Network activated frames: content sharing and perceived polarization in social media.

The article focuses on how users' sharing behavior on social media shapes the frequency and type of content—referred to as Network Activated Frames (NAFs)—that peers observe, influencing perceptions of polarization and social media bubbles. Through original image-based conjoint experiments conducted in Argentina, Brazil, and Mexico, the study finds that partisan users share more content and different types of partisan frames than nonpartisans, leading to an overrepresentation of partisan perspectives in social media feeds despite diverse network connections. This amplification effect explains why users perceive ideological homogeneity and polarization (social media bubbles) even in networks with cross-cutting ties, highlighting the role of sharing frequency and content selection rather than network sorting. The research contributes a novel theoretical framework and methodological approach to understanding framing and polarization in social media across multiple countries.