Welcome to the Nexus of Ethics, Psychology, Morality, Philosophy and Health Care

Welcome to the nexus of ethics, psychology, morality, technology, health care, and philosophy

Sunday, October 6, 2019

Thinking Fast and Furious: Emotional Intensity and Opinion Polarization in Online Media

David Asker & Elias Dinas
Public Opinion Quarterly
Published: 09 September 2019
https://doi.org/10.1093/poq/nfz042

Abstract

How do online media increase opinion polarization? The “echo chamber” thesis points to the role of selective exposure to homogeneous views and information. Critics of this view emphasize the potential of online media to expand the ideological spectrum that news consumers encounter. Embedded in this discussion is the assumption that online media affects public opinion via the range of information that it offers to users. We show that online media can induce opinion polarization even among users exposed to ideologically heterogeneous views, by heightening the emotional intensity of the content. Higher affective intensity provokes motivated reasoning, which in turn leads to opinion polarization. The results of an online experiment focusing on the comments section, a user-driven tool of communication whose effects on opinion formation remain poorly understood, show that participants randomly assigned to read an online news article with a user comments section subsequently express more extreme views on the topic of the article than a control group reading the same article without any comments. Consistent with expectations, this effect is driven by the emotional intensity of the comments, lending support to the idea that motivated reasoning is the mechanism behind this effect.

From the Discussion:

These results should not be taken as a challenge to the echo chamber argument, but rather as a complement to it. Selective exposure to desirable information and motivated rejection of undesirable information constitute separate mechanisms whereby online news audiences may develop more extreme views. Whereas there is already ample empirical evidence about the first mechanism, previous research on the second has been scant. Our contribution should thus be seen as an attempt to fill this gap.