Let's approach the amount of EA causes as finite. We must prioritize issues and approaches to maximize relevant altruism.
Psychedelics have a not insignificant history and period of use, and a limited set of initial research that shows some promise and some bias.
Highlighted by the recent crystallization of the binding of LSD to SERT, the serotonin transporter, we generally can categorize the mechanisms of psychedelics so far to match pathways we currently have agents on, and that have some limitations.
Pursuing psychedelics as a strong target may come at the potential cost of not finding new mechanisms or avenues to go after mental disorders. We would be pursuing agents with known deleterious factors- e.g. MDMA shows serotonergic neurotoxicity in some models at certain doses, exacerbations of disorders and psychotic breaks - that represent a dose ceiling or limited room for treatment. A comparison to say, optimizing courmarin use, with its risks, rather than finding a new agent (like a factor Xa inhibitor).
There's a particular subset of experiences and association in psychedelic use that is decidedly non-scientific and resistant to experimentation. Meta-physicial explanations and decidedly 'hippie' mentalities show resistance to some authorities and funding sources.
Psychedelics comprise a loud, distracting avenue that, while it may bring promise to some areas, may drain funding from novel approaches. Choosing psychedelics as a cause also defines a substance area as a cause, rather than humanistic approaches (one can argue about the psychedelic mentality). Does it truly match other EA causes or is it choosing an 'hot topic' because it comes from alternative point of view?
Note- these are mainly just random sentences rather than an argument