An Initial Study on Human Emotional States in Video Games

Abstract

Nowadays, playing video games has become more than an entertainment element in our everyday life. From education, sport, marketing or clinical practice to aerospace, the video games industry has a relevant role in our society as element of entertainment, learning and social interaction. Due to that importance, the e orts to improve visual realism and games mechanics have grown considerably during last decades. However, the emotional and narrative aspect is not deeply analyzed yet, creating a knowledge gap about their usefulness on the user engagement. Aware of the importance of the emotional states of the player in the their performance when are playing, we present an experimental methodology to analyze and measure the in uence of emotional states in the game performance and nal user experience. This methodology will consist on the induction of di erent emotional states to an experimental group and, immediately after, they will play a video game. While the player is using the video game, his/her performance metrics will be extracted and saved on the computer, translating them into numerical data. This data will be used lately to look after statistical di erences between the di erent conditions. This paper provides a complete description of the methodology used to understand how emotions are related to the performance and opinion of a group of users.

Keywords

• A ective gaming
• Video games
• Emotion
• Psychology

-

games go further from entertainment, as it is usually considered a new art form, and a powerful technology for human being interaction. Even though their short existence (it's commonly accepted that this industry started in 1980), video games are currently a trending technology which only in 2016 [ 1 ] was used by 2.2 billion people around the world. Nowadays, they are present in almost every person's life, specially in western countries, due to its versatility and high demand from society. These products are constantly involved in more daily activities, such as learning and education [ 3 ], sports [ 7 ], clinical practice [ 11 ], training [ 19 ], marketing [ 14 ], or aerospace [22], among others.

The rst decades of this industry were focused on improving the graphic quality, the mechanics, as well as the character aspect and other physical elements of the game. The development of the gaming engines, which facilitates the creation of a game (providing with a software base which reduces the time of software development and testing) [20], represented a critical point in the trend history of video games. This critical point made possible the creation of video games with a more realistic appearance.

The graphic and games mechanics improvement that video games have experienced, have usually originated a gap between the a ective and social realism of game interactions, and the players [ 9 ]. However, these social and human interactions play an important role during the gaming process, to allow the generation of positive emotions that will be directly related to the activity engagement of the players [ 17 ]. This led video game companies and their associated research groups to start working on assessing how the emotions seem to be related with the player general satisfaction with the game [24]. The aim of this research eld is to understand how the emotions of the player are related to the game features, and also to understand how that emotions have an impact (positive or negative) on his/her engagement and performance with it.

Here we present an experimental methodology to to measure how varied emotional states of the player a ects his/her performance when playing a game, and also how this performance and emotional states are related to the user experience.

The rest of the paper has been structured as follows: Section 2 provides a description on how the emotion has been traditionally studied in video games contexts. Section 3 explains what A ective Games are, providing the theoretical context where the supporting methodology will be presented. Section 4 presentes the hypothesis which will be tested with the methodology. Section 5 describes the methodology proposed in this work for a future research related with a basic approach to an emotion research in video games. Finally, section 6 provides a nal conclusion for this proposal. 2

Emotion relevance in video games

Although the main objective of video games is the entertainment, their design is carefully focused on engaging the player into the \gameplay" [ 4 ]. This design is conducted taking into account several strategies to maintain the attention of the player. There are plenty of these strategies (focused on narrative, rewards, dynamics, cooperation vs competition, etc.), and they all have something in common: they are increasingly studied from a cognitive science perspective. These sciences (specially psychology), which were focused until recent years in analyzing the consequences of video game playing [ 16 ], are currently changing their focus to in-game variables and the behavioral impact before, during and after the play [ 6 ]. Here is where the emotion research appears to have a critical role.

As the psychological research in video games has evolved, there have appeared two di erent ways of taking into account the emotions in a video game context [ 8 ]: { Emotions as a "side e ect": the emotion is not important, as long as the individual enjoys the game. The e orts are not focused on providing any emotion, but on avoiding frustration and generating success sensations on the player. { Emotions as an engagement tool: the emotion is a critical point, as it controls the player engagement with the game. This approach can also be divided in two approaches: the emotion observed as an open loop (it is not necessary for the computer or console to analyze the emotion) and the emotion observed as a close loop (the computer or console analyzes the emotions and adapts to it).

As Johnson and Whiles defend, \if a game does not generate positive emotions in the user it is unlikely to succeed" [ 10 ]. The key motivating factor for a video game player is to experience general positive emotions. There are, of course, several strategies to generate these positive emotions (depending on the person, the genre of the game, the motivation of the player, etc.), but the aim of all of them is to generate a positive emotional feedback.

Following this idea, and focusing on the close loop approach, we nd a proposal which is inherited from A ective Computing [ 15 ], a discipline which develops the computer capacities of relating, arising or in uencing emotions. As part of the introduction of the scienti c study of emotion in the computer science world, it has also be applied to the speci c video games industry on the last years. Thus, the emotion research in video games, combined with the current techniques derived from A ective Computing, arises a new discipline: A ective Games. 3

A ective Games

As a sub eld within A ective Computing, A ective Games represents an e ort to adapt the video games to the user's emotional state, both screening these emotions and replaying them with changes in the dynamics of the game [ 12 ]. Once the player feels an emotion, the video game retrieves this information from di erent extracting sources, and adapts itself to generate a positive emotion on the user.

Player expresses

emotions Player's game experience improved

System detects and adds emotions

to the game Gameplay adaption according to

emotions { The extraction of the user information, normally using physiological measurements, external (i.e. body position, button pressure) and internal (i.e. number of enemies defeated in the game) behavioral measurements. { The design of the game, which includes the adaptation of the game to the user emotions.

These elds, which are normally combined with the User Experience research (UX), concentrates most of the research currently published [ 12 ]. Furthermore, most of the research has been concentrated on a physiological perspective to obtain emotional feedback from the user [ 13 ], while the behavioral patterns of the user related to the gameplay (the intern measurement) have been brie y studied. Moreover, there are just few studies assessing how the di erent emotional states in uence the performance of the player [ 18 ]. This basic gap in emotional research applied to video games must be ful lled in order to understand exactly how emotions, performance and user experience are related.

Therefore, as a way to understand how emotions are related with the game performance, this research proposal aims to test how di erent emotional states impact on the player performance playing a game, also analyzing how the emotional state and performance inside the game is related with the gaming user experience. 4

Objectives

The aim of this research is to analyze the in uence of emotional states in the game performance and nal user experience. To do so, di erent emotional states will be induced to the experimental group across a rst phase of the experiment. Immediately after that, they will play a video game where several variables related to their performance will be measured using in-game indirect indicators [ 12 ]. Finally, a nal UX test will be applied to them.

The main hypothesis are presented below.

H1. The users with a positive emotional state (happiness) will need less time to end the game.

H2. The users with a negative emotional state (sadness, fear or anger) will need more time to end the game.

H3. The users with an angry emotional state will show the most aggressive behaviors inside the game, defeating more enemies.

H4. The users will report higher levels of interest on the video game with positive valence emotional states. 5

Proposed methodology

The sample for the experiment will be extracted from a university student population, from the Universidad Autonoma de Madrid. The sample will be divided in ve groups, each of the rst four assigned to a di erent experimental condition, in this case related to an emotion (happiness, sadness, anger and fear) and one assigned to a control condition (no emotional induction will be applied to them).

As a rst part of the experiment, an emotional induction will be conducted with each user. The induction of emotions using videos has been reported to be the most e ective method in prior research [ 5, 23 ]. Thus, and following this methodology, the groups will be asked to have a look for a 5 minutes long video depending on the emotional state objective. Every video will be related with the group objective emotion, and will be extracted from previous research [21]. These lms are presented on table 1.

Once they have stated the video, they will ful ll a questionnaire related to their current emotional state (as a way to measure if the emotional induction was e ective). The questionnaire chosen will be the Analogical Emotional Scale (AES), which has been previously used a a fast psychological tool to measure emotional states [ 2 ].

Immediately after, the user will play the video game chosen for this research. In this study, the video game will be \In nite Mario Bros", an open source game based on the classical saga \Super Mario", which has the same 2D design and mechanics than his formal counterpart. The players will play the same level of the game, in order to avoid possible biases. The game itself will include several measurement features, such as: { Time used to end the level. { Times defeated until the end of the level. { Number of enemies killed. { Number of bricks destroyed. { Frequency of use of each button.

While the player is using the video game, his/her performance metrics will be extracted and saved on the computer, translating them into numerical data. This task will be performed modifying the source code of the video game (in this case, an open code) to obtain this metrics. The data obtained will be used lately to look after statistical di erences between the di erent conditions.

Finally, and once ended the video game level, the users will answer again to the AES questionnaire, in order to check if the emotional state is still present on the person. This test will be applied along with some User Experience questions, where they will be questioned about their performance, their sensations about the video game and about their general opinion about it. 6

Conclusion

The eld of A ective Gaming rises as a promising application of cognitive sciences to the video game industry. To ensure the impact of emotions in the video game industry, this research has been designed as a rst basic approaching to this new discipline. The main contribution of this empirical study is getting a better understanding of how emotions are related to the performance and opinion of a group of users. Also, this will open di erent research questions, once the outcomes are nally prepared, in relation with the impact of di erent emotions on the user behavior, and also how the speci c game will relate to that results.

Acknowledgements

This work has been supported by the next research projects: RiskTrack project, under the European Union Justice Action Grant JUST-2015-JCOO-AG-723180, EphemeCH project (TIN2014-56494-C4-4-P) and DeepBio (TIN2017-85727-C4-3P) by the Spanish Ministry of Economy and Competitivity and European Regional Development Fund FEDER, and the CIBERDINE project "Cybersecurity, Data and Risks" under the CAM grant S2013/ICE-3095.The contents of this publication are the sole responsibility of their authors and can in no way be taken to re ect the views of the European Commission. 20. Richard M Ryan, C Scott Rigby, and Andrew Przybylski. The motivational pull of video games: A self-determination theory approach. Motivation and emotion, 30(4):344{360, 2006. 21. Alexandre Schaefer, Frederic Nils, Xavier Sanchez, and Pierre Philippot. Assessing the e ectiveness of a large database of emotion-eliciting lms: A new tool for emotion researchers. Cognition and Emotion, 24(7):1153{1172, 2010. 22. Robert J Stone, Peter B Pan lov, and Valentin E Shukshunov. Evolution of aerospace simulation: From immersive virtual reality to serious games. In Recent Advances in Space Technologies (RAST), 2011 5th International Conference on, pages 655{662. IEEE, 2011. 23. Rainer Westermann, GUNTER Stahl, and F Hesse. Relative e ectiveness and validity of mood induction procedures: analysis. European Journal of social psychology, 26(4):557{580, 1996. 24. Georgios N Yannakakis and Ana Paiva. Emotion in games. Handbook on a ective computing, pages 459{471.

References

1. Global games market report. Newzoo Magazine , 2016 .

2. Pilar Carrera and Luis Oceja . Drawing mixed emotions: Sequential or simultaneous experiences? Cognition and Emotion, 21 ( 2 ): 422 { 441 , 2007 .

3. Alessandro De Gloria, Francesco Bellotti, and Riccardo Berta . Serious games for education and training . International Journal of Serious Games , 1 ( 1 ), 2014 .

4. Michele D Dickey . Engaging by design: How engagement strategies in popular computer and video games can inform instructional design . Educational Technology Research and Development , 53 ( 2 ): 67 { 83 , 2005 .

5. Kristen K Ellard , Todd J Farchione , and David H Barlow. Relative e ectiveness of emotion induction procedures and the role of personal relevance in a clinical sample: a comparison of lm, images, and music . Journal of Psychopathology and Behavioral Assessment , 34 ( 2 ): 232 { 243 , 2012 .

6. Isabela Granic , Adam Lobel, and Rutger CME Engels. The bene ts of playing video games . American psychologist , 69 ( 1 ): 66 , 2014 .

7. Juho Hamari and Max Sjoblom. What is esports and why do people watch it ? Internet research, 27 ( 2 ): 211 { 232 , 2017 .

8. Eva Hudlicka . A ective computing for game design . In Proceedings of the 4th Intl. North American Conference on Intelligent Games and Simulation , pages 5 { 12 , 2008 .

9. Eva Hudlicka . A ective game engines: motivation and requirements . In Proceedings of the 4th international conference on foundations of digital games , pages 299 { 306 . ACM, 2009 .

10. Daniel Johnson and Janet Wiles . E ective a ective user interface design in games . Ergonomics , 46 ( 13 -14): 1332 { 1345 , 2003 .

11. Pamela M Kato. Video games in health care: Closing the gap . Review of general psychology , 14 ( 2 ): 113 , 2010 .

12. Raul Lara-Cabrera and David Camacho . A taxonomy and state of the art revision on a ective games . Future Generation Computer Systems , 2018 .

13. Lennart E Nacke and Regan L Mandryk . Designing a ective games with physiological input . In Workshop on Multiuser and Social Biosignal Adaptive Games and Playful Applications in Fun and Games Conference (BioS-Play ) , 2010 .

14. Stephanie O'Donohoe and Ilya Vedrashko . Game-based marketing: inspire customer loyalty through rewards, challenges , and contests, 2011 .

15. Rosalind Wright Picard et al. A ective computing . 1995 .

16. Andrew K Przybylski , C Scott Rigby , and Richard M Ryan. A motivational model of video game engagement . Review of general psychology , 14 ( 2 ): 154 , 2010 .

17. Amy L Reschly , E Scott Huebner , James J Appleton , and Susan Antaramian . Engagement as ourishing: The contribution of positive emotions and coping to adolescents' engagement at school and with learning . Psychology in the Schools , 45 ( 5 ): 419 { 431 , 2008 .

18. Leticia Grossi Gomes Ribeiro , Victor Hugo Magalhes Pereira, Victor Rezende Costa, Raphael Viana Corujo de Oliveira Moura , and Alex Fernandes da Veiga Machado. A ective computing in electronic games: in uence of emotional states in player performance . In 31st European Simulation and Modelling Conference , 2010 .

19. James C Rosser , Paul J Lynch, Laurie Cuddihy, Douglas A Gentile, Jonathan Klonsky, and Ronald Merrell. The impact of video games on training surgeons in the 21st century . Archives of surgery , 142 ( 2 ): 181 { 186 , 2007 .