New research: when you smoke you might not yet have psychosis. People with psychosis smoke. More people who smoked before psychosis [was diagnosed nb] had psychosis (57%).Therefore smoking might cause psychosis.
See the flaw that turns the correlation into a cause? It has to do with the tendency to be black and white, without shades of grey (following the logical "law" of the excluded middle, which can be a big problem in Western logic).
First there is a confusion of phenomena and thing. "Having psychosis" actually means "being diagnosed with psychosis." You can be diagnosed while you are having it, which means that you already had it. So the fact that you are smoking when you are diagnosed proves nothing.
Then there is the black and white thinking. Either cigarettes cause psychosis, or they medicate psychosis. Since people with psychosis were already smoking, cigarettes must have caused psychosis. But remember, they "have" psychosis in the sense that they have been seen to have psychosis.
See the problem? Psychology scholars are the last to admit confirmation bias, namely the fact that we are all correlationists, more or less, and that science works that way.
I know people with psychosis. Like my brother. He was diagnosed with psychosis at a certain point. But he was building up to it for years and years. And was perhaps smoking to self-medicate, for instance to control the anxiety. Before he was diagnosed.
Assuming psychosis actually exists...
Anyway, whatever is happening, there are shades of it. You can be more or less psychotic. You can become very upset, then you can become psychotically upset as a defense against how upset you are. You smoke to plug the hole in your being. This hole was already there--in fact you know, if you follow someone like Heidegger, that you kind of are that hole, so the possibility of wanting to plug it with something (a fantastic job, a cigarette, suicide) is quite high.
Come on, scientists! This is a classic example of how an old not well proved logical "law" is screwing you.