Information Consumption and Social Response in a Segregated Environment: the Case of Gab

TL;DR

This study analyzes how users on Gab interact with COVID-19 content, revealing that they do not differentiate between reliable and questionable sources, which has implications for understanding misinformation spread and social polarization.

Contribution

It provides a detailed characterization of interaction patterns on Gab regarding COVID-19, highlighting the lack of user differentiation between source reliability and insights into information operation early-warning.

Findings

No significant difference in social response to questionable and reliable content.

Questionable and reliable sources differ in hashtag usage and topical focus.

User commenting behavior shows similar sentiment and lifetime for both content types.

Abstract

Most of the information operations involve users who may foster polarization and distrust toward science and mainstream journalism, without these users being conscious of their role. Gab is well known to be an extremist-friendly platform that performs little control on the posted content. Thus it represents an ideal benchmark for studying phenomena potentially related to polarization such as misinformation spreading. The combination of these factors may lead to hate as well as to episodes of harm in the real world. In this work we provide a characterization of the interaction patterns within Gab around the COVID-19 topic. To assess the spreading of different content type, we analyze consumption patterns based on both interaction type and source reliability. Overall we find that there are no strong statistical differences in the social response to questionable and reliable content, both following a power law distribution. However, questionable and reliable sources display structural and topical differences in the use of hashtags. The commenting behaviour of users in terms of both lifetime and sentiment reveals that questionable and reliable posts are perceived in the same manner. We can conclude that despite evident differences between questionable and reliable posts Gab users do not perform such a differentiation thus treating them as a whole. Our results provide insights toward the understanding of coordinated inauthentic behavior and on the early-warning of information operation.