Superstition
Correlation fail
I've looked at how machine learning systems that can work with vast amounts of data can find correlations that work, but nobody can see just how they work in a causal sense. They are black boxes that we have learned to trust.
Many people who celebrate the old scientific method of hypothesize, predict, test find this to be profoundly unsatisfying. One wants to know WHY something works, not just THAT it works.
Truth be told though; the cutting edge of science is always pretty opaque in terms of the why question. I mean, I've looked at things like quantum mechanics or relativity to some depth and find it reasonable - but there is lots of material that I just have to accept as a given because I'm not trained to interpret it . I trust that somebody someplace actually does understand it all - but maybe not.
A big problem with correlations is that we are so good at detecting patterns. We detect patterns in masses of data all the time without even thinking about it. But we know that we often find patterns where there is none. I'm thinking of how we can see faces in clouds.
I suggest that superstition can be thought of as a false signal from our pattern matching system.
The idea that Friday the 13th brings bad luck is a superstition. But what if that correlates to your experience? It probably does for someone in the world.
Behaviorists can train pigeons to do all sorts of things by giving a reward when the bird does what the trainer wants.
The bird learns that a particular behavior is rewarded.
So the bird starts paying attention to what they were doing just before the reward was offered.
It works.
But what happens if the reward is just given at random without being associated with a reward?
This can lead to very strange behavior depending on what random event immediately preceded the reward.
This has been called superstitious behavior.
And it's caused by a failure of correlation - finding a correlation when there is none.
Let's think of this in the context of our polarized culture. For instance, I don't think vaccines magnetize people. The idea is absurd. And yet many people think the idea makes sense. Their information sources are giving them lots of anecdotes correlating vaccines with magnetism. And they don't know enough about how either vaccines or magnetism work to be looking for a causal explanation. So they see the correlation (wrong as it may be) and feel entitled to act on their perception as citizens of a free country.
One superstition that I see as influential in America has to do with the magical power of guns to solve problems. I think having a gun provides a satisfying feeling of personal power. I might put that as a correlation - have gun - feel power. But does it actually fit with the facts? It seems to be more a case of - have gun - get into lots of trouble.
When dealing with correlation the direction of causality gets lost. I mean - Trump and right wing craziness are certainly correlated. But did Trump cause the craziness or did the craziness cause Trump?
What do you think? I open the floor.
I present regular philosophy discussions in a virtual reality called Second Life.
I set a topic and people come as avatars and sit around a virtual table to discuss it.
Each week I write a short essay to set the topic.
I show a selection of them here.