tDCS may reduce tinnitus loudness slightly: a 2022 meta-analysis of 14 trials found a small effect that held only for stimulation over the left temporoparietal area, and a 2026 analysis could not rule out no effect.
Transcranial direct current stimulation is the cheaper, more portable cousin of transcranial magnetic stimulation: a weak constant current passed between two electrodes on the scalp, rather than a magnetic pulse. It has been tried for tinnitus for well over a decade, and the accumulated evidence is now large enough to be worth summarising precisely.
What the pooled evidence shows
A 2022 systematic review and meta-analysis searched eight databases through July 2021 and included 14 controlled trials with 1,031 participants. Every included study had a sham or control group.
Overall:
- Loudness: significantly reduced, standardised mean difference −0.35 (95% CI −0.62 to −0.08, p = 0.01).
- Distress: significantly reduced, standardised mean difference −0.50 (95% CI −0.91 to −0.10, p = 0.02).
- Psychiatric symptoms: no significant effect.
A standardised mean difference of −0.35 is a small effect; −0.50 is conventionally moderate. Both confidence intervals stay on the beneficial side of zero, but neither comfortably.
A 2026 meta-analysis found the same effect size and called it null
This is worth putting directly beside the numbers above, because it is the clearest illustration on this site of what a confidence interval is for.
A 2026 systematic review and meta-analysis in The Laryngoscope covered randomised trials of neuromodulation for chronic tinnitus between 2015 and 2025 — 26 RCTs, 1,576 participants, of which 11 trials and 372 participants were transcranial electrical stimulation. Its pooled result for tDCS:
From the sourceSMD −0.36 (95% CI −0.75 to 0.02), I² = 51%.
Compare that with the 2022 figure above: −0.35 for loudness, reported as significant. The point estimate is, for practical purposes, identical. What changed is that this interval crosses zero, so the same apparent effect is now reported as non-significant.
Nothing contradictory has happened. Two analyses drew on different trial sets with different inclusion rules, and an effect this small sits close enough to the threshold that a modest change in which studies are pooled moves it across. That is the honest characterisation of tDCS for tinnitus: an effect small enough that competent analysts pooling overlapping literatures disagree about whether it exists at all.
The same paper found rTMS at SMD −0.15 (−0.37 to 0.07) with I² = 0% — eight trials, no heterogeneity, no effect. And it found the one modality that did hold up: bimodal stimulation, with reductions often of 10–20 points on the THI or TFI and durability to twelve months. Bimodal neuromodulation is where that evidence sits.
The subgroup finding that matters more than the headline
The review split the trials by where the current was applied. Six studies targeted the dorsolateral prefrontal cortex, six the left temporoparietal area, and two both.
From the sourceThe subgroup analysis showed a significant effect only for tDCS over LTA for loudness (SMD = −0.46), and no other area resulted in significant change.
That is the sentence to carry away. The overall positive result rests on one target and one outcome. Stimulation over the prefrontal cortex — used in nearly half the studies — did not reach significance for anything.
This is the same shape as a lot of neuromodulation research, and it cuts both ways. It is a reason to take the left temporoparietal result more seriously than a bare pooled average, because it identifies where the effect lives. It is also a reason to treat "tDCS helps tinnitus" as too loose a statement to act on.
The negative result is informative
No significant effect on psychiatric symptoms is worth pausing on, because almost everything else with good evidence in this field moves those numbers.
CBT, internet-delivered CBT, Progressive Tinnitus Management — all of them shift anxiety and depression alongside tinnitus distress, because that is the pathway they work through. tDCS apparently does not. Whatever it is doing, it is doing it somewhere else, which is at least consistent with a direct effect on auditory processing rather than on the reaction to it.
What the authors themselves conclude
Not that it is ready. Their words: tDCS may improve tinnitus loudness and distress with a small to moderate effect size; despite the overall positive effect, only left temporoparietal stimulation yielded a significant result; and further well-controlled studies with larger sample sizes and broader exploration of montages and doses are warranted.
That is a research agenda, not a treatment recommendation, and it should be read as one. The 2024 VA/DoD tinnitus guideline reached the same place from its own review of the evidence: insufficient evidence to recommend for or against tDCS for tinnitus management.
What this means if you are considering it
- In a trial or a specialist centre, tDCS is a reasonable thing to be offered, with realistic expectations — a small effect on loudness, at one specific target.
- As a home device, no. The trials used defined montages, currents and schedules under supervision. Approximating that with consumer hardware on the basis of a subgroup finding is not the intervention that was studied.
- If you have already tried it and nothing happened, that is entirely consistent with the evidence. An SMD of −0.35 across a population means a lot of individuals got nothing.
- If someone offers it as a cure, the meta-analysis they are implicitly citing asks for better studies in its own conclusion.
For where this sits relative to everything else, the treatment comparison puts the neuromodulation approaches side by side with the ones that have stronger evidence and lower cost.