Why Do We Share Disinformation? A Survey Experiment Using Physiological Measures

• In Event: Lyin' Eyes: Perceptions of Misinformation and its Consequences

Abstract

Exposure to online information is a part of daily life for almost everyone. People search for, create, disseminate, and exchange information online with unprecedented access. Inconveniently, engagement with most types of mis and disinformation seems to be higher than with truthful stories. In other words, false news spread farther, faster, deeper, and more broadly than the true facts. Democracies around the world have been conducting unsuccessful attempts to combat this phenomenon providing citizens with more information, labelling false or low quality information online or promoting media literacy campaigns. Recent developments in psychology might allow us to better understand why these attempts are not enough and what to do instead. Emotions shape attention, decision-making, attitudes, and behavior. So, why has the role of emotion been under explored when it comes to explaining why people share low quality information?

Of course having accurate information, both in substance and in style available is important. Even if there is no consensus on how well informed citizens need to be for the optimum functioning of a democracy, evidence suggests that: a) higher levels of political knowledge are better predictors of consistent political choices according to preferences and b) ignorance or lack of information is not as harmful for political decisions as false beliefs. Regarding the first point, highly informed citizens, for instance, do better at choosing political representatives in line with their own attitudes, they are more capable of linking their interests with their attitudes and show higher levels of political participation. Conversely, citizens holding false beliefs "can distort public debate and undermine people's ability to form meaningful opinions". Without disregarding these findings, it is important to point out that the way in which information is presented and the emotional responses it provokes play a significant role in information processing and motivation for action.

On a daily basis human beings are exposed to a myriad of stimuli. Either if they are consciously processed or not, all of them elicit a physiological reaction, both at a central and autonomic level. Physiological measures offer the possibility to tap into subconscious responses to stimuli such as news articles in real-time. High-arousal emotions seem to operate as motivational triggers for political engagement and political participation. Hence, they might play a similar role in the online environment as well. Experts have identified social transmission as one of the mechanisms involved in the diffusion of false facts. When it comes to the online environment, it is mostly humans and not bots who spread false content. This suggests there is a need to identify the psychological mechanisms behind the differential engagement produced by false information. A growing body of literature in the topic builds on the assumption that so called “fake news” are more likely to diffuse than hard news or corrective articles because they trigger strong emotional reactions (high arousal). However, even when there are sound reasons to believe this is true, there is not much empirical evidence supporting the assumption.

In this paper we explore whether the use of uncivil language increases the likelihood of social transmission of news through arousing readers, regardless of perceived accuracy. To this end, the study conducts an innovative survey experiment relying on new generation sensors to measure levels of arousal through physiological responses to news articles. Affective reactions are processed faster than conscious elaboration, making it difficult to capture them through self report. This innovative approach overcomes this problem and taps into the levels of arousal generated by each article. Preliminary results suggest that uncivil language increases engagement with news articles.

Author

• Paula Szewach, University of Essex