The effect of self-disclosure in celebrity tweets on message processing, parasocial relationships, attitudes, and behavioral intentions

Abstract

This study examined the role of self-disclosure in celebrity tweets. Thirty-six fictional tweets created by the researcher were used as stimuli in a single-factor (disclosure: low v. moderate v. high) fractionated experiment. The 55 participants were told the purpose of the experiment was to better understand the use of Twitter by up and coming celebrities in the entertainment genre. After taking an entertainment preference questionnaire and reading each celebrity's biography, participants were instructed to view each tweet presented in a random order while psychophysiological measures were recorded to index real-time emotional processes engaged while viewing the tweets. After initial exposure to the tweets, participants were briefly shown each tweet again and answered various self-report items. Results show that the reading of one individual tweet is not motivationally relevant enough to influence arousal or valence. However, selfreport results indicate that moderate and high disclosure tweets, which are tweets conveying a celebrity's likes and dislikes or personal information, lead to higher parasocial interaction, which effects attitude and behavioral intentions. Practical implications, construction of celebrity tweets, and limitations of the study are discussed.