifically emphasized, exemplified by Crysis where targeted enemies are highlighted in red. Interface features can be integrated, such as the exclamation marks in Diablo 2; while being added for the player’s information and not corresponding to anything we know from the physical world, these features behave as if they are part of the game world by following a particular character. The interface can also be added as an overlay that provides the player with the ability to interact with the game world, such as the traditional superimposed frames that we see in strategy and simulation games like Command & Conquer 3 and The Sims 2. We can also identify the metaphorical interface, which has the ability to affect the game world even though it appears external to it; this is typically exemplified by enemy music that signals approaching danger.

With the exception of iconic interface features that are completely integrated into the game world, these interface examples are all in different ways located on the threshold between the game system and the game world, combining them functionally and thereby posing the question where the interface ends and the game world begins.

Challenges and Ideals in Game UI Design

According to Schell, good game UI design is critical for the player’s experience. He claims that “it is crucial for us to understand how our game interface works, and to make it as robust, as powerful, and as invisible as we can” (Schell, 2008, p. 222). For Schell, this is connected with the ability of the interface to communicate relevant and necessary communication (2008, p. 234), a view that is on par with the UI designers interviewed for this study. They see the game interface as a vital element in game development, as it includes everything that provides information that guides interaction with the game world. This means that the interface also is responsible for communicating the game system beyond in an effective manner. The interface is a simplification of the game system, a communications channel that makes the system clear:

(…) So the interface actually starts with the controller, collapsing things down to something that is much more easy to be processed. And then the game interface is communicating that in a way that is easily learned and managed. (…) (Harmonix, respondent A)

These principles are central for the virtual and the physical interface alike. According to Schell, it is easy to ignore this aspect when it comes to the physical interface, and instead copy existing interfaces instead of customizing controllers for a given game (Schell, 2008, p. 226). As a developer of games such as Rock Band where the physical interface is particularly important, Harmonix is in particular attentive to this issue. Their goal in developing a physical interface that emulates musical instruments is about making “people feel like they’re rock stars” through intuitive physical interfaces (Respondent A). Intuitivity for Harmonix is therefore about whether or not the physical controllers are easy to understand in terms of how real-world musical instruments work.

Today the ideal for game interface design is to make it as intuitive and as involving as possible (Schell, 2008, p. 227). In many genres, and in particular on video game consoles, there is a trend toward including as much as possible of the UI into the game world (Andrews, 2010; Wilson, 2006). According to Bolter and Grusin, an “invisible” or “interfaceless” interface means that “there will be no recognizable electronic tools—no buttons, windows, scroll bars, or even icons as such. Instead the user will move through the space interacting with the objects “naturally”, as she does in the physical world” (2000, p. 23). The design ideal is to make the medium transparent, thereby providing the media users with the impression that they are no longer interacting with a medium, but instead directly with its contents (Bolter & Grusin, 2000, p. 24). This desire of new media to erase any trace of mediation is also present in the developers’ minds, not least because the two companies interviewed are both developers of games that must rely on information-heavy interfaces. Massively multiplayer online game developer Turbine expresses worry about the information-heavy overlay interfaces that are being used in that genre, and argues that it necessarily will maintain a certain distance between the player and the game world:

I guess a user interface can only fight against immersion at the end of the day. I mean, to be completely immersive is that you’re there and you’re doing it, so any kind of interface that separates you from actually being in the gameworld, is taking you out of the immersive experience. So I guess (…) a good interface will immerse you when it’s time to be immersed, and then when it’s time, to pull out a little. (Turbine, respondent A)

According to the developer respondents, however, intuitive and involving do not necessarily mean transparent or fully implemented into the game world. This corresponds to Schell’s description of transparency, which focuses not on “invisibility,” but on whether or not the UI provides relevant information to the player without feeling intrusive (Schell, 2008, p. 227). Prioritizing functionality before aesthetics is therefore paramount, and following established genre conventions is important in order to fulfill player expectations. When the primary objectives of functionality and usability are within reach, however, the developers aim for an aesthetic interface that is associated with the game world thematically and stylistically, or by being implemented into it. Fox focuses on this by writing extensively about how aesthetic principles such as color harmony and visual organization may support usability (2005, pp. 43–60). Saunders and Novak list aesthetics as a secondary interface design goal after the primary functional goals (2006, pp. 20–30).

Neither Turbine nor Harmonix has the opportunity to implement all interface into the game world. As a complex multiplayer genre, Massively Multiplayer Online Games (MMOs) require the player to monitor of a lot of processes simultaneously. Although Harmonix has the benefit of creating the illusion of nonmediation through the intuitive physical interface, the focus on player precision and attention puts specific demands on the clarity and efficiency of the virtual interface. The two developers are therefore bound to use overlay and integrated interface elements to a great degree. Although they lack the opportunity to implement information seamlessly into the game world, the benefit is that they can rely on the safety of being able to present information through dedicated channels, and may focus on making the interface as intuitive and unobtrusive as possible (Turbine, respondent C) within the limits of the genres they are working with.

Although being representatives of two different viewpoints with respect to transparent UI design, the two developers agree that aesthetics and the sense of nonmediation always must be downgraded when compared to functionality. In practice, they both take a relativist view where they emphasize that functionality always will have to win, but where aesthetics come into play through a clean, clear, and intuitive interface design.

A fully integrated iconic or emphasized interface is not an ideal for presenting essential information in an unhindered manner, but it is still important that the UI does not take up the player’s attention or stand in the way of gameplay, but is “comfortable for you to instantly look at” (Turbine, respondent B). Aesthetics is therefore an ideal that the developers always screen functionality against, and one of Turbine’s UI designers states that “a good interface would also be aesthetically attractive”:

If you can sit down and you can figure out what exactly what you want to do and how you want to do it, and it’s not impeding your enjoyment of the game world, that, at the end of the day, is a successful interface. And on top of that it obviously should be aesthetically attractive. But you can have the most beautiful interface in the world, and if it just impedes your enjoyment and your ability to do things in the gameworld, then it’s a poor user interface. (Turbine, respondent A)

A way to present usability information without interfering with the visual aspects of the game world is to use auditory features. One of Turbine’s UI designers calls audio a “nice second layer” that can be used when visual information is limited:

A classic thing that just popped into my head, (…) Mega Man (…) has three different kinds of power-ups, and half the time you don’t even see what they are when you get them. But each one has a very distinct sound effect, so you just keep playing the game, and you don’t even really notice, but you know exactly what you got based on the sound. (…) By just having a sound effect you can really tell the person what they’ve got in a non-glaring way. (Turbine, respondent C)

The developer respondents are also careful to emphasize that different genres have different conventions regarding the presentation of information and the UI design. Different genres will have different needs, and the positioning of the player and the system’s real estate have important influences on the realization of the interface (Andrews, 2010). As the Harmonix’ UI designer states, “your interface is only as intuitive as the game design allows.” One of the respondents from Turbine explains:

A lot of the more recent games have gone for the super-minimalist UI where there’s almost nothing on screen at any time until you need to see it. And that’s great for a lot of the more recent kind of horror games where you just try and have it so immersive that you’re not thinking about anything but what’s happening in the moment. (…) But for something more like our type of game where there’s just so much going on at any given moment, you need so much information. So in our case invisible isn’t necessarily ideal. (Turbine, respondent C)

Interface Acceptance

Although the player respondents showed different preferences regarding the design of virtual interfaces and the integration of UI elements into the game world, there was a tendency among the respondents, regardless of what game they were playing, to accept the UI as long as it provided the necessary information. In this section, I will present those views that were most pronounced among the player respondents regarding the integration of UI elements into the game world. Most of the claims presented here are of general relevance, but where necessary I will refer to how the interface of the individual games is interpreted. In accordance with the arguments proposed by Schell (2008, p. 227) and Andrews (2010), it is clear that players tend to accept the presence of UI elements, regardless of its visualization. This seems to hold for all games used in this study, although there is tendency to see the minimalist interface of Crysis as more aesthetically attractive compared to the others.

Players and developers agree that usability and the need for information, together with established conventions in computer games, are the most important reasons why apparently “intrusive” interfaces are accepted. Functionality is at the center of acceptance: If a feature provides the player with relevant information and supports gameplay, the player is likely to accept it without hesitation (Andrews, 2010). When asked about the overlay interface in Command & Conquer 3, Stuart states,

(…) In a game like this (…), it is not natural that things like these exist, but at the same time it is accepted because you couldn’t play the game without them.” (Stuart, 24 years) ¹

Another reason why we accept game interfaces is that they are conventional to the game medium. Due to the informative and functional necessity of interfaces, computer games have made physical and virtual interfaces parts of their mode of communication. Players therefore expect them to be there to present system information. The players have become so accustomed to these conventions that it is not something they contemplate when playing (Bolter & Grusin, 2000, p. 42). As a convention and a part of the communicative toolset of computer games, we are not affected negatively by the presence of the interface. Being motivated by the computer game medium itself, the UIs are established as functional conventions that cannot be seen as separate from the content they are providing access to (Manovich, 2001, p. 67). This is not any different from how we have come to accept the conventions of older media: the intrusive interface of the novel, consisting of pages and letters, never hindered readers to become absorbed by the fictional worlds within (Murray, 1997, p. 104; Ryan, 2001, p. 127). During the focus group, Oliver explains,

It’s like cartoons, right. (…) We accept the speech bubbles because that’s the way cartoons work. (Oliver, 27 years)

From the above, we see that computer game players accept the UIs as conventional to games and do not find them disturbing for the involvement in the game world. But what about those UI elements that are not clearly defined as system information features? I pointed out above that there are many interface elements that lie on the border between the universe of the game and the informative layer that communicates to the player. These features are typically partly implemented into the game world as emphasized, integrated, or metaphorical interface, apparently belonging to two dimensions at once. Similar to overlay interface elements, players do not reflect upon the ontology of such borderline features much, but when confronted with them, the respondents present different responses. Some categorize them together with the traditional overlay interface and state that they are a conventional feature of games, sometimes explaining them as abstractions of real-world features or representations of the game system. Others, however, try to rationalize and present fictional explanations for their presence. Others acknowledge the fact that the borderline features are there, and see it either as a curiosity, or do not contemplate on their existence or spatial belonging.

The players’ attitudes toward borderline interface features seem to be closely associated with the UI design conventions within a game genre. When interpreting borderline features in the first-person shooter Crysis, for instance, the respondents tend to explain them as part of the game world. This is connected to the first-person perspective and the fact that the game uses a minimalist UI partly explained by the game fiction as a technologically advanced feature implemented into the avatar’s helmet. When the focus group is discussing the red markings that inform the player about acquired target in the same game, Steve has no problems accepting that these are also a part of the fictional universe:

This might be a further development of this HUD that he has. He may track targets and things with it. So it’s not unrealistic that this is a part of the game world. (Steve, 24 yrs)

However, not all genres have the luxury of being able to explain the interface thus. The First-person shooter (FPS) genre presents the player’s visual perspective as equal to the avatar’s, and may for that reason explain everything on screen as something also seen by the avatar. Games with a third-person perspective, on the other hand, cannot use this explanation since the avatar and the player does not share perspective. This is clear in Diablo 2, where the respondents do not accept the borderline elements quite as readily. When John hears the response “I’m overburdened!” as he tries to pick up items on the ground, he states that the response appears to originate from the avatar as a character existing in the game world, but emphasizes that this is not a satisfactory explanation:

Well, it is the character’s voice saying this. But I don’t get the feeling that it is the character that speaks. It’s like the game narrator’s voice provides the player with a hint that, okay, you should check your inventory. (John, 30 years)

Informative game features that balance on the threshold, not completely ready to fall down on either side of the boundary between interface and game world, are called intraface by Galloway (2008, p. 944). The intraface is “an interface internal to the interface,” located “within the aesthetic” instead of as a window between the interior and the exterior of the artwork. Galloway defines the intraface as an “internal interface” that is “contained within the image.” He provides computer games as the example of an intraface, as they mix together a fictional environment with different kinds of interface elements. For Galloway, the intraface signals that the fictional environment is demoted in terms of importance to the benefit of system information, and “ultimately determined by a very complex nondiegetic mode of signification” (Galloway, 2008, p. 946). However, in my view, this does not mean that the fictional or represented environment loses its value because of the presence of functional elements, but that the presence of such elements changes the very nature of the game worlds compared to other fictional worlds.

Interpreting the Interface–Game World Relationship

According to Manovich, the presence of a computer interface always influences the media content that it presents: It “shapes how the computer user conceives of the computer itself [and] determines how users think of any media object accessed via a computer” (Manovich, 2001, p. 65). This view is in concert with how many of the player respondents describe the relationship between the UI and the game world. Oliver puts it like this:

On the one hand you have this kind of distanced level with the interface and stuff. But there are those exclamation marks as well, and they are placed in a kind of … I think the world of the game doesn’t work in the same way as the real world does. (Oliver, 27 years)

While Juul (2005) has argued that computer games are real rules in fictional worlds, others have pointed out that game worlds are qualitatively and functionally different from the traditional fictional worlds (Aarseth, 2005; Jørgensen, 2010, p. 87; Klevjer, 2007, p. 58). This is in concert with Oliver’s view that game worlds tend to work in a different way than other environments, as well as the respondents’ emphasis on the idea that it does not fit the nature of game worlds to ask about the origin of UI and other functional elements. I have argued elsewhere (Jørgensen, 2010, pp. 87–88) with basis in the same empirical material that game worlds for this reason must be seen as qualitatively different from other fictional worlds because they are environments carefully designed with gameplay in mind. Game worlds operate on other premises than other fictional worlds, and one of their characteristics is that they need an information system that supports gameplay. That said, it is important to notice that I still see game worlds as designed environments with content meant for the player’s imaginary involvement and that they must be understood as a particular kind of fictional worlds. What triggers the players’ involvement, however, is the particular gameplay as it exists in the particular representational game world (Salen & Zimmerman, 2004, p. 451).

The relationship between the game world and the UI is also complicated by incoherence. Juul argues that a game world is incoherent if it “contradicts itself or prevents the player from imagining a complete fictional world” (2005, p. 123). As an example, he points out that the practice of having more than one life is an example of an incoherent game world when there is nothing in the fiction that explains why people come back to life. When we cannot explain features like this by referring to the fiction, however, players instead explain it by referring to the rules of the game (Juul, 2005, p. 130). It is indeed true that players may fill in the gaps when confronted with incoherence, but incoherence does not seem to prevent the players from imagining the world. Instead incoherence is explained, if not as part of the game rules, then as shortcuts in the presentation of a world. Game worlds are described by the player respondents as stylized, simplified, abstract, or representational worlds that are not ruled by coherence, but by functionality. Fred explains:

We’re talking about a stylized world. Everything is represented by something else. (…) When you see a ring lying on the ground, it’s really huge compared to how a ring is [in reality]. They have magnified it a little so that you’re able to see it. (…) When you pick it up it, it jingles, it makes a little more noise than it would normally do, but it’s still part of that world. (Fred, 28 years)

For this reason, simplification and a lack of correspondence to the real world do not seem to ruin the gameplay experience or the understanding of the game world. What seems to clearly disturb them, however, is breaking the conventions and premises that have been set up for the game. Cases where the avatar expresses autonomy, either by starting to communicate to the player or by breaking the control link and starting to act on its own may in some genres create negative attention toward the communicative situation and therefore feel awkward. This seems to be associated with genre and perspective. In avatar-based games like Diablo 2, it may raise questions who the avatar is addressing when an avatar appears to speak to the player, although it is generally accepted as system information supporting intuitiveness, and a convention inspired by the rhetoric of fiction. In a first-person view game like Crysis, however, the player may feel alienated and powerless if the avatar suddenly starts acting as if there never was a control link in the first place. Eric explains how cinematic cut scenes in Crysis may ruin the sense of coherence if the game conventions are broken:

It’s very clear when the widescreen appears that I lose control into a cutscene. Suddenly there’s music and all. (…) [It] communicates a tailored experience of the events, but at the same time you are removed of control, you lose some of the involvement, suddenly you are powerless. (Eric, 26 years)

An interesting finding is that perspective is crucial to the players’ interpretation of the ontological status of the interface. In avatar-less games, including first-person perspective games, players do find it necessary to question whether any interface elements exist in the game world. In such games, there is no need for the players to distinguish between information addressed toward the player as opposed to that addressed toward the avatar. With reference to Command & Conquer 3, Carl explains:

The game makes you part of the army you are controlling; you are supposed to lead it. The troops are responding directly to you, they answer you, at the same time as you are tied into it through cutscenes, where you’re directly addressed as the person who is the commander. (Carl, 29 years)

Discussion

So why do the respondents accept elements and modes of address that appear intrusive to the game experience? It may be tempting to refer to the magic circle in explaining this and to claim that we accept all interface elements because we have put ourselves into the mind-set of the game—it is a lusory attitude (Salen & Zimmermann, 2004, p. 97; Suits, 1990). While it is not incorrect to explain this by referring to a specific lusory mind-set, the metaphor of the magic circle suggests clear and distinct internal/external perspective (Consalvo, 2009, pp. 414–415) that does not open for an understanding of the game boundary as flexible.

As we have seen, interface information may be located on different layers between the screen and the game world. One way of explaining this may be by referring to Goffman’s frames. Framing is related to how human beings interpret social situations, and how we distinguish different situations from each other based on the conventions for that situation (Goffman, 1974). This is on par with how Linderoth understands the gaming situation. He sees gameplay as a multifaceted and changing activity where frames may change over very small time periods (Linderoth, 2004, p. 232). This means that the player needs to keep a range of potential frameworks active at the same time and that frameworks often may blend together (Linderoth, 2004, p. 222). Viewed in this light, different frames of reference will be available for the player when interpreting a specific game feature. In this situation, it does not matter if information lies on the border between the game world and the interface as long as the player may view it through an available framework. Our ability to always move between the different informative frames of reference when playing computer games explains why we accept the UI, even when its spatial status is unclear.

In his discussion of self-referentiality in games, game designer Matthew Weise states that “the line between reality and fiction in games does not function as it does in traditional media” (2008). He believes that the dynamic between fantasy and reality is complicated by the player, and that one should see the boundary between player and fiction as “an elastic membrane (…) rather than a wall.” This perspective provides a more open interpretation of the relationship between the interface and the game world: There is no absolutism between different frames of information, as they are porous and permeable and may leak into each other. For this reason, features located on the border between the game world and the interface do not follow the rules of traditional fiction; instead, they follow the logics of games and their elastic membranes. These features are based on the above mentioned design principle from Team Fortress 2 (Andrews, 2010)—to utilize whatever means there are for the sake of clarity and functionality. This is a principle that works because people are used to negotiate different frames of reference continuously when playing games.

According to Murray, “once the illusory space is created, it has such a psychological presence that it can almost divorce itself from the means of representation” (1997, p. 104). This means that once we have accepted the premises of the game world and play along with these, we are willing to accept it regardless of its shortcomings. This explains why we accept game worlds that are fairly similar to traditional fictional elements, but still have striking qualities that separate them clearly from them. Klevjer follows this argument by emphasizing game worlds as a functional environment meant for playing games and states that game fictions are not delineated by a “fourth wall.” Instead, game fictions have extended the boundaries of the fiction to accept more than we would accept in other fictional worlds (Klevjer, 2007, p. 59). This is on par with how this article views the relationship between interface elements and the game world. As a game world, its borders are by definition flexible. At some times, it may make sense to include a UI element as part of the world, at other times it may not. And when there are features that cannot be clearly identified as being on the inside or the outside, these elements must be interpreted as indicative to how this specific game world functions.

Conclusions

We have seen that game interfaces are not disturbing for involvement and gameplay, regardless of whether they are implemented into the game world or situated as superimposed windows on the level of the screen. However, the discussions above do suggest that interfaces implemented into the game world are viewed as a more elegant approach to UI design and that the players have a preference for this when it is done consistently and as a natural part of the game world without losing any information. While players accept any UI that promotes gameplay and supports functionality in a consistent manner, they have different preferences according to genre, and are in particular positive toward elegant interfaces. In any case, the players do not contemplate the presence of any interface elements regardless of their integration into the game world. The players never question why interface elements are present as long as they are able to provide relevant information (Andrews, 2010; Schell, 2008, p. 227). These features are accepted because they are there for a functional purpose with respect to gameplay. Since the interface is what enables agency and embodiment in the game world, it seems to enhance the sense of involvement. These observations also partly explain the recent popularity of “natural” physical interfaces that focus on intuitive functionality through immediate recognition.

As demonstrated, interfaces that are implemented into the game world appear in a range of different formats, and a completely integrated or “invisible” interface is not always desirable. Often overlay interfaces may be optional for the sake of clarity, and for the same reason iconic interfaces may often be too vague, at least for novice players. Good game UI design maintains awareness of these layers and has a clear intention with how they decide to associate the UI with the game world. It is important to use interface features that fit the genre and game mechanics, and evaluate the priority of different pieces of information. Information that always needs to be available may benefit from an overlay interface, while information with lower priority or which is aimed toward more experienced players may benefit from being iconic.

According to Bolter and Grusin, interacting with new media is an oscillation between its content and the interface (2000, p. 33). They claim that the user moves back and forth between interactive and static content. This assumption suggests that there exist different incompatible spheres of logic. However, this is not on par with the argument of this article, which states that the interface and the game world coexist and operate together in a functional whole, and teaches the player about the game system and its mechanics. Juul claims that fiction in games “plays an important role in making the player understand the rules” (2005, p. 163), but I would like to make this claim more specific and state that the system and the world work in tandem in providing the player with a comprehension of both. The game world cannot be understood as separate from the game system; it is a part of the game system. This is also why and how game worlds must be seen as a special, if not a different, case compared to traditional fictional worlds. In this sense, this article shows that Juul’s argument that computer games must be seen as real rules in fictional worlds is challenged by the existence of the interface.

As emphasized, making the game interface elegant and appropriate to the environment is paramount to UI design, and aesthetics should be used to enhance the functionality and signal effect of game system information. Utilizing the different levels of world integration described above in order to emphasize and downplay different features is important, as the players do not find the switching between the different frames confusing or intrusive. As long as usability and functionality with respect to gameplay are in focus, the player does not contemplate the level of fidelity of independent features. Any degree of integration of the interface into the game world should be seen as an enhancement of usability; something that provides the player with more options for interaction in the game world, and not something that removes the sense of presence and spatiality.