I wonder how one would go about actually quantifying any of this? The article tries to use two different studies to do this, but this approach seems flawed for

I wonder how one would go about actually quantifying any of this? The article tries to use two different studies to do this, but this approach seems flawed for the very simple reason that participants are not the same across both studies. Imho this whole problem is very comparable to being biased in general. We are all biased in some way or another and I like to think most of us are aware of this reality. But being aware of it does not simply translate to being aware of the full extent of the biases or lead to an outcome where the bias is "negated", if it is as easy as that, then barely anybody should be biased. In a way, it's like trying to define what goes on inside a blind spot, which by definition, is kinda impossible: If you'd be aware of your blind spot, it wouldn't be a blind spot anymore. I see the same problem with "filter bubbles", which in a way are just systemic biases very comparable how different social circles have different Overton windows. Most of these seem not to be forced upon people but are rather consequences of choices made based on personal biases/the Overton window often heavily influenced by in ones peer groups. While one might be aware of these choices, and how one reasoned them, that does not entail knowledge about the reality of other, maybe equally valid, choices. In that context, a lot of this feels more like a philosophical problem than an actual social one. People will, for the most part, rationalize their own choices as the "most valid ones". If they wouldn't, then they wouldn't have made those choices because only very few people make irrational choices on purpose for the sake of acting irrational.