John Timmer writes:
Greetings -I've read a number of your papers regarding how people's cultural biases influence their perception of expertise. I was wondering if you were aware of any research on the converse of this process – where people read material from a single expert and, in the absence of any further evidence, infer their cultural affinities. I'm intrigued by the prospect of a self-reinforcing cycle, where readers infer cultural affinity based on objective information (i.e., acceptance of the science of climate change), and then interpret further writing through that inferred affinity.Any information or thoughts you could provide on this topic would be appreciated.Thanks,John
Am hoping others might have better answer than me-- if so, please post them! -- but here is what I said:
Hi, John. Interesting. Don't know of any.Some conjectures:a. I would die of shock if there weren't a good number of studies out there, particularly in political science, looking at how position-taking creates a kind of credibility aura or spillover or persuasiveness capital etc -- & how about how durable it is.b. There is probably some stuff out there on how citizens simultaneously update their beliefs when they get expert opinions & update their views on experts' knowledge & credibility as they get information from those experts that contradicts their beliefs. Pretty tricky to figure out the right way to do that even from a "rational decisionmaking" point of view!I wish I could say, oh, "read this this & this" -- but I haven't seen these things specifically or if I have I didn't make note of them. But there's so much stuff on confirmation bias, bayesian updating, & source credibility that it is just inconceivable that these issues haven't been looked at. If I see something (likely now I'll take note), I'll let you know.c. There's lots of stuff on in-group affinities & credibility & persuasion. Our stuff is like that. But I *doubt* that the interaction of this w/ a & b -- & the contribution of this feedback effect in generating conflict over things like societal risks has been examined. That's exactly what your interested in, of course. But I'd start w/ a&b & see what I found!--Dan
Here is something relevant to (b) in my response. It's pretty cool. Not sure whether there is more recent version or in fact whether it has been published or is under review. BTW, more answers still welcome!
Lauderdale, B.E. Bayesian Social Learning: A Model of Citizen Learning with Implications for Analyzing Survey Response. (unpublished, Apr. 24 2008), available at http://qssi.psu.edu/files/psunf-lauderdale.pdf