SOPhiA 2022

Salzburgiense Concilium Omnibus Philosophis Analyticis

SOPhiA ToolsDE-pageEN-page

Programm - Vortrag

Ideology-Based Trust in Information Sources A Bayesian Analysis
(Epistemology, Englisch)

Why/How does ideological misalignment create disagreement on factual issues?

It seems that people who are on opposite sides of an ideological spectrum disagree not only on ideological issues, but also on a. which sources of information are trustworthy and b. on factualissues. For instance, in the US, 75 percent of Republicans trust Fox News, while 77 percent of Democratsdistrust them (Gentzkow, Wong, and Zhang 2018, p. 2). On a factual matter, 85 percent of conservative Republicans do not believe that humans are responsible for climate change, while 79 percent of liberal Democrats do. (Kaplan, K. (2016, October 4). When it comes to views on climate change, liberals and Conservatives Are Still Worlds Apart. Los Angeles Times. Retrieved May 2, 2022).

What to make of this phenomenon? Much has been written in the literature on political episte-mology about so-called "deep disagreement" and the (arguably) resulting polarization.

In my thesis, I build and analyze a Bayesian model to make sense of it (specifically, I draw on Bovens and Hartmann 2004 and Duijf 2021). I find that trust in an information source can be influenced by ideological alignment. I refer to this effect as ideological (dis-)trust, in which trust in an information source is "piggybacked" by ideological (mis-)alignment. For the Bayesian agent, ideological trust erodes and emerges from the reception of testimony in a way that is strongly reminiscent of the confirmation bias. However, my model does not presuppose overtly irrational biases. Hence, my simple model can be interpreted as a rational reconstruction of the phenomenon: Ideologically misaligned people disagree about facts because they trust different sources. And this (ideologically based) trust can arise through rational belief updating.


Bovens, Luc, and Stephan Hartmann. 2004.Bayesian epistemology. OUP Oxford.

Duijf, Hein. 2021. "Should one trust experts?" Synthese,1-24.

Gentzkow, Matthew, Michael B Wong, and Allen T Zhang. 2018. "Ideological bias and trust ininformation sources." Unpublished manuscript.

Chair: Cristian Vulpe
Zeit: 10:00-10:30, 09. September 2022 (Freitag)
Ort: SR 1.003

Leon Assaad
(LMU Munich, Deutschland)

Testability and Meaning deco