Fake news zealots: Effect of perception of news on online sharing behavior
Abstract
Why do we share fake news? Despite a growing body of freely-available knowledge and information fake news has managed to spread more widely and deeply than before. This paper seeks to understand why this is the case. More specifically, using an experimental setting we aim to quantify the effect of veracity and perception on reaction likelihood. To examine the nature of this relationship, we set up an experiment that mimics the mechanics of Twitter, allowing us to observe the user perception, their reaction in the face of shown claims and the factual veracity of those claims. We find that perceived veracity significantly predicts how likely a user is to react, with higher perceived veracity leading to higher reaction rates. Additionally, we confirm that fake news is inherently more likely to be shared than other types of news. Lastly, we identify an activist-type behavior, meaning that belief in fake news is associated with significantly disproportionate spreading (compared to belief in true news).
Study specs
- Authors
- F t'Serstevens,G Piccillo,A Grigoriev
- Institution
- Maastricht University
- Discipline
- Social Science
- Year
- 2022
- Human Data Platform
- Prolific
- Source
- View Source DOI Google Scholar
Peer Review & Critical Discussion
Potential Selection Bias in 2023 Cohort
The participant pool shows a concerning overrepresentation of users from high-income demographics. Looking at Table 3, we can see that 78% of respondents had annual incomes above $75k, which significantly limits the generalizability of these findings to broader populations.
Non-naive Participants Issue
I've noticed a methodological concern regarding participant naivety. Given that Prolific users often complete multiple studies, there's a real risk that participants had prior exposure to similar experimental paradigms, which could confound the results.
RLHF Applicability to This Study Design
The implications for RLHF training pipelines are understated. If we accept the authors' conclusions about preference stability, this has direct consequences for how we should structure reward model training. The temporal decay effect described in Section 4.2 is particularly relevant.
Verify your expertise to join discussion
Create an account and verify your credentials to participate in peer discussions.