We showed respondents news excerpts from TVP and the private outlet TVN, randomizing the inclusion of the source label. This let us measure whether audiences prefer the content of one outlet over the other even without knowing the source, and how the brand label itself shifts trust in that content.
Post-reform, the TVP brand cue was null for both partisan groups, and blind condition preferences for aligned content disappeared as well. Polarization around TVP had collapsed — but trust didn't simply flip. PiS supporters didn't come to distrust TVP the way opponents once had.
Ashley Blum
new paper!
🚨New review🚨 Do warning labels on misinformation actually work?
In Current Opinion in Psychology, @dgrand.bsky.social & I show that the answer is YES: warnings can reduce belief & sharing by ~25-50%!
We also review moderators (eg more visible=bigger effect)
www.sciencedirect.com/science/arti...
Ashley Blum
Ashley Blum
🚨WP🚨
Conspiracy beliefs famously resist correction, right?
WRONG: We show that brief convos w GPT4 reduce conspiracy beliefs by ~20pp (d~1)!
🡆Tailored AI response rebuts specific evidence offered by believers
🡆Effect undiminished after 2mo
🡆Works even on entrenched beliefs
osf.io/preprints/ps...
Background: Under the PiS government (2015-2023), Poland's public broadcaster TVP became a widely acknowledged pro-PiS propaganda outlet. By 2021, 76% of Poles believed TVP favored the ruling party. Attitudes toward it split sharply along partisan lines.
Pre-reform, the TVP label had large, polarized effects on PiS supporters’ and opponents’ trust in content, with effect sizes much larger than what we had found in a similar US study on Fox News and CNN. Even absent the label, news audiences preferred the content of aligned sources.
