What drives gamer toxicity? Essays from players Bastian Kordyaka 1, Samuli Laato 2, 3, Juho Hamari 2, Tobias Scholz 4 and Björn Niehaves 1 1 University of Bremen, Bibliothekstraße 1, 28359 Bremen, Germany 2 Gamification Group, Korkeakoulunkatu 10, 33720 Tampere, Finland 3 University of Turku, Kiinamyllynkatu 13, 20500 Turku, Finland 4 University of Siegen, Adolf-Reichwein-Straße 2a, 57076 Siegen, Germany Abstract Negative online behaviors, such as toxicity, continue being issues in several popular multiplayer online games. Related research suggests that there are individual differences in how players understand the concept, and that various interconnected variables are relevant in understanding the emergence of toxicity. To explore this topic further, in this study, we gathered 16 essays from gamers regarding their experiences of toxicity in online games. Using the Gioia method for qualitative analysis, we divided the concepts described in the essays broadly into characteristics related to (1) the socio-technological setting in which the playing takes place; (2) the stakeholders' individual disposition including personality and player relationships; and (3) situational drivers, meaning events and actions that transpire during gameplay. As an important meta-level implication, our findings raise concerns regarding the lack of a universally shared view on toxicity, which were visible even with the rather homogenous sample of participants in this study. Keywords 1 Gamer toxicity, toxic behavior, League of Legends, video games, multiplayer online games 1. Introduction of Tahm Kench’s abilities, which is an indicator of toxic escalation). Using R (the most important ability of champions in the game), Fynn teleports After a long day of work, Fynn comes home, them both directly behind the enemy turret, killing takes off their jacket and riles up an old desktop both players almost instantly (illustrating sincere computer. They click open the Riot client and start behavioral toxicity). After this ordeal, Fynn reads playing League of Legends (one of the most a new message in the chat. It is the midlaner: popular eSports titles at the moment). While in the “fucking donkey”. matchmaking queue (a pre-game environment to The above description is a typical example of decide what champion to play), Fynn envisions toxicity in League of Legends. Perhaps starting dominating the game with their favorite from a misunderstanding or a minor provocation, champion, Galio, and naturally, Fynn expresses to team members end up spoiling each other’s game their teammates intention to pick this champion. through both in-game actions and messages in the But oh no - a player from their own team bans chat. Industry stakeholders as well as academic Galio x(due to a communicative researchers have studied this phenomenon misunderstanding)! Angry, frustrated and extensively (see, e.g. [5, 7, 24]), and designed disappointed by this, Fynn starts plotting revenge various counter measures for curbing such picking Tahm Kench. When the game starts, Fynn negative behaviors, including both (1) proactive levels up top lane Tahm Kench normally, until measures, such as removing certain interaction reaching level 6 (reaching a relevant power spike opportunities or offering players the option to of champions within the game). Fynn then walks shield themselves from unwanted actions [28], to the midlaner (who banned Galio in the and (2) post hoc measures, such as allowing matchmaking queue) and eats him up (using one players to report malicious actors [20, 22]. 7th International GamiFIN Conference 2023 (GamiFIN 2023), April 18-21, 2023, Lapland, Finland EMAIL: Kordyaka@Uni-bremen.de (A. 1) ORCID: 0000-0003-3495-6855 (A. 1) ©️ 2023 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0). CEUR Workshop Proceedings (CEUR-WS.org) 86 Despite these extensive efforts, toxicity remains a systematically observed what are the most huge challenge in not only multiplayer online pertinent components of drivers of gamer toxicity, games such as League of Legends, but also and whether there are outstanding fundamental discussion forums and other online platforms epistemic or ontological differences between the where people meet each other. A good example of players’ thinking. In order to guide this research a recent development is the Zero Harm in Comms including data collection and analysis, we thus project, an industry-driven initiative that seek to propose the following research question (RQ): develop AI tools among other solutions for mitigating gamer toxicity2. RQ: What factors from a gamer’s perspective The first step to solving a problem is lead up to the occurrence of toxicity in multiplayer accurately defining it. Rooted in theories of online games? cyberbullying [2–4] and nurtured by newly arisen technological opportunities to interact with others Through answering the RQ, we demonstrate in real-time [13], online toxicity (or toxic how gamers perceive the various factors behavior) is characterized as a prominent, yet still influencing the emergence of toxicity. We also unresolved challenge in a variety of video games, look at the differences between players in their such as multiplayer online battle arena games thinking and show that there is subjectivity (MOBAs). Toxicity is generally understood as an involved in the interpretation of toxic intent. umbrella term for negative behaviors in These findings have important implications on multiplayer video games [1]. In contrast to better both academia and industry, such as highlighting established and understood concepts such as the importance of communication for neglecting cyberbullying and online harassment, toxicity is false positives in toxic intent interpretation. The of short duration, non-systematic and fueled by rest of this study is structured as follows. First, we situational frustration and anger and the high present our research methodology followed up by levels of real time competition [16]. The toxic the findings. We then discuss the key results and behavior has various forms of expression such as position our work back to real life situations in insulting, criticizing, resource stealing, and which gamer toxicity takes place. We conclude external attribution which are dependent on the the study by discussing the limitations and future perpetrator’s actions, the players’ subjective research directions. interpretation of these actions, and the affordances of the online platform where the interactions take 2. Methodology place. Within these tensions, toxicity is generally accepted as negative and the umbrella of toxic behaviors are associated with decreased positive As a methodological guide for our data player experience and game atmosphere, and in collection and analysis, we selected the Gioia method [12]. This method makes a few the worst cases, enduring toxicity can even affect players’ mental health [21]. assumptions that are important to clarify. First, the While previous research has looked at toxicity method assumes that the participants are experts in various settings and through multiple on the topic, and as such, their views and opinions theoretical lenses, deriving insights related to are not critically evaluated in the analysis. This is relationships between social exclusion, and group a distinction over alternative methods (e.g. [8]), norms [10, 11], the role of social identity [18, 29], where the participants’ views are debated, team composition [25], measurement instruments challenged and reflected against existing [15] and many more, it remains unclear to what knowledge bases. In our case, since we were degree the academic understanding of the concept specifically interested in discovering the matches with players’ lived experiences and the participants’ views on toxicity, the assumption of conceptions that gamers have regarding toxicity. participants as knowledgeable agents was To address this research gap, in this study we sensible. Second, the Gioia method is an inductive gathered structured essays from gamers, where method, where the data is coded, the codes are they explain on a deep level how they understand then grouped together, and finally connected to the emergence of toxicity in online video games. theory-guided aggregate dimensions. Because of Through the analysis of these essays, we then this streamlined approach, the method has been called “template-based” and “procedurally 2 Zero Harm in Comms industry research project for mitigating ubisoft-tackling-toxicity-in-games-with-new-project, visited gamer toxicity: https://www.riotgames.com/en/news/riot-games- January 8, 2022 87 rigorous”, but also criticized for the lack of for research. Those students who did not give interpretive rigor [23]. In our case, the clear permission were assessed for the course, but not analysis procedure provided a framework within included in this study. Participants were explained which we could compare individual differences that declining to partake in the research had no between the participants. Despite qualitative impact on their grade. research being inherently interpretive, the Gioia As the content of the course from where we method helped bring structure and hence collected the essays was designed around objectivity in the otherwise multi-layered and examples from the game League of Legends, we iterative sense-making process. Regardless, the suggested the students also use the game as an analysis process was iterative, and the authors example in their essays, but this was not debated and refined the data structure multiple mandatory. After collecting the essays and times through reasoning, interpretation and grading them, the essays which students had given discussion along with increasing familiarization permission to use in research were anonymized with the data. Continuing with the Gioia method and shared with the rest of the research team for [12], we next describe our data collection, analysis. Altogether out of 18 participants, 16 introduce profiles of the research participants, and gave permission to use their responses in research. describe the analysis process. Half of the students (n=8) were female, and the age range of participants was between 20-38 (M 2.1. Data collection =26.06, SD = 4.78). All participants were familiar with video games, were third year students, and had been exposed to academic definitions of In order to address our RQ, we collected data toxicity during the university course. All of the 16 from a sample of university students in the form participants also received a passing grade, with no of written essays (three pages or ~2000 words). signs of plagiarism or computer-generated The advantages of having a sample of university responses detected in their essays. Students were students over anonymous samples were the given the choice to write the essays in either following. First, as the assignment was evaluated German or English, and we received essays in and participants were scored based on their both languages. essays, they had an additional incentive to provide thoughtful and thorough essays. This is an important distinction to alternative data collection 2.2. Data analysis methods such as Prolific or MTurk samples, where the users are incentivized to simply return The data analysis proceeded following the passing works as fast and efficiently as possible Gioia method (Gioia et al., 2013) as follows. First, with emphasis on producing not more than we labeled the essays with a number P1-P16. We passing quality. Second, the participants of our shared the essays with the research team and study were exposed to teaching about toxicity, proceeded with familiarizing ourselves with the which gave them time and tools to conceptualize material by reading the essays. In this step we the phenomena and potentially also express using made notes of interesting remarks, potential codes the scientific theories and understanding of the or concepts related to the RQ. Next, we went topic. Simultaneously this strength could also be through the essays, coding passages that discussed a limitation, as the teaching the students received a specific concept related to the RQ, such as related to e.g., the online disinhibition effect could frustration, provocation, social norms, self- have also guided their thoughts to a more regulation and losing. At this stage we were not narrowed direction. worried about looking at individual differences, The instructions for the essays that the students but our concern was on identifying all unique wrote were as follows. After a lecture on using concepts mentioned in the essays. Some of the gamification to address toxicity in online codes were not clear, as students did not explain environments, we asked students to write about their thoughts in a way that could be condensed their personal experiences and understanding of into a few words or into one. In these cases, we online toxicity, and to enumerate what they highlighted complete sentences, or in a few cases thought causes online toxicity. Students were even paragraphs. Altogether, in the first step we required to write at least three pages and were identified 30+ codes that describe the asked whether they would provide us the participants’ understanding of gamer toxicity, and permission to use their responses anonymously factors leading to the emergence of it. The coding 88 process was done by the first author due to a themes of game affordances, game context, social language barrier, and key quotes were translated norms, and real-world environment. and shared with the rest of the team. In the second step of the analysis, continuing 3.1.1. Game affordances to follow the Gioia method, we grouped the 1st level concepts together based on similarity to form 2nd order themes. This was done together by The first theme that showed itself described affordances of the game located on a level of the first three authors, who discussed the data technology design. Specifically, several of the structure and framework on multiple occasions to form themes that best describe the data. This participants mentioned manifestations such as the process was iterative, and the authors adjusted the chat function and pinging during games, where themes and the grouping multiple times. As an sometimes no clear distinction can be made here outcome, we ended up with ten 2nd order themes, between normal communication and toxicity. The which are described in Figure 1. two subsequent passages from P3 and P7 describe As the third and final step, Gioia et al. (2013) corresponding instantiations: describes that the authors should take their “It's always a dilemma in ranked games to choose between more communication by not findings towards a more theoretical direction and muting the chat and more toxicity or less toxicity connect the 2nd order themes to abstract and worse communication by muting the chat. I aggregate dimensions. For this step, we looked at factors related to (1) the setting, meaning things don’t really have an appropriate answer to this related to the game or platform, social norms or challenge.” (P3). the real world environment in where players sit “…another challenge is that there is often no when accessing online content; (2) individual’s consistent use of the ping command, which leads disposition, meaning things such as personality, to a variety of misunderstandings and ultimately motivation to play and possible relationships with to irritation and toxicity.” (P7). other players; and (3) situational drivers, describing things such as emotions that spark 3.1.2. Game context during gameplay, in-game events (winning or losing) and between-players interactions. All ten Another relevant theme here was the game 2nd order themes could be connected to one of context comprising concepts such as the ranked these three dimensions. game mode and its competitive environment that had an impact on the likelihood of experiencing 3. Findings toxicity in different roles during gameplay, which showed itself in statements such as the following: Through the analysis process of the Gioia “…since the ranked game mode is very method, we discovered multiple drivers of competitive by nature, the stakes are high as toxicity, which we ultimately sorted into ten 2nd players invest a lot of time and effort into order themes and further into three aggregate improving their gameplay and climbing the ranks. dimensions. As we discuss the emerging themes, This high-pressure environment can lead to we do so under the three above-mentioned players becoming more toxic.” (P1). aggregate dimensions. We present some Another relevant notion that emerged were illustrative passages from the participants’ essays, characteristics of the solo queue game mode, which was mediated by the present anonymity in which are direct quotes in case the essay was written in English, or translations made by the the game. authors in case the essay was written in German. “In solo-queue players always get frustrated if they do not get the role they want during the champ selection process before the games. As a 3.1. The toxic setting consequence, the perpetrate toxicity before the game has even started.” (P10) The first aggregate dimension that emerged was the toxic setting, which sets the boundaries of 3.1.3. Social norms the game and events within it, and consequently, also toxicity. This dimension refers to events that are taking place before playing the game. These References to the social surrounding were events are rather static, and influenced by the frequent in the essays. This was expected, as 89 gamer toxicity remains an inherently and dimension, but this quote also introduces the idea holistically social phenomena. In relation to the that real world social and ethical norms are less theme at hand (the setting where toxicity occurs), relevant, or not relevant at all, in certain online P11 expressed their thoughts about the social environments. For example, in League of Legends influence as follows: the developer takes a strong stance in dictating “That [the online disinhibition effect] what kind of behavior is acceptable in their game, potentially leads to social and ethical norms being becoming the ultimate arbiter of socially ignored online. Another important factor is the acceptable behavior in the online environment. absence of education about online behaviour and Here we noticed that some participants were communication.” (P11) against the idea that platform owners would have “In my own experience the lack of such power over people (P3, P7), while others felt consequences for toxic behavior is a sincere that it was necessary for the developer to take a problem that can be even considered an accepted stance and interfere with toxicity, even more part of the game related culture.” (P12) strongly than what they do currently (P2, P13). The communication here refers to phenomena discussed further in the third aggregate Figure 1: The results of the qualitative analysis encouraged by Gioia et al. (2013) physical world (such as network latency issues, 3.1.4. Real world environment lighting of the room, interference by roommates) can translate into emotions and actions that players experience in the online environment. In Furthermore, the dimension encapsulates the the essays participants discussed various ways real-world environment. As people go online, they consider the environment before playing, to they are still simultaneously present in the reduce interruptions that may lead to toxicity, but physical world, and events happening in the 90 to also provide them with the adequate tools to personality that players carry with them to games, deal with toxicity if it were to arise during a and which are not subject to change in the short match. For example, P7 and P 14 wrote the term. Related to this theme a substantial part of following: participants wrote about the influence of the “During matchmaking I always use a process personality of players affecting toxicity as P10 consisting of three steps: first, I make sure to pick and P14 stated: a suitable champion in relation to the opponent “Players have different personality and the own team; second, I make sure I have the characteristics that hurt or make other players right runes selected; third, I select the mad. As an example, if you are a very extroverted appropriate summoner spells.” (P7) person this might increase the likelihood of “…before every game session I mute my phone portraying toxicity during games.” (P10) to make sure I don’t get interrupted.“ (P14). “Some players just lack resiliency to deal with Another example comes from P1, who wrote challenging moments of conflicts during games, about how they prepare for games by checking which oftentimes leads to toxic behavior.” (P14) their settings: “To avoid the problem [of having to endure 3.2.3. Social relationships toxicity], I make sure that my chat- and ping settings are accurate in relation to if I play normal Furthermore, social relationships occurred as or ranked.” (P1) another relevant concept that occurred. Accordingly, participants mentioned that social 3.2. Individual pre-dispositions that relationships are one relevant predisposition as guide actions and reactions well, regarding the likelihood of the occurrence of toxicity and the potential to deal with negative The second aggregate dimension that emerged situations. Interestingly, some even stated that were individual pre-dispositions that guide player they experienced higher levels of toxicity playing actions and reactions during games that may lead with friends (opposed to strangers): to toxicity. In accordance with the first dimension, “As I played these games with my friends, we events are rather static, and influenced by the steadily improved and with that my ambition themes of playing motivation, personality, and grew. In this situation and similar situations, it is social relationships. easier to be toxic, as you know the other players.” (P15) However, we found the complementary 3.2.1. Playing motivation relationship in our data as well: “One of my former boyfriends introduced me The first theme, playing motivation, that had to the game and we played hundreds of hours in an impact on the likelihood of experiencing duo queue together. Since I knew him quite well, toxicity in different roles during gameplay, which it was much easier to avoid misunderstandings in showed itself in statements such as the following: the game and it happened very rarely one of use “The motivation before a game is a complex carried out toxic perpetration.” (P16) topic but definitely has an influence how sensitive I will react in relation to situations that drive me 3.3. Situational factors triggering mad.” (P13). “During the end of every season I want to toxicity improve my Elo level. As a consequence, my motivation is much more achievement related and The third aggregated dimension, situational I react to losses much more sensitive, which factors, referred to events that happen during the (probably) shows in my own toxicity game. These events were highly dynamic and perpetration.” (P5) comprised the 2nd order themes in game events, emotions, and perceived interactions. 3.2.2. Personality 3.3.1. In-game events The second theme, players’ individual predispositions, described a rather static pre-given Multiple participants expressed in their essays characteristics of individuals such as their how frustrating in-game events such as losing a 91 match, dying, others not following could simply be the result of the team losing communication or being provoked by the enemy (which happens roughly 50% of the time). The team were often the catalysts for toxicity. As a participants also talked about insulting, a specific rationale, participants stated that players feel form of toxicity, which was one of the most often greater pressure to perform and can become mentioned expressions of toxicity. The following frustrated when their team does not perform as two quotes highlights this: well as they would like due to events during the “Another well-known way of Insulting is game, which can lead to higher levels of toxicity (obviously) insulting the enemy team if they killed in communication between players, such as someone or won the game itself or even if one of blaming others for mistakes. the enemies or the whole team got outplayed in an “The sad thing about ranked games in League unexpected way” (P8). of Legends is that the outcome often depends on “What really drives me mad is behavioral just a few key moments. For example, a baron toxicity I experience during gameplay such as if fight after 30 minutes is often game-changing. others steal my experience by stealing camps in Accordingly, it's hard to understand why players my jungle.” P3 don't listen to communication when preparing the target, but just farm somewhere on the map.” (P9) 4. Discussion “As a top laner it is really annoying if you have three AP champions on your team and the 4.1. Key findings opponent still buys lots of armor. As a consequence, you’re pretty useless then and need Through our analysis of student essays from a to burn off some steam.” P12 rather homogenous sample of gamers (n=16) we identified ten 2nd order themes that are relevant 3.3.2. Emotions in the emergence of toxicity in online multiplayer games, which we then connected to three aggregate dimensions that all show references to Another important theme that was ubiquitous previous work dealing with toxicity: (1) the in the essays related to the situational drivers were setting in which toxicity occurs comprising game players’ emotional states. Triggered by the above- related affordances and game content, social discussed frustrating in-game events, or possibly norms, and real-world environment [6, 19, 26]; (2) things that occur offline such as a boyfriend individual dispositions consisting of motivation, nagging or having poor internet, participants players’ personality, and social relationships [14, connected the resulting negative sentiment to 17]; and (3) situational drivers such as in-game triggers of toxicity and subsequent malicious events and interactions that transpire between actions. The following two quotes’ passages players such as in-game events, emotions, and illustrate these ideas: interactions [9, 27]. “People get easily frustrated if the game does We now return to the illustrative story not go how they expected it to go. That happens presented in the Introduction section. In Figure 2, especially in higher ranked competitive games we show how the initial perpetrator of the story which can have very long queue times and losing may have banned Galio from Fynn out of (a) such games multiple times in a row because of malicious intent, (b) simply being clueless someone else’s (they themself always play regarding the situation, or (c) through another perfectly!) is frustrating and that frustration can reason which Actor 1 failed to communicate to turn into anger” (P8) Fynn. The action of banning Galio can be “The possibilities of spreading toxic behavior interpreted by Fynn (Actor 2) in multiple ways. via an anonymous account and thus letting out For example, they can give Actor 1 the benefit of frustration, stress and suppressed feelings are the doubt and assume a positive interpretation of manifold.” (P4) the action such as that Actor 1 banned Galio as they were afraid the opposing team would steal it. 3.3.3. Perceived interactions On top of the intention and interpretation, the actors can choose to suppress or commit to their Connected to the negative emotions was the impulses for actions. In Figure 2, we show how idea that toxicity was provoked in some way or the three aggregate dimensions (to which our another due to interactions during the game. The second order themes, and consequently the 1st provocation did not have to be intentional and order concepts relate to) can be used to explain 92 this situation. First, we have the setting (e.g., the human interactions and relationships inherently game and the affordances) that dictates the indicate behavior. Participants agreed that interactions at a high level. Nested inside this are relationships and personality were critical factors the actors and their interactions, which are in explaining gamer toxicity. Furthermore, these impacted by the individual dispositions. There are factors are by large out of the developers’ control, then the events and situational drivers that meaning that developers need to compensate in transpire during games, that all ultimately their platform things that are fundamental human contribute to the actions (toxic or not) that players issues by imposing rules and regulations for fair take during the game. play and behavior. They also need to reinforce those rules, which may lead to various issues. For example, even in our homogenous sample not all participants agreed on what was toxic and what was not (see the first dimension). Furthermore, games such as League of Legends are played globally, with players coming from various cultural background and having potentially very different behavioral expectations and understandings on what sort of behavior is Figure 2: Relationships aggregated dimensions allowed. All these factors combined; this dimension showcased aspects related to 4.2. Implications for research and individuals’ predisposition and factors prima facie practice disconnected from the technology platform, that still need to be accounted for and dealt with by the developer. In this study we sought out to better understand The quotes regarding the third aggregated drivers of gamer toxicity through an analysis of 16 dimension suggest that players are creative in essays that provide some added value for research making use of various affordances in behaving in and practice. Our purpose was not to produce a a toxic fashion. For the victims, this is a difficult new definition, but rather, to map and elucidate situation as it is almost impossible to shield the various circumstances that are relevant in the oneself from all the possible expressions of emergence of toxicity. Through this approach, we situational toxicity. Even if the developer were able to elucidate 10 themes which could be punishes perpetrators retroactively, many of the broadly divided into three dimensions. toxic actions are not necessarily done with The quotes regarding the first aggregate malicious intention, hence punishing for such demonstrate the participants’ lived experiences behaviors would result in false positives. As when playing League of Legends, where they are players learn which malicious actions are actively preparing themselves for situations where punished and which are not, they gravitate toxicity may occur. While participants have some towards those actions that are not punished. For leeway in controlling the environment (e.g., example, currently we are seeing the chat being through arranging the offline environment and heavily regulated in League of Legends, which tweaking in-game settings), and even on a meta- has simply moved the toxic expression more and level selecting which game they play, when more to the in-game actions. committing to a match of League of Legends there Summarizing, through Figure 2, we are countless of environmental factors that are demonstrate how the discovered framework can beyond the participants’ control, such as who be used to explain the occurrence of toxicity in happen to be their teammates, what in-game League of Legends. These findings contribute to affordances there are and what are the social the literature on online toxicity [1,5,6,12,13,14] as norms and expectations of their teammates. Thus, follows: while there is personal responsibility involved in First, the findings suggest that as there is combating toxicity in terms of the setting where subjectivity involved in the interpretation of toxicity takes place, we cannot rule out the toxicity. To counter this, stakeholders should influence of other factors such as the game investigate strategies for improving player developer. communication, and to also identify situations in The given examples in relation the second which misunderstandings happen in the first place aggregate dimension highlight how fundamental (such as Champion selection screen in League of 93 Legends) to break the cycle of toxicity at an early toxicity?” According to our findings, it is the stage. interplay of the three dimensions a) the game Second, the findings show that much of the related setting, b) dispositions of players, and c) factors leading up to toxicity are beyond the situational factors that lead to actions that cause control of the developer. Furthermore, the current negative emotion and sentiment to other players. measures of developers (very strict chat rules, Participants in our data emphasized these interaction disabling, judgement and report dimensions to varying degrees, highlighting systems), may in fact overcompensate and step individual differences in understanding the beyond the boundaries of what the developer drivers of toxicity. We encourage future research should do, interfering with the territory of social addressing gamer toxicity to focus on dimensions norms and other broader characteristics of culture of drivers of toxicity rather than individual which arguably should be beyond the control of displays of actions such as swearing, stealing a individual tech companies. resource or leaving the game. Third, the findings illustrate that toxicity occurs in various places throughout even an 6. References individual match, and to various degrees, and that the actions and reactions of individuals contribute [1] Adinolf, S. and Turkay, S. 2018. Toxic to a complex dance of player interactions nested Behaviors in Esports Games: Player inside the game setting and influenced by individual predispositions. This suggests that Perceptions and Coping Strategies. instead of punishing individual acts of toxicity, Proceedings of the 2018 Annual Symposium malicious online behavior should be looked at on Computer-Human Interaction in Play more broadly. Companion Extended Abstracts - CHI PLAY ’18 Extended Abstracts (Melbourne, VIC, Australia, 2018), 365–372. 4.3. Limitations and future work [2] Ang, R.P. and Goh, D.H. 2010. Cyberbullying among adolescents: The role The empirical data collected for our research of affective and cognitive empathy, and consisted of 16 essays from a heterogenous group gender. Child Psychiatry & Human of League of Legends players, and accordingly, Development. 41, 4 (2010), 387–397. the final list of characteristics should not be [3] Baldry, A.C. et al. 2017. School bullying considered exhaustive. Despite this, we still and cyberbullying among boys and girls: identified differences in characteristics and views Roles and overlap. Journal of Aggression, that the participants expressed in their essays. Maltreatment & Trauma. 26, 9 (2017), 937– However, due to the limitations of the sample, 951. future steps of this research will include refining [4] Barlett, C.P. 2017. From theory to practice: the essay instructions and expanding the essay Cyberbullying theory and its application to recruitment to a larger audience. Furthermore, intervention. Computers in Human alternative strategies such as player interviews or Behavior. 72, (2017), 269–275. ethnographic observations could be used to [5] Beres, N.A. et al. 2021. Don’t you know that support and triangulate the findings of our you’re toxic: Normalization of toxicity in approach. Another limitation relates to the online gaming. Proceedings of the 2021 research setting being tied to the game League of CHI Conference on Human Factors in Legends. For the purpose of deriving a holistic Computing Systems (2021), 1–15. conceptualization of the factors impacting the [6] Beres, N.A. et al. 2021. Don’t you know that emergence of toxicity we encourage critical you’re toxic: Normalization of toxicity in studies between various environments that seek to online gaming. Proceedings of the 2021 identify which factors are specific to the context CHI Conference on Human Factors in (such as League of Legends), and which are more Computing Systems (2021), 1–15. universal. [7] Blackburn, J. and Kwak, H. 2014. STFU NOOB!: predicting crowdsourced decisions 5. Conclusion on toxic behavior in online games. (2014), 877–888. [8] Braun, V. and Clarke, V. 2019. Reflecting To conclude, we return to the title of this work, on reflexive thematic analysis. Qualitative and address the question of “What drives gamer 94 research in sport, exercise and health. 11, 4 [21] Kwak, H. et al. 2015. Exploring (2019), 589–597. Cyberbullying and Other Toxic Behavior in [9] Deslauriers, P. et al. Assessing Toxic Team Competition Online Games. Behaviour in Dead by Daylight : Proceedings of the 33rd Annual ACM Perceptions and Factors of Toxicity Conference on Human Factors in According to the Game’s Official Subreddit Computing Systems - CHI ’15 (Seoul, Contributors. Republic of Korea, 2015), 3739–3748. [10] Embrick, D.G. et al. 2012. Social exclusion, [22] Kwak, H. 2014. Understanding toxic power, and video game play: New research behavior in online games. Proceedings of in digital media and technology. lexington the 23rd International Conference on World books. Wide Web - WWW ’14 Companion (Seoul, [11] Gabbiadini, A. and Riva, P. 2018. The lone Korea, 2014), 1245–1246. gamer: Social exclusion predicts violent [23] Mees-Buss, J. et al. 2022. From templates to video game preferences and fuels aggressive heuristics: how and why to move beyond the inclinations in adolescent players. Gioia methodology. Organizational Aggressive behavior. 44, 2 (2018), 113– Research Methods. 25, 2 (2022), 405–429. 124. [24] de Mesquita Neto, J.A. and Becker, K. 2018. [12] Gioia, D.A. et al. 2013. Seeking qualitative Relating conversational topics and toxic rigor in inductive research: Notes on the behavior effects in a MOBA game. Gioia methodology. Organizational Entertainment computing. 26, (2018), 10– research methods. 16, 1 (2013), 15–31. 29. [13] Hamari, J. and Sjöblom, M. 2017. What is [25] Ong, H.Y. et al. 2015. Player behavior and eSports and why do people watch it? optimal team composition for online Internet Research. 27, 2 (Apr. 2017), 211– multiplayer games. arXiv preprint 232. arXiv:1503.02230. (2015). [14] Kordyaka, B. et al. 2023. Exploring the [26] Sengün, S. et al. 2019. Exploring the relationship between offline cultural relationship between game content and environments and toxic behavior tendencies culture-based toxicity: a case study of in multiplayer online games. ACM league of legends and MENA players. Transactions on Social Computing. (2023). Proceedings of the 30th ACM Conference [15] Kordyaka, B. et al. 2019. Perpetrators in on Hypertext and Social Media (2019), 87– League of Legends: Scale Development and 95. Validation of Toxic Behavior. (2019), 10. [27] Shen, C. et al. 2020. Viral vitriol: Predictors [16] Kordyaka, B. et al. 2020. Towards a unified and contagion of online toxicity in World of theory of toxic behavior in video games. Tanks. Computers in Human Behavior. 108, Internet Research. (2020). (2020), 106343. [17] Kordyaka, B. et al. 2022. Understanding [28] Stoop, W. et al. 2019. Detecting harassment toxicity in multiplayer online games: The in real-time as conversations develop. roles of national culture and demographic Proceedings of the Third Workshop on variables. (2022). Abusive Language Online (2019), 19–24. [18] Kordyaka, B. and Hribersek, S. 2019. [29] Tang, W.Y. and Fox, J. 2016. Men’s Crafting Identity in League of Legends – harassment behavior in online video games: Purchases as a Tool to Achieve Desired Personality traits and game factors. Impressions. (2019), 10. Aggressive behavior. 42, 6 (2016), 513– [19] Kordyaka, B. and Kruse, B. 2021. Curing 521. toxicity–developing design principles to buffer toxic behaviour in massive multiplayer online games. Safer Communities. (2021). [20] Kou, Y. 2020. Toxic behaviors in team- based competitive gaming: The case of league of legends. Proceedings of the annual symposium on computer-human interaction in play (2020), 81–92. 95