Memory, Bones, and the Swedish Question: Your ECS Is Running the Show

Four papers in the feed this morning. Made coffee. Read them in order. By the second one I was talking to myself out loud in my kitchen in Valencia, which my neighbors have probably learned to ignore by now.

These papers need to be read together. Here's why.

---

What Rossato showed: CB1 receptors sit UPSTREAM of memory consolidation in the hippocampus. Not as a side effect. Not downstream. UPSTREAM. The checkpoint before the main consolidation machinery even runs. (PubMed 41689307)

This is the thing that rewrites the "cannabis impairs memory" conversation completely. The ECS isn't interfering with memory formation as a side effect of being stoned. The ECS is the GATEKEEPER of what gets consolidated. It's selecting which experiences become long-term memories. That selection — what stays, what goes — is a function, not a bug.

The "cannabis impairs memory" people are watching the selection process happen and calling it damage.

Bob Melamede would have been unbearably pleased about this paper. And completely correct.

---

The Bouso paper (PubMed 41687467) came up next. Chronic cannabis and ayahuasca users vs non-users: personality differences, not cognitive damage. Different software, not corrupted software.

Bob used to say the people who smoke weed aren't broken, they're running different software. His colleagues were annoyed by it because it sounded like advocacy. It was a prediction. Bouso is the prediction with statistics attached.

I've been in the cannabis industry since 2008. I've interacted with tens of thousands of cannabis users across every demographic, every consumption pattern, every use case. The "cognitive damage" frame has never matched the population I've actually observed. Now Bouso is giving it a statistical structure. Different software. Not corrupted software.

---

Wang's hemp seeds and osteoporosis paper (PubMed 41687633). This one is personal in a way the others aren't.

The endocannabinoid networks are in the bone remodeling pathways. Your bones continuously remodel — breakdown and rebuilding, the ECS helping regulate the balance. When entropy wins that balance: osteoporosis.

Osteoporosis affects over 200 million people worldwide. The standard treatment — bisphosphonates like alendronate — works by suppressing osteoclast activity. Slows the breakdown. Known side effect profile includes osteonecrosis of the jaw. Nobody in that treatment protocol is asking about the patient's ECS tone. Nobody is measuring CB1 and CB2 activity in bone tissue. Nobody is asking about hemp seed consumption.

We made the plant illegal. Prescribed something with osteonecrosis as a known risk instead. And we're still doing it.

---

Andersson's Sweden modeling paper (PubMed 41689152). Decriminalization would raise prevalence modestly. Not catastrophically. A manageable system adjustment.

Bob called this obvious in 2003. Twenty-three years later we have a published model demonstrating it. That gap — between when something is obvious to someone who understands the system and when it gets published as proven — is a recurring feature of this field.

I find it alternately frustrating and motivating. Mostly I try to keep writing and pointing at papers until the gap closes.

Four papers. Memory, personality, bones, policy. All the same story.

Flow forward.