Temporal Interference Stimulation Increases Field Potential Power at the Interference Frequency, and Affects Single Unit Activity in Humans
DOI: Pending DOI
Synopsis
Temporally interfering electric fields (TIEF) refers to the interaction of two or more high -frequency (kHz) electric fields to generate a low-frequency amplitude modulation ‘envelope’, proposed as a means of non-invasive deep-brain modulation. However, direct neurophysiological evidence of its effects in humans remains non-existent. We recorded intracranial field potentials (FPs) and single-unit activity in epilepsy patients undergoing stereo-EEG while applying TIEF stimulation. TIEF produced spatially localized increases in FP power at the interference frequency that aligned with modeled electric field distributions and showed frequency-specific temporal dynamics. These findings provide the first intracranial characterization of TI-induced neural dynamics in humans.
Background
Temporal interference (TI) stimulation is a non-invasive neuromodulation method that applies two high-frequency (kHz) alternating currents with a frequency offset so that their superposition creates a lower-frequency amplitude modulation envelope, hypothesized to be capable of modulating neural activity in arbitrarily targeted and deep brain regions, offering the potential for deep brain neuromodulation without surgery. While TI has been widely studied in computational models and animal experiments, direct neurophysiological evidence of its intracranial effects in humans remains non-existent.
Investigations in minipigs and mice using implanted electrodes reported TI-related modulation of FP activity in vivo, providing initial evidence that TI may modulate FP activity1,2. This suggests that envelope-frequency fields can influence aggregate synaptic currents. Non-human primate experiments show that TI may alter the timing, but not the rate of spiking activity3. Despite these findings, replication in humans, with rigorous paradigms that provide high statistical power, and exclude all possible artifacts are needed. Systematic characterization of TI-induced LFPs and single-unit responses in humans is absent, and the spatial topography and temporal dynamics of such modulation remain unexplored. Intracranial recordings such as stereo-EEG provide a unique opportunity to fill this gap by mapping the anatomical distribution and frequency-specific dynamics of TI effects with high spatiotemporal resolution, and the possibility of tetrodes for single unit recording.
Methods
We performed TIEF stimulation in adult patients with medically refractory epilepsy undergoing stereo-EEG (sEEG) for presurgical evaluation. sEEG provides high spatiotemporal resolution to measure intracranial field potentials and single-unit activity through the use of deployable tetrode or Behnke-Fried microwires. Participation is voluntary with UC Davis IRB oversight and informed consent. TIEF stimulating electrode placement is replicated from Missey et al. 20264, to target the anterior temporal lobe TIEF protocol: One session using scalp electrodes delivering two high-frequency carrier signals to create low-frequency envelopes at targeted sites. Four stimulation envelopes are evaluated [Δ5Hz, Δ10Hz, Δ40Hz, Δ130Hz], using 20 epochs of 5s of envelope stimulation, and 25s of equal frequency carriers with 180° phase shift. Data were recorded with a Neuralynx Atlas, and frequency domain analyses were conducted to recover electric fields and FP activity. Single units were identified with spike sorting and peri-stimulation spike-rate changes were examined.
Results
Seven adult participants undergoing stereo-EEG (sEEG) monitoring for presurgical evaluation were included. Across participants, Temporal Interference electric field (TIEF) stimulation produced frequency-specific modulation of intracranial local field potentials (FPs) at the interference (envelope) frequency, in the anatomical distribution of peak amplitude modulation, and as predicted by finite element modeling 5. Envelope frequencies Δ10 Hz, Δ40 Hz, and Δ130 Hz, stimulation produced focal increases in spectral power at the offset frequency relative to sham (equal-frequency carriers delivered with a 180° phase shift). The spatial distribution of contacts exhibiting the largest power increases closely matched regions of maximal envelope electric field predicted by finite element simulations 5. Across subjects, this region formed a cone-like spatial profile extending from the cortex current path toward deeper temporal lobe structures. The spatial pattern of modulation was consistent across Δ10 Hz, Δ40 Hz, and Δ130 Hz conditions. In contrast, Δ5 Hz stimulation did not produce a reliable increase in FP theta band power, and therefore did not yield a stable spatial map of modulation.
Time-resolved spectral parameterization using Spectral Parameterization Resolved in Time (SPRiNT) revealed distinct temporal dynamics of LFP power modulation during the 5-second stimulation epochs. Δ10 Hz stimulation produced brief and spatially spreading increases in alpha-band power that appeared transient and propagated beyond the region of maximal field exposure. Δ40 Hz stimulation produced more sustained increases in gamma-band power localized primarily to temporal lobe contacts, accompanied by intermittent pulsatile coupling with surrounding regions. In contrast, Δ130 Hz stimulation generated a relatively stable increase in high-frequency power confined to a more spatially delimited region, with reduced evidence of pulsatile coupling or propagation.
Feasibility of concurrent single-unit recordings during TIEF was demonstrated in one participant with tetrode recordings. After spike sorting, eight putative neurons were identified. During Δ130 Hz stimulation, these units exhibited a significantly reduced firing rate relative to sham stimulation (Δ0 Hz, carriers with 180° phase offset). Ongoing analyses are evaluating spike timing relationships to the interference envelope, including spike-field coherence and phase locking to the interference frequency.
These results demonstrate that TIEF stimulation can produce spatially localized and frequency-specific modulation of human intracranial LFP activity that aligns with modeled electric field distributions, while also enabling simultaneous single-unit recordings to probe cellular-level responses to temporal interference stimulation.
Discussion
This study provides direct intracranial evidence that Temporal Interference electric field (TIEF) stimulation modulates human neural activity in a frequency-specific manner. Across participants, increases in FP power at the interference frequency spatially aligned with regions of maximal electric field predicted by finite-element modeling, supporting the mechanistic premise that the low-frequency envelope generated by kHz carriers can drive measurable neural responses in the human brain. Importantly, the temporal dynamics of the response varied systematically with envelope frequency, suggesting that different frequency regimes may engage distinct network mechanisms.
These results extend prior work in animal models by providing the first intracranial characterization of TI-induced FP modulation in humans. Furthermore, we demonstrate the feasibility of recording single-unit activity during TI stimulation, an important step toward understanding the cellular mechanisms underlying envelope-frequency neuromodulation. Preliminary single-unit observations suggest that high-frequency envelopes may influence firing rate and potentially spike timing, though larger datasets are required to confirm these effects.
Some limitations should be noted. The cohort size is modest and recordings are constrained by clinically determined electrode placements.
Together, these findings support the ability of TI stimulation to produce localized, frequency-dependent modulation of human brain activity and provide a framework for mechanistic studies of noninvasive deep brain neuromodulation.
Disclosures
RCA is the founder and CEO of Candel Medical Company, an electrical stimulation medical device development company.
All other authors declare no financial conflicts.
References
1. Lin, H.-C., Wu, Y.-H. & Ker, M.-D. Modulation of Local Field Potentials in the Deep Brain of Minipigs Through Transcranial Temporal Interference Stimulation. Neuromodulation 28, 434–443 (2025).
2. Zhu, X. et al. Temporal interference stimulation evoked neural local field potential oscillations in-vivo. Brain Stimul. 16, 341–342 (2023).
3. Vieira, P. G., Krause, M. R. & Pack, C. C. Temporal interference stimulation disrupts spike timing in the primate brain. Nat Commun 15, 4558 (2024).
4. Missey, F. et al. Non-invasive temporal interference stimulation of the hippocampus suppresses epileptic biomarkers in patients with Epilepsy: biophysical differences between kilohertz and amplitude modulated stimulation. Brain Stimul 19, 102981 (2026).
5. ID# 1906835 Planning Tool for Personalized Temporal Interference Stimulation Optimization. Neuromodulation: Technology at the Neural Interface 28, S171 (2025).