The
Dispatch
OPEN CHANNEL · FILE AN INQUIRY · DIEGETIC VOICE
The Dispatch is the open channel. File an inquiry. The Compiler reads everything and replies in one compilation window per month.
The Evidence
The Position argues that the interior has become legible to instruments. Below is what that rests on. Each entry gives the claim as we make it, the paper, the number, and what the paper does not show. The last field is the important one. A result reported without its limit is not evidence, it is advertising.
Stimulation aimed at a circuit mapped in each patient’s own brain put most of a small trial group into remission from treatment-resistant depression after a five day course, and the FDA cleared the device.
Cole EJ, Phillips AL, Bentzley BS, et al. Stanford Neuromodulation Therapy (SNT): A Double-Blind Randomized Controlled Trial. American Journal of Psychiatry. 2022;179(2):132–141. doi:10.1176/appi.ajp.2021.20101429
Eleven of fourteen in the active arm met remission criteria in the four weeks following a five day course, against two of fifteen on sham. Mean MADRS reduction 52.5% active, 11.1% sham. Each participant’s stimulation site was located from their own resting-state scan.
Magnus Neuromodulation System with SAINT Technology. FDA 510(k) K220177, cleared 1 September 2022. Cleared means found substantially equivalent to an existing device. It does not mean approved.
Fourteen people in the active arm, reported at a planned interim analysis. Remission was measured across the four weeks after treatment, not on the fifth day.
Focused ultrasound reached the subcallosal cingulate through an intact skull, with the target confirmed on a scan, and more than halved depression scores in a sham-controlled pilot.
Riis TS, Feldman DA, Kwon SS, et al. Noninvasive Modulation of the Subcallosal Cingulate and Depression With Focused Ultrasonic Waves. Biological Psychiatry. 2025;97(8):825–834. doi:10.1016/j.biopsych.2024.09.029
Depression scores fell 55% at twenty-four hours and 52% at seven days in the active group, against 22% and 29% on sham. Target engagement was confirmed: subcallosal cingulate activity dropped during stimulation, p = 0.028. No surgery, no opening of the skull.
Nine active participants in the per-protocol sample. In the full sample of twenty-two the difference did not reach significance. The sham group improved too. The part of this result we would defend hardest is not the depression score. It is that the beam arrived where it was aimed.
The live signature of self-referential thought has been put on a screen, and people trained in introspection can steer it.
Garrison KA, Scheinost D, Worhunsky PD, et al. Real-time fMRI links subjective experience with brain activity during focused attention. NeuroImage. 2013;81:110–118. doi:10.1016/j.neuroimage.2013.05.030
Participants watched a live graph of their own posterior cingulate cortex, a default-mode hub tied to mind-wandering and self-referential thought, and reported that the graph corresponded to what they were experiencing. Experienced meditators could deliberately drive the signal down.
Non-meditators could read the graph but could not steer it. A single session, so nothing here shows that the skill can be taught. This is why we say trained in introspection rather than anyone.
In a laboratory a closed loop already runs that reads attention from a brain in real time and reshapes what it shows to pull that brain back toward a chosen state.
deBettencourt MT, Cohen JD, Lee RF, Norman KA, Turk-Browne NB. Closed-loop training of attention with real-time brain imaging. Nature Neuroscience. 2015;18(3):470–475. doi:10.1038/nn.3940
A decoder read attentional state from the scan within seconds and continuously altered the image on screen, making the task harder as attention drifted. Attention improved in the feedback group and not in matched controls who received someone else’s feedback. Group difference p = 0.04.
Scangos KW, Khambhati AN, Daly PM, et al. Closed-loop neuromodulation in an individual with treatment-resistant depression. Nature Medicine. 2021;27(10):1696–1700. doi:10.1038/s41591-021-01480-w The clinical counterpart, in which a personalised brain marker triggers stimulation rather than a display. One patient.
Sixteen participants, one session, inside a research scanner. The system nudges a state, it does not hold one. The benefit did not survive when the feedback came from reaction time instead of the brain signal. Nothing here is a deployed technology, which is why we say in a laboratory.
A chemist read one line in a formulary sixteen centuries old as a technical instruction, changed her extraction temperature because of it, and got a working antimalarial out of a plant.
Tu Y. Artemisinin, a Gift from Traditional Chinese Medicine to the World. Nobel Lecture. Angewandte Chemie International Edition. 2016;55(35):10210–10226. doi:10.1002/anie.201601967 Nobel Prize in Physiology or Medicine, 2015.
Ge Hong, Zhouhou Beiji Fang, a handbook of prescriptions for emergencies, composed around 340 CE. It says to soak the plant in water and wring out the juice. It does not say to boil.
Tu read the absence of heat as an instruction, moved to a low temperature ether extraction, and on 4 October 1971 sample 191 reached 100% inhibition of the malaria parasite in rodents. Earlier preparations had managed between 12% and 68%.
The text held a clue, not a compound. Artemisinin was isolated, purified and structurally characterised by modern chemistry, and the first pure crystals came from other teams working with her method. An old book is a hypothesis generator. It is not a pharmacy. We read our own corpus the same way.
Four of these five results come from groups of fewer than twenty people. We are not claiming the case is closed. We are claiming the case is now testable, which it was not within living memory, and that the reasonable response to a testable question is to build the instrument and run it.
If any entry here is wrong, or has been superseded, write to the Dispatch and it will be corrected in the open.