Normalization of Toxicity (Publication)
“Don’t You Know That You’re Toxic: Normalization of Toxicity in Online Gaming
Project Summary
Purpose: To understand how players rationalize and justify toxicity as acceptable, and which players are most at risk of normalizing toxic behaviors in online games.
Methodology: We used a mixed-methods approach to assess how player traits (moral disengagement, online disinhibition, and aggression) are associated with perceptions of toxicity, and conducted a thematic analysis to explore their reasonings for reporting or not reporting toxic behaviors.
Definitions: * ‘Moral Disengagement’ (a process by which players rationalize engaging in immoral behaviors to avoid feelings of guilt), * ‘Online Disinhibition’ (the perceived freedom an individual feels in online environments to express themselves in ways they would refrain from exhibiting in offline settings due to decreased behavioral inhibitions).
My Contribution
Led my research team in developing the research questions and the study design.
Conducted a comprehensive literature review to provide the theoretical background for our study.
Found appropriate validated scales to measure our constructs and adapted them to fit our online gaming context.
Sourced toxic audio clips from Overwatch gameplay on Twitch (following a specific criteria) to use in our study investigating player perceptions of toxicity.
Composed the write-up of the results and discussion sections.
Presented our work as the lead speaker at the CHI 2021 conference (an A* rank conference with a 25% acceptance rate with over 5000 attendees, comprised of researchers and industry folk).
Process
Step 1: Deployed an online study where filtered participants (N=106) answered questionnaires about their demographics, gaming preferences and history, as well as trait inventories.
Step 2: Participants were asked to listen to the full duration of three audio clips (presented in random order) sampling social interactions from Overwatch, and were prompted after each clip to rate perceived toxicity.
Step 3: Participants stated whether they would report the behaviors if they were in that match (‘Yes’, ‘No’, ‘Unsure’), and were prompted to explain the rationale behind their decision to report or not report.
Step 4: Hierarchical regressions analyses (controlled for age, gender, Overwatch knowledge, and gamer identification) were conducted to explore relationship between traits and perceptions of toxicity.
Step 5: A thematic analysis was conducted to identify themes reflecting how players rationalize reporting or not reporting toxic behaviors.
Results
There is so much to see in the paper, but the key takeaways are that:
Moral Disengagement in Games and Toxic Online Disinhibition both contribute to reduced perceptions of toxicity in a competitive gaming context.
Players normalize toxicity as banter, as typical of games, as acceptable, and as not their circus.
Players with higher Moral Disengagement in games may be more susceptible to the cycle of normalization.