Keywords

Introduction

This chapter uses Calleja’s [3] Player Involvement Model to organize an analysis of affect in games. The Player Involvement Model dissects player involvement with games, identifying six main dimensions of involvement (i.e. kinesthetic, spatial, shared, narrative, ludic and affective involvement) plotted on two temporal phases: the immediate moment of game-play (i.e. micro-involvement) and off-line engagement with games (i.e. macro-involvement). Since the model was developed through qualitative research, we decided that the next step in its evolution would be to develop it further by quantitatively testing it in an experimental set-up. Consequently, we conducted three experiments in order to elaborately examine how different components of digital gameplay (i.e. game story, social setting, game control) can affect players’ involvement on the micro level, and how the different dimensions of player involvement relate to each other. As this collection focuses on emotion and affect in games, the current chapter especially investigates the relationship between affective involvement and the other five dimensions of the Player Involvement Model.

In order to arrive at the investigation of affective involvement in combination with the other dimensions, we will first give a brief description of the different layers of player experience. Next, we will give an overview of the Player Involvement Model, outlining its six dimensions and two temporal phases. In the empirical portion of the chapter, we will then investigate how varying a specific component of a digital game (e.g. its story) affects players’ involvement on all of the dimensions, and how these dimensions relate to and combine with each other.

The Bottom-Up Experience Triangle

One of the challenges in discussing player experience, at least within Game Studies, is the lack of differentiating between different forms of engagement with the game [3]. Authors often use terms such as engagement, involvement, attention, absorption, and sometimes even immersion, interchangeably (e.g. [2, 6, 7, 10, 15]). This makes it challenging to know what aspect of experience each author is actually referring to. In order to clarify this issue, we start from a layered model of player experience, adopted from the bottom-up experience model in cognitive psychology (see Fig. 3.1).

Fig. 3.1
figure 1

The bottom-up experience model of player experience

Full size image

Beginning from the bottom and working our way up, each layer of the triangle acts as a pre-requisite for those that follow. We cannot, for example, experience involvement in a game without first paying attention to it. The first and most basic layer is wakefulness. Wakefulness is the basic state of being conscious. This is a biological, not a cognitive function. If we are unconscious, we obviously cannot be playing. Next, we have attention. Attention is the willed or automatic direction of our awareness to certain stimuli in the environment. Whereas attention deals solely with cognitive functions, involvement deals with the nature and quality of the thing we are directing our attention to. Involvement considers the emotional spectrum and thus the whole, embodied experience. Incorporation [3] refers to the experience of inhabiting the game environment, sometimes referred to as “presence” or “immersion”. This phenomenon occurs as a result of the blending and internalization of involvement and is thus the most elusive of the layers as it is experienced at a subconscious level.

This chapter will deal primarily with the involvement layer of the experience triangle. While attention will also play a part in our discussion of affect, its utility is primarily in describing the structure of the cognitive resources that we have at our disposal and how we direct these resources during game-play. It is involvement that describes the quality of the game-playing experience and thus its affective dimension. The rest of the chapter will therefore describe the role of affect within player involvement in greater detail. First of all, however, we will give a brief overview of involvement as a whole.

The Player Involvement Model

In his book-length treatment of player involvement, Calleja [3] uses his model as a foundation upon which to build further investigations into player experience. His Player Involvement Model identifies six dimensions of involvement in digital games (i.e. kinesthetic, spatial, shared, narrative, ludic and affective involvement), each considered relative to two temporal phases (i.e. the macro-involvement phase and the micro-involvement phase) (see Fig. 3.2).

Fig. 3.2
figure 2

The player involvement model

Full size image

The six dimensions concern involvement related to (1) control and movement in the game environment (kinesthetic involvement), (2) the exploration, navigation and learning of the game’s spatial domain (spatial involvement), (3) players’ awareness of and interaction with other agents and/or players in the game environment (shared involvement), (4) story elements that have been written into a game, or those that emerge from the player’s interaction with the game (narrative involvement), (5) the various rules, goals and choices provided by a game (ludic involvement) and (6) the emotions that are generated during gameplay (affective involvement) [3].

As mentioned earlier, these dimensions can be considered relative to two temporal phases: the macro-involvement phase and the micro-involvement phase. The macro phase of the model deals with the off-line, long-term involvement of players with a game; their desire to engage with a specific game, the desire to return to it afterwards, and all off-line thinking that occurs in between game-playing sessions [3]. While the macro phase does constitute a crucial part of the player experience, we will limit ourselves here to analyzing the micro-involvement phase due to the limited scope of this chapter. Micro-involvement, then, describes players’ moment-to-moment (and thus immediate) engagement while playing a game. This temporal phase thus deals with the imminent quality of involvement during game-play [3]. As we discussed earlier, a pre-requisite to experiencing micro-involvement is the direction of attention to the game.

These phases tend to be experienced in a combinatory manner and should thus be seen as layered and transparent in nature. This means that one phase influences how another is experienced and interacted with. The dimensions of the Player Involvement Model similarly combine in experience, with the inclusion or exclusion of a dimension affecting how others are experienced [3]. The combinatorial aspect of the model makes it particularly useful for developing a structural approach to understanding game affect.

For instance, with Fig. 3.3 as a visual aid, we can think of the maximum attentional resources players have at their disposal as the outer line of each dimension’s triangle. If players are dedicating all of their attention to a single dimension, then they will not be able to combine dimensions. If all of the attentional resources of the players are directed towards learning a game’s controls, for example, the only dimension the players will interact with is kinesthetic involvement. As the controls are learnt, less attentional resources are demanded by kinesthetic involvement, allowing the players to pay attention to, for example, the actions of other members on their team (i.e. combining kinesthetic involvement with shared involvement), the story of the game (i.e. combining kinesthetic involvement with narrative involvement), etcetera.

Fig. 3.3
figure 3

Internalisation of involvement dimensions

Full size image

The rest of the chapter will put into action this combinatorial nature of the Player Involvement Model [3], especially focusing on combinations with affective involvement in order to give a robust account of the forms of emotional types that games can elicit in our moment-to-moment engagement with them. Combining the affective involvement dimension with the other involvement dimensions will thus yield an experiential framework of game emotions that is unique to the gaming situation.

A Quantitative Perspective

As we mentioned earlier, Calleja’s [3] Player Involvement Model was developed through extensive qualitative research. Building on this theoretical framework, a necessary subsequent step is to test the model in an experimental context, in order to validate, disprove or complement the model in a quantitative set-up. This will provide a new foundation for future research studying the player experience. Therefore, the aim of the current study was to test the micro phase of Calleja’s [3] Player Involvement Model experimentally, investigating how the different dimensions of player involvement relate to, react and combine with each other. In light of the current chapter, we pay specific attention to affective involvement.

In order to be able to quantitatively examine the combinatorial nature of the Player Involvement Model, we conducted three experiments. The aim of these experiments was to test how varying one or two dimension(s) of player involvement – by manipulating a specific component of a digital game (e.g. game story, social setting, game control) – affects the other dimensions, and to what degree the different dimensions of player involvement are related to affective involvement.

Experimental Design

Experiment 1: Game Story

In our first experimental study, we focused primarily on narrative involvement (i.e. the player’s involvement with the story elements that have been written into a game, and those that emerge from the player’s interaction with the game) and ludic involvement (i.e. the dimension of player involvement concerned with the various rules, decision-making processes, goals and rewards of the game). These two dimensions are strongly connected: game stories do not only provide the player with background information, they outline the context of the game and often point out the player’s objectives and the rewards he will get upon completing them. In turn, the goals of a game do not only help players find their position in the game, but also guide them in understanding the importance of their actions and choices within the narrative setting [3].

In order to vary these dimensions of player involvement, we manipulated the story of the game used in the experiment as a between-subjects factor, resulting in two experimental conditions. People in one experimental condition played a game level that had an elaborate story and in which the player had to perform an emotionally engaging task (i.e. save children from vicious raiders; elaborate story condition), while the people in the other experimental condition played a game level that had a minimal story and in which the player only had to perform a simple, unemotional task (i.e. collecting objects; minimal story condition).

We used the PC version of the action role-playing game Fallout: New Vegas (Bethesda Softworks, 2010) for this purpose. Using the game’s editor, we created an experimental level that suited the purpose of the study. Only the game’s story was manipulated between the experimental conditions, meaning that apart from the implementation of the game’s narrative, participants in the different conditions played the exact same game level.

Sixty-two experienced gamers (57 male, 5 female) between 18 and 37 years old (M = 22.32, SD = 3.21) eventually participated in the experiment (i.e. 31 people per experimental condition). During the experiment, participants first played a tutorial level explaining the basics of the game, after which the actual experimental level would start. The experimental level took approximately 12 min to finish.

Experiment 2: Social Setting

In our second experiment, we addressed shared involvement (i.e. player involvement related to the awareness of and interaction with other agents or players in a game environment).

In order to vary this dimension of player involvement, we manipulated the social context in which participants played a digital game as a between-subjects factor, leading to four experimental conditions. People either played the experimental game alone (single-player condition); together with one other person (multiplayer condition); or alone while another person watched (public play condition). Because this last condition consists of both players and observers, it is further divided into two groups (public play: player and public play: observer groups).

For this purpose, the PlayStation 3 puzzle-platformer game LittleBigPlanet 2 (Sony Computer Entertainment, 2010) was used. By using the editor of the game, we created our own experimental level. Only the social setting in which the game was played differed between conditions, meaning all participants played the exact same level.

One hundred twenty-one gamers (82 male, 39 female), 18–24 years of age (M = 20.69, SD = 1.79) participated in the experiment. The single-player condition included 31 participants, while the three other conditions all contained 30 participants. During the experiment, participants first played a tutorial level that explained the basics of the game, after which the experimental level started. The experimental level had an average play time of 8 min.

Experiment 3: Game Controller

For the third experimental study we looked at kinesthetic involvement (i.e. player involvement related to control and movement in the game environment) and spatial involvement (i.e. player involvement with the exploration, navigation and learning of the game’s spatial domain). Again, these two dimensions are strongly intertwined: players have to control a game in order for their character(s) on-screen to be capable of movement and actually navigate through the game space. Until players learn to move in the world, they cannot engage with its spatial dimensions [3].

In order to vary these dimensions, we performed a within-subjects experiment in which we manipulated the game controller that was used in two experimental conditions. This means that participants played the same game twice, once with a PlayStation 3 gamepad controller (i.e. traditional controller), and once with a PlayStation Move racing wheel controller (i.e. motion-based controller).

The PlayStation 3 kart-racing game LittleBigPlanet Karting (Sony Computer Entertainment, 2012) was employed in the experiment. By using the editor of the game, we created an experimental level for use in the study. Only the game controller that was used differed between conditions, meaning participants played the exact same game level twice, each time with a different controller.

Thirty-one gamers (24 male, 7 female), 18–30 years of age (M = 22.61, SD = 2.99) participated in the within-subjects experiment. During the experiment, participants first played the official tutorial level of the game, after which the experimental level started. This experimental level had an average play time of 6 min.

Measures

After participants finished playing the games in the experiments, they were asked to fill out a questionnaire asking them how involved (on all dimensions) they felt during gameplay.

Narrative involvement was measured by a combination of narrative engagement scales from Busselle and Bilandzic [1], de Graaf et al. [5] and Green and Brock [8], adapted for use in a game context where necessary. The scale includes statements regarding the player’s focus on and interest in the story (e.g. “My attention was fully captured by the story”, “I was interested in the game’s story”; 6 items, Cronbach’s α = 0.85). Agreement with these statements was measured on 5-point intensity scales ranging from 0 (“not at all”) to 4 (“extremely”).

The ludic involvement scale was based on a combination of items from IJsselsteijn et al.’ [9] Game Experience Questionnaire and de Graaf et al.’s [5] narrative engagement scale. It includes statements assessing the player’s interest in and focus on the game’s mission and goals (e.g. “I was fully concentrated on reaching the goals of the mission”, “I was interested in the mission of the game”; five items, Cronbach’s α = 0.68). Agreement was measured on 5-point intensity scales ranging from 0 (“not at all”) to 4 (“extremely”).

Shared involvement was measured by making use of the Social Presence module of IJsselsteijn et al.’s [9] Game Experience Questionnaire. The scale includes statements regarding empathy with the other and behavioral involvement (e.g. “I felt connected to the other”, “My actions depended on the other’s actions”; 14 items, Cronbach’s α = 0.83) to which agreement is measured on a 5-point intensity scale ranging from 0 (“not at all”) to 4 (“extremely”).

Kinesthetic involvement was measured by making use of a combination of items from previously validated (kinesthetic) involvement questionnaires such as Witmer and Singer’s [17] Presence Questionnaire, Jennett et al.’s [10] Immersion Scale and Vanden Abeele’s [16] Perceived Control Scales. The utilized scale includes statements regarding the player’s perceived amount of control over the actions and movements of the avatar in-game and the difficulty of the game controls (e.g. “I felt that I was able to control what happened in the game environment”, “The game controls were easy to learn”; nine items, Cronbach’s α = 0.94). Agreement with these items was measured on a 5-point intensity scale ranging from 0 (“not at all”) to 4 (“extremely”).

Spatial involvement was measured by making use of items from IJsselsteijn et al.’s [9] Game Experience Questionnaire, in combination with items based on descriptions of the dimension given by Calleja [3]. The scale includes statements regarding the exploration of the game level and the navigation through the game world (e.g. “I felt that I could explore and discover things”, “The navigation through the game level was difficult”; five items, Cronbach’s α = 0.67) to which agreement is measured on a 5-point intensity scale ranging from 0 (“not at all”) to 4 (“extremely”).

Affective involvement was measured by taking into account both general emotions (i.e. largely uncontrollable and spontaneous emotional reactions that are continuously present to some degree), and more specific player experiences.

When people play digital games, they experience several general emotions, such as pleasure, arousal and dominance. Pleasure refers to the pleasantness or enjoyment of a certain experience [14], arousal gives an indication of the level of physical and mental activation associated it [14], and dominance concerns the feeling of control and influence over others and surroundings [11]. The general emotions participants felt while playing the game were measured by using Lang’s [12] Self-Assessment Manikin. This scale uses 9-point visual scales ranging from 0 to 8, with ascending scores corresponding to higher pleasure, arousal and dominance ratings.

Apart from general emotions, digital games have the potential to evoke a wealth of specific player experiences, such as challenge, competence, tension, positive affect and negative affect [10, 13]. These player experiences were also measured using IJsselsteijn et al.’s [9] Game Experience Questionnaire. The Game Experience Questionnaire includes statements to which agreement is measured on a 5-point intensity scale ranging from 0 (“not at all”) to 4 (“extremely”). Challenge measures the stimulation players perceive and the amount of effort they have to put into the game (e.g. “I felt challenged”, “I felt stimulated”; five items, Cronbach’s α = 0.64–0.82). Competence refers to how successful and skilful people feel while playing a game (e.g. “I felt successful”, “I felt skilful”; five items, Cronbach’s α = 0.87–0.96). Tension measures the degree to which players feel frustrated and annoyed (e.g. “I felt frustrated”, “I felt irritable”; three items, Cronbach’s α = 0.66–0.83). Positive affect probes players’ fun and enjoyment of the game (e.g. “I felt good”, “I enjoyed it”; five items, Cronbach’s α = 0.79–0.88). Negative affect is concerned with the degree to which players are feeling bored and distracted (e.g. “I felt bored”, “I found it tiresome”; four items, Cronbach’s α = 0.64–0.74).

Results

Impact of Manipulations on Player Involvement

To be able to analyze the impact of several components of digital gameplay on the dimensions of player involvement, as well as the relationships that exist between these dimensions, we conducted three experiments.

In what follows, we will first examine how varying one or two dimension(s) of player involvement (i.e. by manipulating a game’s story, social setting or game controller) influences the other dimensions of involvement as well, with a particular focus on affective involvement.

Experiment 1: Game Story

Our first experiment manipulated the story of the game participants had to play as a between-subjects factor, resulting in two experimental conditions: an elaborate story condition and a minimal story condition. In order to test the impact of this manipulation on player involvement in its narrative, ludic and affective dimensions, we performed one-way analyses of variance (ANOVAs).

Results of these analyses show that there are indeed significant differences in narrative involvement between our story conditions (F(1, 60) = 5.13, p = 0.03). Playing the game with the elaborate story led to significantly greater interest in and focus on the story (M = 2.12, SD = 0.97) compared to playing the game with the minimal story (M = 1.63, SD = 0.71).

However, contrary to our expectations, our results do not show significant differences in ludic involvement (F(1, 60) = 0.04, p = 0.84) or affective involvement (pleasure: F(1, 60) = 2.59, p = 0.11; arousal: F(1, 60) = 0.02, p = 0.88; dominance: F(1, 60) = 0.12, p = 0.73; competence: F(1, 60) = 0.12, p = 0.73; tension: F(1, 60) = 0.14, p = 0.71; challenge: F(1, 60) = 0.04, p = 0.84; negative affect: F(1, 60) = 0.01, p = 0.93; positive affect: F(1, 60) = 2.10, p = 0.15) between conditions. Players were not more or less absorbed in the ludic and affective dimensions of the game when playing with the elaborate versus minimal story.

We believe that, for this study, our manipulation of game story may not have been extensive enough to lead to differences in other dimensions between experimental conditions. An alternative explanation may be that some players are simply more focused on the narrative of the game than others. In a game, it is often up to the players themselves to decide whether they want to engage with the entirety of the story or simply focus on the goal-oriented tasks that push the game forward. It may be that in the current experimental setting, the narrative disposition of each of the players had a more important part to play in affecting player involvement than the between-subjects manipulation of the game story. In the section on Combining affective involvement (see further), we will take a look at potential relationships between narrative, ludic and affective involvement across conditions.

Experiment 2: Social Setting

The second experiment manipulated the social setting in which participants played a game as a between-subjects factor, resulting in four experimental conditions: a single-player condition, a multiplayer condition, a public play: player condition and a public play: observer condition). To be able to analyze the impact of this manipulation on player involvement, we again conducted one-way ANOVAs.

The results of these analyses demonstrate that there are significant variations in shared involvement between conditions (F(2, 86) = 11.43, p < 0.001), with people in the multiplayer group experiencing the highest sense of shared involvement (M = 2.51, SD = 0.53), followed by people in the public play: player (M = 2.16, SD = 0.57) and public play: observer groups (M = 1.82, SD = 0.55). This seems logical, since the multiplayer setting can be seen as the most social condition, with players having to actively interact with each other and synchronize their behavior in order to be able to work together. The public play: player setting can be considered as less social; although participants in this group played the game in a social setting (i.e. while being watched), the actual gameplay was experienced solo. However, this setting is still more susceptible to social influence compared to the public play: observer condition, since participants in the latter group did not have to actually perform the task (i.e. controlling the game).Footnote 1

Moreover, our results demonstrate that the variations in social setting had a significant impact on affective involvement, especially affecting arousal (F(3, 117) = 3.90, p = 0.01), dominance (F(3, 117) = 6.83, p < 0.001) and negative affect (F(3, 117) = 3.61, p = 0.02). Participants in the multiplayer context felt more aroused (M = 4.53, SD = 1.59) than participants in public play: player (M = 3.83, SD = 1.42), public play: observer (M = 3.73, SD = 1.46) and single-player context (M = 3.16, SD = 1.79). However, people in the single-player condition reported higher levels of dominance (M = 5.26, SD = 1.41) than people in the public play: player (M = 4.63, SD = 1.30), multiplayer (M = 4.27, SD = 1.39) and public play: observer conditions (M = 3.73, SD = 1.31). Moreover, negative affect, although low in all conditions, was highest in the public play: observer condition (M = 0.33, SD = 0.41), differing significantly from the experienced negative affect in the other conditions (single-player: M = 0.14, SD = 0.33; multiplayer: M = 0.09, SD = 0.28; public play: player: M = 0.12, SD = 0.22).

A possible explanation for these results may be that players in the social play conditions (and especially the multiplayer condition) were more excited to be playing with/in the presence of another person, but did not have enough time to balance their individual playing and/or communication styles during the short period in which the experiment took place, resulting in a clumsy collaboration and the players feeling less dominant. Moreover, observers experienced less dominance than players, and more negative affect. This result can be attributed to the fact that observers are not in control of what happens in the game and therefore experience emotions that are less intense (i.e. dominance) on the one hand, and more boredom (i.e. negative affect) on the other.

Experiment 3: Game Controller

The third and final experiment manipulated the type of game controller that was used to play a game as a within-subjects factor, resulting in two experimental conditions: a traditional game controller condition and a motion-based game controller condition. In order to analyze the impact of this manipulation on player involvement, we performed repeated measures ANOVAs.

Our results show that manipulating the game controller has a significant impact on kinesthetic involvement (F(1, 30) = 81.78, p < 0.001): the controls of the traditional controller were perceived as easier to learn and handle (M = 3.04, SD = 0.76) than those of the motion-based controller (M = 1.55, SD = 0.70), allowing the players more precise control over their movements and actions in the game world.

Further, the findings also reveal a significant impact of the manipulation of game controller on spatial involvement (F(1, 30) = 27.48, p < 0.001); when playing the game with the traditional controller, the exploration and navigation of the game’s spatial domain was perceived to be more easy and uncomplicated (M = 2.87, SD = 0.51) compared to playing with the motion-based controller (M = 2.21, SD = 0.76).

Finally, the game controller that was used to play the game was shown to significantly influence affective involvement. First of all, the manipulation of game controller significantly affected the players’ sense of dominance (F(1, 30) = 34.70, p < 0.001): playing with the traditional game controller resulted in people feeling more dominant and in control (M = 5.58, SD = 1.54) compared to playing with the motion-based game controller (M = 3.45, SD = 1.43). Moreover, results show a significant effect on competence (F(1, 30) = 71.02, p < 0.001), tension (F(1, 30) = 25.49, p < 0.001), challenge (F(1, 30) = 65.48, p < 0.001) and positive affect (F(1, 30) = 14.86, p = 0.001). Playing with the traditional controller resulted in more competence (M Traditional = 2.90, SD = 0.82; M Motion-based = 1.47, SD = 1.02) and positive affect (M Traditional = 2.81, SD = 0.60; M Motion-based = 2.34, SD = 0.85), and less tension (M Traditional = 0.53, SD = 0.59; M Motion-based = 1.33, SD = 1.03) and challenge (M Traditional = 1.01, SD = 0.72; M Motion-based = 2.17, SD = 0.75).

Combining Affective Involvement

The previous section demonstrates that manipulating a specific component of digital gameplay (i.e. game story, social setting, game control) has a significant impact on player involvement, not only affecting the obviously linked dimensions (e.g. social setting was expected to influence shared involvement), but in two out of three cases also reflecting on affective involvement.

In the current section, we look at the relationships between affective involvement and the other dimensions of player involvement across conditions, in order to get a more detailed look at the interrelatedness of the six dimensions. The first experiment provides us with data regarding relations between affective, narrative and ludic involvement; the second experiment reveals relations between affective and shared involvement; and the third experiment exposes relations between affective, kinesthetic and spatial involvement.

Experiment 1: Game Story

In the first experiment, we found that manipulating the story of a game (i.e. elaborate story versus minimal story) led to significant differences in narrative involvement between conditions. Against our expectations, however, ludic and affective involvement were not significantly influenced. By making use of correlation analyses, we now take into account relationships between the respective dimensions of involvement across conditions.

Narrative Involvement and Affective Involvement

The results of these correlation analyses reveal significant relationships between narrative involvement and several components of affective involvement. First, we observe moderate to strong positive relationships between narrative involvement and pleasure (Pearson’s r = 0.32, p = 0.01), arousal (Pearson’s r = 0.31, p = 0.02), competence (Pearson’s r = 0.43, p < 0.001), positive affect (Pearson’s r = 0.45, p < 0.001) and challenge (Pearson’s r = 0.47, p < 0.001). Moreover, narrative involvement is shown to be moderately and negatively related to negative affect (Pearson’s r = −0.33, p = 0.01).

Ludic Involvement and Affective Involvement

Further, correlation analyses also show significant relationships between ludic involvement and affective involvement, with ludic involvement being moderately and positively related to feelings of competence (Pearson’s r = 0.38, p = 0.003) and positive affect (Pearson’s r = 0.30, p = 0.02).

Experiment 2: Social Setting

The results of the second experiment showed us that varying the social setting in which a digital game is played (i.e. single-player setting, multiplayer setting, public play: player setting, public play: observer setting) can significantly affect shared involvement, as well as lead to differences in player affect. Correlation analyses further provide us with a greater insight into the relationships between the two dimensions.

Shared Involvement and Affective Involvement

The results of correlation analyses demonstrate that shared involvement is significantly related to various aspects of affective involvement. More specifically, we record moderate to strong positive relationships between shared involvement and pleasure (Pearson’s r = 0.38, p < 0.001), arousal (Pearson’s r = 0.30, p = 0.01) and positive affect (Pearson’s r = 0.48, p < 0.001), while moderate negative relationships are observed between shared involvement and dominance (Pearson’s r = −0.23, p = 0.03) and negative affect (Pearson’s r = −0.33, p = 0.002).

Experiment 3: Game Controller

Finally, the results of the third experiment demonstrated that manipulating the type of game controller that is used to play a game (i.e. the traditional PlayStation 3 gamepad controller versus the motion-based PlayStation Move racing wheel controller) has a significant impact on kinesthetic, spatial and affective involvement. Correlation analyses give us a more detailed look concerning the relationships between these dimensions.

Kinesthetic Involvement and Affective Involvement

The findings of these correlation analyses show that there are significant relationships between kinesthetic involvement and a variety of components of affective involvement. Kinesthetic involvement seems to be strongly and positively associated with pleasure (Pearson’s r = 0.52, p = 0.003), dominance (Pearson’s r = 0.54, p = 0.002), competence (Pearson’s r = 0.78, p < 0.001) and positive affect (Pearson’s r = 0.46, p = 0.01). Moreover, strong negative relationships are registered regarding kinesthetic involvement and arousal (Pearson’s r = −0.62, p < 0.001), tension (Pearson’s r = −0.45, p = 0.01) and challenge (Pearson’s r = −0.57, p = 0.001).

Spatial Involvement and Affective Involvement

Finally, the correlation analyses reveal significant relationships between spatial involvement and affective involvement. Spatial involvement is strongly and positively related to pleasure (Pearson’s r = 0.66, p < 0.001), competence (Pearson’s r = 0.52, p = 0.003) and positive affect (Pearson’s r = 0.61, p < 0.001); and strongly and negatively related to negative affect (Pearson’s r = −0.45, p = 0.01) and tension (Pearson’s r = −0.49, p = 0.01).

Conclusion

The current chapter takes as its main object of study the Player Involvement Model, an analytical framework designed to understand player experience developed through qualitative research by Calleja [3]. The Player Involvement Model identifies six dimensions of involvement in digital games, namely kinesthetic, spatial, shared, narrative, ludic and affective involvement. As the current collection focuses on emotions in games, it was our goal to elaborately investigate affective involvement and its relationship with the other dimensions of player involvement in a quantitative context.

In order to do this, three experimental studies were set up. Each of these experiments aimed to vary one or two dimension(s) of player involvement by manipulating a specific component of a digital game participants had to play. In the first experiment, we manipulated the story of a game in order to study narrative and ludic involvement. The second experiment manipulated the social setting in which a game was played to influence shared involvement. Finally, the third experiment manipulated the game controller that was used to play a game in order to affect kinesthetic and spatial involvement.

The results of these experiments show that the manipulations have a significant impact on player involvement, not only affecting the intended dimensions, but in two out of three cases also the emotions that participants experienced during gameplay (i.e. affective involvement). Subsequently, we further investigated the combinations of the dimensions of player involvement with affective involvement across conditions. Our findings suggest that players’ affective involvement can be influenced by their interaction with each of the other dimensions of involvement.

In the context of the first experiment, several components of affective involvement are significantly related to both narrative and ludic involvement.

When players were more focused on and involved in the story of the game (i.e. narrative involvement), they were more aroused, they perceived the game to be more challenging and felt more competent. Moreover, they experienced more positively-valenced emotions (i.e. pleasure and enjoyment) and less negatively-valenced emotions (i.e. boredom). These results support previous findings by Calleja [4], who similarly describes the relationship between narrative and affective involvement. It is worth noting that this relationship works both ways: affective involvement makes narrative moments more memorable and significant, whether these moments arise from the story pre-scripted by the game’s designers or generated through interaction with the game [3, 4]; while simultaneously engaging narrative creates a context of meaning that allows for more positive emotions to be experienced.

Furthermore, greater involvement with the goals and rewards of the game (i.e. ludic involvement) was associated with players feeling more competent and skillful during gameplay, as well as experiencing more positively-valenced emotions (i.e. enjoyment). Again, this is in line with Calleja’s [3] qualitative research where participants commented on the sense of satisfaction they experienced from attaining game goals and reaping the related rewards.

The results of the second experiment demonstrate that shared involvement is also related to affective involvement. When players experienced more involvement because of a higher awareness of, and more interactions with other players (i.e. shared involvement) in the study, they were more aroused and experienced more positively-valenced emotions (i.e. pleasure and enjoyment) on the one hand, while also feeling less dominant and experiencing less negatively-valenced emotions (i.e. boredom) on the other. The presence of these negative emotions associated with shared involvement is also highlighted by Calleja’s [3] research. Here, participants reported how collaboration with others can be intensely exciting and satisfying, if the members in the relevant team(s) manage to work together well, or very frustrating if things go wrong. For teams of players to collaborate properly they would have needed to play together for a considerable amount of time. In our experiment, players only had a short time to play the game and synchronize their individual playing and communication styles; it is thus understandable that collaboration in our experiment led to both negative and positive emotions related to shared involvement.

Finally, the third experiment shows that the dimensions of kinesthetic and spatial involvement are significantly associated with affective involvement as well.

When players experienced more involvement because of a higher sense of control over the actions and movements of their game character (i.e. kinesthetic involvement), they perceived themselves to be more dominant and in control, more competent and skillful in general, and they experienced more positively-valenced emotions (i.e. pleasure and enjoyment). Further, they also perceived the game to be less challenging, and experienced less arousal and negatively-valenced emotions (i.e. frustration). Participants from Calleja’s [3] studies previously linked these positively-valenced emotions in reaction to mastering and internalizing the game controls with a strengthening of the link between player and avatar, which yielded a stronger sense of virtual environment habitation or incorporation [3], an experience that was described in positive terms by his participants.

Lastly, we found that higher involvement with the exploration, navigation and learning of the game’s spatial domain (i.e. spatial involvement) was associated with players feeling more competent, more positively-valenced emotions (i.e. pleasure and enjoyment), and less negatively-valenced emotions (i.e. frustration and boredom). This relationship between spatial and affective involvement is reminiscent of Calleja’s [3] argument that the internalization of spatial involvement turns abstract virtual space into familiar place. Participants in Calleja’s research commented on the positive emotions that stem from this transition:

I am a pretty visual person. I drive by landmarks not by road names, so, I can visualize it pretty vividly. Though some of the in-between spaces are hazy cause I fly from Undercity to Tarren Mill now. It doesn’t matter if it’s the real world or digital, as I travel around, and learn new areas, I naturally seek certain kinds of landmarks to help me keep my bearings. A twist in the road here, a tree there. I find it comforting when I start to get the lay of the land in an MMO. For others, I have no idea. It’s not just comfort, but also a bit of pride. I know where I am, I know where I want to go, and I know how to get there. (Rheric, quoted in [3])

Based on these findings, we can provide initial support for Calleja’s [3] Player Involvement Model in a quantitative setting. Our experiments show the existence of the six dimensions of player involvement, and demonstrate that a manipulation of certain components of digital gameplay can result in significant effects on these dimensions. Moreover, our results reveal that the dimensions of player involvement are indeed interrelated, and that affective involvement can react to and combine with each of the other dimensions in different gaming situations.

Nevertheless, further research is still essential in order to come to a better understanding of players’ involvement in digital games. Our studies have examined involvement in games such as action role-playing game Fallout: New Vegas (Bethesda Softworks, 2010), puzzle-platformer LittleBigPlanet 2 (Sony Computer Entertainment, 2010) and kart-racing game LittleBigPlanet Karting (Sony Computer Entertainment, 2012), in a variety of digital game situations. However, a lot of different game genres, platforms, controllers, social modes,… exist, all possibly affecting the player in different ways, both cognitively, affectively and conatively. It is therefore of high importance that future studies continue to investigate player involvement in a multitude of digital games and gaming situations.

Finally, it is important to note that there is not yet a reliable, multidimensional scale of player involvement available for use in quantitative research. The experiments described in this chapter made use of a combination of previously validated (player) experience scales, adapted for use in the current studies. Although this suited the purpose of our initial and explorative studies, the development of a reliable, valid, sensitive and multidimensional measure based on Calleja’s [3] Player Involvement Model would be very valuable for game researchers studying player experience. In that regard, following Yannakakis and Martinez [18], a rank-based questionnaire instead of a rating-based questionnaire to tap into player involvement could be considered. This would overcome some serious shortcomings (e.g. ordinal values treated as numerical, non-linearity of the scale) inherent to the use and analysis of the self-report items in the current studies.