DRMacIver's Notebook
A simple model of depression and “depressive realism”
A simple model of depression and “depressive realism”
I was talking to Vathy about depressive realism, and when I expressed skepticism about its existence they asked me to say more. I did, and it seemed an unusual enough perspective that I thought I’d quickly write it up in a blog post.
So here’s what I think, roughly. This is a simplified version of what I actually think (because what I actually think is a giant tangle of stuff I haven’t fully figured out), and what I actually think is probably no more than roughly right, but I still think this is a good model that is helpful to bear in mind.
Basically it’s this: Humans have all sorts of biases in their thinking. No matter what anyone tells you, this isn’t bad. Biases are good, as long as you’re expecting to make mistakes (and, I hate to break it to you, humans make a lot of mistakes and are going to keep doing that), because not all mistakes are created equal. If you’re going to make mistakes, you want to bias in the direction of mistakes that are safe to make.
(See e.g. There’s no single error rate).
One of the ways that depression differs from healthy cognition is that it has a much stronger bias in the direction of inaction than healthy cognition does. If you see a problem, you can decide to do something about it or not. A depressed person is much more likely than a non-depressed person to choose inaction, and as a result is more likely to believe things that justify that inaction (this latter feature isn’t specific to depression - a non-depressed person who chooses action is more likely to believe things that justify that action. Motivated reasoning is a whole big thing).
As a result, a depressed person is more likely to get the right answer in situations where the right answer is inaction, and less likely to get the right answer in situations where the right answer is action.
In situations where the right answer is typically inaction (which can happen if you’re in a dangerous environment which you have little control over, certainly, and as a matter of numbers this is probably more likely for depressed people) this will tend to mean that depressed people are more likely to be right than people who are neutral or biased towards action, but it’s not that the depressed person is more realistic than the non-depressed person, it’s that their biases are currently well adapted to their situation, and if the situation changes they would retain those biases and be less well adapted. I’m no more inclined to call this realism than I am to call a stopped clock accurate because it happens to be stopped at the current time.