Perceptual realism and immersion in video games : a study on the effects of perceptual realism in virtual environments on player immersion in video games

(1)

Perceptual realism

and immersion in

video games

A study on the effects of perceptual realism in virtual environments on player immersion in video games

— Anne van der Zanden Student nr.: 10305769

Universiteit of Amsterdam

Graduation Project: Entertainment Communication Supervisor: Monique Timmers

(2)

CONTENTS

Abstract 2

Introduction 3

Conceptual framework 5

Immersion 5

Realism in video games 5

Perceptual realism in video games 6

Perceived realism of graphics 7

Perceived realism of sound 8

Method 9

Participants 9

Stimulus 9

Procedure 10

Manipulation check: Perceived realism 10

Dependent variables: Immersion 11

Control variables 11

Results 12

Manipulation checks 12

Immersion in the “high-” and “low quality” condition 13

Regression models 13

Control variables 15

Conclusion 17

Limitations and further research 18

References 19

(3)

2

ABSTRACT

Realism in virtual environments has been linked to immersion in video games by several scholars. In this study, I focus on how players of video games perceive realism in a video game setting and how this affects immersion. Because presence is often considered as a dimension of immersion literature on the effects of realism on presence have also been taken into account. Two conditions of a video game were played by two groups of participants and a questionnaire on perceived realism and immersion was given. The results show that the condition with high quality graphics and sound are more immersive than the condition with low quality graphics and sound. However, participants in the condition with high realism did not perceive the virtual environment as a whole as more realistic than participants in the condition with low realism. Only the graphics were perceived as more realistic in the high realism condition. The main effects of perceived realism was only significant for overall perceived realism: Participants who perceived the virtual environment as more realistic overall felt more immersed in the virtual environment of the video game.

(4)

3 INTRODUCTION

Enjoyment is seen as one of the core motivations for people to choose certain types of media entertainment. Therefore, the enjoyment of video games should evidently be one of the main goals to aim for when designing games. Scholars studying player enjoyment have identified several elements in game design that lead to enjoyment of video games. Sweetser and Wyeth (2005) developed a model based on flow theory as used in research on the enjoyment of media entertainment, where flow is described as a combination of elements that cause “a sense of deep enjoyment so rewarding that people feel that expending a great deal of energy is worthwhile simply to be able to feel it” (Sweetser & Wyeth, 2005). The GameFlow model distinguishes a total of eight elements that cause the enjoyment of video games specifically, to wit: concentration, challenge, skill, control, clear goals, feedback, immersion and social interaction. The element of concentration is dependent of individual player characteristics – the ability of the player to concentrate on the game – while all other elements can be achieved by various approaches in game design. Sweetser and Wyeth (2005) suggest several criteria to achieve enjoyment. For the first element, a game should provide a lot of stimuli that require a high amount of concentration from the player. Challenges in the game should match the player’s skill level and the level of challenge should increase as the player progresses through the game. In effect, player skill development should be supported by the game. Furthermore, players should feel a sense of control over their actions in the game, which is achieved by control over the game’s interface and input devices. Games should provide clear goals and players must receive appropriate feedback at appropriate times. To approach immersion, game designers should create an environment in which players feel a deep but effortless

involvement in the game. Finally, Sweetser and Wyeth (2005) add that games should support and create opportunities for social interaction, both inside and outside the game.

Because of technical limitations, time constraints and other factors, it is not always possible to effectively implement all elements and criteria to achieve player enjoyment. It is therefore of importance to study what specific approaches in game design affect elements of the

GameFlow model. In this thesis, I will focus on the element of immersion and more specifically how perceptual realism in the game’s virtual environment affects player

immersion. Video games that are perceived as more realistic give an accurate representation of the world around us and are often experienced as more immersive in popular media. This is

(5)

4

argument is backed by several scholars (Cheng & Cairns, 2005; Nichols, Haldane & Wilson, 2008), but more research is needed. In this study, I will try to measure the effect of perceptual realism in games on immersion by comparing an original video game with a remastered version of the same game, which is said to be more realistic, while keeping all other elements of game design consistent.

(6)

5

CONCEPTUAL FRAMEWORK

Immersion

Immersion can be described as the degree of involvement that players have while playing a game (Brown & Cairns, 2004). Brown and Cairns (2004) state that immersion is not a static experience, but that players experience different levels of immersion when playing video games. Total immersion can be difficult to achieve, but designers can make an effort to overcome barriers that Brown and Cairns (2004) identified for each level of immersion. Removing these barriers allows for immersion to occur, however, it is stressed that it does not necessarily guarantee it. The stages of immersion that Brown and Cairns (2004) mention are engagement, engrossment and finally, total immersion. The level of engagement is reached when barriers of human activity are removed: The player must invest time, effort and attention to become engaged with the video game. In this stage, players say that they feel like they lose track of time. Engrossment can be reached when removing barriers that are embedded in the game’s design: Game design elements that were mentioned to achieve engrossment are graphics, interesting tasks and plot. During this stage, players are experiencing a high level of emotional investment in the game and they feel less aware of their surroundings. Total immersion occurs when the barriers of empathy and atmosphere are overcome. Empathy is linked to the growth of attachment to the game and its characters, while atmosphere is a combination of game design elements, such as graphics, sound and plot. Brown and Cairns (2004) suggest that the stage of full immersion is similar to the concept of presence. Other scholars also link the concept of immersion with presence, with presence being the experience of “being” in the virtual environment of a game (Cummings & Bailenson, 2015). This is why in this study we also look at literature and studies that focus on effects on presence.

Realism in video games

Scholars suggest that realistic video games have the properties to affect player experience by creating higher degrees of immersion for players and in turn, leading to enhanced enjoyment of the video game itself (Cheng & Cairns, 2005; Nichols, Haldane & Wilson, 2008; Ribbens, Malliet, Van Eck & Larking, 2016). To create immersive virtual environments, game designers should aim for that are “salient, both in appearance and object behaviour” (Nichols, Haldane & Wilson, 2008). What makes a game realistic to players is not limited to the audio-visual qualities of a game. There is a wide variety of conceptualisations for realism in video games. First of all, McMahan (2003)

(7)

6

devides the concept of realism in video games into social realism – to what extent social

interactions in the virtual environment match interactions in the real world – and perceptual realism – to what extent objects, environments and events in the virtual environment match those in that exist in the real world (McMahan, 2003). In their study on perceived realism and presence, Skalski and Whitbred (2010) point out that social realism has more to do with the social plausibility of content rather than its perceptual appearance and therefore social realism were not considered when studying perceived realism. Ribbens, Malliet, Van Eck and Larking (2016) take more recent literature on the conceptualisation of realism in video games into account and argue that the concept of realism can be expanded in a six-dimensional structure as illustrated in Table 1.

Simulational realism The degree to which the programmed rules and the different types of behavior that are possible within these rules credibly simulate the real world, thereby making the game potentially instructive for life.

Freedom of choice The degree to which the choices in a video game reflect the nature of choices one has in real life.

Social realism The degree to which events and characters in a video game are considered similar to events in real life.

Perceptual pervasiveness The degree to which a text creates a compelling audio-visual illusion, independent of the degree to which the content of the text may relate to real-world experience.

(Character) involvement The degree to which a player feels embodied in the video game world through the engagement with an avatar and the video game world.

Authenticity The degree to which the players have belief in the game designers’ intention and ability to convey an authentic, emotionally

convincing, or consistent message.

Table 1: Six-dimensional structure of perceived realism (Ribbens, Malliet, Van Eck & Larkin, 2016)

Perceptual realism in video games

The dimension of perceptual pervasiveness as mentioned in Table 1 is linked to the audio-visual qualities of a video game is one of the core dimensions of perceived realism. For players to perceive audio-visual qualities as realistic, graphics and sound in the virtual environment should be consistent to the player’s real world expectations of concepts as portrayed in the virtual environment. In their study on presence and immersive virtual environments, Slater and Wilbur (1997) propose that for a

(8)

7

system to be immersive it should offer high fidelity simulations through multiple sensory modalities. Cummings and Bailenson (2015) also conclude that the improvement of visuals and sound quality is linked to presence and immersion. These improvements are said to be in both manipulations such as better resolution, but also general level of detail and overall realism. With this in mind, I expect that video games with realistic virtual environments should therefore offer a more immersive experience for players.

H1a: Overall perceptual realism of the virtual environment is increased by high quality

graphics and sound in a video game

H1b: Overall perceptual realism has a positive effect on player immersion.

The effect of audio-visual qualities and its perceived realism on immersion and presence have been studied by several scholars. Skalski and Whitbred (2010) conducted a study with two conditions: One with high quality sound and graphics and one with low quality sound and graphics. They found no significant difference on perceived realism and immersion on the two conditions of graphics in their study: Increasing the quality of graphics did not increase how real participants perceived the virtual environment and in turn there was no effect found on immersion. They did find a significant positive effect of high quality sound on perceived realism and immersion: Sound in the high quality condition was perceived as significantly more realistic than sound in the low quality condition, which led to a more immersive experience for players of the video game. Skalski and Whitbred (2010) see perceptual realism as a dimension of immersion. The individual effects of perceived realism in graphics and sound on immersion will be discussed next.

Perceived realism of graphics

Graphics have by far been studied the most in recent work on the effects of realism on immersion. Realism in graphics can be divided in two game design components – geometric realism (a virtual object closely representing an object as perceived in real life) and illumination realism (referring to the fidelity of the lighting model) (Slater, Khanna, Mortensen & Yu, 2009). With their experiment, Slater, Khanna, Mortensen & Yu (2009) showed that while geometric realism had little effect on immersion, the condition where lighting was realistic (the character model had a shadow) was perceived as significantly more immersive. However, comparing the experimental design in both conditions with games and its realistic graphics today, both could be perceived as unrealistic. It is

(9)

8

therefore of importance to keep studying the effects of realism in graphics on immersion.

Graphics in video games have evolved drastically in history. Bracken and Skalski (2009) compared a video game with and without high quality graphics and found a significant improvement of immersion in the condition with high quality graphics. However, other scholars suggest that even though graphics will continue to improve, player experiences such as immersion and enjoyment might not increase significantly. Slater, Khanna, Mortensen and Yu (2009) expected to find an Uncanny Valley effect: in a highly realistic virtual environment imperfections of this environment could be breaking player immersion. Lin and Peng (2015) also did not find a significant effect when comparing a sport game with realistic graphics versus a sports game with unrealistic graphics. I therefore expect that even though higher quality graphics are perceived as more realistic, perceived realism will not have a significant effect on player immersion.

H2a: Perceived realism of graphics increases with higher quality graphics

H2b: Perceived realism of graphics does not have a significant effect on player immersion.

Perceived realism of sound

Game designers make use of different types of sound to embed the player in the virtual environment of the game. When looking purely at its manifestation, sound in a virtual environment has the potential to be more realistic than graphics. Grimshaw (2007) points out that “images on screen can only be a part of a virtual environment as they are a 2-dimensional representation of 3-

dimensionality. Sound, though, exists and operates both in reality and in virtuality; it has a real volume and dimensionality that is a 3-dimensional representation of the 2-dimensional

representation and events of the 3-dimensional world of the game.” The way players are exposed to sound also affects immersion. Grimshaw (2007) points out that wearing headphones while playing video games is more immersive than hearing sound through speakers. Skalski and Whitbred (2010) found a significant effect of sound quality on perceived realism of sound and on immersion. I expect to find similar results in this study.

H3a: Perceived realism of sound increases with higher quality sound in a virtual environment H3b: Perceived realism of sound has a positive effect on player immersion.

(10)

9

METHOD

To test the hypothesises I conducted an experiment with two conditions – a condition with high quality graphics and sound and a condition with low quality graphics and sound. Participants were randomly assigned to the two conditions and were asked to complete a questionnaire afterwards. The questionnaire consisted of 24 statements measuring perceived realism, immersion, control variables and demographics.

Participants

For the experiment I selected participants that were familiar with games and playing games themselves. The participants (N = 90) were selected by means of convenience sampling: The questionnaire was posted on Facebook groups aimed at gamers (The Dutch Gamers, GirlGamer NL) and the link was sent to individual gamers via several servers on Discord aimed at video game communities and Twitch streamer communities. To receive additional data, the questionnaire was sent out internally to selected game developers in a game development studio in Amsterdam. These game developers were selected based on their availability to fill out the questionnaire during the time that the questionnaire was available online. Furthermore, all participants were asked to share the link to the questionnaire with friends and family who also play video games. The age range of the participants was between 2 (14 – 17 years old) (N = 7), 3 (18 – 24 years old) (N = 44), 4 (25 – 34 years old) (N =29), 5 (35 – 44 years old) (N = 7), 6 (45 – 54 years old) (N = 1) and 10 (85 years old or older) (N = 1) (Mdn = 3; IQR = 3-4). 38.9% of participants were male (N = 35) and 60% were female (N = 54), 1.1% was defined as missing because no gender was indicated (N = 1). From the participants, 25.6% works or worked in the game industry or studies or studied are game related study (N = 23). The amount of time that the participants spent playing in a regular week varied from between 1 and 5 hours (N = 6), 6-20 hours (N = 46), 21-40 hours (N = 26) and more than 40 hours (N = 11). One participant indicated that they do not play any video games during a regular week.

Stimulus

The participants were asked to watch a short video of gameplay from two different games. The two conditions in the experiment were a condition with low quality graphics and sound – for which a video of gameplay of the original version of Resident Evil (1999) for PlayStation 1 was used – and a condition with high quality graphics and sound – for which a video of gameplay of the remastered version of Resident Evil (2015) for PlayStation 4 was used. Both games follow the same story, but

(11)

10

the remastered version of the game was updated with higher quality graphics and sound and more realistic looking graphics and sound effects. Because difference in gameplay (the control scheme in the remastered version differs from the original version) and input mode (PlayStation 1 controller versus PlayStation 4 controller) could also affect immersion, I chose to ask participants to imagine playing themselves while watching a video, over having them play the game themselves. The video shows gameplay from the point of view that participants would see when they would control the game themselves. Furthermore, the video’s show gameplay in the same area, where the player solves a puzzle to get to an item and fights one of the enemies in the game. The conditions in this study will be referred to as the “high quality” condition (N = 47), in which participants were exposed to the Resident Evil HD Remaster (2015) and the “low quality” condition (N = 43), in which participants were exposed to Resident Evil (1999).

Procedure

The questionnaire was online between November 29, 2016 and December 3, 2016. On the first page of the questionnaire, participants were explained the purpose of the experiment and asked to imagine themselves playing the game while watching the video of gameplay. The video showed gameplay in a way that participants would also see if they would control the character themselves. In an attempt to eliminate effects of different screen sizes and audio output, they were also asked to watch the video’s in full screen on a PC while wearing headphones. After the video, participants were redirected to the questionnaire.

Manipulation check: Perceived realism

To check if the two conditions were also perceived as their intended level of realism, a total of nine statements were given. With the use of a 5-point Likert scale, participants could indicate to what extent they agreed or disagreed with these statements. The statements used are modifications of statements taken from earlier studies on perceived realism in virtual environments (Witmer & Singer, 1998; Schubert, Friedmann & Regenbrecht, 2001). One statement was aimed to measure perceived realism of the virtual environment as a whole (“How much did the virtual environment seem consistent with the real world?”), three statements measured perceived realism of the graphics of the virtual environment (e.g. “The graphics in this game match how the world looks in real life”) and five statements measured perceived realism of the sound of the virtual environment (e.g. “The sound in this game match how the world sounds in real life”). All statements and how they were measured can be found in Appendix A.

(12)

11

Dependent variables: Immersion

Immersion was measured using modifications of statements used in the presence questionnaire as used by Schubert, Friedmann & Regenbrecht (2001). Immersion was measured directly after watching the video and before all other questions, so participants could immediately recall their experience. Participants were asked whether they agreed with a set of statements on a 5-point Likert scale. A total of eight statements to measure immersion (e.g. “How aware were you of the real world surrounding while playing the game”; “I did not feel present in the virtual environment”) were included in the questionnaire, which can be found in Appendix A.

Control variables

The control variables used in the questionnaire are whether or not the participant has already seen or played one of the versions of Resident Evil that were used in the experiment. If participants already played the game, their experience of immersion might be different from participants who never played the Resident Evil games. Furthermore, the amount of time spent on playing games was asked. The amount spent playing games could affect how easily participants are able to imagine oneself in the position of the player while watching a video of someone else playing. Finally, I asked if participants worked in the game industry or studied a game related field. Prior knowledge of how realism can affect immersion in games could explain differences in experience of immersion between these participants and other participants. In addition, participants were asked their age and gender.

(13)

12

RESULTS

The items that together created the scale for immersion was tested for its reliability. The scale consists of seven items and has a Cronbach’s Alpha of 0.79 and is therefore a reliable scale to measure immersion. However, the scale can be improved by deleting one item. This item coming from the statement “I felt like I was just perceiving a video” was deleted from the immersion scale. The Cronbach’s Alpha for the new scale measuring immersion is 0.80. The scale for perceived realism of graphics consists of 2 items and has a Cronbach’s Alpha of 0.88. The scale for perceived realism in sound consist of 4 items, but did not create a reliable scale with a Cronbach’s Alpha of 0.69. In a factor analysis, two components were identified to create two new scales that together account for 78% of the variance. The first scale consists of the items “The sound in this game is very realistic” (component loading after Varimax rotation = 0.85) and “The sound in this game match how the world sounds in real life” (component loading after Varimax rotation = 0.91) and has a Cronbach’s Alpha of 0.76. The other two items did not create a reliable scale as its Cronbach’s Alpha was 0.62, and therefore these items were deleted from the analysis. The means and standard deviations for the scales for the two conditions in the study are defined in Table 2.

Condition

Low quality (N = 43) High quality (N = 47)

Immersion 3.12 (0.80) 3.52 (0.70)

Perceived realism Virtual Environment 2.86 (1.06) 3.23 (1.03)

Perceived realism Graphics 2.07 (1.07) 2.83 (1.12)

Perceived realism Sound 2.81 (1.02) 3.20 (1.02)

Table 2: Means and standard deviations for scales and variables

Manipulation checks

I checked if the two conditions that the participants were exposed to in this study were perceived as significantly more realistic in the “high quality” condition in comparison with the “low quality” condition. First I checked if the virtual environment in the “high quality” condition is perceived as more realistic than the virtual environment in the “low quality” condition. On an average, participants in the “high quality” condition scored 3.23 (SD = 1.03) on perceived realism in the virtual environment and participants in the “low quality” condition scored 2.86 (SD = 1.06) on average. The difference between the means are not significant, t(88) = 1.70, p = 0.093, CI = [-0.06, 0.81]. This means that the virtual environment in the “high quality” condition was not perceived as significantly more realistic than the “low quality” condition by participants in both conditions. I

(14)

13

then checked if the graphics in the “high quality” condition used in this study were perceived as more realistic than the graphics in the “low quality” condition. Participants in the “high quality” condition scored 2.83 (SD = 1.12) on average for perceived realism in graphics and participants in the “low quality” condition scored 2.07 (SD = 1.07) on average. The difference between means is significant, t(88) = 3.28, p = 0.002, CI = [0.30, 1.22] and the effect size is medium (Cohen’s d = 0.69). This means that the graphics in the “high realism” condition were perceived as more realistic than the graphics in the “low realism” condition. Finally, I checked if the sound in the “high quality” condition was perceived as more realistic than the sound in the “low quality” condition. Participants in the “high quality” condition scored 2.81 on average for realism in sound (SD = 1.02) and participants in the “low quality” condition scored 3.20 (SD = 1.02). The difference between means is not significant, t(88) = 1.78, p = 0.082, CI = [-0.05, 0.80]. This means that the “high quality” condition was not perceived as significantly more realistic than the “low quality” condition for sound.

Immersion in the “high-” and “low quality” conditions

Participants in the “high quality” condition score significantly higher on the immersion scale, which means that people in this condition felt more immersed in the video game environment than people in the “low quality” condition. The difference between means is significant, t(88) = 2.50, p = 0.014,

CI = [0.08, 0.71] but the effect size is medium (Cohens d = 0.53).

Regression models

The regression model with immersion as dependent variable and perceived realism in the virtual environment, perceived realism of graphics and perceived realism of sound as independent variables is significant, F(3,89) = 10.24, p < 0.001. The model is useful to predict immersion, but only 26% of the differences in immersion can be predicted by perceived realism in the virtual environment, perceived realism in graphics and perceived realism in the sound of the video game (R² = 0.24). Perceived realism in the virtual environment has a significant effect on immersion, b* = 0.28, t = 3.59, p = 0.001, 95% CI = [0.13, 0.44]. When the virtual environment is perceived as more realistic, participants score 0.28 higher on the immersion scale. As shown in Table 3, perceived realism of graphics and perceived realism of sound did not have a significant effect on immersion.

Immersion

b* p

Constant 2.06 < 0.001

Perceived realism virtual environment 0.28 0.001

Commented [MT2]: Het is denk ik ook interessant om per conditie en regressive-analyse toe doen.

(15)

14

Perceived realism graphics 0.10 0.185

Perceived realism sound 0.05 0.567

R² 0.26

F 10.24 < 0.001

Table 3: Regression model predicting immersion (all conditions)

The regression model for the “low quality” condition is significant, F(3,46) = 4.41, p = 0.009. The model is useful to predict immersion, but only 24% of the differences in immersion can be predicted by perceived realism in the virtual environment, perceived realism of graphics and perceived realism of the sound of the video game (R² = 0.18). Only perceived realism in the virtual environment has a significant effect on immersion, b* = 0.26, t = 2.51, p = 0.016, 95% CI = [0.05, 0.47]. When the virtual environment in the “high quality” condition is perceived as more realistic, participants score 0.26 higher on the scale of immersion. Table 4 shows that perceived realism of graphics and perceived realism of sound in the “high quality” condition did not have a significant effect on immersion.

Immersion

b* p

Constant 2.30 < 0.001

R² 0.18

F 4.41 0.009

Table 4: Regression model predicting immersion (“High quality” condition)

The regression model for the “low quality” condition is also significant, F(3,42) = 3.80, p = 0.018. The model is useful to predict immersion, but only 23% of the differences in immersion can be predicted by perceived realism in the virtual environment, perceived realism in graphics and perceived realism in the sound of the video game (R² = 0.17). When looking at the model for the “low quality” condition, only perceived realism in the virtual environment has a significant effect on immersion, b* = 0.29, t = 2.41, p = 0.021, 95% CI = [0.05, 0.54]. When the virtual environment in the “high quality” condition is perceived as more realistic, participants score 0.29 higher on the scale of immersion. Table 5 shows that perceived realism of graphics and perceived realism of sound in the “low quality” condition did not have a significant effect on immersion.

(16)

15

Immersion

b* p

Constant 1.98 < 0.001

R² 0.17

F 3.80 0.018

Table 5: Regression model predicting immersion (“Low quality” condition)

Control variables

I checked for several control variables if they had an interaction effect on immersion in the two conditions, means and standard deviations can be found in Table 6.

Condition

Low quality High quality

Gender Female 3.05 (1.51) 3.61 (0.14)

Male 3.21 (0.18) 3.35 (0.18)

Works in game industry Yes 3.11 (0.22) 3.32 (0.23)

No 3.11 (0.14) 3.57 (0.13)

Time spent playing video games in a regular week

1-5 hours 2.88 (0.38) 4.00 (0.54)

6-20 hours 3.08 (0.17) 3.43 (0.15)

21-40 hours 3.12 (0.20) 3.57 (0.22) 40+ hours 3.39 (0.44) 3.54 (0.27) Table 6: Means and standard deviations for control variables on scale of immersion

For females, the “high quality” condition led to a significantly higher score on immersion (F(1,85) = 7.33, p = 0.008). For males, the two conditions had no significant effect on immersion (F(1,85) = 0.33, p = 0.568). Participants who do not work in the game industry in the “high quality” condition scored significantly higher on the immersion scale than participants who do not work in the game industry in the “low quality” condition (F(1,86) = 6.00, p = 0.016). For people who do work in the game industry, the condition that they were exposed to had no significant effect on immersion (F(1,86) =

(17)

16

0.45, p = 0.511). There was no significant interaction effect of time spent playing for the two conditions on immersion, 1-5 hours (F(1,81) = 2.92, p = 0.091), 6-20 hours (F(1,81) = 2.47, p = 0.120), 21-40 hours (F(1,81) = 2.39, p = 0.126), 40+ hours (F(1,81) = 0.09, p = 0.767).

(18)

17 CONCLUSION

The purpose of this study was to build on existing theory on perceived realism and its effects on immersion. It is interesting to point out that in the “high quality” condition, where participants played the remastered version of Resident Evil (2015), participants felt significantly more immersed than participants in the “low quality” condition. Because all other game design elements that could interfere as variables that can affect immersion were kept constant, the difference in immersion can be explained by the improvement of quality for the graphics and sound in the video game. In addition to these results, I will now discuss if the results of the experiment back the hypothesises that were proposed in this study.

First of all, I proposed that high quality graphics and sound in the “high quality” condition would result in participants valuing the virtual environment of the game as more realistic (H1a). The first manipulation check shows that the virtual environment in the “high quality” condition was not perceived as significantly more realistic than the “low quality” condition. Therefore, this hypothesis is rejected: High quality graphics and sound does not improve how realistic participants perceive a virtual environment. The second manipulation check was to test if the graphics in the “high quality” condition were perceived as more realistic (H2a). Graphics in the “high quality” condition were perceived as significantly more realistic than graphics in the “low quality” condition. The hypothesis was backed by these results: High quality graphics increase how realistic participants perceive graphics in the virtual environment. The final manipulation check showed if quality of sound affects how realistic participants perceive sound in the virtual environment. The hypothesis that high quality sound would be perceived as more realistic (H3a) was not backed by the results in this study: Sound in the “high quality” condition was not perceived significantly more realistic than sound in the “low quality” condition.

I also tested the main effect of overall perceived realism in the virtual environment on immersion. All regression models show that the more realistic a virtual environment is perceived, the more immersed participants felt in the video game. These results back the hypothesis that overall perceived realism increases immersion (H1b). The hypothesis that the more realistic the participants perceive the graphics of the virtual environment does not significantly affect immersion (H2b) was also backed by the results of the regression models. Finally, the hypothesis that increased perceived realism of sound has a positive effect on immersion (H3b) was not backed. The results of this study show that how real participants perceived sound in the virtual environment did not significantly affect immersion.

(19)

18 Limitations and further research

While Skalski and Whitbred (2010) found a significant effect of sound quality on perceived realism of sound as well as a significant positive effect on immersion, this was not the case in this study. Further research is needed to back whether improvements of sound quality and perceived realism of sound has a positive effect on immersion. In this study I used existing scales for perceived realism, however, the scales could be expanded and improved. The four items that should measure perceived realism in sound did not create a reliable scale in this study. New items to measure how real participants perceive sound in video games should be developed to create scales to use in further research on the effects of perceived realism on immersion and other effects of video games.

As mentioned in the conceptual framework, a lot of components of video game design can affect how immersed players feel while playing a video game. It is possible that not all of the components have effectively been controlled in this study, so effects of immersion that are registered for the two conditions could also be the results of other components of game design that were changed in the remastered version of the video game that was used in this study. Since gameplay has also changed in the remastered version of the video game, I decided that participants should watch a video of gameplay, instead of playing the game themselves. However, actually playing the game could have an impact on immersion and in further research, a different approach on how participants are exposed to the virtual environment should be considered.

During the collection of the data, some of participants pointed out that the video was not available to watch in full screen. This was due to participants using different browsers and some did not support playing video’s full screen directly from the Qualtrics website. Not all participants were exposed to the stimulus in the way that it was described in the operationalisation of this study, which could have an effect on how immersed people felt while watching the video. In further studies, the way that participants are exposed to the virtual environment should be controlled more tightly.

(20)

19

REFERENCES

Brown, E., & Cairns, P. (2004). A grounded investigation of game immersion. Paper presented at CHI 2004, Vienna, Austria, 24-29 April (pp. 1297–1300). New York: ACM.

Bracken, C. C., & Skalski, P. (2009). Telepresence and video games: The impact of image quality.

PsychNology Journal, 7(1), 101-112.

Cheng, K., & Cairns, P. A. (2005). Behavior, realism and immersion in games. Paper presented at CHI 2005, Portland, USA, 2-7 April (pp. 1272-1275). New York: ACM.

Cummings, J. J., & Bailenson, J. N. (2015). How immersive is enough? A meta-analysis of the effects of immersive technology on user presence. Media Psychology, 19(2), 272-309. doi: 10.1080/15213269.2015.1015740

Grimshaw, M. (2008). Sound and immersion in the first-person shooter. International Journal of

Intelligent Games & Simulation, 5(1), 119-124.

Lin, J. H., & Peng. W. (2015). The contributions of perceived graphic and enactive realism to enjoyment and engagement in active video games. International Journal of Technology and

Human Interaction, 11(3), 1-16. doi: 10.4018/ijthi.2015070101

McMahan, A. (2003). Immersion, engagement and presence. In M. J. P. Wolf & B. Perron (Eds.)

The Video Game, Theory Reader (pp. 77-78). New York, NY, US: Routledge, Taylor &

Francis Group.

Nichols, S., Haldane, C., & Wilson, J. R. (2000). Measurement of presence and its consequences in virtual environments. International Journal of Human-Computer Studies, 52(3), 471-491. doi: 10.1006/ijhc.1999.0343

Ribbens, W., Malliet, S., Van Eck, R., & Larkin, D. (2016). Perceived realism in shooting games: Towards scale validation. Computers in Human Behavior, 64, 308-318. doi:10.1016/ j.chb.2016.06.055

Schubert, T., Friedmann, F., & Regenbrecht, H. (2001). The experience of presence: Factor analytic insights. Presence: Teleoperators and Virtual Environments, 10(3), 266-281. doi:

10.1162/105474601300343603

Skalski, P., & Whitbred, R. (2010). Image versus sound: A comparison of formal feature effects on presence and video game enjoyment. PsychNology Journal, 8(1), 67-84.

Slater, M., Khanna, P., Mortensen, J., & Yu, I. (2009). Visual realism enhances realistic response in an immersive virtual environment. IEEE Computer Graphics and Applications, 29(3), 76-84. doi: 10.1109/MCG.2009.55

Slater, M., & Wilbur, S. (1997). A framework for immersive virtual environments (FIVE): Speculations on the role of presence in virtual environments. Presence Teleoperators and

Virtual Environments, 6(6), 603-616. doi: 10.1162/pres.1997.6.6.603

Sweetser, P., & Wyeth, P. (2005). GameFlow: A model for evaluating player enjoyment in games. Computers in Entertainment: Theoretical and Practical Computer Applications in Entertainment, 3(3), 1-24. doi: 10.1145/1077246.1077253

Witmer, B. G., & Singer, M. J. (1998). Measuring presence in virtual environments: A presence questionnaire. Presence: Teleoperators and virtual environments, 7(3), 225-240. doi: 10.1162/105474698565686

(21)

20

APENDIX A: QUESTIONNAIRE

Thank you for taking the time to participate in this study. It should take about 5 to 10 minutes to finish. Taking part in this study is voluntary. You may choose to discontinue your participation at any time.

In this study you will be watching a video of gameplay of a popular console game. It is important that while watching this video, you imagine playing the game yourself. Try to imagine that you are controlling the character the way that it is played in the video.

It is important for this study that you watch the video in full screen and wear headphones – Thank you for your cooperation.

[NEW PAGE: PARTICIPANTS RANDOMLY ASSIGNED TO ONE OF TWO CONDITIONS]

[NEW PAGE: ALL PARTICIPANTS]

-

How aware were you of the real world surrounding while playing the game (i.e. sounds, room temperature, other people, etc.)

Not aware at all Slightly aware Moderately aware Very aware Extremely aware

-

I had a sense of being present in the virtual environment of the game

Strongly agree Somewhat agree Neither agree nor disagree

Somehwhat disagree

Strongly disagree Condition 1: High Quality

Video: https://www.youtube.com/watch? v=TlEcB0pkjOY

Length: 1m54s

Condition 2: Low Quality

Video: https://www.youtube.com/watch? v=h2XoRiALkTg

(22)

21

-

I was not aware of my real environment

Somehwhat disagree

Strongly disagree

-

I was focused on the game

Somehwhat disagree

Strongly disagree

-

I was interested in seeing how the game would progress

Somehwhat disagree

Strongly disagree

-

Somehow I felt that the virtual environment surrounded me

Somehwhat disagree

Strongly disagree

-

I felt like I was just perceiving a video

Somehwhat disagree

Strongly disagree

[NEW PAGE: ALL PARTICIPANTS]

-

How much did the virtual environment seem consistent with the real world?

Extremely consistent Somewhat consistent Neither consistent nor inconsistent Somehwhat inconsistent Extremely inconsistent

-

I like the graphics in this game

Somehwhat disagree

Strongly disagree

-

The graphics in this game are very realistic

Somehwhat disagree

(23)

22

-

The graphics in this game match how the world looks in real life

Somehwhat disagree

Strongly disagree

-

I like the sound in this game

Somehwhat disagree

(24)

23

-

The sound in this game is very realistic

Somehwhat disagree

Strongly disagree

-

The sound in this game match how the world sounds in real life

Somehwhat disagree

Strongly disagree

-

How well could you identify sounds?

Extremely well Very well Moderately well Slightly well Not well at all

-

How well could you localize sounds?

Extremely well Very well Moderately well Slightly well Not well at all

Before participating, did you ever play any of the following games?

-

Resident Evil (1996)

Yes, I have played this game No, I have not played this game, but I have heard of it before

No, I have not played this game and I have not heard of it before

-

Resident Evil Remake (2002)

-

Resident Evil HD Remake (2015)

(25)

24

- How much time do you spend playing video games in a regular week? 1-5 hours

6-20 hours 21-40 hours 40+hours None at all

- Do you work in the game industry or follow a game-related study? Please also fill in yes if you've previously worked in the game industry or followed a game-related study.

Yes No - Age Under 14 14 -17 18 - 24 25 - 34 35 - 44 45 - 54 55 - 64 65 - 74 75 - 84 85 or older -Gender Male Female Other