There's evidence that the reason for the split is that psychosis is itself damaging. Once someone has a psychotic episode, they are much likelier to have more. But the earlier a psychotic episode is interrupted, by anti-psychotics or therapy, the better outcomes are. So the implication is that many people have some degree of vulnerability to schizophrenia, and only those who have psychotic breaks develop the full condition.
That's a possible causal explanation for a link between marijuana and schizophrenia. Someone who's got some vulnerability to psychosis, but wouldn't otherwise have developed schizophrenia, might be pushed over the edge by heavy drug use. This jives also with anecdotal reports of people who've had really terrible reactions to other psychedelics.
There's a fascinating account by a psychologist who's also schizophrenic. He didn't have his first psychotic break until he was 33, far later than average. He talks about the personality traits of schizophrenics, how he noticed something was off and got early therapy, and how he credits that with delaying a psychotic break and improving his outcomes.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2632294/
Not a doctor or scientist, just regurgitating research I read for personal interest.