Eleanor Crane: There's a line in the Imperial News announcement — June 15th, 2026 — that I keep rereading. It describes the hippocampus as the brain's primary memory centre, and it says the stimulation drove neural regeneration there. In an Alzheimer's model. Ben, hi, by the way.
Ben Okonkwo: Ha — hey. Yeah, and I think — wait, the reason that line lands is because neurogenesis in a diseased hippocampus isn't what TI stimulation was supposed to do. The 2023 Nature Neuroscience result showed memory modulation in healthy volunteers. That's functional. This is regenerative. Those are not the same claim.
Eleanor Crane: Right, so the jump feels — almost too fast?
Ben Okonkwo: That's exactly the tension. Dr. Nir Grossman's group at Imperial College London, the UK Dementia Research Institute — they're claiming this is the first time TI stimulation has driven endogenous neural regeneration in the central nervous system. And the Advanced Science paper is the vessel for that claim.
Eleanor Crane: No drugs. No implanted electrodes. Which, if you know what conventional deep brain stimulation involves surgically — that distinction is not small.
Ben Okonkwo: Not small at all.
Ben Okonkwo: And to make that concrete — imagine two high-frequency electrical signals, both around ten kilohertz, beamed through the scalp from separate electrodes. Neither one, on its own, does anything to neurons. The tissue just lets it pass. But where those two fields overlap — deep inside, at the hippocampus — they beat against each other. And that beating creates a low-frequency envelope. Maybe a hundred hertz. And *that* is what neurons actually respond to.
Eleanor Crane: So the skull isn't a barrier — it's almost like a lens?
Ben Okonkwo: That's — yeah, that's actually a really good way to put it. The interference pattern is steerable. You adjust the geometry of the electrodes, you move the target.
Eleanor Crane: And this isn't new — I mean, the principle. When did this actually get validated?
Ben Okonkwo: 2017. Published in Cell. Animal validation — that's the foundational paper, the one the entire trajectory runs from. So when the 2026 Advanced Science result lands, we're not talking about a sudden claim. It's nearly a decade of work building toward it.
Eleanor Crane: And the conventional alternative — Deep Brain Stimulation — requires surgically implanted electrodes. Which is, I mean — that's not a minor footnote. That's a completely different risk calculation for a patient.
Eleanor Crane: But here's what I keep getting stuck on. If you grow new neurons inside an Alzheimer's brain — a brain that's inflamed, that's full of amyloid plaques and tau tangles — are you growing hope, or are you growing neurons that will die the same way?
Ben Okonkwo: That's — yeah, that's the actual question. And the 2026 Advanced Science paper doesn't answer it. What it shows is neurogenesis in a mouse model. New neurons appearing. But integration into damaged circuits? Whether they survive the amyloid environment? That's a different claim entirely.
Eleanor Crane: And the 2023 Nature Neuroscience result — healthy volunteers, improved episodic memory accuracy — that's not evidence for this, is it. I mean, a healthy hippocampus and an Alzheimer's hippocampus are not—
Ben Okonkwo: Completely different substrates. You can't carry a functional result across that gap.
Eleanor Crane: Now — and this is the part that surprised me — Imperial received nearly £15 million from ARIA, the Advanced Research and Invention Agency, as part of a £69 million UK government package. But that money is framed around interfacing with the human brain. Not specifically therapeutic neurogenesis for Alzheimer's.
Ben Okonkwo: Hm. That's — interesting. Because that's a structural question, not a character one. Funding shapes what gets prioritized next. And TMS — transcranial magnetic stimulation — is also non-invasive, also targets brain circuits, and hasn't resolved neurodegeneration. TI stimulation needs to be evaluated within that landscape.
Eleanor Crane: So the competitive field already exists. And the bar is higher than just — non-invasive and promising.
Ben Okonkwo: And that's — I mean, that's where I want to just name the arc plainly. 2017, Cell, animal validation. 2023, Nature Neuroscience, healthy humans, episodic memory. 2026, Advanced Science, neurogenesis in a diseased mouse hippocampus. Each step is real. Each step is also not the next step. The gap between 'new neurons appeared in a mouse' and 'a person with Alzheimer's remembers something they'd lost' — that gap still requires a human brain that can actually integrate those neurons into damaged circuits. Whether they survive in an amyloid environment. Whether they wire in correctly. We can't see any of that yet.
Eleanor Crane: What clicked for me — and I think this is what I'll hold onto — is that the finding doesn't say we've reversed Alzheimer's. It says the brain might not be as finished as we thought it was. That's a different claim. Smaller, maybe. But actually harder to argue with.
Ben Okonkwo: Yeah. The door isn't sealed. That's — honestly, right now, that's the honest version of it.
