Russ Roberts: recommendation

Transcript context

…I think I should have said it’s a bimodal distribution, that you go one way or another. Look at it this way: In the simplest Bayesian model, your views should be a random walk, that the recent evolution of your views shouldn’t predict where you’ll end up tomorrow. But that’s not the case, really, with anyone that I’ve ever met. There’s some kind of trend in your views. You’re either getting more fanatical, getting more moderate, getting more religious, more or less something. And that, to me, is one of the most interesting facts about human belief, is how hard it is to find belief as a random walk. So, what’s wrong with all of us? If you’re getting more moderate all the time, that’s wrong too. That’s a funny kind of, you could say, almost fanaticism, where you ought to say, “Well, I see the trend so I’m just going to leap to where I ought to be.” Then the next day, maybe 50 percent chance I’ll take a step back toward being more dogmatic or less moderate. But again, that’s not what we see from the moderates either. I wonder how much of it is the fact that it’s really convenient to have a system, gives you something to shove into the box. You’ve got this black box that you take the world’s events and you’ve decided how they should be processed. Then something new comes along, and you know how to deal with that because you’ve got this box; you’ve got all these great examples from the past. At some point for me, I just started thinking that maybe the box doesn’t work all the time. I think a lot of people love the box. It’s a great source of comfort, whether it’s religion or ideology or other things. Maybe there’s just something peculiar about me. When you’re younger, certainty is deeply comforting because the world’s a bit too complicated to deal with. It still is, but I’m just less certain. There’s also a more charitable interpretation of what you’re describing. Think of yourself as working through problems, which is fine. Working through problems takes some time. You can’t every day pick up a new problem. The problems you’re working through as you — I wouldn’t say solve them, but as you somewhat make progress on them — that’s going to give you some persistence in the deltas of how your beliefs change. I’m not sure — the pure Bayesian model might just be wrong. It’s so far from actual human practice. Maybe we shouldn’t just damn humans for not meeting it, but realize there are structures to how you work through things, and they are going to imply certain trends that go on for periods of time.…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.