【公开学术报告】Do Expressive Behaviors Matter?

Do Expressive Behaviors Matter? The Effects of Broadcasters’ Visual Expressiveness and Emotion Transitions on Viewership Engagement in Social Mobile Live Streaming

Guest Speaker:     Yutong Guo (NUS)

Date & Time:        14:00-15:30 (Beijing Time), Wed. 16th, Nov. 2022

Zoom Meeting:     83042342540（Password: 216475）

Click the Link:      https://us02web.zoom.us/j/83042342540

ABSTRACT

A critical challenge faced by live streaming platforms and broadcasters is how to increase viewership engagement and monetization performance in terms of virtual gifts. This paper aims to investigate the underexplored but critical role of broadcasters’ expressive behaviors in influencing viewership engagement in terms of gift-sending behaviors in the context of social mobile live streaming. Drawing on the social psychological theories of interpersonal communication, we specifically focus on two aspects of broadcasters’ expressive behaviors – visual expressiveness and emotion transitions. By utilizing live streaming videos and viewers’ gift-sending data from a mobile-based social live streaming application, we algorithmically extract the features of human visual displays and facial emotions with computer vision techniques and empirically quantify their effects on virtual gifts given using multi-level mixed-effects model estimations. Our empirical analyses find that although broadcasters’ hands display does not relate to gift-sending behaviors, there exists a positive association between broadcasters’ face display and gifts given. Our result implies that a one-percent increase in the proportion of time with a face displayed is related to increased gift values by 0.026%. Moreover, we find that a one-percent increase in the weighted frequency of broadcasters’ positive (negative) emotion transitions is associated with a 0.29% (0.14%) increase (decrease) in the monetary value of gifts. Notably, these relationships primarily stem from emotion transitions to long-lived positive emotions and short-lived negative emotions. Our study provides novel empirical findings and associated theoretical explanations, as well as important practical implications for live streaming platforms and broadcasters.

Keywords: live streaming, viewership engagement, visual expressiveness, emotion transitions, econometric analysis