Unveiling the Positivity Bias on Social Media: A Registered Experimental Study on Facebook, Instagram, and X

Social media has transformed how people engage with the world around them. The positivity bias on social media, in particular, warrants in-depth investigation. This is particularly true as previous research has concentrated on one specific platform, Facebook. Based on a pilot study of 279 university students, this pre-registered experimental research used a mixed design to examine the positivity bias on Facebook, Instagram and Twitter/X. After recalling a personal event, 312 participants were randomly assigned to imagine telling this event to a group of friends and sharing it on social media (Facebook vs. Instagram vs. Twitter/X). Several characteristics were examined through repeated measures ANCOVAs including the texts’ valence and the usage of emoji. Contrary to the pilot results, we found no significant differences in how events were reported. We, however, found significant differences between platforms. Specifically, messages posted on Twitter/X had a more negative valence than those posted on Instagram and Facebook. The positivity bias, often associated with social media, may therefore not necessarily represent a bias per se, but only a form of positive self-presentation that is consistent across contexts. These findings highlight the importance of focusing on platform characteristics, such as architecture, affordances, and socio-cultural context. This study advances our understanding of the positivity bias across multiple platforms through a pre-registered experimental approach.