Anyone Can Become a Troll: Causes of Trolling Behavior in Online Discussions (1702.01119v1)

Abstract: In online communities, antisocial behavior such as trolling disrupts constructive discussion. While prior work suggests that trolling behavior is confined to a vocal and antisocial minority, we demonstrate that ordinary people can engage in such behavior as well. We propose two primary trigger mechanisms: the individual's mood, and the surrounding context of a discussion (e.g., exposure to prior trolling behavior). Through an experiment simulating an online discussion, we find that both negative mood and seeing troll posts by others significantly increases the probability of a user trolling, and together double this probability. To support and extend these results, we study how these same mechanisms play out in the wild via a data-driven, longitudinal analysis of a large online news discussion community. This analysis reveals temporal mood effects, and explores long range patterns of repeated exposure to trolling. A predictive model of trolling behavior shows that mood and discussion context together can explain trolling behavior better than an individual's history of trolling. These results combine to suggest that ordinary people can, under the right circumstances, behave like trolls.

• The paper shows that situational factors like negative mood and discussion context, rather than fixed personality traits, trigger trolling behavior.
• It employs both controlled experiments and analysis of 16 million posts to demonstrate how prior trolling increases the likelihood of further trolling.
• The findings indicate that incorporating context-based moderation strategies can effectively mitigate the spread of online antisocial behavior.

Analysis of "Anyone Can Become a Troll: Causes of Trolling Behavior in Online Discussions"

The paper "Anyone Can Become a Troll: Causes of Trolling Behavior in Online Discussions" by Cheng et al. explores the situational factors contributing to trolling behavior in online discussions. This research challenges the notion that trolling is an inherent trait attributed solely to a select group of individuals with antisocial tendencies. Instead, it posits that situational factors like mood and discussion context significantly impact the likelihood of an ordinary individual engaging in such behavior.

Experimental and Observational Analysis

The paper employs a mixed-method approach, combining controlled experiments with large-scale observational data, to dissect the mechanics of trolling.

Controlled Experiment:

The experimental component investigated two key variables: mood and discussion context. Participants were subjected to a mood manipulation task via quizzes of varying difficulty levels, and then engaged in simulated online discussions seeded with either troll-like or non-troll comments. The findings reveal that both negative mood and exposure to prior trolling significantly increase the propensity for participants to post troll-like comments, demonstrating the significant role of situational triggers.

Observational Study:

Complementing the experimental results, the authors conducted a longitudinal analysis based on 16 million posts from CNN.com. This analysis confirmed that mood variations (correlating with diurnal patterns) influenced trolling behavior. Furthermore, the paper highlighted how negative mood effects can persist over time and spill over across discussions.

Key Findings and Model Development

The authors present significant insights into how trolling not only arises from internal dispositions but also from external situational factors. These include:

• Mood Influence: Negative mood states, often influenced by external factors such as time of day and day of the week, are key predictors of trolling incidents.
• Contextual Influence: The presence of troll posts in a discussion significantly raises the probability of additional trolling, indicating that negative social norms can rapidly spread and be reinforced.
• Predictive Modeling: The development of a logistic regression model emphasized that discussion context (e.g., previous posts) and recent user history are more influential in predicting trolling behavior than historical user data, suggestive of trolling's situational nature.

Implications

Practical: The findings suggest actionable insights for designing online platforms. Moderation strategies could be enhanced by incorporating behavioral signals indicating mood and context-triggered trolling potential, enabling more proactive moderation interventions. Features such as rate-limiting and context-based comment filtering could mitigate the spread of negative behavior.

Theoretical: The evidence challenges and extends theoretical frameworks of online behavior by emphasizing the fluid nature of trolling, shaped significantly by environmental variables. This contributes to a more nuanced understanding of online interactions beyond innate personality traits.

Future Directions

The paper opens avenues for further research into the dynamics of online antisocial behavior, particularly in examining the interaction between individual traits and situational contexts. Future work could explore deeper dynamics such as the role of anonymity, differing platforms, and cultural variations in trolling propensity.

Overall, Cheng et al. provide a comprehensive examination of trolling behavior, highlighting the complex interplay between individual disposition and situational context. This research underscores the transformative impact of environmental factors in shaping online behavior, promoting a paradigm shift towards understanding and managing antisocial conduct in digital communities.