| Abstract: |
The confirmation bias is a well-known form of motivated reasoning that serves
to protect an individual from discomfort. Hearing opposing viewpoints or
information creates cognitive dissonance, and so avoiding exposure to, or
discounting the validity of, dissonant information are rational strategies
that may help avoid or mitigate negative emotion. Because there is often a
systematic thought process involved in generating the confirmation bias,
deliberation tends to promote this behavioral bias. Nevertheless, the
importance of negative emotion in triggering the need for this bias is an
underappreciated facet of the confirmation bias. This paper addresses this gap
in the literature by examining mood and the confirmation bias in the political
domain. Using data from two studies and three distinct decision tasks, we
present data on over 1100 participants (Study 1, n=611; Study 2, n=503) that
document the confirmation bias in different settings. Specifically, task 1
(Study 1) examines perceptions of opposing argument strength in a classic
confirmation bias task. Task 2 (Study 1) is a novel task that measures the
change in one’s perceptions and normative preferences regarding political
issues after receipt of a random issue-specific informational message.
Finally, Task 3 (Study 2) administers a Bayesian decision task to examine
one’s belief-updating regarding the truthfulness of factual political
statements after receipt of a noisy signal about whether the statement is true
or false. All methods (recruitment and sample size, hypotheses, variables,
analysis plans, etc.) were preregistered on the Open Science Framework. Our
data show evidence of a confirmation bias across the variety of tasks
administered, which covered distinct dimensions of belief and preference
formation. As hypothesized, the data show a strong increase in self-reported
negative mood states after viewing political statements or information that
are dissonant with one’s political ideology. Finally, while not as robust
across tasks, we report evidence that supports our hypothesis that negative
mood will moderate the strength of the confirmation bias. Together, these
results highlight the importance of mood response in understanding the
confirmation bias, which helps further our understanding of how this bias may
be particularly difficult to combat. Key Words: confirmation bias; sleep;
deliberation; cognitive reflection; motivated reasoning |