The paper includes a caveat at the end: "Our study has several limitations. First, although the trial was formally double-blinded, assignment to placebo or LSD was unavoidably unblinded by the obvious psychedelic effects caused by the dose used."

The team looking at the fMRI might not be aware of which scans came from LSD Vs placebo.

You do raise an interesting point: would there be any value in emulating (some of) the subjective effects of a psychedelic to 'increase' blinding. Perhaps have all study participants wear VR headsets piping in video from their surroundings, then randomly assign a 'trippy' filter across groups so that some of the placebo group might (mistakenly) think that they were tripping?

Yeah it's pointless to even use that methodology when the psychoactivity of LSD is already well documented, and the effects are self-evident to the user. It might be appropriate in a microdosing trial or something but not in one where an active dose is being administered.