Context Impacts on Confirmation Bias: Evidence from the 2017 Japanese Snap Election Compared with American and German Findings

Abstract

Much concern exists about individuals' tendency to favor attitude-consistent messages (confirmation bias) and the consequences for democracy; yet, empirical evidence is predominantly based on U.S. data and may not apply to other cultural contexts. The current three-session online experimental study unobtrusively observed Japanese participants' (N = 200) selective exposure to political news articles right before the 2017 Japanese snap general election. The research design paralleled an earlier U.S. study and a German study, which allowed direct comparisons of confirmation biases among the three countries. Japanese exhibited a confirmation bias, but it was smaller than the confirmation bias among Americans, though comparable to that of Germans. The extent of the confirmation bias among Japanese participants was influenced by individual media trust, which provides new insight into causes of these cross-country differences. Attitudinal impacts resulted from selective exposure, in line with message stance, and persisted for two days.