Andrew Huberman· PhD
so if you're just randomly sprinkling mutations you know in you know a genome that they're more likely to sort of manifest in you know maybe neuros pych atric conditions
The headline is broadly defensible, but the qualifications matter. Effect sizes vary by population, the strongest claims rest on shorter trials, and credible voices push back on how it's typically framed.
so if you're just randomly sprinkling mutations you know in you know a genome that they're more likely to sort of manifest in you know maybe neuros pych atric conditions
Every Sunday: the week’s new conflicts and verdict changes — and nothing else.
Native comments, Twitter mentions, and Reddit threads about this claim — surfaced together so the conversation isn't fragmented across platforms.
Bookmarking — the dossier-vs-overview split is the right call. Most of the time I want overview; sometimes I want receipts.
Would love a "what would change this verdict" RSS feed. Sign me up if it exists.