Ask HN: How do you deal with confirmation bias in your thinking? | Hacker News Reader

Ask HN: How do you deal with confirmation bias in your thinking?

4 pointshto_tam4 comments

When I ask myself where I might be biased, I get no answer and assume all my current beliefs are unbiased. I understand that I may be wrong on some issues, but not because of bias, rather because of a lack of information. Science says that the essence of confirmation bias is exactly not knowing where you are biased.

I thought, maybe I have no biased beliefs, maybe I have somehow overcome them. But then I thought that to overcome bias on issue "x", I would first need to realize that I am biased on issue "x". In that case, I should be able to recall moments when I recognized my bias, but I can't. The only case I recalled was when I was in love with a girl and probably biased about whether she liked me.

I checked, there is a scientific consensus that confirmation bias is real, and the more important the issue, the higher the likelihood of bias.

View on HN

Interesting, I assume the opposite, that all of my beliefs are biased.

Now this is a huge area where I'm biased:

I assume that it is better to assume one is biased in all areas, and that this approach leads to less bias (because I'm challenging my assumptions) or leads me to accept certain places where I am biased. Part of the goal is to be detached from the need to feel unbiased:

I don't think this is true in all situations, but a belief I have based on observation is that the more that you care that you are unbiased, or the more tied your ego is to being unbiased, the more vulnerable you are to confirmation bias, which creates an interesting paradox.

So I guess my approach is, I try not to care too much, and here's maybe the ultimate paradox or logical fallacy: You're values don't have to match reality, and if you try to get reality to match your values you will be super vulnerable to confirmation bias.

The obvious version of this is in religion, if you believe that the world is 6000 years old, and seek to prove the world is 6000 years old, you can find lots of evidence to support the idea.

As I understand it, confirmation bias can occur even when a person challenge his assumption, in which case he will challenge it in a way that confirms it without realizing it. That's why always challenging one's assumptions is certainly valuable, but I think does not significantly help with this bias.

Yeah, this was sort of the crux of the paradox that make it hard to explain, you want to challenge your assumptions, but you want to detach yourself from your ego's' need to be right, so not challenge them too much. This is sort of like the "Don't be so open minded your brain falls out" thing, or maybe the opposite of that.

You're on to something with 'lack of information'. I'm not perfect in this (when I first started commenting on HN, an alarming proportion of replies said that I had confirmation bias or 'correlation is not causation'), but I've learned to find data and (most importantly) evidence to support alternate conclusions. It's not obvious until something doesn't have the intended results, but by the time I've investigated and figured out why, I'm too thrilled to see it work out that I don't check to see whether the initial choice had confirmation bias.

Reply on news.ycombinator.com