Loading...
Thumbnail Image

Date

2024-08-01

Journal Title

Journal ISSN

Volume Title

Publisher

Creative Commons
Except where otherwise noted, this item's license is described as Attribution 4.0 International

Objective: I evaluated 1) how well people can objectively discriminate fact-checked true and fake news headlines, 2) whether the order headline stimuli are presented to participants influences aspects of performance, and 3) what individual differences predict participants confidence judgments, discrimination, and response bias. Background: There are multiple verbal theories explaining what makes people susceptible to misinformation. However, many of these studies focus on perceived accuracy rather than objective accuracy. Accuracy and discrimination should be influenced by other stimulus effects involving the order in which headlines are presented to participants and the degree to which headlines are true or fake. Method: Stimuli used in this study consisted of fact-checked true and fake headlines obtained from Politifact.com – a fact-checking website. In study 1, the original headline source was removed and was included in study 2. In two studies, participants were first randomly assigned to stimuli order conditions (true-fake, fake-true, or random) and then rated sharing likelihood and confidence in judgments for 30 fact-checked true and fake headlines using a Likert-style scale. Finally, participants completed individual difference measures associated with misinformation susceptibility. Results: I found more similarities than differences in my studies. In general, participants were able to discriminate true and fake headlines, but not very well, when subjected to signal detection analyses. Interestingly, I found that participants were better calibrated in their confidence judgments and accuracy for fake headlines than true headlines. I suspect this was driven by decreases in discrimination participants exhibited for headlines labeled as definitely true. Mixed effects models found that order only predicted participants’ confidence judgments when the degree to which stimuli were true of fake was considered. Individual differences in conspiracy belief and political orientation consistently explained the variance in participants’ performance when in competition with other measures. Moreover, individual differences in participants’ cognition supported a classical cognitive account of misinformation susceptibility in study 1, while study 2 found support for an integrated motivational account. Conclusion: Findings revealed that while participants struggled to discriminate true from fake headlines, they were better calibrated and accurate with fake headlines. The effect of stimulus order on confidence judgments was dependent on the stimulus class. Individual differences in conspiracy beliefs and political orientation consistently influenced performance. Methodological considerations for future research are discussed. Keywords: misinformation, fake news, signal detection theory, individual differences, confidence, calibration, order, fact-checked, source, mixed effects modeling

Description

Keywords

Psychology, Cognitive., Psychology, Experimental., Psychology, Social.

Citation

DOI

Related file

Notes

Sponsorship