Mechanisms and clinical potential of combined tDCS and virtual reality in psychiatric disorders: a systematic review

Abstract

Background

Transcranial direct current stimulation (tDCS) and virtual reality (VR) have emerged as promising non-invasive interventions in treating psychiatric disorders. Despite their individual efficacy in improving symptoms of various psychiatric conditions, the understanding of the combined use of tDCS and VR is limited. This review aims to evaluate the clinical effects and mechanisms of combined tDCS and VR in treating psychiatric disorders.

Methods

We conducted a PRISMA 2020–compliant systematic review, searching major databases (PubMed, Web of Science, Scopus, PsycINFO, ScienceDirect, Cochrane Library, Google Scholar, medRxiv and ClinicalTrials.gov) for studies from January 2000 to July 2025 that evaluated combined tDCS–VR in psychiatric populations. Eligible clinical trials were screened, with tDCS/VR parameters and clinical outcomes extracted, and randomized controlled trials appraised using the Cochrane Risk of Bias 2 tool.

Results

Fourteen studies met inclusion criteria: seven reviews and seven empirical trials (five randomized controlled trials, two pilot/feasibility studies) using mainly 1–2 mA prefrontal tDCS paired with disorder-congruent VR. In post-traumatic stress disorder (PTSD) and specific phobias showed short-term symptom reductions, with some PTSD benefits maintained up to 12 months. Evidence for social anxiety and mild cognitive impairment-related depression was limited to single small RCTs with transient or inconsistent improvements. Overall confidence in the evidence is limited by small sample sizes, variable protocols, and risk‑of‑bias concerns.

Conclusion

Although seven small, heterogeneous studies indicate that combined tDCS–VR is feasible and shows preliminary therapeutic promise—most consistently in PTSD and, to a lesser extent, in specific phobias—the overall evidence base remains limited. Mechanistic findings suggesting modulation of medial and ventromedial prefrontal–amygdala circuits are still exploratory. Given substantial methodological heterogeneity, small sample sizes, and risk of bias, tDCS–VR should be regarded as experimental. The larger, well‑designed, disorder‑tailored randomized controlled trials using standardized stimulation/VR protocols, mechanistic outcome measures, and efforts to identify predictors of response are required before routine clinical implementation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12991-025-00621-6.

Background

Psychiatric Disorders, such as depression, anxiety, and post-traumatic stress disorder (PTSD), are characterized by disturbances in mood, cognition, and behavior, and arise from complex genetic, biological, psychological, and environmental factors [1, 2]. Conventional pharmacological treatments often have limited efficacy and notable side effects [3], driving the exploration of neuroscience- and technology-based alternatives. Transcranial direct current stimulation (tDCS) is a portable, flexible neuromodulation technique that enhances neural plasticity, particularly within prefrontal cortical regions implicated in various psychiatric conditions [4–6]. However, while tDCS alone has demonstrated efficacy across various psychiatric disorders, its therapeutic effects are often unstable in severe or chronic cases [7–9]. This has led to growing interest in combination therapies—for example, pairing tDCS with pharmacological interventions can enhance certain outcomes, although results with psychological therapies remain inconsistent [10, 11].

Recent advances highlight the potential of combining tDCS with immersive virtual reality (VR), offering a combined approach to psychiatric treatment. VR creates ecologically valid, disorder-specific scenarios that engage patients in cognitive and emotional processing, while tDCS may enhances neuroplasticity in prefrontal circuits relevant to symptom expression [5, 12, 13]. Among psychiatric disorders, the strongest short-term evidence for combined tDCS–VR interventions has emerged in phobias and post-traumatic stress disorder (PTSD). In these conditions, VR-based exposure therapy targets disorder-related cues (e.g., feared situations or trauma reminders), and concurrent prefrontal tDCS aims to strengthen emotion regulation and fear-extinction networks involving medial and ventromedial prefrontal cortex regions (mPFC and vmPFC) [14–19; Fig. 1]. However, no systematic review has yet evaluated the clinical efficacy of combined tDCS–VR therapy. This review therefore aims to clarify the current state of research in this emerging field by examining clinical outcomes, identifying target disorders, outlining putative mechanisms, and discussing future directions.

Fig. 1.

Putative mechanisms of tDCS-VR combined treatment for psychiatric disorders

Methods

Protocol and reporting standards

This systematic review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA 2020) guidelines [14]. The review protocol was registered in the Open Science Framework (https://osf.io/g6ceu). A PRISMA-style flow diagram of the study selection process is presented in Fig. 2, and the completed PRISMA checklist is provided in Supplementary Table S1.

Fig. 2.

PRISMA flowchart of the literature search and screening procedure of the tDCS + VR reviews and studies

Search strategy

We systematically reviewed studies on combined tDCS–VR interventions for psychiatric disorders. A comprehensive search was conducted for articles published between January 2000 and July 2025 in PubMed, Web of Science, Scopus, PsycINFO, ScienceDirect, the Cochrane Library, Google Scholar, medRxiv, and ClinicalTrials.gov. English-language records were identified using Boolean combinations of terms such as “tDCS”, “transcranial direct current stimulation,” “virtual reality,” and “psychiatric disorders,” supplemented by broader concepts (e.g., “neuromodulation,” “mental illness,” “technology-supported therapy,” “clinical trials,” “immersive technology,” “brain stimulation,” “digital therapeutics,” “telemedicine,” “artificial intelligence,” “personalized medicine”). Although the search strategy relied on English terms and records, one non‑English full text (Korean) was identified via its English abstract and was screened; its full text was reviewed with translation assistance and data were extracted from the translated article.

Study selection

Eligible studies were randomized controlled trials (RCTs), cohort or experimental/quasi-experimental designs that evaluated combined tDCS–VR interventions and reported clinical or neurocognitive outcomes (e.g., symptom change, cognitive performance, or brain activity). We excluded: (1) studies of tDCS or VR alone; (2) non-clinical (animal or theoretical) work; (3) research on non-psychiatric conditions; and (4) reports with incomplete or unclear outcome data. At the time of this review, all included studies provided complete analyzable data. The only previously registered combined-treatment trial without results (https://clinicaltrials.gov/study/NCT03372460#study-overview, first posted December 13, 2017) published its findings in 2024 andthe published article (not the registry entry) was included in the review . No additional combined-intervention studies with missing data were identified. One reviewer (XRG) independently screened titles/abstracts and full texts against the eligibility criteria, and subsequently extracted data using a piloted form while the second (ZMZ) checked all entries; discrepancies at either stage were resolved by discussion or consultation with a senior author (HTJ), and no automation tools were used. For each study, we extracted information on design, sample size, intervention characteristics, outcomes and reporting between-group differences effect sizes where available (see Table 1). For tDCS, this included current intensity (mA), session duration, number and frequency of sessions, and electrode montage (anode/cathode locations). For VR, we recorded scenario content and purpose (e.g., combat, public speaking, height exposure), VR session duration, total number/frequency of sessions, and timing relative to tDCS (simultaneous vs. sequential). We then qualitatively compared trends and differences in outcomes across disorders (phobias, PTSD, anxiety, depression), protocols and treatment duration.

Table 1.

7 VR + tDCS clinical experimental studies

Diagnosed type	Studies (year)	Subjects and Design and RoB 2	tDCS and VR treatment protocols (Intensity, Duration, Combined modes)	tDCS + VR Combination therapy outcomes and effect sizes	Follow-Up Period and effect
PTSD	van’t Wout-Frank et al. (2019, 2021, 2024)	54 participants (treatment group: 26) RCT (Low RoB 2)	tDCS (vmPFC) Intensity: 2 mA Duration: six 25-mins/10days VR: six 25-mins VR warzone exposures Modes: synchronous	Superior reduction in self-reported PTSD symptom Effect size: d = − 0.82	1 month, 1 year (the symptom relief effect is still significant)
Phobias	Bulteau et al. (2022); Hui et al. (2024)	28 and 64 (treatment group: 11 and 30) participants RCT (Low and Some Concerns RoB 2)	tDCS (vmPFC/mPFC) Intensity: 1 and 1.5 mA Duration: two 20-mins/week and two 20-mins/day VR: four 10/20 mins and two 50-mins height exposure Modes: synchronous/sequential	Significant reduction in the fear of height without noticeable adverse effects and acrophobia Effect size: d = 0.11/0.83	None
Social anxiety	McDonald et al. (2024)	31 (treatment group: 16) participants RCT (Low RoB 2)	tDCS (vmPFC) Intensity: 2 mA Duration: one 20-mins/day VR: six 3-mins speeches exposure Modes: synchronous	Only produced a temporary lower expected threat and anxiety ratings in-group Effect size: d = 0.01	1 month (increased in-group favoritism, dissipating by 1 month)
Mild cognitive impairment with depressive symptoms (neurocognitive condition)	Kim et al. (2021)	50 (treatment group: 25) participants RCT (High RoB 2)	tDCS (DLPFC) Intensity: 1 mA Duration: five 50-mins/six weeks VR: five 50-mins training program Modes: synchronous	Significantly improved the cognitive function of mild cognitive impairment patients and decreased the depression of them. Effect size: no report	None

^※RCT: randomized controlled trial; RoB 2: Cochrane Risk of Bias 2.0. Study-level risk of bias was assessed with RoB 2 (see Supplementary Table S2). DLPFC: Dorsolateral prefrontal cortex

van’t Wout-Frank et al. [24, 25, 26] represent a sequential program of work: the 2019 study was a small pilot RCT, the 2021 article a methodological protocol paper, and the 2024 JAMA Psychiatry trial the confirmatory randomized clinical trial, and stimulation and VR parameters shown in this row refer to the 2024 trial

Quality assessment

Methodological quality was assessed for all empirical studies. Risk‑of‑bias assessments were performed independently by two reviewers (XRG and ZMZ), with disagreements resolved by discussion and, when necessary, consultation with a senior author (HTJ). Randomized controlled trials were assessed using the Cochrane Risk of Bias 2 tool (RoB 2) [15]. Domain-level judgements were combined into overall risk-of-bias ratings according to RoB 2 guidance (Table 1), with detailed study-level ratings provided in Supplementary Table S2.

Results

Despite growing interest in tDCS–VR integration for psychiatric conditions, the screening process identified only 14 eligible publications: 7 review articles (2 systematic reviews and 5 narrative or perspective reviews) and 7 empirical studies (5 RCTs and 2 small pilot or feasibility trials; Fig. 2). To avoid conflating evidence levels, we first summarize the conclusions of existing reviews and then synthesize findings from the primary trials.

Overview of secondary evidence

The review articles suggest a synergistic potential between VR and.

Non‑invasive brain stimulation (NIBS)—such as repetitive transcranial magnetic stimulation (rTMS) and transcranial direct current stimulation (tDCS)—for cognitive, emotional and motor rehabilitation [16–22]. They report that VR–NIBS can enhance cognitive training in immersive, ecologically valid settings improving risk assessment, attention and executive function [16, 17] and has been proposed as a way to enhance cognition and potentially improve psychological well-being in mild cognitive impairment (MCI) [18]. Several reviews also note promise for PTSD, phobias and treatment‑resistant conditions and suggest combined approaches may be more effective [19–22]. However, the empirical base is small and heterogeneous, and reviewers call for rigorous RCTs before clinical recommendations. Most reviews conflate rTMS and tDCS, making modality‑ and disorder‑specific effects hard to isolate. In contrast, this review focuses specifically on tDCS–VR for PTSD, phobias, depression and anxiety, integrating recent trials to identify disorder‑specific benefits and mechanisms.

Primary empirical evidence

The seven empirical studies employed diverse tDCS–VR protocols across disorders (Table 1). tDCS was typically delivered at 1–2 mA with anodal electrodes over prefrontal regions (Dorsolateral prefrontal cortex (DLPFC), vmPFC, mPFC) and cathodal electrodes over contralateral prefrontal, occipital or inion sites. VR scenarios were disorder-congruent: combat or warzone environments for PTSD, height exposure for acrophobia, virtual public speaking for social anxiety, and cognitive training program for older adults with MCI. Sessions lasted 20–50 min, with 1–6 sessions per trial. In most studies, tDCS and VR were applied simultaneously; one pilot used a sequential design with tDCS immediately before VR exposure [23]. Sample sizes ranged from 16 to 30 participants per arm (28–64 per study). Follow-up ranged from none to 12 months, and outcomes included self-reported and clinician-rated symptoms and cognitive or behavioural measures. A structured comparison of montages, VR scenarios, timing and outcomes is provided in Table 1.

Risk of bias in the six randomized controlled trials, assessed with the Cochrane RoB 2 tool, was overall low in three studies, of “some concerns” in two (includes a pilot study) and high in one, mainly due to issues in outcome measurement (Table S2). Despite heterogeneity and small samples, disorder-specific patterns emerged. Four PTSD studies reported significant reductions in self-reported symptoms after tDCS–VR, with benefits maintained up to 12 months in some trials [24–27]. In specific phobias, combined tDCS–VR reduced fear and avoidance, but effect sizes were modest and follow-up data sparse [18 (some concerns), 19]. One RCT in older adults with MCI and depressive symptoms found improvements in both cognition and mood [28], however this trial was rated as high risk of bias; because the intervention explicitly targeted depressive symptomatology, we classified this trial under depressive disorders while acknowledging the co-occurring neurocognitive condition. In social anxiety, a single pilot RCT showed transient reductions in threat and anxiety ratings, increased in-group favoritism that normalized by 1-month follow-up, and no sustained symptom improvement [29]. Overall, these seven studies suggest that tDCS–VR is feasible and potentially beneficial, with relatively consistent short-term benefits in PTSD and specific phobias (albeit with modest effect sizes and limited follow-up in the latter), and weaker or inconsistent effects in social anxiety and depression.

Discussion

The present review integrated findings from both secondary literature and a small number of primary trials. Existing reviews of VR–NIBS highlight the theoretical promise of combining immersive exposure with neuromodulation across cognitive, emotional and motor domains, but also emphasize that empirical evidence remains scarce and heterogeneous [16–22]. Our conclusions therefore rest mainly on seven clinical tDCS–VR studies, among which two are pilot or feasibility trials [24, 25]. At this stage, tDCS–VR should thus be considered an experimental rather than an established treatment. It is plausible that combined tDCS–VR facilitates fear extinction by modulating medial prefrontal regions (including vmPFC) and their interactions with the amygdala (Fig. 1). Neuroimaging and meta‑analytic evidence in PTSD and specific phobias show exaggerated amygdala responses to disorder‑relevant cues together with hypoactivation of mPFC/vmPFC [30–32], providing targets for intervention. VR exposure reliably engages these dysfunctional circuits with ecologically valid trauma‑ or fear‑related scenarios, while prefrontal tDCS can bias plasticity within the same networks: tDCS has been reported to modulate prefrontal–amygdala coupling, reduce threat‑related amygdala responsivity, and alter emotional processing [33]. Consistent with this framework, the tDCS–VR trials reviewed here most consistently report improvements in fear reduction and emotional control in PTSD and specific phobias [23, 26, 34]. However, direct neuroimaging evidence from combined tDCS–VR protocols in clinical samples remains scarce and is currently limited to preliminary data in veterans with PTSD showing modulation of ventromedial prefrontal–basolateral amygdala connectivity [35].

Disorder-specific differences in efficacy likely reflect both underlying pathophysiology and the extent to which VR scenarios capture core symptoms and elicit sufficient cognitive–emotional engagement. PTSD and simple phobias involve relatively circumscribed fear networks triggered by discrete external cues; here, highly congruent VR environments (e.g. combat scenes, height exposure) may provide potent exposure contexts that interact favorably with prefrontal neuromodulation, yielding more robust and durable benefits [23–26, 34]. By contrast, social anxiety and depressive disorders are characterized by diffuse interpersonal fears, negative self-beliefs and pervasive rumination, which are more difficult to reproduce and modify in brief VR scenarios. This may contribute to the weaker or short-lived clinical effects observed in social anxiety and MCI-related depression [28, 29]. Moreover, the only depression study was conducted in older adults with MCI, limiting generalizability to primary mood disorders [28].

Methodological heterogeneity remains a major constraint, but some patterns emerge. As shown in Table 1, multi‑session (1–2 mA) anodal prefrontal stimulation paired with disorder‑congruent VR within the same sessions (e.g., combat scenes for PTSD, heights for acrophobia) tended to yield larger, more durable symptom reductions [23–26, 34] than single‑session or low‑dose protocols and studies using generic VR tasks or brief social scenarios, which produced smaller or transient effects [28, 29]. Timing of of tDCS relative to VR exposure is unresolved but important: most trials applied tDCS online (concurrently with VR), whereas one acrophobia pilot used pre‑exposure stimulation [23]. From a neuroplasticity perspective, online tDCS stimulation is hypothesized to enhance state‑dependent modulation during VR [36], while pre‑exposure stimulation may influence consolidation [37]; however, existing data are too sparse to favor one approach. This review also has limitations. We restricted relied primarily on published reports, and the small number and heterogeneity of studies precluded meta-analysis and formal assessment of reporting bias or certainty of evidence. Given the small, heterogeneous samples and several domains with high or unclear risk of bias, current findings are preliminary and do not support adoption of standardized protocols [23–26, 28, 29, 34]. There is no consensus on stimulation or VR parameters: studies differ widely in electrode montage, current intensity, session duration and number, as well as in VR content and exposure design, which hampers cross‑study comparison and precludes reliable quantitative synthesis. This methodological heterogeneity raises bias risk and limits generalizability. Most trials are underpowered and have short follow‑ups, restricting assessment of durability and relapse prevention [23, 28, 34]. Reliable blinding and placebo control remain challenging in neuromodulation and immersive VR and may increase susceptibility to expectancy effects. Finally, pragmatic barriers — including equipment costs, need for technical support, and variable patient adherence — may constrain real‑world scalability. To advance the field we recommend: (1) harmonized, adequately powered RCTs that systematically manipulate timing, dose, montage, session number, and VR design while using standardized clinical and mechanistic outcomes and rigorous blinding procedures; (2) development of personalized, biomarker‑informed protocols that tailor stimulation and VR parameters to individual characteristics (e.g., age, symptom profile, neurobiology); and (3) evaluation of long‑term efficacy, relapse prevention, and broader applications (e.g., primary depression, generalized anxiety), including testing adjunctive combinations with psychotherapy or pharmacotherapy [27]. If these constraints are addressed, the portability of tDCS and VR could enable safe, scalable at‑home interventions [8].

Conclusion

In conclusion, evidence from seven small, methodologically heterogeneous studies support the feasibility of combined tDCS–VR and suggest preliminary therapeutic promise—most consistently for PTSD and more modestly for specific phobias—while any superiority over single‑modality treatments (particularly for social anxiety and depression) remains unproven. Early mechanistic data point to modulation of mPFC activity and vmPFC–amygdala coupling, but these findings are exploratory. Given the limited, protocol variability, and residual risk of bias, tDCS–VR should be considered experimental; larger, well‑designed, disorder‑tailored RCTs with standardized protocols and efforts to identify response predictors are required before clinical adoption.

Abbreviations

tDCS

Transcranial direct current stimulation

rTMS

Repetitive transcranial magnetic stimulation

NIBS

Non-invasive brain stimulation

VR

Virtual reality

PTSD

Post-traumatic stress disorder

mPFC

Medial prefrontal cortex

vmPFC

Ventromedial prefrontal cortex

DLPFC

Dorsolateral prefrontal cortex

RCTs

Randomized controlled trials

RoB

Cochrane Risk of Bias

MCI

Mild cognitive impairment

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

Author contributions

H.T.J and X.R.G Proposed the theme idea and main framework, X.R.G wrote the main manuscript text. X.R.G., Z.Z.M. and H.T.J. contributed to the Methods and Results sections. H.T.J and L.Y.C mainly revised the manuscript. H.Q.J, Z.M.Z and J.H.Z assisted in revising the manuscript. All authors reviewed the manuscript.

Funding

This work was supported by STI2030-Major Projects (2022ZD0212400， 2021ZD0200404), National Natural Science Foundation of China (82371453), Key R&D Program of Zhejiang(2024C03006, 2024C04024, 2024ZY01010 ), Fundamental Research Funds for the Central Universities (2025ZFJH01-01), Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China (JYB2025XDXM605) and the Construction Fund of Key Medical Disciplines of Hangzhou (2025HZGF10).

Data availability

All data used in this review are derived from published studies cited in the reference list. Further details can be obtained from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.