If two things (A and B) are correlated with each other, there are four possibilities:
1) A leads to B.
2) B leads to A.
3) A and B are both related to some other factor C.
4) A and B are unrelated; their correlation is an accident.
If we observe that some great accomplishments in science were achieved by people who used psychedelic drugs, we might wonder whether the drugs helped. But we should also note that intelligent, curious, novelty-seeking people are likely to be the ones behind discoveries… as well as likely to experiment with perception-altering drugs.
If there’s an obvious property C that we might rationally expect A and B to be related to, and it’s present, we have no reason to presume a correlation between A and B implies causality between them.
I have no idea precisely how dangerous drugs like LSD are (and are not)… but I don’t think they have much potential for furthering scientific discovery.