The way information is processed by decision makers is frequently at odds with predictions of theory. In well studied issues like climate change, evolution, GMOS, etc., decision makers' beliefs may fail to converge (and in fact may continue to polarize) despite exposure to an arbitrarily high number of signals. The first piece of information received may have an excessively high weight in posteriors (i.e., anchoring). Decision makers' beliefs may continue to shift after repeated signals like with advertisement and propaganda, although no learning occurs. Based on recent findings on working memory, this paper proposes a cognitively less demanding yet ultimately incorrect way to update beliefs to account for these cognitive biases. Specifically, if a signal is informative on multiple dimensions, and if individuals can hold only a limited number of items in working memory; then the order with which beliefs are updated across different dimensions may influence the posteriors, leading to some of the more ubiquitous and salient cognitive biases social scientists have encountered. The unifying model advanced by this paper suggests new avenues of research for a better understanding of how information is processed by individuals, as well as how persuasion occurs.
The use of social media as a tool for spreading political information is accentuated when traditional media outlets are under pressure from governments. One such case was the 2013 protests in Turkey; largely ignored by mass media, protesters took to Twitter to get their voice heard, despite threats and attempts at suppression. We analyze this interaction on social media in a global game setting. Our results show that an increase in connectedness of a country serves to bring in line the information revealed by the mainstream and the alternative media. An increase in connectedness may free the mainstream media by making their capture more costly for the incumbent, thus allowing all media outlets to share relevant information with the public. Conversely, it may also lead the government to capture the alternative media outlets as well as the mainstream media, thus preventing dissemination of information in the country completely. The relationship between press freedom (as well as voter welfare) and social media usage rates depends on the incentives faced by the incumbent, who may either increase or decrease pressure on traditional media in response. Finally, we examine a cross-country dataset to test the predictions of our model. We show that our theoretical hypotheses are corroborated by data, and are robust to various specifications and empirical methods.