On Biases

To begin, biases scare the shit out of me.

It's scary to think that humans aren't rational agents. It's scary to think that humans aren't rational agents even when they are trying to be rational. But what I think is scariest is that people have cognitive biases they don't even know about!

I make it my goal to learn about the mental shortcuts we humans tend to take. For example, people tend to believe that sources of information that agree with their internal views are more factual.

Humans also tend to take shortcuts in observation and perception. My favorite example is something that more economists should consider. Humans are risk-averse when it comes to gains and risk-seeking when it comes to loses. Let me demonstrate.
You have a choice between a 100% chance of getting 50$ or a 50% chance of getting a 100$ and a 50% of getting nothing. Most humans choose the guaranteed 50$ because it's a safe bet.
Now we have the same situation again but this time you have a choice between a 100% chance of loosing 50$ (read: you have to give up the money) or a 50% of loosing 100$ and a 50% of having to loose nothing. Most humans end up choosing the second option.
Why? Because you are likely to take risks when there's a possibility of losses and avoid risks when it comes to gains. It's all part of prospect theory and it's really fascinating stuff.

We also have a tendency to be really awful at math. Estimating probabilities in our modern world is something we weren't designed for. Anecdotes tend to cause people to up their estimates of the likelihood of an event occurring. How many people are terrified of strangers sexually abusing children when only 10% of child abusers can be considered "strangers" from the point of view of the abusee? source

I guess it's a good thing that I'm afraid of biases, at least it helps me keep an eye out for them. It's not a perfect solution, but it will tide me over until our perfect robot overlords arrive.

No comments: