From Attention to Dialogue: Does Audience Engagement Reinforce Constructive Cross-Party Communication?

2026-03-20

AI Summary

This study investigates whether audience engagement on social media fosters constructive cross-partisan communication among political elites, thereby mitigating political polarization. Leveraging a dataset of 1.1 million cross-partisan interactions between U.S. state legislators and the public on Twitter/X from 2020 to 2021, the research combines large-scale text analysis, natural language processing, and causal inference methods. Findings reveal that highly visible cross-partisan interactions significantly increase legislators’ subsequent engagement and lead them to adopt discourse frames characterized by causal reasoning, subjective expression, and positive sentiment. These results uncover a positive feedback loop between public responsiveness and elite-level constructive dialogue, challenging the prevailing assumption that social media inevitably exacerbates polarization and offering new empirical insights into the dynamic evolution of political communication in the digital era.

Technology Category

Natural Language Processing: Discourse, Pragmatics & Argument MiningApplication Domains: Humanities & Computational Social ScienceCognitive Modeling & Cognitive Systems: Social Cognition And Interaction

Application Category

Social Networks and Social Media: Media and governance, opinion dynamics, filter bubbles, polarizationWeb Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web dataResponsible Web: Algorithmic accountability and transparency on the web

Abstract

While existing works have emphasized how elites shape mass opinion, we ask whether the reverse also holds: do audience reactions on social media actively shape elite behavior? We examine this question through the lens of cross-partisan interactions (CPIs), which can either foster deliberation or deepen polarization. Using a dataset of over 1.1 million cross-party retweets, replies, and mentions between U.S. state legislators and their audiences on Twitter/X (2020-2021), we first establish baseline patterns of engagement: Democrats gain modest engagement in replies and mentions, while Republicans often face penalties in direct cross-party interactions. Building on this, we show that audience engagement produces a feedback loop that conditions future elite behavior. Following highly visible CPIs, legislators are not only more likely to engage again in cross-talk, but also shift their rhetorical strategies. Engagement consistently promotes causal reasoning, subjective language, and positive-emotion framing in subsequent CPIs. These findings suggest a positive association between audience engagement and constructive cross-party discourse among elites, challenging overly simplified interpretations in the literature that emphasize social media as a primary driver of rising or falling polarization.

Problem

• audience engagement
• cross-partisan interactions
• elite behavior
• social media
• political polarization

Innovation

• audience feedback loop
• cross-partisan interaction
• elite behavior adaptation
• constructive discourse
• social media engagement