Effects of transcranial focused ultrasound stimulation to human lateral geniculate nucleus on visual perception and steady-state visual evoked potentials
Martin TW Scott, Patti N Limon, Morteza Mohammadjavadi, Benjamin R. Kop, Nai-Feng Chen, Eva A Feredoes, Vladimir Vildavski, Gerald R Popelka, Anthony M. Norcia, Kim Butts Pauly, Ryan T Ash
2026 · 10.64898/2026.07.30.741804
Abstract
Transcranial ultrasound stimulation (TUS) is an emerging tool to non-invasively modulate neural activity in deep brain areas. A key need in accelerating TUS into cognitive neuroscience and neuropsychiatry is to better understand how different sonication parameters relate to neuromodulatory effects. Here we assess the role of pulse repetition frequency (PRF), a key TUS parameter thought to determine the relative contribution of molecular displacement and acoustic radiation force effects on neural tissue using the human subcortical visual pathway as a testbed. We combined frequency-tagged steady-state visual evoked potential (SSVEP) measures of contrast-response with contrast increment detection psychophysics as neural and behavioral readouts of visual pathway function. We used structural MRIs and acoustic simulations to target the lateral geniculate nucleus (LGN). Concurrent with visual stimulus presentation, the left LGN or a more superficial control site were stimulated with a neuronavigated depth-steerable 4-element TUS transducer at a range of PRFs with 68 W/cm 2 free-water I SPPA , and a 10% duty cycle. An effective white-noise auditory mask blinded participants to stimulation conditions. Recordings from 25 neurotypical participants failed to detect any impact of TUS on SSVEP response amplitude, SSVEP response latency, or perceptual behavior. Analysis of simulations generated from the measured transducer positions grant reasonably high confidence that the LGN was within the TUS focus in most participants, with no correlation between targeting accuracy and changes in activity during TUS. Our results provide a cautionary note about the effect size of neuronavigated TUS for online causal manipulations in cognitive and clinical neuroscience.
Abstract via europepmc.
Exposures
Exposure 1: LGN-TUS (70 mm depth)
Target: lateral geniculate nucleus — “left LGN”
Device: NeuroFUS · Sonic Concepts (sold by Brainbox) · NeuroFUS Pro CTX-500, 4-element annular transducer ✓
| Waveform | pulsed | |
|---|---|---|
| Fundamental frequency (kHz) | 500 | ✓✓✓ |
| Pulse duration (ms) | not reported | ⚑ |
| Pulse repetition frequency (Hz) | 4.88, 48.75, 487.5swept | ✓✓✓ |
| Duty cycle (%) | 10 | ✓✓✓ |
| Sonication duration (s) | 12 | ✓✓✓⚑ |
| Free-field pressure (kPa) | 1,400 | ✓✓✓ |
|---|---|---|
| Free-field Isppa (W/cm²) | 68 | ✓✓✓ |
| Free-field Ispta (W/cm²) | not reported | |
| In-situ estimate | simulationmean or range across subjects | |
| In-situ pressure (kPa) | 690 | ✓✓✓⚑ |
| In-situ Isppa (W/cm²) | 16.2 | ✓✓✓⚑ |
| In-situ Ispta (W/cm²) | not reported |
12 s trains at 10% duty cycle, PRF randomised trial by trial among 4.875, 48.75 and 487.5 Hz, with 5% linear ramps on the pulses at the two lower PRFs; synchronised with visual stimulus onset; eight 16-trial TUS blocks interleaved with no-TUS trials, depth 70 mm (LGN) or 30 mm (control).
Exposure 2: Control-TUS (30 mm depth)
Target: white matter, auditory cortex — “a more superficial control site in the auditory cortex and subcortical white matter”
Device: NeuroFUS · Sonic Concepts (sold by Brainbox) · NeuroFUS Pro CTX-500, 4-element annular transducer ✓
| Waveform | pulsed | |
|---|---|---|
| Fundamental frequency (kHz) | 500 | ✓✓✓ |
| Pulse duration (ms) | not reported | |
| Pulse repetition frequency (Hz) | 4.88, 48.75, 487.5swept | ✓✓✓ |
| Duty cycle (%) | 10 | ✓✓✓ |
| Sonication duration (s) | 12 | ✓✓✓ |
| Free-field pressure (kPa) | 1,400 | ✓✓✓ |
|---|---|---|
| Free-field Isppa (W/cm²) | 68 | ✓✓✓ |
| 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 |
12 s trains at 10% duty cycle, PRF randomised trial by trial among 4.875, 48.75 and 487.5 Hz, with 5% linear ramps on the pulses at the two lower PRFs; synchronised with visual stimulus onset; eight 16-trial TUS blocks interleaved with no-TUS trials, depth 70 mm (LGN) or 30 mm (control).
Flags from extraction
n_subjects— Text says n=25 and 'Twenty-five subjects were retained with 2 rejected', implying 27 enrolled; 25 recorded.exposures[0].timing.sonication_duration_s— Trains stated as 12 s, but trial text also says 11 s trials and 10.667 s of TUS per 13.333 s trial.exposures[0].in_situ.isppa_w_cm2— Results give mean simulated in situ Isppa 16.2 W/cm2 (range 8.5-20.5) while the Discussion says about 8 W/cm2 on average at the target.exposures[0].in_situ.pressure_kpa— 0.69 +/- 0.33 MPa is a peak negative pressure, with range 0.5-0.78; the SD looks inconsistent with the range.exposures[1].target.terms— Control site is described as auditory cortex and subcortical white matter at 30 mm depth; no separate dosimetry reported.exposures[0].timing.pulse_duration_ms— Pulse duration not stated; duty cycle and PRF are given but no arithmetic done.sham_type— Mask-only no-TUS trials were also interleaved; whether the transducer was powered off is not stated.