This study investigates how social media, specifically Twitter, shapes political discourse and facilitates communication between communities with different political orientations. The researchers analyzed two networks of political communication on Twitter, comprising over 250,000 tweets from the six weeks leading up to the 2010 U.S. midterm elections. They found that the network of political retweets exhibits a highly segregated partisan structure, with limited connectivity between left- and right-leaning users. In contrast, the user-to-user mention network is dominated by a single politically heterogeneous cluster, where ideologically-opposed individuals interact more frequently. The study attributes this difference to politically motivated individuals injecting partisan content into information streams, which are primarily aimed at ideologically-opposed users. Statistical evidence supports this hypothesis, showing that users who use hashtags with neutral or mixed valence are more likely to engage in cross-ideological communication. The findings highlight the role of hashtags in shaping the network topology and suggest that while cross-ideological interactions may occur, they do not significantly reduce political polarization.This study investigates how social media, specifically Twitter, shapes political discourse and facilitates communication between communities with different political orientations. The researchers analyzed two networks of political communication on Twitter, comprising over 250,000 tweets from the six weeks leading up to the 2010 U.S. midterm elections. They found that the network of political retweets exhibits a highly segregated partisan structure, with limited connectivity between left- and right-leaning users. In contrast, the user-to-user mention network is dominated by a single politically heterogeneous cluster, where ideologically-opposed individuals interact more frequently. The study attributes this difference to politically motivated individuals injecting partisan content into information streams, which are primarily aimed at ideologically-opposed users. Statistical evidence supports this hypothesis, showing that users who use hashtags with neutral or mixed valence are more likely to engage in cross-ideological communication. The findings highlight the role of hashtags in shaping the network topology and suggest that while cross-ideological interactions may occur, they do not significantly reduce political polarization.