Role of “Likes” and “Dislikes” in Influencing User Behaviors on Social Media

Two recent changes on social media platforms include (1) allowing users to mask the number of "likes" on others' posts, aimed at reducing social comparisons and consequent technology-mediated dangerous behaviors (TMDBs), such as disclosing private information, and (2) adding a "dislike" reaction, aimed at increasing engagement. Nevertheless, the effects of these changes on user behaviors are unclear. In this paper, we seek to address this gap by integrating risk-sensitivity theory (RST) and prospect theory. First, we explain that while masking others' "likes" may reduce social comparison, based on the "homeostatic violation" concept, people also make internal comparisons to expectations. Undesirable deviations from users' expectations of "likes" and "dislikes" (e.g., too few "likes" or too many "dislikes") can motivate TMDBs that users believe can alleviate the undesirable deviations. Thus, we argue that, in addition to social comparison mechanisms, there is an internal comparison mechanism that can motivate TMDBs. We test these claims via five randomized controlled experiments (total n = 1,594). Results show that, beyond social comparison mechanisms, receiving too few "likes" or too many "dislikes," compared to internal expectations, can motivate TMDBs. Moreover, we found that losses loom larger than gains as users are more likely to engage in TMDBs to avoid excess "dislikes" than to avoid deprivation of "likes." These findings make novel contributions to social media research and practice by pointing to an internal comparison mechanism as a potent motivator of TMDBs beyond social comparison processes and to the higher "toxicity" of "dislikes" than "likes" in terms of inciting TMDBs on social media.