Troll game

Department of Communication & Cognition, Tilburg School of Humanities & Digital Sciences, Tilburg University
Department of Cognitive Science and Artificial Intelligence, Tilburg School of Humanities và Digital Sciences, Tilburg University
Department of Communication and Cognition, Tilburg School of Humanities and Digital Sciences, Tilburg University
Department of Communication and Cognition, Tilburg School of Humanities và Digital Sciences, Tilburg University
Journal of Computer-Mediated Communication, Volume 24, Issue 6, November 2019, Pages 293–318,
Navbar Search Filter This issueAll Journal of Computer-Mediated Communication All ICA JournalsAll Journals sản phẩm điện thoại Microsite Search Term Search
search filter This issue All Journal of Computer-Mediated Communication All ICA Journals All Journals tìm kiếm input Search


The present study aimed khổng lồ expand our understanding of trolling interactions by examining 10,025 community-reported trolling incidents in the online game League of Legends to determine what characterizes messages sent by trolls, their teammates, & their opponents. To do this, we used a novel method blending content analysis and topic modelling. Contrary khổng lồ extant literature, our study of complete trolling interactions found striking similarities between teammates’ và trolls’ chats, with both displaying the negative traits (e.g., exclusionary language) typically attributed khổng lồ trolls. Findings also suggest that the transition from victyên ổn khổng lồ perpetrator can occur extremely rapidly. This has important implications for the labelling of actors in trolling interactions, for future studies inkhổng lồ the trolling cycle, và for theories of computer-mediated communication.

You watching: Troll game

As time progresses, our world is becoming increasingly digitalized. In 2013 alone, there were 145 million people globally who were self-described gamers and who played online games as often as 45 to 107 minutes per day (Digital Strategy Consulting, 2013). When this becomes problematic is when one discovers that not all of these players have sầu good intentions. These mal-intentioned people are often called “trolls,” và their behavior “trolling” (Buckels, Trapnell, & Paulhus, 2014; Cook, Schaafsma, & Antheunis, 2018; Fichman và Sanfilippo, 2014; Thacker & Griffiths, 2012). Recently, academia has begun to take an interest in this online phenomenon, providing definitions (Buckels et al., 2014; Fichman và Sanfilippo, 2014; Thacker và Griffiths, 2012), dissecting early cases of the behavior (Herring, Job-Sluder, Scheckler, & Barab, 2002; Luzón, 2011), & surveying or interviewing various parties involved in the act, from the trolls themselves (Cook et al., 2018; Thacker & Griffiths, 2012) to the moderators of the online communities in which they operate (Shachaf & Hara, 2010).

As a field of study, trolling is known for its multiplithành phố. Researchers from multiple disciplines have used myriad methods (e.g., vignette studies, surveys, interviews, case studies), và examined numerous populations (e.g.,, gamers, bloggers, general Internet users) to understvà the phenomenon (e.g., Cheng, Bernstein, Danescu-Niculescu-Mizil, và Leskovec, 2017; Luzón, 2011). Few studies, however, have examined how other people interact with trolls, và the majority of those that have sầu took an indirect approach, either by asking participants what they would vày in a trolling situation or asking them to reflect on previous trolling experiences (see Maltby et al., 2015; Thacker và Griffiths, 2012). The few trolling interaction case and corpus studies that exist suggest that what both bystanders và victims choose to lớn say has a major impact on the troll’s choices (Hardaker, 2010; Herring et al., 2002), but we still vì chưng not even know specifically what characterizes the messages of a troll versus the messages of anyone else in the interaction.

See more: Tai Game 5V5 Mobile Games Ranking, Tai Game 5V5 Games

To begin to lớn fill this gap, we looked at actual trolling interactions to lớn see how all of the actors involved behaved in real-life trolling interactions, by examining community-reported trolling incidents. We procured a data set of over 10,000 reported cases of trolling), ranging from assisting the opponent team to lớn using offensive sầu language, from the immensely popular online game League of Legends (Riot Games, 2014). Using this data mix, we searched for verbal characteristics of trolling interactions (features) which were inherent in the data and examined these khổng lồ see whether they matched the features identified by previous researchers. In this way, we compared & contrasted a multidisciplinary literature lớn victim-reported trolling situations in order lớn answer the following two questions:

RQ1: Do the features portrayed in trolling literature exist in actual trolling interactions?

RQ2: How are these features distributed ahy vọng the actors (trolls, victims, & bystanders) in the interaction?

To determine what was & was not considered trolling, we relied on the victim’s perspective sầu. If the person was reported, the behavior was perceived as trolling. Since we could not determine intent, we followed O’Sullivan and Flanagin’s (2014) flaming definition method, and categorized trolling behavior based on victlặng perceptions. Because of this and the exploratory nature of the study, our definition of trolling was quite broad & included both verbal & behavioral trolling types, defined by Riot Games (2012) as a negative attitude, offensive sầu language, verbal abuse, and assisting the opposing team. For the purpose of the present study, trolling was thus defined as direct or indirect verbal or behavioral aggression that was reported by a League of Legends player under Riot Games’ earliest trolling nomenclature (circa 2012), provided this aggression type had also been previously called trolling in gaming-context trolling literature (see Cook et al. <2018> và Thacker và Griffiths <2012> for complete lists).

See more: Until Dawn Global Game Awards Nominee 2015, Game Debate Far Cry 5

Existing trolling research

General features of trolling interactions

As explained earlier, trolling research has taken many forms, crossing disciplines, populations, and methods (see Table 1 for an overview). However, it has focused heavily on the person of the troll instead of trolling as a behavior. As such, even when looking at a wide variety of studies, many of the features present in the literature—personality constructs, motivations, emotions, tactics, and more—highlight only the troll, both personally & as a thành viên of the interaction. The current study looked at the messages of all members of the interaction in a gaming context—the reported troll, the members of their team (teammates, composed of one or more victims và one or more bystanders; typically four actors total), và the members of the opposing team (opponents, composed of bystanders; typically five sầu actors total)—to lớn see whether and how the features identified in the literature manifested in a real-life trolling interaction.