In 1974 a paper was published by Tversky and Kahneman entitled "Judgement under Uncertainty: Heuristics and Biases". It's an interesting study that deals with how people come to decisions when they're presented with information, and how inherent biases tend to leak into those decisions in predictable ways.
One example is asking people to judge a sequence of heads or tails generated by flipping a coin. Consider the two sequences HHHTTT and HTHTHT. On average, people judge the second sequence to be "truly" random, whereas the first sequence seems too organized. The trick, of course, is that these sequences are equiprobable. On the flip side (get it), if you present people instead with simply "3 heads and 3 tails" without specifying an order, they think that the coin is fair.
An even more fun (and paraphrased) example: imagine that there are 50 farmers and 50 engineers in a room, and I pick one at random. What is the chance that I picked an engineer? The answer is of course 50%, and that's what most people think when asked. Here's the part I find so fascinating - let's say I add some totally irrelevant piece of information to the problem statement: imagine that there are the same 50 farmers and 50 engineers in the room. I've chosen one at random, and the person is soft spoken and wears glasses. What are the odds that I picked an engineer? People no longer guess 50%. For whatever reason, that piece of information has skewed their perspective even though it's actually useless inferentially.
The other day I read over a nice blog post called "What the hell is going on", which deals with the consequences of information asymmetry in the marketplace. As I was reading it I was reminded of the lesson of Tversky and Kahneman - especially with respect to information/news outlets. When people receive information, even useless information, in the context of something that they're being asked to make a judgement about, they incorporate it.
A great example of this (in my opinion) came up recently during the whole "border emergency" in early 2019. It was widely reported that actually, the vast majority of apprehensions of both drugs and people occurred at ports of entry. From those on the left, this was clear evidence that a border wall was worse than useless - it was a misappropriation of resources that should be targeted at ports of entry. From the right, this was damning evidence that there was a massive tide of illegal activity coming across the border and we must do something about it. Disentangling the politics is of course not possible because both sides have an agenda, but here's the thing: it's actually evidence of neither.
The fact that there are a large number of apprehensions at one location doesn't necessarily say anything about the number of crossings at another, of course. And in fact it's not clear what it does say anything about, because we have absolutely no clue what the denominator is - how many got through? And on top of that, we have absolutely no information (at least from that statement) about the spatial distribution of the denominator, or the relative efficiency of apprehension after detection at a port vs over open land, or etc etc etc.
In other words, it's not useful information. It's saying that the person I drew at random is wearing glasses. That said, I think this is an interesting lens through which to view political polarization. We're constantly presented with information, only a small subset of which tells us anything, and put in a position of making some judgement. But what it really does (on average) is provoke our biases.