Optimizing tDCS cognitive outcomes in schizophrenia: the role of dopaminergic tone and antipsychotic load

Abstract

Working memory deficits are a central cognitive impairment in schizophrenia and a key therapeutic target. Transcranial direct current stimulation (tDCS) has emerged as a potential cognitive enhancer, yet considerable interindividual variability limits its clinical utility. Pharmacological modulation of dopaminergic signaling, particularly through antipsychotic medications, may influence neuroplasticity and thereby alter responsiveness to tDCS. This study investigated whether antipsychotic dose or pharmacodynamic profile affects tDCS-related working memory improvements in schizophrenia. In a randomized, double-blind, placebo-controlled trial, 120 clinically stable outpatients received active or sham bifrontal tDCS delivered over 10 consecutive weekdays. Working memory performance was assessed using the MATRICS Consensus Cognitive Battery. Antipsychotic regimens were categorized as full D2 antagonists or partial D2 agonists, and all doses were converted to olanzapine equivalents. Results showed a significant negative correlation between antipsychotic dose and tDCS-induced working memory enhancement (r = –0.197, p = 0.031), indicating diminished cognitive gains at higher doses. In contrast, no significant differences in working memory outcomes emerged between patients receiving full antagonists versus partial agonists. These findings suggest that overall dopaminergic blockade, rather than specific receptor-binding profiles, modulates cognitive responsiveness to tDCS. High antipsychotic doses may reduce neuroplastic capacity, thereby limiting the therapeutic potential of neuromodulation. The results highlight the need to consider antipsychotic load in the design and interpretation of cognitive enhancement trials and support the use of pharmacological burden as a stratification variable in precision-psychiatry approaches.

Introduction

Cognitive dysfunction, particularly impairments in working memory, is a hallmark of neuropsychiatric disorders, especially schizophrenia [1–3]. Transcranial direct current stimulation (tDCS), a non-invasive neuromodulation technique, has emerged as a promising intervention to enhance cognitive processes by modulating cortical excitability and promoting synaptic plasticity [4–6]. Meta-analytic evidence suggests that tDCS can produce small but significant improvements in working memory across diverse brain disorders, highlighting its potential role as a cognitive enhancer [7, 8]. Nevertheless, the clinical efficacy of tDCS exhibits considerable variability, likely reflecting underlying neurobiological differences among individuals [9].

Multiple factors may influence the ability of tDCS to induce neuroplasticity. Among those, the role of medication is particularly critical, as pharmacological agents can significantly enhance or hinder the neuromodulatory effects of tDCS [10, 11].

Dopaminergic neurotransmission plays a critical role in cognitive functioning, with particular relevance to working memory, executive functions, and attentional control [12, 13]. Dopamine exerts its effects mainly through D1 and D2 receptor-mediated pathways within prefrontal cortical circuits, influencing key synaptic plasticity processes such as long-term potentiation (LTP) and long-term depression (LTD) [12]. In schizophrenia, dysregulation of dopaminergic signalling, notably within the dorsolateral prefrontal cortex (DLPFC), leads to impaired cortical efficiency and disrupted cognitive performance [14, 15]. As tDCS aims to modulate cortical excitability and plasticity, its effectiveness may be highly dependent on the integrity of dopaminergic modulation [16, 17]. Recent evidence further indicates that dopamine-related mechanisms, particularly involving D1 and D2 receptor expression, are crucial for regulating large-scale brain state transitions that underlie flexible working memory performance and are altered in schizophrenia [18].

Antipsychotic medications, through their effects on D2 receptor occupancy, represent a major factor influencing the dopaminergic environment [19]. Full D2 antagonists and partial D2 agonists differ significantly in their pharmacodynamic profiles, with the degree of receptor blockade critically determining dopaminergic tone [20]. Emerging evidence indicates that excessive D2 receptor blockade may hinder neuroplastic responses to tDCS, while lower levels or partial agonism of receptor occupancy might preserve or even enhance plasticity [21, 22]. Consequently, antipsychotic treatment characteristics, including dosage and receptor profile (partial agonism versus antagonism), may act as mediators of cognitive outcomes following neuromodulation.

Given the increasing demand for precision in psychiatric services [23, 24], where treatments must be efficiently targeted, identifying biological predictors of tDCS responsiveness is of main importance. Thus, understanding how antipsychotic treatment characteristics influence tDCS-induced cognitive improvements could help stratify patients and optimize intervention strategies.

Emerging evidence from both experimental neurophysiological studies and clinical research suggests that the cognitive effects of tDCS may be critically influenced by concurrent dopaminergic modulation. Experimental studies in healthy participants have demonstrated that dopaminergic agents can enhance and prolong tDCS-induced plasticity, whereas D2 receptor blockade may attenuate or abolish these effects, indicating a dose and receptor dependent modulation of cortical excitability and LTP/LTD-like mechanisms [12, 18]. In clinical populations, observational and interventional studies in schizophrenia have reported that antipsychotic treatment characteristics may moderate cognitive and neurophysiological responses to tDCS, with higher levels of dopaminergic blockade being associated with reduced plasticity-related effects [25]. Together, these findings suggest that both the magnitude of D2 receptor occupancy and the pharmacodynamic profile of antipsychotic medication may influence the direction and extent of working memory improvement following tDCS.

Accordingly, the primary objective of this study is to investigate the relationship between antipsychotic dosage and working memory improvements induced by tDCS. The secondary objective is to investigate whether differences exist between patients treated with full D2 antagonists and those treated with partial D2 agonists in terms of working memory improvement following transcranial direct current stimulation (tDCS).

Methods

This study is a secondary, post hoc analysis derived from a previously published double-blind, randomized, placebo-controlled trial conducted at San Juan University Hospital (Alicante, Spain), including 120 clinically stable outpatients diagnosed with schizophrenia according to DSM-5 [26] criteria confirmed by the SCID-5-CV [27].

Eligible participants were aged 18–50 years, with clinical stability defined by the absence of symptom exacerbations, pharmacological treatment modifications, or hospitalizations in the previous month. Key exclusion criteria included significant neurosensory deficits, intellectual disability (IQ < 70), recent electroconvulsive therapy, organic brain disorders, or contraindications to tDCS. Participants underwent baseline assessments including the Positive and Negative Syndrome Scale (PANSS) [28] and the MATRICS Consensus Cognitive Battery (MCCB) [29, 30]. Participants were randomly assigned to receive either active or sham bifrontal tDCS (anode over F3, cathode over F4), delivered at 2 mA for 20 min over 10 consecutive weekdays. Sham stimulation involved brief initial stimulation to mimic sensory effects without delivering a sustained current. No structured cognitive tasks were performed during stimulation (for an extended version of the clinical trial, see García-Fernández et al. [6]).

Antipsychotic treatment categorization

Patients were stratified into two groups based on their prescribed antipsychotic treatment regimen. The first group included individuals receiving second-generation antipsychotic (SGA) primarily considered D2 receptor antagonists (n = 61). The second group comprised patients treated with third-generation antipsychotic (TGA) exhibiting partial agonist activity at D2 receptors (n = 59). The patient flow diagram detailing screening, randomization, and inclusion is available in Supplementary Material, Figure 1.

To ensure comparability across pharmacologically diverse treatments, all antipsychotic dosages were converted to olanzapine equivalents using established conversion guidelines [31, 32].

The distribution of individual antipsychotic agents across pharmacodynamic groups and stimulation conditions is presented in Supplementary Table 2. Subgroup sizes were comparable between the active tDCS and sham conditions within both pharmacodynamic classes, indicating adequate balance across treatment arms.

The study was conducted in accordance with the Declaration of Helsinki and approved by the Research Ethics Committee on Medicines of the General University Hospital of Elche. All participants provided written informed consent prior to participation.

Statistical analysis

First, Pearson correlation analyses were performed to explore the relationship between changes in cognitive performance within the WM domain of the MCCB and clinical variables, including antipsychotic dose and treatment type. The assumption of normality was verified prior to conducting the Pearson analyses. Additionally, the influence of antipsychotic treatment type (full D2 antagonists vs. partial D2 agonists) and stimulation group (active tDCS vs. sham) on cognitive outcomes was specifically examined for the working memory (WM) domain of the MATRICS Consensus Cognitive Battery (MCCB), as this was the domain in which a significant improvement between active tDCS and sham was observed in our previous study. A two-way ANOVA was conducted with antipsychotic type (full D2 antagonists vs. partial D2 agonists) and group (tDCS vs. sham) as fixed factors, and the WM domain score of the MCCB as the dependent variable. In line with our a priori hypotheses, we examined (i) the association between cognitive performance changes in the working memory (WM) domain of the MATRICS Consensus Cognitive Battery (MCCB) and clinical variables (including antipsychotic treatment characteristics), and (ii) the potential moderating effect of antipsychotic medication on tDCS-related cognitive outcomes.

The analyses examining the relationship between antipsychotic treatment and tDCS-related cognitive outcomes are exploratory and not prespecified in the original trial protocol. Therefore, these findings should be considered hypothesis-generating and interpreted with caution regarding causal inference

Results

Participant characteristics by antipsychotic treatment dose and type

Baseline demographic and clinical characteristics are presented in Table 1, stratified by antipsychotic type (D2 antagonists vs. D2 partial agonists) and stimulation condition (active tDCS vs. sham). Group sizes were balanced across conditions (n = 28–31 per subgroup), allowing for appropriate comparisons.

Table 1.

Baseline characteristics by antipsychotic type and stimulation condition.

		D2 antagonists	D2 partial agonists	P value
	Total
n	120	61	59
Age	33.3 (11.2)	3.3 (10.6)	33.3 (11.8)	0.997
Sex (female, n, %)	40 (33.3%)	16 (26.2%)	24 (40.7)	0.137
Age at onset (years)	22.4 (3.9)	22.5 (3.9)	22.3 (4.1)	0.833
Years of education	13.3 (3.5)	13.3 (3.6)	13.4 (3.4)	0.884
PANSS score	76.1 (18.2)	76.2 (20.3)	76 (15.9)	0.953
Positive symptoms	14.9 (5.8)	15.2 (6.1)	14.6 (5.3)	0.347
Negative symptoms	21.1 (5.3)	20.8 (6.3)	21.8 (4.7)	0.597
General psychopathology	39.7(9.3)	40.1 (10.4)	39.4 (8.1)	0.497
Working Memory MCCB*	43.9 (12.2)	44.7 (11.1)	43.2 (13.2)	0.506
Olanzapine equivalent dose, mg	11.8 (4.6)	12.0 (5.5)	11.6 (3.5)	0.592

Data are presented as mean ± standard deviation unless otherwise indicated. Group comparisons were conducted using one-way ANOVA for continuous variables and chi-square tests for categorical variables.

PANSS positive and negative syndrome scale, MCCB MATRICS consensus cognitive battery, tDCS transcranial direct current stimulation. Antipsychotic doses are expressed in olanzapine equivalents.

*Working Memory MCCB shows T-score corrected by age and sex.

No statistically significant differences were found between groups with respect to sex, age, age at illness onset, years of education, PANSS scores, or baseline working memory performance, supporting baseline equivalence across experimental conditions. Antipsychotic doses, expressed in olanzapine equivalents [30, 31], did not significantly differ across antipsychotic type groups (p = 0.592), although a non-significant trend toward higher doses in the sham condition was observed.

Supplementary Table 1 displays the working memory outcomes following tDCS vs sham intervention, as described in our earlier study [6], providing detailed data relevant to all the cognitive effects observed.

Impact of antipsychotic dosage on tDCS-induced working memory improvement

Pearson correlations analyses, conducted after confirming the assumption of normality, revealed a significant negative correlation between antipsychotic dose and the change in working memory performance following the intervention (r = –0.197, p = 0.031), indicating that higher antipsychotic doses were associated with smaller working memory gains.

Impact of antipsychotic group on working memory response to tDCS

Working memory (WM) T scores in the D2 antagonist group increased from a pre-intervention mean of 44.66 (SD = 11.13) to a post-intervention mean of 50.48 (SD = 12.37). Similarly, the partial agonist group showed an increase from a pre-intervention mean of 43.17 (SD = 13.22) to a post-intervention mean of 47.71 (SD = 12.79). No significant differences were observed between the groups at baseline [t(118) = –0.667, p = .506].

No significant differences were observed in working memory improvement between patients treated with full D2 antagonists and those receiving partial D2 agonists following tDCS intervention. ANOVA results did not reveal a significant main effect for the type of antipsychotic medication [F(1,115) = 2.67, p = 0.105], nor was there a significant interaction between the type of antipsychotic and the intervention (tDCS or sham) [F(1,115) = 2.36, p = 0.128]. Although the assumption of homogeneity of variance was formally violated, the factorial ANOVA was retained due to the balanced group sizes and the non-significant findings, as variance heterogeneity primarily affects Type I error rates.

Discussion

The present findings support the hypothesis that targeted dopaminergic modulation, particularly at lower doses, may enhance the cognitive efficacy of tDCS in individuals with schizophrenia. Notably, we observed a significant negative correlation between antipsychotic dose and gains in working memory performance, suggesting that a higher dopaminergic blockade burden may attenuate the neuroplastic and pro-cognitive effects of tDCS. However, contrary to our initial hypothesis, no significant differences in cognitive outcomes were found between patients treated with full D2 receptor antagonists and those receiving partial D2 agonists. These results imply that the overall magnitude of dopaminergic blockade, rather than the specific pharmacodynamic profile of the antipsychotic agent, may play a more decisive role in modulating the cognitive impact of neuromodulatory interventions such as tDCS.

These findings offer important insights into the intricate interplay between dopaminergic modulation, antipsychotic pharmacotherapy, and neuroplasticity-enhancing interventions such as tDCS. They align with the well-established inverted-U model of dopaminergic function in the prefrontal cortex (PFC), which posits that both hypo and hyperdopaminergic states are associated with impaired cognitive performance, whereas an optimal intermediate level of dopaminergic activity facilitates executive functioning and working memory [33, 34].

The lack of differential effects between full D2 antagonists and partial D2 agonists further could suggest that the primary determinant of responsiveness to tDCS may not lie in the qualitative pharmacological mechanism of D2 receptor engagement per se, but rather in the overall dopaminergic postsynaptic dopaminergic modulation related to D2 receptor blockade. This postsynaptic dopaminergic modulation appears to be influenced by the cumulative antipsychotic dose, with higher doses potentially suppressing the plasticity-dependent cognitive benefits of tDCS through excessive dopaminergic blockade.

This interpretation aligns with preclinical models demonstrating that an optimal dopaminergic state is necessary for the induction of activity-dependent plasticity mechanisms, such as long-term potentiation (LTP) and long-term depression (LTD), within the prefrontal cortex [35, 36]. Accordingly, these findings support the hypothesis that excessive D2 receptor blockade may impair intracellular signaling cascades critical for synaptic modifications, thereby attenuating the facilitatory effects of tDCS on cortical networks [37, 38].

Importantly, our findings are consistent with a growing body of neuromodulation research indicating that pharmacological status is a critical modulator of brain stimulation efficacy [39, 40]. In particular, dopaminergic tone appears to play a central role: dopaminergic agents have been shown to enhance and prolong the after-effects of tDCS on cortical excitability, while dopaminergic blockade, such as that mediated by D2 receptor antagonists, can significantly attenuate these neuroplastic responses [41]. Within this framework, our observation that higher antipsychotic dose negatively predicts cognitive improvement reinforces the view that antipsychotic load constitutes a key biological stratification variable for identifying optimal candidates and predicting individual responsiveness to neuromodulatory interventions. These results further underscore the need for a nuanced and individualized approach to dopaminergic pharmacotherapy in schizophrenia. Uniform dosing strategies are unlikely to yield optimal cognitive outcomes; instead, the lowest effective dose should be carefully and systematically considered, particularly when cognitive enhancement and neuroplasticity-based interventions such as tDCS are pursued in clinical practice.

Taken together, these findings highlight the critical importance of integrating dose-dependent pharmacological variables into both the design and interpretation of tDCS clinical trials targeting cognitive dysfunction in schizophrenia. Rather than focusing solely on antipsychotic drug class or receptor subtype, our results point to the total dopaminergic blockade burden as a more biologically meaningful moderator of neuromodulatory efficacy. This shift in perspective aligns with and reinforces the relevance of the minimum effective dose paradigm: while adequate D2 receptor occupancy is essential for maintaining control over positive symptoms, surpassing this threshold may inadvertently compromise cortical plasticity by saturating dopaminergic circuits integral to synaptic modulation and learning. At the same time, it is important to acknowledge that neuroplasticity itself may contribute to variability in the effects of tDCS, since individual differences in functional reorganization can mean that the same anatomical “target” may also function as a “window” into a broader network rather than a fixed locus of action.

The principal strength of this study lies in its rigorous double-blind, randomized, placebo-controlled design, which substantially enhances the internal validity and interpretability of the findings. Furthermore, the study adopts a comprehensive approach to examining the dose-dependent effects of antipsychotic pharmacotherapy on tDCS-induced cognitive enhancement in schizophrenia, a topic on which, to our knowledge, no prior studies have been published.

Nonetheless, certain limitations should be acknowledged. While the use of antipsychotic equivalency conversions provides a standardized framework for dose comparison, these calculations may not fully capture the nuanced differences in receptor-binding affinities, intrinsic activity, or pharmacokinetic profiles among the various compounds. Moreover, the sample size within antipsychotic subgroups, although adequate, remains relatively modest and may constrain statistical power. Importantly, the partial dopamine agonist subgroup included cariprazine, despite its minimal representation in the sample. Given cariprazine’s distinct pharmacodynamic characteristics, particularly its high affinity for D3 receptors and its putative pro-cognitive properties [42], its inclusion may introduce a degree of heterogeneity within this subgroup that should be interpreted with caution. Overall, in the present study, the term “dopaminergic tone” primarily reflects postsynaptic dopaminergic modulation related to D2 receptor occupancy and antipsychotic dose, rather than direct measures of presynaptic dopamine release capacity. Although both mechanisms are known to influence cortical plasticity, our design does not allow their dissociation. Therefore, our findings should be interpreted mainly in the context of postsynaptic receptor-mediated modulation.

Conclusion

In summary, our findings indicate that dopaminergic tone, partly influenced by cumulative antipsychotic load, may play a limited role in modulating the cognitive response to interventions such as tDCS in schizophrenia. This study highlights the necessity of moving beyond categorical drug classifications toward a more quantitative, biologically-informed understanding of pharmacological modulation in neuromodulation research. Future studies should aim to develop stratification models that incorporate pharmacodynamic load alongside neurobiological markers, such as cortical excitability profiles, genetic variants, or neuroimaging-based indicators, to guide personalized antipsychotic dosing and enhance individual responsiveness to tDCS and related neuroplasticity-based interventions.

Author contributions

LGF: Conceptualization, Data curation, Investigation, Methodology, Project administration, Writing – original draft, Writing – review & editing. VRF: Formal analysis, Writing – review & editing. APMG: Writing – review & editing. SCB, DS, IRA, AMR: Investigation. MM-BG, RW: Writing – review & editing. RRJ: Conceptualization, Investigation, Methodology, Writing – review & editing. No artificial intelligence (AI) tools or AI-assisted technologies were used in the drafting, data analysis, or preparation of this manuscript.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

The data supporting the findings of this study are available upon request from the corresponding author.

Competing interests

RRJ has been a consultant for, spoken in activities of, or received grants from: Instituto de Salud Carlos III, Fondo de Investigación Sanitaria (FIS), Centro de Investigaci ´on Biomédica en Red de Salud Mental (CIBERSAM), Madrid Regional Government (S2010/BMD-2422 AGES; S2017/BMD-3740), JanssenCilag, Lundbeck, Otsuka, Pfizer, Ferrer, Juste, Takeda, Exeltis, Casen-Recordati,Angelini, Rovi. All other authors have no conflict of interest.

Supplementary information

The online version contains supplementary material available at 10.1038/s41398-026-04160-w.