Arguing against confirmation bias: The effect of argumentative discourse goals on the use of disconfirming evidence in written argument

This study explores the impact of argumentative discourse goals on confirmation bias in young adults. All participants were presented three types of graphical evidence: data supporting their initial view, challenging their initial view and ambiguous data that could be interpreted either way. They were asked to use the evidence to write argumentative essays before and after engaging in a chat-based dialogue with a partner who held an opposing view. Dyads were assigned to one of two argumentative discourse goal conditions: Argue to persuade or argue to reach consensus. At the posttest, participants in the persuasion condition were more likely to misinterpret evidence and less likely to reference their dialogue than peers in the consensus condition. Educational implications are discussed.