Verified by 
"I don't know a single reasonable person on the left." I've read this more than once on Twitter. More than once, I've also read: "I don't know a single reasonable person on the right."
Yet I seem to have conversations with more or less reasonable people on the left and right almost every day. Curious. How do we make sense of the fact that many folks sincerely (I assume) think everyone on the other side is unreasonable?
One explanation is that these folks are themselves unreasonable. They think that everyone who disagrees with them is unreasonable, simply because they disagree. But when I engage with these folks who think that everyone on the other side is unreasonable, they themselves don't seem to be extraordinarily unreasonable or irrational. They seem to be able to engage in some give and take in arguments, and even admit that they might be wrong on some points.
So the question becomes: How do fairly reasonable and rational people come to believe that everyone on the other side is unreasonable?
The possibility that I wish to explore here is that they have gotten themselves into an echo chamber. And the echo chamber is responsible for their skewed perception about the reasonableness of people on the other side.
Many have suggested that echo chambers are at least partly responsible for the political polarization we seem to see these days. And the idea has some intuitive appeal. If people consume only partisan news and opinion sources, they might not be able to see the other side of an argument very well. And if they only interact with like-minded folks, positions might drift apart.
But echo chambers aren't always as bad as we might imagine. In "The Wisdom of Partisan Crowds," Josh Becker, Ethan Porter, and Damon Centola find that, under some conditions, the social influence within echo chambers nudges group members toward more accurate and less polarized views.
Their study was limited. They considered only numerical, factual questions (such as how many undocumented immigrants currently reside in the U.S.), provided a financial incentive for accuracy, and did not explore the effects of argumentation. Instead, they studied what happened when participants saw an updated average opinion of members of their group in a series of rounds.
But, for the sake of argument, let's assume that even when people are not financially incentivized for accuracy, even when it is a matter of values rather than facts, and even when the form of social influence is argumentation, the members of echo chambers can drift toward more accurate and less polarized positions.
Even if that's true, echo chambers can still radically affect our impression of the other side—especially on a social media network, such as Twitter. And one of the main mechanisms for this is a practice many call "nut-picking".
Here's how that works:
Suppose (for the sake of argument) that 90 percent of people on the left are reasonable, and only 10 percent are unreasonable (by your own standards). Also assume that 90 percent of people on the right are reasonable, and only 10 percent are unreasonable.
And suppose that many people consume only their-side partisan news and opinion sources and interact on social media only with people who are on their side of the left-right continuum. That is, many people, on both sides, have gotten themselves into very tight echo chambers.
Also, assume that reasonable opinion on the other side is completely boring and not worth sharing, and unreasonable opinions on the other side are super exciting, downright threatening, and definitely worth sharing.
These assumptions might not match reality especially well. Perhaps more than 10 percent of the people on one or both sides are unreasonable, and perhaps very few people are as cut off from the other side as I'm asking us to assume. And maybe not every reasonable opinion is completely boring. But I'm considering an extreme case first, so we can see how the nut-picking dynamic works.
What would we expect, if all of these assumptions held?
Well, we would expect that almost all of the opinions on the right which members of a left-leaning echo chamber share with each other would be the craziest, most threatening stuff. And almost all of the opinions on the left that members of a right-leaning echo chamber would share with each other would be the craziest, most threatening stuff.
And that means, if you are in a tight echo chamber like this, every person on the other side that you are exposed to will come from the most unreasonable 10 percent of people on the other side.
Even if 90 percent of the other side is reasonable, we can come away with the impression that every single person on the other side is unreasonable. And that's because, when our side chooses which opinions to share, we dependably "pick the nuts."
Now we can relax the assumptions a bit, allow that more people on each side are actually unreasonable, allow that some reasonable positions get shared across echo chambers, and allow that almost no one's echo chamber is perfectly tight. And we would expect that the effect would be attenuated accordingly. But even with leaky echo chambers, people might still be left with the impression that the other side is far more unreasonable than they actually are.
In light of this, is probably good to ask ourselves the following question:
"How often do I directly encounter extremely unreasonable opinions on the other side? And how often do I encounter them because someone on my side has shared or quote-tweeted tweets, or linked to articles expressing such opinions?"
It might also be a good idea to explicitly include a disclaimer when we do want to share some nutty thing someone on the other side has said. Something like this:
"Of course this isn't representative of the whole other side, but this is something some people on that side believe."
And, if we're feeling especially diplomatic, we might even go out of our way to share some of the reasonable things people on the other side say.
