Focal brain stimulation sites that modify autonomic arousal map to a convergent brain circuit and potential therapeutic target
Ryan D. Webler, Clemens Neudorfer, Eva S. A. Dijkstra, Summer B. Frandsen, Nicole Chiulli, John Miller, Ru Kong, Sanaz Khosravani, Garance M. Meyer, Alexandre Boutet, Jurgen Germann, Gavin J.B. Elias, Nicholas L. Balderston, Andrew Strohman, Wynn Legon, Jordan H. Grafman, Martijn Arns, Andres M. Lozano, Shan H. Siddiqi
2026 · 10.64898/2026.01.08.698446
Abstract
Causal modulation of autonomic outflow could yield new therapeutic targets for autonomic hyperactivation. We employed three natural experiments in which different brain regions were targeted using transcranial magnetic stimulation (TMS) (n=139 sites, n=14 individuals), deep brain stimulation (n=392 sites, n=58 individuals), or low-intensity focused ultrasound (n=46 sites, n=23 individuals) with subsequent autonomic measurements. Using a human connectome database (n=1000) as a wiring diagram, we identified a convergent brain circuit that, when focally modulated, transiently reduces autonomic arousal. This circuit significantly resembled previously reported causal circuits for posttraumatic stress disorder (PTSD) and anxiety. In independent datasets, TMS to the autonomic arousal circuit reduced laboratory startle in healthy volunteers (n=28), lesions to this circuit reduced exaggerated startle in PTSD (n=193), and TMS to this circuit reduced anxiety-related autonomic symptoms in patients with clinically significant anxiety (n=30). Thus, the convergent circuit may serve as a potential neuromodulation target for autonomic hyperactivation.
Abstract via europepmc.
Exposures
Exposure 1: LIFU to individualized anterior/posterior insula targets (heart rate variability circuit dataset)
Target: anterior insula, posterior insula — “individualized anterior and posterior insula targets”
Device: Sonic Concepts · Sonic Concepts · H-281 ✓
| Waveform | pulsed | |
|---|---|---|
| Fundamental frequency (kHz) | 500 | ✓✓✓ |
| Pulse duration (ms) | not reportedimplied by duty cycle ÷ PRF: 0.36 ms (not stated by the paper) | ⚑ |
| Pulse repetition frequency (Hz) | 1,000 | ✓✓✓ |
| Duty cycle (%) | 36 | ✓✓✓ |
| Sonication duration (s) | 1 | ✓✓✓ |
| Free-field pressure (kPa) | not reported | |
|---|---|---|
| Free-field Isppa (W/cm²) | not reported | |
| Free-field Ispta (W/cm²) | not reported | |
| In-situ estimate | not reported | |
| In-situ pressure (kPa) | not reported | |
| In-situ Isppa (W/cm²) | not reported | |
| In-situ Ispta (W/cm²) | not reported |
One second pulse trains of 500 kHz (1 kHz pulse-repetition frequency, 36% duty cycle) LIFU were delivered to insular targets using a single-element focused transducer (Sonic Concepts H-281). LIFU was applied 200ms before a painful stimulus (i.e., contact heat evoked potential) for 1-second with a 36% duty cycle.
Flags from extraction
sham_type— The reused LIFU dataset description does not mention a sham or control condition; classified as 'none' by default and should be checked against the original Legon/Strohman source paper.exposures[0].timing.pulse_duration_ms— Only PRF (1 kHz) and duty cycle (36%) are stated for the pulse train; individual pulse duration would require arithmetic and was left unreported.n_sessions_per_subject— Not stated whether anterior and posterior insula targets were stimulated in the same or separate sessions.exposures[0].free_field— No pressure or intensity values are reported anywhere in this paper for the LIFU dataset.