Only three of the six were randomized, and the advantage over controls is short-term.
Weekly Science Digest, 3 Oct 2026 to 9 Oct 2026. 24 papers read, 12 preprints.
1. Alcohol use disorder gets its first meta-analysis
Pooled across six trials, psilocybin's edge on drinking is real but moderate and short-lived.
Six trials totalling 202 patients, only three randomized, give standardized mean differences of roughly 0.4-0.5 for drinks per day, percent drinking days and heavy drinking days versus control. Within-group reductions held out to 28 weeks, but those have no comparator.
How much weight it carries: Pooled estimate from six trials, only three randomized, 202 patients total; adequate to size a short-term signal, not to establish durability. (Humans, Meta-analysis, Systematic review, RCT, Small sample, n = 202)
2. Most psilocybin use is happening outside the clinic
One in 14 US adults with a past-year depressive episode reported using psilocybin.
In the 2024 National Survey on Drug Use and Health, 7.3% of 6,793 people with a past-year major depressive episode had used psilocybin, and the study asks whether that tracks with skipping conventional care. RAND's 2025 survey adds parallel data on veterans' use and their views on VA policy.
How much weight it carries: Nationally representative cross-sectional survey; good for prevalence, cannot establish whether psilocybin use causes or follows changes in treatment seeking. (Humans, Survey, Observational, n = 6,793)
3. LSD speeds rhythms up rather than just damping them
MEG shows LSD shifts alpha and beta peak frequencies upward, not merely weakens oscillations.
Source-resolved recordings under LSD versus placebo separate genuine power loss from peak-frequency drift, and add a flattened 1/f slope plus increased signal fractality. The changes concentrate in sensory, language and imagery networks and spare motor cortex.
How much weight it carries: Within-subject placebo-controlled MEG in a small healthy sample; characterises the acute drug state, says nothing about therapeutic effect. (Humans, Placebo-controlled, Brain imaging, Mechanism, Small sample)
4. Two compound reviews worth reading for the caveats
A DT120 profile insists 100 micrograms of LSD is a psychedelic dose, not microdosing.
The drug profile calls the phase 2b anxiolytic signal credible after a single 100 µg dose without mandated psychotherapy, while rejecting the "microdose" and "modified release" framing used around it — controlled microdosing work has used 5-20 µg.
How much weight it carries: Narrative drug profile plus a systematic review of 51 publications; both summarise existing trials rather than adding data. (Review, Systematic review)
