Confirmation biases in selective exposure to political online information: Source bias vs. content bias

ABSTRACT

The present work examines the role of source vs. content cues for the confirmation bias, in which recipients spend more time with content aligning with preexisting attitudes. In addition to testing how both source and content cues facilitate this biased pattern of selective exposure, the study measures subsequent attitude polarization. An experiment (N = 120) presented messages with opposing political stances, associated with unbiased or slanted sources. Software tracked selective exposure in seconds, and attitudes were measured before, immediately after, and two days after message exposure. Further, information processing styles were assessed. The confirmation bias emerged regardless of source quality. Information processing styles moderated the confirmation bias as well as selective exposure to messages from unbiased vs. slanted sources. Selective exposure reinforced attitudes days later.

KEYWORDS:

• Selective exposure
• confirmation bias
• source cues
• partisan media
• attitudes
• political communication
• online search
• polarization

Acknowledgements

The authors would like to thank Nick Polavin and Shan Xu for their help with data collection.

Notes

1 For example, Johnson and Kaye (2013, p. 1864) built on respondents’ lay ideas to capture a confirmation bias (selective avoidance) by asking survey participants “On a scale of 1–10 where 1 indicates not at all likely and 10 indicates extremely likely, how likely are you to purposely connect to online political sources that SHARE [CHALLENGE] your point of view on political issues?”

2 For example, the survey data that Garrett et al. (2013, p. 120) utilized to examine a confirmation bias in selective exposure included question wordings such as “politically liberal organizations, such as People for the American Way or MoveOn.org,” “politically conservative organizations, such as the Christian Coalition or the American Enterprise Institute,” “politically liberal news organization or blog, such as Alternet.org or DailyKos.com,” and “politically conservative news organization or blog, such as NewsMax.com or Townhall.com.”

3 Indeed, sequence effects were evident in the data. The lower an article lead was shown on the page, lesser time was spent on it, r = −.09, p < .001, and the less likely it was to be selected, r = −.07, p = .001. Hence, positions or sequence of messages also serve as a cue that guides selective exposure, which future work might examine further. Article selection and article reading times were correlated at r = .76, p < .001.

4 A topic such as within-subjects factor yielded an interaction with the within-subjects factor of attitude consistency because for the first displayed topic of health care, the confirmation bias fell short of significance with p = .15, while it was significant for all three other topics, with p < .039.

5 Aside from impacts relevant to hypotheses, a topic such as within-subjects factor had an impact because for the first displayed topic of health care, participants generally spent significantly more time on the first search results overview page to orient themselves (M = 24 s, SD = 12) than for the other topics (with averages ranging between 13 and 16 s for the other three topics). Thus, overall article pages’ reading levels differed by topic because initial examination of the overview took longer.