New Search

Export article
Open Access

Finding uncommon ground: Extremist online forum engagement predicts integrative complexity

Andrew L. Gregory, Paul K. Piff
Published: 19 January 2021
PLOS ONE , Volume 16; doi:10.1371/journal.pone.0245651

Abstract: How do interactions with an ideologically extreme online community affect cognition? In this paper, we examine whether engagement with an online neo-Nazi forum is associated with more one-sided, “black and white” thinking. Using naturalistic language data, we examined differences in integrative complexity, a measure of the degree to which people acknowledge and reconcile conflicting ideas and viewpoints, and contrasted it with Language Style Matching, a measure of group cohesion. In a large web scraping study (N = 1,891), we tested whether two measures of engagement and interaction with the community are associated with less complex, balanced cognition. Using hierarchical regression modeling, we found that both individuals who had been community members for longer and those who had posted more tended to show less complexity in their language, even when accounting for mean differences between individuals. However, these differences in integrative complexity were distinct from group cohesion, which actually decreased with our measures of engagement. Despite small effect sizes, these findings indicate that ideologically extreme online communities may exacerbate the views of their members and contribute to ever-widening polarized cognitions.
Keywords: Linear regression analysis / internet / language / emotions / semantics / Social influence / Social communication / cognitive linguistics

Scifeed alert for new publications

Never miss any articles matching your research from any publisher
  • Get alerts for new papers matching your research
  • Find out the new papers from selected authors
  • Updated daily for 49'000+ journals and 6000+ publishers
  • Define your Scifeed now

Share this article

Click here to see the statistics on "PLOS ONE" .
References (39)
    Back to Top Top