Hacker News Reader

Comment by dlkf | Hacker News Reader

IMO the conclusion "facts don't change our minds" is a stronger conclusion than the first two experiments show. On my reading, the first two experiments show that:

1. if I have a uniform/undefined prior (how the fuck should I know how risky/conservative firefighters are?)

2. and then I'm given an anchor

3. and then told the anchor is bunk

4. the anchor still affects me

But I suspect this hinges very heavily on the fact that our initial prior is basically non-existent. By contrast, if you:

1. picked a topic where I actually have some prior belief (What country is colder: Sweden or Germany?)

2. gave me some information "Germany is actually colder on average than Sweden because of a weird atmospheric thing that affects the nordics"

3. told me that 2 was BS

I highly doubt you'd be able to replicate 4.

View on HN

Specific facts are orthogonal to the actual underlying positions held, which are presented outwardly as other positions for the sake of political cover, hence the illusion of facts not changing minds. What's needed is an understanding of the actual underlying, usually hidden, positions, then present facts to disrupt those positions.

This is unclear to me. Can you explain what you mean in the context of my alternative example?

Not OP, but

You might not know anything about risk-taking behavior of firefighters but you may already have some vague belief like "firefighters are heroes" that obliquely colors your impression of their behavior.

Or alternatively you might hold onto the info that Germany is colder because your underlying belief is more like, you don't like cold and you don't like Germany, so you'd like to also believe that Germany is colder than other places.

This entails two things. One, your apparent position on firefighter behavior or the weather in Germany can change depending on what in the context of the conversation is being construed as good or bad. Second, trying to inform you with specific facts on these issues is unlikely to change your mind because the drivers of your positions are your more general beliefs about firefighters and Germany.

In politics, I think partisanship often degenerates in this way. Arguing the issues is often just a facade for arguing for your party's position or arguing against an opposing party's position, regardless of merit. Facts won't work here to change minds.

You're just making same claim as the authors in the study, but adding a proposed mechanism. But just as they don't have evidence, neither do you. You have to actually do the study to prove it. If you believe that the experiment would work this just means you and I have different priors on the matter.

Why not? There are many strange facts in the world. Some of them are even true. It is very difficult, when recalling a strange facts, to remember whether it was one of the true ones or not. So naturally, things we've heard will tend to exert a pull, even if we later found out they were wrong.

In virtue of the experimental design, you were just now told that the "strange fact" is bullshit. If you had a clear opinion prior to receiving the strange fact, you'd revert to it

Reply on news.ycombinator.com