Insights on New Information Consumption: Evidence from an Online Video-Sharing Platform
(A2020-63467)

AUTHORS

Amir Sepehri, Western University; SeyyedNasir HaghighiBardineh, Washington State University; Rod Duclos, Western University

KEYWORDS

Online content virality; Topic modeling; Information consumption

ABSTRACT

Interdisciplinary ideas are becoming more and more popular as they tend to push the edges of existing scientific knowledge. These ideas span across a variety of fields and feature several topics. Contrary to the intuitive account that the more interdisciplinary an idea the better it will be received, we show it may not always be the best to incorporate several topics when presenting an idea. We take a mixed method approach including field data, dictionary-based text analysis, topic modeling, and a lab experiment, to test our hypothesis. We find that the more topics an idea is comprised of the less views it will receive online. We show that perceived cognitive complexity mediates this effect and that the use of analytical language when presenting the idea buffers against this adverse effect. The effect is reversed for individuals high in Need for Cognition, emphasizing the dependence of this effect on the audience type. Implications for public speakers and online platforms are discussed.