The Third Neurotransmitter

We can say that with some confidence, because we watched it happen once already. "Chemical imbalance" entered ordinary English on the back of SSRI marketing and stayed thirty years — long enough that when researchers audited the serotonin-deficiency account in 2022 and reported it couldn't carry the weight put on it, the finding landed as a scandal rather than a footnote. (That review was contested; the dispute is live.) The phrase had outlived its evidence by decades, and managed it because it was never really resting on the evidence. It was resting on a product.

So: centanafadine, sold as Simtriyo, is now cleared in the US for ADHD in adults and children six and up. Four phase-3 trials. Once-daily capsule. Side effects are the familiar shelf — reduced appetite, insomnia, dry mouth, headache — plus the suicidality warning standard to the category. And the headline distinction, reported nearly everywhere: the first ADHD medication to slow the reuptake of three neurotransmitters rather than two. Dopamine, noradrenaline, and now serotonin.

Set that sentence beside the binding data and it comes apart in your hands. Centanafadine's IC50 is roughly 6 nM at the noradrenaline transporter, 38 nM at dopamine, 83 nM at serotonin — a ratio near 1 : 6 : 14. It is a noradrenaline drug with a dopamine sideline and a serotonin rumor. "Targets three neurotransmitters" is true the way "contains eleven real fruits" is true.

Which is the whole story, arriving a level earlier than expected. The title of this piece is a marketing artifact, and so is the frame around most of the coverage. The reverse-engineering doesn't begin when a patient reaches for an explanation. It begins in the press kit, and the patient inherits it.

Now the older layer, the one the new drug is being fitted onto. The popular dopamine story — my dopamine is broken — was itself an inference run backwards from a pill. Nobody located a shortfall in inattentive brains and commissioned a molecule to top it up. Stimulants helped; someone asked what stimulants do; the answer became the explanation for the condition. That's a respectable method, and it has earned its keep across medicine — but it yields an explanation shaped like the available treatment.

The obvious objection is that independent evidence exists: the DRD4 and DAT1 associations, the dopamine-transporter imaging. It's a fair objection, and it deserves a real answer, which is that both fold back into the pill when you follow them. Those genes became candidates in the first place because stimulants act on those targets — the same reverse inference, one floor down, wearing a lab coat. Genome-wide studies, which don't get to pick their candidates in advance, haven't replicated them. The imaging is worse: striatal transporter density does run higher in ADHD groups, but a meta-analysis found the effect tracked prior stimulant exposure, with medication-naive patients showing lower density, not higher. Much of the abnormality was an adaptation to the treatment.

What survives is real and points elsewhere entirely. ADHD is strongly heritable, and the loci that have replicated implicate early neurodevelopment — neurite outgrowth, synaptic plasticity — not a tank running low on a chemical. The genetics never said low dopamine. The pharmacy said it, and we repeated it back.

Hold that next to what these trials establish, which is less than the framing implies. Centanafadine was tested against placebo, not against the drugs people are already taking. That it's gentler than the alternatives is a bank shot off two separate datasets, only partly public. "Fewer side effects" is a sentence in a lab coat, not a result.

The multi-channel picture is still an improvement — not because three is nearer the true number, but because attention was never one dial. A network out of phase is a more honest description of a nervous system than a tank out of stock. It also means more calibration surface, and the person doing that calibrating is you, at your kitchen table, with a symptom diary and fifteen minutes of prescriber a quarter.

Then there's the part we're all very good at not mentioning. Every approved intervention sits on one side of the equation. Attention isn't a property of a brain; it's a relationship between a brain and a room. We spent twenty years building the most systematically attention-fragmenting environment in human history — engineered, funded, optimized against your ability to stay with anything — and then routed the entire treatment response to the individual standing inside it.

I used to think that was mainly about patents. It's worse than that, and more boring. Individuals can be randomized; environments can't. You can run a trial on a tablet, and you cannot run one on twenty years of everyone's phone. The evidence base is structurally lopsided before anybody's commercial motive shows up — which is why the tilt survives every reform aimed at the money.

And that's the thing underneath all of it. It isn't only that treatment gets individualized while the cause stays ambient. It's that the vocabulary you use to understand your own interior arrives as a byproduct of a commercial pipeline, with no competing supplier, because the funding that produces folk-psychological categories is the same funding. The distortion doesn't enter at the level of facts, where a fact-check could reach it. It enters at the level of which concepts are available to think with.

So keep the pill if it works — a person drowning takes the rope, and refusing the rope to make a point about the water is a stupid way to die. Just notice, next time you reach for "my dopamine," whose sentence that is.

Hold the story loosely enough to be revised. It just was.