Contextual Engagement to Enhance Effective Transcranial Direct Current Stimulation: A Theoretical and Mechanistic Perspective

Abstract

Transcranial direct current stimulation (tDCS) is a form of non-invasive brain stimulation that delivers a low amount of electrical energy to targeted brain regions. Recently a tDCS device has been approved by the US Food and Drug Administration for use in major depressive disorder, and tDCS has significant potential for the treatment and rehabilitation of other psychiatric disorders due to its portability and favorable safety profile. Although this approval is an important step forward, the observed effect size was modest. Thus, finding ways to improve the efficacy of tDCS is important. Here we suggest that one way to do so is by leveraging contextual dependence of tDCS effects. Importantly, tDCS modulates neuronal excitability in a state-dependent manner, meaning that concurrent cognitive, emotional, or behavioral engagement might be important to consider. We outline the theoretical rationale for integrating therapeutic engagement with tDCS and the need for longer follow-up to assess efficacy, drawing on mechanistic insights. We then describe prior use of tDCS concurrent with a virtual reality context to underscore this use. To this end, while currently approved tDCS is efficacious, stimulation applied concurrently with behavioral engagement within a therapeutic context might induce further improve acute and longer-lasting outcomes.

1. Introduction

Transcranial direct current stimulation (tDCS) is a non-invasive neuromodulation technique that applies low-level (typically up to 2mA) electrical currents to the scalp to alter cortical excitability (Figure 1) [1]. Over the past two decades, tDCS has been studied across a wide spectrum of conditions, including psychiatric disorders such as depression, anxiety, and trauma-related disorders as well as neurological disorders such as post-stroke motor deficits and cognitive decline. However, despite promising results in laboratory studies, larger-scale randomized clinical trials have produced mixed outcomes, often with modest effect sizes [2].

Figure1.

A schematic of transcranial direct current stimulation. Typical use includes three main elements: 1) a power source that delivers a low level of current, typically up to 2mA (comparable to that obtainable by a 9V battery); 2) the anode, where electricity will flow into the brain, 3) the cathode, where electricity flows out of the brain. This set up creates a simple circuit delivering a modest amount of energy into underlying brain regions. From Zandvakili et al., [3].

In December 2025, the US Food and Drug Administration approved the first tDCS system (Flow FL-100) for the treatment of “moderate to severe major depressive disorder (MDD) in the current episode, either as monotherapy or as an adjunctive treatment, in patients 18 years and older who are not considered treatment refractory to medication.” This was based on a large remotely-supervised trial in MDD that demonstrated statistically significant symptom improvements with home-based tDCS relative to sham stimulation [4]. The FDA’s PMA listing underscores both the indication and boundaries (including exclusion of treatment-resistant depression in trial eligibility), and FDA’s Summary of Safety and Effectiveness Data describes a home-based schedule with 30-minute sessions of 2 mA stimulation five times a week for three weeks and then three sessions per week for seven weeks across 10-weeks.

Despite the importance of this trial for FDA approval, the difference between the active and sham groups was modest. Patients treated with active tDCS achieved an average of ~9 point reduction in their Hamilton Depression Rating Scale, compared with a ~7 point reduction in those treated with sham. While this outcome was statistically significant, a 2-point difference on this rating scale may not be considered clinically meaningful; the result represents a small-to-medium effect size (Cohen’s d of 0.37; number needed to treat of about 6) for clinical response (defined as a 50% reduction in symptoms from baseline). There were no serious adverse events, and side effects were consistent with the known safety profile of tDCS, with more participants in the active group reporting skin irritation, redness, and difficulty concentrating. Although two participants in the active group reported mild skin burns, these did not result in lesions or scarring. As an important point, there were no incidences of systemic side effects (weight gain, diabetes, sexual side effects) from this nonpharmacological intervention.

Thus, the field is positioned for a decisive shift: at-home tDCS is no longer experimental, and the utilized protocol offers a reasonable starting point for clinicians to implement tDCS as an accessible treatment option for non-treatment-resistant depression. Nonetheless, considering the relatively modest effects on symptoms, the field continues to search for ways to improve tDCS efficacy. One possible way to do this is through the utilization of combined approaches in which tDCS is combined with other contextually relevant manipulations.

Although the delivery of tDCS “at-rest”, meaning that no other adjunctive cognitive behavioral therapies were provided concurrently at the time of stimulation, reflects the current standard, the investigation of combined approaches can be framed as a next step in further optimization to maximize clinical benefit and durability. Such combined approaches, where the brain’s functional state is manipulated through direct behavioral context engagement during stimulation, may augment the efficacy of tDCS-based interventions. Below we will discuss the theoretical and mechanistic rationale for why the brain’s functional state and the behavioral context during stimulation might be relevant. A second goal of this paper is a discussion of the timing of efficacy testing. To this end, the clinical effects of tDCS might be delayed, becoming more apparent following the acute stimulation phase. Efficacy testing occurring too soon following the end of stimulation intervention might miss medium-or longer-term clinical benefits.

2. Theoretical Framework for State-Dependent Neuromodulation

The low level direct electrical current used in tDCS modulates resting membrane potentials. This induces subthreshold polarization of neuronal populations. Generally, anodal stimulation is thought to depolarize neurons, thereby increasing excitability, while cathodal stimulation tends to hyperpolarize and reduce excitability. However, to characterize anodal stimulation to be solely excitatory and cathodal stimulation to be inhibitory would be a misnomer, and the effect of current on direction of excitability depends on a variety of factors [5], including the type of neurons stimulated and their specific orientation [6]. Moreover, these subthreshold level effects are not necessarily sufficient to generate action potentials. Rather, tDCS biases neuronal firing probability and alters the gain of neural circuits engaged at the time of stimulation.

With this in mind, we refer to the concepts of anatomical and functional specificity as described by Bikson et al. [7]. Briefly, anatomical specificity refers to the extent to which stimulation selectively targets a particular cortical region or neural structure based on electrode placement and resulting current flow. From an anatomical perspective, tDCS effects are assumed to arise primarily from modulation of the stimulated area (e.g., dorsolateral prefrontal cortex), with behavioral outcomes attributed to changes in excitability within that anatomically defined region. However, biophysical modeling and empirical data consistently demonstrate that conventional tDCS produces diffuse and spatially broad electric fields, with substantial inter-individual variability driven by anatomy [8], tissue conductivity, and electrode montage [9], and which be different for clinical populations [10]. As a result, strict anatomical focality is limited, and stimulation-induced effects cannot be readily mapped onto a single cortical locus.

In contrast, functional specificity refers to modulation of targets that are actively engaged at the time of stimulation, regardless of the spatial extent of the induced electric field. Consequently, stimulation effects emerge preferentially within task-relevant networks, reflecting state-dependent interactions between externally applied electric fields and endogenous neural activity [11]. Critically, functional specificity provides a parsimonious explanation for the context- and task-dependence of tDCS effects and supports the observation that stimulation at rest provides modest clinical effects.

To underscore a key point, the principle of Hebbian plasticity [12] – “neurons that fire together, wire together” – provides the theoretical foundation for why context and functional engagement matters. After all, tDCS alone does not provide the energy needed for long-term potentiation (LTP) or long-term depression (LTD). Instead, it modulates the threshold for plasticity induction, effectively gating synaptic changes that are driven by endogenous neural activity during learning or therapy.

When tDCS is paired with a cognitive task or (therapeutic) behavioral process, the ongoing synaptic activity provides the correlated firing patterns that allow tDCS to bias synaptic strengthening (or weakening) and amplify learning-related plasticity within engaged circuits [13]. Under this view, electrode placement constrains the network space within which modulation occurs, but functional engagement determines which synapses and circuits are ultimately affected. The impact of tDCS thus depends on ongoing neural activity: networks that are active during stimulation are more susceptible to modulation. It should be recognized that individual variability in the ability to active stimulated neural networks can be expected to impact the effect of tDCS [14–16]. This principle of state-dependence suggests that tDCS is a facilitator of plasticity within task-relevant circuits, and that combining tDCS concurrent with a cognitive-behavioral driven activation might ultimately enhance tDCS’ effects.

3. Conceptual Model: Example of tDCS as a “Plasticity Primer”

Following the above, one way to conceptualize tDCS is as a plasticity primer. This is not a novel concept and in fact, rather at the heart of tDCS. Yet despite a growing awareness of the importance of context, to date investigators and clinicians alike may not sufficiently consider this quality of tDCS. The ability of tDCS to bias synaptic modification in response to behaviorally relevant input aligns with models of meta-plasticity, where the history and current state of neural activity determine the direction and magnitude of subsequent plastic changes. To underscore this idea, we will describe our team’s use of tDCS combined with virtual reality (VR) exposure for posttraumatic stress disorder (PTSD) as a case in point for how the field may consider enhancing clinical efficacy. We are mindful that other examples exist and are included in the latter portion of the manuscript.

To establish the pipeline to deliver tDCS with VR exposure, our initial studies examined the effect of tDCS combined with extinction learning of conditioned threat [17, 18]. This was done as extinction learning can be considered a laboratory analogue to exposure therapy [19]; the key component being repeated exposure to feared stimuli (trauma cues or conditioned stimuli) without occurrence of the expected, negative consequences. These studies demonstrated that neuromodulation could be delivered concurrently with a relevant cognitive process and provided an initial signal of effect in both healthy individuals [17] and in patients with PTSD [18]. In translating this work to the clinic, an early pilot study in which tDCS was applied during VR exposure, suggested that active tDCS paired with VR was associated with greater reductions in psychophysiological arousal across sessions compared with sham, consistent with accelerated habituation and potentially enhanced extinction learning [20]. These findings provided a mechanistic signal that tDCS might influence learning trajectories as well as feasibility of immersive exposure without compromising engagement or safety.

Building on these observations, we completed a double-blind randomized clinical trial that evaluated whether these mechanistic advantages translated into clinically meaningful and durable symptom improvements. Veterans with warzone-related PTSD were randomized to receive active or sham tDCS during a standardized course of VR exposure sessions, typically six sessions over 2–3 weeks [21]. At the 1-month primary outcome timepoint, participants receiving active tDCS+VR demonstrated significantly lower PTSD symptom severity compared with the sham+VR group, with a large effect size on self-reported PTSD symptoms (PCL-5; Cohen’s d = 0.82). Importantly, symptom reduction in the active tDCS+VR group did not rebound after cessation of treatment, suggesting that gains were not dependent on ongoing stimulation. A substantial proportion of participants in the active tDCS+VR condition maintained clinically meaningful improvement (e.g., ≥10-point reduction on the PCL-5) over three months, whereas the sham+VR group demonstrated greater variability and partial symptom return.

The persistence of benefit beyond the stimulation window supports the hypothesis that tDCS, when paired with an appropriately engaging context such as VR exposure, may enhance the underlying targeted process (in this example, extinction retention and safety memory consolidation). The pattern of improvement extended out to 1-year, assessed through medical chart review [22], as those in the active tDCS+VR group had longer relapse-free time (315±22 days) compared those in the sham+VR group (231±28 days). Participants randomized to active tDCS+VR were also less likely to relapse (defined as hospitalization, suicide attempt, and or other psychiatric events; OR 3.02; 95% CI 1.1–8.1) and were also more likely to have medication reductions in the year following stimulation.

This case is especially relevant in the current regulatory landscape. With the recent FDA approval of a home-use tDCS for major depressive disorder, the field has moved beyond the question of whether tDCS can be delivered effectively, safely, and at scale, and can now shift to how to deploy tDCS in ways that maximize efficacy and durability.

4. Changing the scientific question: Translating tDCS as a plasticity primer beyond PTSD

Historically, tDCS has been burdened by the binary debate “does it work?” The recent FDA approval of at-home tDCS for MDD demonstrates tDCS can act as a standalone antidepressant agent. With the efficacy question addressed, this approval reframes the debate toward questions such as “for whom does it work best (phenotypes, comorbidity, baseline network state)”, “under what conditions” and “how to enhance long lasting clinical response.”

The PTSD data argue for a more nuanced interpretation: stimulation alone may be beneficial but can be optimized; greater leverage comes from pairing stimulation with a context that makes the desired neural computation “online” when tDCS is delivered. A comparison can be made to the use of tDCS during working memory training; while in a meta-analysis in healthy individuals tDCS effects were relatively small, tDCS facilitated transfer of learning to untrained working memory tasks as well as resistance to forgetting over time [23]. As expected from the theoretical framework on the role of intrinsic neural activation, the cognitive demand during task impacts the effects of tDCS [24]. Similarly, tDCS can augment cognitive-emotional reappraisal, regardless of whether the goal was to decrease or increase emotional responsiveness [25]. The beneficial effect of combining tDCS with a cognitive control task also extends to individuals with depression, and tDCS ameliorated the negative effect of emotional pictures on accuracy and response times [26]. As such, combining tDCS with a therapeutically relevant contextual manipulation (VR or otherwise) becomes a rational next step – especially if the goal is to reduce variability and enhance effect sizes by synchronizing stimulation with the intended therapeutic, emotional learning process.

Next, we offer a translational step by mapping the PTSD exposure-learning framework onto depression-relevant, cognitive-behavioral therapeutic processes and beyond. The following section thus reflects extrapolations from our prior work and how the principle of context engagement (and which does not need to be VR) transfers to other disorders. Currently this is hypothetical and requires additional scientific validation, as there are currently no commercially available or manualized programs clinicians can integrate with tDCS.

For example, whereas the tDCS+VR combination for PTSD targeted extinction learning and context-dependent safety retrieval, depression treatment is plausibly driven by different learning and control processes, i.e., reward learning and approach behavior (behavioral activation), cognitive control over negative bias (reappraisal/working memory), social reinforcement learning, and affective regulation in response to stress. Yet the theory behind the proposed combined-intervention principle remains the same: 1) choose a context that reliably targets the maladaptive state, 2) train the adaptive alternative while stimulation is applied, and 3) consider the therapeutic “dose” to be the coupling of “stimulation parameters + learning dose,”. In depression, scalable example context engines during stimulation might be:

• Behavioral activation tasks delivered via smartphone or web platforms (e.g., structured activity scheduling, in-the-moment prompts, reward prediction tracking) during stimulation.

• Cognitive control training (working memory, inhibitory control, cognitive bias modification, emotion regulation tasks) that recruits dorsolateral prefrontal and frontoparietal networks, and which could be calibrated to individual performance, at the time of stimulation.

• Social/affiliative contexts (guided social interaction modules, prosocial tasks, or VR social environments) for patients whose depression is dominated by social withdrawal and threat sensitivity during stimulation.

• Rumination interruption protocols (brief induction + redirection) to create a reproducible “state” that can be shifted during stimulation.

This framing might also have relevance for home use. For example, adherence may improve when sessions are purposeful, guided, and intrinsically structured, and effectiveness may improve when the brain state is aligned with target engagement. Moreover, we believe that this theoretical framing can generalize across disorders when considering interactions between cognition, context and stimulation.

Applied broadly, this theoretical framework supports combination tDCS development for various psychiatric disorders, including anxiety (e.g., interoceptive/threat exposure + stimulation), obsessive-compulsive disorder (e.g., exposure/response prevention + stimulation), substance use (e.g., cue reactivity as a form of extinction/reappraisal + stimulation), and chronic pain (e.g., movement/catastrophizing + stimulation). Although these are currently mere suggestions for future directions, it is interesting to note that noninvasive brain stimulation studies for substance use disorder already often pair stimulation with cue reactivity [see 27 for a review].

In sum, the combined tDCS+VR PTSD use case provides a translational playbook for the current era of FDA-approved at-home tDCS. As at-home tDCS becomes clinically accessible for depression, the time is ripe to test whether the next substantial gains might come from embedding stimulation within scalable contexts that recruit the circuits and learning mechanisms that drive durable recovery. For example, in a small pilot study in acutely depressed individuals [28], five sessions of 2 mA active tDCS + cognitive control training were compared to five sessions of sham tDCS + cognitive control training, and five sessions of active tDCS + sham cognitive control training (n= 9 per group). Results demonstrate that at the 3-week follow-up timepoint, clinician-rated severity of depression severity was substantially lower compared to baseline for the group who received the “double-active” group, indicating delayed effects of tDCS. Both other control conditions demonstrated a return of symptoms at 3 weeks. However, a larger trial in which tDCS was applied during group behavioral therapy for depression, combined tDCS application did not outperform group therapy or tDCS alone [29].

5. Timing as a determinant of tDCS efficacy for psychiatric treatment

The observation of a delayed effect of active tDCS + cognitive control training [28] brings up another important point on the optimal timing to assess efficacy. Despite research demonstrating that tDCS alters cortical excitability and connectivity [30], knowledge about the optimal time to assess the effect of tDCS on psychiatric symptoms is modest. In fact, as a field we have no clear guidelines on the optimal time window for efficacy testing, even though we recognize the many variables that influence stimulation efficacy (e.g., brain target, frequency, dose, context) [31]. This might have important implications for efficacy testing in clinical trials.

Generally, therapeutic efficacy in clinical trials is measured close to when the to-be tested intervention ceases. The idea being that once the intervention ends, any beneficial effects will (at best) be sustained over time. This leads to the situation where primary endpoints to determine efficacy are set proximal to the end of the intervention. However, this may be poorly suited to tDCS because stimulation-induced neural changes and symptom improvements might take time to become apparent (or even increase over time).

The presence of delayed stimulation effects is not unique to tDCS, as the maximum effect of other non-invasive stimulation techniques like transcranial magnetic stimulation also takes time to build [32]. The presence of after-effects following tDCS are well-known [5, 33–35], and even a single tDCS session is associated with after-effects [34]. Yet, following the pattern of most clinical trials, tDCS efficacy is typically tested at the end of the intervention (as was done in the study that led to FDA approval described above). Thus, not considering possible delayed tDCS effects, by setting primary endpoints at the end of the intervention, may miss its peak effects. In other words, not deliberately measuring outcomes during post-stimulation follow-up may have led to “failed” clinical trials, contributing to the earlier idea that tDCS is ineffective.

In addition to the study on active tDCS + cognitive control training, a different clinical trial assessing the efficacy of escitalopram and tDCS for depression, escitalopram produced a significant reduction in depressive symptoms beginning at week 3 compared to placebo. tDCS diverged from placebo at both weeks 8 and 10. By week 10, remission rates for escitalopram and tDCS were comparable [36]. Another study of tDCS monotherapy [37] found significant improvements from baseline in depression, associative learning, and social cognition two weeks after tDCS. Similarly, another study [38] observed greater depression improvements 12 weeks following tDCS combined with cognitive behavioral therapy versus sham stimulation plus therapy (note: in this trial tDCS was not applied concurrent with therapy but was instead provided asynchronously).

These delayed tDCS effects are not limited to depression. In a randomized, double-blind, sham-controlled crossover trial, only patients with schizophrenia spectrum disorder who received active tDCS during a first visit showed improved executive performance and emotion recognition, including introspective ability, at their second visit when they received sham [39]. Those who were assigned to sham during the first visit and active tDCS during the second visit did not show these improvements. Had these order effects not been tested, the overall conclusion would have been that tDCS was ineffective. Comparable cross-over effects of tDCS on cognitive performance have been observed in individuals without psychiatric diagnoses.

Returning to our experience using tDCS+VR for PTSD, we detected a clearly delayed effect – in both the pilot study and subsequent confirmatory randomized controlled trial, we observed the greatest clinical effect one month after cessation of stimulation. To be specific, given that we observed emerging clinical effect sizes at one month post tDCS+VR in the pilot study (n=12), we set the primary endpoint of the randomized controlled trial at one month and added an exploratory 3-month outcome. After the randomized controlled trial was complete, we observed clear superiority of active stimulation at 1-month and continued gains at the 3-month timepoint. Interestingly we also observed improvements in social/occupational function in the active group that only emerged 3 months after the intervention [21], again underscoring that changes can take time to emerge. The recommendation to assess delayed effects of tDCS on psychiatric symptoms should be balanced with the practicalities of clinical trials and the possibility that effects could decrease over time. Yet, overlooking delayed effects can result in premature conclusions about efficacy, and by directly testing these effects the field will be better poised to clarify tDCS’ temporal dynamics in the treatment of psychiatric disorders.

6. Other considerations

The contribution of contextual factors has other ramifications. First, tDCS (and other forms of stimulation) have been combined with other elements such as cognitive training and physical task performance, in the hope the combination will provide superior outcomes than either alone, see also [40–43] for experimental support and theoretical rationale. Moreover, this approach has also begun to be explored in rehabilitation domains such as chronic stroke, auditory rehabilitation [44, 45] and beyond [46], indicating these general ideas are not constrained to the psychiatric space.

That stated, not all findings from the combined use of tDCS and other forms of training have been beneficial, which underscores the need for careful future study. For example, one trial with a unique electrode montage (2 prefrontal anodes, two occipital cathodes) delivered 1 mA tDCS during a 5-day working memory training protocol and yielded impaired working memory; a follow-up experiment that relocated the cathodes reduced the negative effects. Nonetheless, even with this revised montage (anodes over frontal electroencephalogram (EEG) sites F3 and F4 – cathodes over supraorbital EEG sites AF7 AF8), working memory training outcomes did not improve relative to sham stimulation [47]. However, it is important to note that these experiments used a relatively low stimulation intensity (1 mA rather than 2 mA) and a four-electrode configuration (two anodes and two cathodes), rather than the more typical two-electrode setup (e.g., anode over frontal EEG site F3 and a cathode over F4).

In a separate study, researchers evaluated the commercially available foc.us tDCS device for at-home use [48]. This device also uses a four-electrode configuration to deliver 1.5 mA stimulation. Although all four electrodes are positioned over the frontal cortex, use of the foc.us device impaired working memory updating compared with sham stimulation. Indeed, the authors concluded that “the use of this device can actually be detrimental and, as such, cannot be regarded as an alternative to CE-certified tDCS devices” [c.f. 48]. This study also compared “online” stimulation (applied concurrently with an n-back working memory task) and “offline” stimulation (applied prior to task performance) and found no significant differences in working memory outcomes between these conditions.

Yet with the recent FDA approval of tDCS for clinical depression treatment, the field should anticipate a renewed interest in do-it-yourself stimulation. Psychiatrists and other clinicians will therefore need to be prepared to inquire about patients’ use of tDCS and related technologies, and to counsel users that although this area warrants further research, it is unlikely to be ready for near-term cognitive enhancement applications without additional testing. Guidance for navigating this issue is provided in [3].

7. Future Directions and What Clinicians Need to Know Now

tDCS is now FDA approved for home use in patients with non-treatment-resistant major depression. While the effect size is modest, side effects are very favorable compared to antidepressant pharmacotherapy. Thus, the next step is to consider novel methods to improve clinical outcomes; theoretical frameworks around the mechanisms of action can guide this trajectory. Specifically, concurrent engagement of the targeted neural systems – through active participation in learning and emotion processing as part of cognitive-behavioral therapy –could augment the efficacy of tDCS on symptoms reduction. As such continued testing of theoretically motivated tDCS montages, with the goal to amplify ongoing, meaningful brain activity, to guide plasticity toward adaptive ends are sorely needed. Although the future of tDCS research and clinical implementation rests partially on determining tDCS parameters (e.g., stimulation intensity, electrode montage, session number and inter-session interval), more precise contextual manipulation to align stimulation with neural dynamics should actively be pursued.

Currently unanswered questions include testing the parameters of synergistic tDCS effects on therapeutic engagement, for example for anxiety, stress, and trauma-related disorders. Specifically, individuals who have difficulty engaging in the therapeutic process might show different beneficial effects than those who are fully engaged or who might find cognitive-behavioral therapy difficult to tolerate. A second relevant question is whether it is possible to detect an early signal for whom combined tDCS might be especially beneficial. Insight into the profile(s) of neural adaptation, and its individual differences, that predict clinical response would further a personalized medicine approach to tDCS. That stated, the emergence of an FDA-approved tDCS protocol for depression means clinicians can begin offering tDCS to appropriate patients today using the FDA-approved protocol.