Online social networks have become essential to modern communication, shaping how individuals share information, express opinions, and engage. Platforms like Reddit facilitate large-scale discussions, enabling millions of users to participate in conversations about various topics. One area of interest for researchers is understanding how these online conversations evolve when users are asked to make moral judgments. This requires analyzing how the network structure of discussions changes over time, especially in communities where users are explicitly invited to assess others’ behavior. The subreddit “AmITheAsshole” (AITA) offers a unique case study for this, as it is centered around users posting personal experiences and asking the community to determine whether they behaved immorally in various situations.
In social judgment platforms like Reddit, interactions often involve conflicting opinions and polarized judgments. However, how these disagreements shape user interaction networks’ overall growth and dynamics is poorly understood. Traditional methods of social network analysis tend to focus on relationship-driven platforms, such as Twitter and Facebook, where users engage based on social ties. On Reddit, a content-driven platform, users interact based on a shared interest in the discussion, often with no direct relationship with one another. This raises the challenge of how to model the evolution of these conversations, especially in threads where moral judgment and disagreement are the main drivers of interaction.
Existing studies have analyzed the structural properties of growing social networks by examining factors such as clustering and average path length. However, these tools have often been applied to more static or slowly increasing networks. In fast-paced, judgment-oriented Reddit communities like AITA, traditional methods fail to capture the dynamic changes in user interactions. Previous research has demonstrated that interactions follow a predictable pattern on social platforms with fixed-length conversations, such as Twitter. Still, Reddit threads, which grow continuously over time, require new approaches for temporal analysis. Current tools for analyzing temporal network evolution need to address the nuances of user behavior on Reddit, where the content rather than relationships drive interactions.
Researchers from Infolab, Department of Information Technology, Uppsala University introduced a new method for analyzing Reddit’s unique conversational structure. Focusing on over 6,000 threads from the AITA subreddit, they aimed to model user interactions as a dynamic network, capturing how these conversations evolve. The team collected data from over 6.3 million comments posted between 2023 and 2024. They found that these networks grew differently from typical social networks, exhibiting structural properties not aligned with existing theories. For example, while most real social networks see a decrease in average shortest path length as the network grows, AITA threads showed the opposite, with path lengths increasing over time. This indicated that users were interacting in a less connected and more isolated manner as the conversation progressed.
The researchers focused on two main subgraphs within the Reddit threads: the “star” subgraph, consisting of users responding directly to the original post, and the “periphery” subgraph, where users engaged in more in-depth discussions with each other. The analysis revealed that the star subgraph grew two to three times faster than the periphery. This was attributed to the community’s guidelines, which incentivized users to post direct responses rather than engage in longer conversations. The team noted that more than 60% of participants in these threads contributed only a single comment, indicating limited engagement beyond the initial interaction. The star subgraph’s faster growth reflected this behavior, as users quickly responded to the post without engaging in deeper discussions.
Further analysis of the proposed method showed that Reddit threads in the AITA community had a clustering coefficient five orders of magnitude lower than what is typically observed in real social networks. This means that very few tightly-knit groups formed during these conversations, and most interactions occurred between users who were not otherwise connected. The average shortest path length increased over time, indicating that users were more likely to interact with only one other user rather than participate in a broader discussion. The analysis also revealed that most voting interactions, where users provided their moral judgment, occurred in the star subgraph, with only about 30% of voters participating in the periphery subgraph, where conversations were more prolonged.
The study provided notable insights into user behavior in judgment-driven communities, particularly highlighting the impact of disagreement on interaction dynamics. In threads with higher levels of disagreement, users were more likely to write multiple comments and engage in discussions rather than simply vote. The team measured this disagreement using Shannon entropy and found that users tended to discuss rather than vote in posts with high entropy, meaning greater uncertainty in the judgment outcome. This led to slower decision-making, with votes taking longer to be cast. In contrast, threads with low disagreement saw users cast votes more quickly and engage less in reciprocal discussions.
In conclusion, this research demonstrated that platforms like Reddit, where user interaction is content-driven, exhibit distinct structural properties that differ from relationship-based networks like Twitter or Facebook. By analyzing Reddit threads as growing networks, the study highlighted how user engagement patterns, driven by community rules and the level of disagreement in the discussions, shape the evolution of these networks. The findings provide important insights into how online communities function, particularly in spaces where moral judgments are central to user interactions. This research contributes to social network analysis and offers an understanding of the dynamics at play in online platforms that facilitate judgment-based discussions.
Check out the Paper. All credit for this research goes to the researchers of this project. Also, don’t forget to follow us on Twitter and LinkedIn. Join our Telegram Channel.
If you like our work, you will love our newsletter..
Don’t Forget to join our 50k+ ML SubReddit
The post Researchers from Uppsala University Analyze the Impact of User Disagreement on the Growth and Dynamics of Reddit Threads: A Case Study of the AITA Subreddit’s Evolving Network Structures appeared first on MarkTechPost.