Real-time neuromodulation - a practical step towards closed-loop

k-Plan TUS suppresses EEG ERP VEPs

fMRI-guided Ultrasound Localized Measurable Repeatable Modulation

Article:

Dunsford, S., Murphy, K., Darrieutort, E., Fouragnan, E., & Ganis, G. (2026). Unilateral online ultrasound stimulation of early visual cortex suppresses responses to contralateral visual stimuli. Brain Stimulation, 19(2), 103023. https://doi.org/10.1016/j.brs.2025.103023

BrainLatam Comments - First-Person Consciousness:

I read this article as an honest attempt to answer a simple question:

“Is ultrasound actually changing the brain in real time… or am I just seeing side effects (sound, vibration, startle) that look like ‘stimulation’?”

Their strategy is elegant: use the visual cortex, because it is organized by sides. If I manipulate the left visual cortex, the most convincing effect appears in the right visual field (contralateral). If the effect were just “noise in the body,” I would expect something more widespread, less lateralized.

TUS suppresses EEG ERP VEPs

What they did (as I see it):

They stimulated one side of the early visual cortex with k-Plan TUS (guided by fMRI) in 19 people.

They showed checkerboards in the left or right visual field and measured VEPs.

The TUS was applied in half of the trials, randomly, so it's possible to compare myself to myself (strong control).

They also performed modeling/estimation of “target engagement” (how much of the acoustic focus actually hit the target, taking into account skull differences).

What appears as a result (without poetry):

The main pattern is:

VEPs decrease when the visual stimulus is in the contralateral field to the stimulated side.

The same suppression does not appear on the “wrong” side (ipsilateral), which helps to separate it from peripheral confounds. And there's a detail that I consider "down-to-earth":

The greater the engagement/dose at the target, the greater the observed suppression. This sounds like a dose-response relationship, and not like a "placebo effect of the noise."

What I feel the article truly delivers:

I don't leave here thinking "ultrasound is a miracle."

I leave thinking:

"They found a convincing way to show an online and spatially specific effect, using brain topography as a control."

This is useful because the TUS field is constantly struggling with the question: "was it the brain or was it the ear/skin?" Here, the experimental design reduces this doubt.

What remains open (and I don't pretend it's closed):

This is VEP modulation. I still want to see more connection with perception/behavior (what does the person notice? improvement? worsening? in which tasks?).

Skull differences and coupling seem to matter a lot (they themselves model this), so personalization may be key.

And I still want to see more about reproducibility and "safety window/parameters" in different laboratories (the field is moving in this direction).

Three BrainLatam 2026 questions (to become the next post/experiment):

If I change the visual task (more attention, less attention), does this VEP suppression remain the same or does it depend on the state?

If I measure pupil/HRV together, can I better separate "load/alertness" from "local neural effect"?

Is it possible to repeat the same topographic trick in other areas (motor/somatosensory) to test online modulation without confusing it with the periphery?