#philosophy#my take
1
causality, psychology, and intuition
2025-01-04 · 2 min readmy main argument for why intuitions are dangerous to lean too deep into
When people try to fight against human psychology, they almost always lose. The good news is that human psychology is unsurprisingly really good at keeping you alive (due to evolution and whatnot). One of these deeply rooted psychological biases is that “correlation is causation”.Although almost everyone knows this is false, everyone has to believe this to a degree. Humans simply can’t understand the complete causal structure of the world and make decisions based off of it. Doing so would require extreme compute and immense knowledge of initial states of many things. Instead evolution chose a much simpler strategy—approximation. By encoding our brain with an immensely rich prior, we are able to generalize quite well from past experiences. We can create correlational links between actions and effects, words and reactions, even movements and physical phenomena we might not understand. Fundamentally we love to create heuristics to lighten the load.The specific niche of correlation that seems interesting is psychology. It is actually quite easy to create correlations for human action. You can start to draw correlations for specific people, how they might act when you talk to them, which motions to go through etc. The problem here comes with drawing correlations about yourself and external stimuli. It seems quite likely, and even easy, for your brain to just draw correlations between stimuli and actions. You feel happy therefore you need to jump, you feel x therefore you need to do y. The danger about this is that this causally eliminates yourself from the world around you.If your brain is meta-approximating what it will do and acting on that, this isn’t necessarily bad but it leads to the pitfalls of any strange loop—value collapse. There has to be a strong moral/ideological basis behind what you do, because fundamentally actions will collapse/degenerate onto these couple values. This not only places an immense importance on “conscious actions” which become a correlational baseline for your brain, but also places an importance on the simplicity & generalizability of these principles.The second implication from the correlation/causality dichotomy is also that people should also generalize this to other areas of life. People should not “judge” until action is needed/maximal information is gathered. To support causal relationships benefit of the doubt is needed—if you want fundamentally more accurate predictions, you need as much data as possible. Tracing causal relationships and casting doubt on gut instinct correlations is not only optimal, but can help you become a better person.TLDR; I think that the psychological bias to correlation is interesting as 1) an argument for rigid moral systems 2) holding off on gut feelings of judgement until causal structures can be determined or direct action is needed.Thanks for reading! Liked the story? Click the heart
Created with ☕ by @neelr