Abstract
Working memory deficits are linked to irregularities in the dorsolateral prefrontal cortex (DLPFC) and the posterior parietal cortex (PPC) in schizophrenia, effective intervention strategies are lacking. We evaluated the differential efficacy and underlying neuromechanisms of targeting transcranial direct current stimulation (tDCS) at the DLPFC and the PPC with concurrent cognitive performance for working memory in schizophrenia. In a randomized and double-blind clinical trial, sixty clinically stable schizophrenic patients with below-average working memory were randomly assigned to active DLPFC, active PPC, and sham tDCS groups. Two sessions of tDCS during N-back task were delivered daily for five days. The primary outcome was changes in spatial span test scores from baseline to week 1. The secondary outcomes included changes in scores of color delay-estimation task, other cognitive tasks, and mismatch negativity (biomarker of N-methyl-d-aspartate receptor functioning). Compared with the active DLPFC group, the active PPC group demonstrated significantly greater improvement in spatial span test scores (p = 0.008, d = 0.94) and an augmentation in color delay-estimation task capacity at week 1; the latter sustained to week 2. Compared with the sham tDCS group, the active PPC group did not show a significant improvement in spatial span test scores at week 1 and 2; however, significant enhancement was observed in their color delay-estimation task capacity at week 2. Additionally, mismatch negativity amplitude was enhanced, and changes in theta band measures were positively correlated with working memory improvement in the active PPC group, while no such correlations were observed in the active DLPFC group or the sham tDCS group. Our results suggest that tDCS targeting the PPC relative to the DLPFC during concurrent cognitive performance may improve working memory in schizophrenia, meriting further investigation. The improvement in working memory appears to be linked to enhanced N-methyl-d-aspartate receptor functioning.
Similar content being viewed by others
Introduction
Working memory is the brain’s ability to hold and manipulate transient information necessary for complex cognitive tasks [1]. Deficits in working memory constitute a primary aspect of cognitive impairment in patients with schizophrenia, often severely affecting their daily functioning [2]. However, effective treatments for the deficits in working memory in schizophrenia are notably lacking [3,4,5]. Transcranial direct current stimulation (tDCS) is capable of modulating subthreshold membrane potentials, thereby altering the likelihood of action potentials, affecting the activation, neural oscillations, and functional interactions of brain regions [6, 7]. Consequently, tDCS may represent a viable therapeutic strategy for ameliorating working memory deficits in schizophrenia.
Both the dorsolateral prefrontal cortex (DLPFC) and the posterior parietal cortex (PPC) are associated with the execution of working memory [8,9,10,11,12]. Although the majority of tDCS research on schizophrenia has targeted the DLPFC, yielding mixed results according to extant meta-analyses [13,14,15], no studies have yet targeted the PPC in patients with schizophrenia, nor have any studies directly compared the effects of tDCS on the DLPFC and the PPC in this patient population. However, a study highlighted the superior effect of targeting the PPC over the DLPFC with single-session tDCS in bolstering working memory performance among healthy young individuals [16]. In contrast, other studies have reported conflicting results [17,18,19]. These inconsistencies might be due to differences in assessment tools and the subtle effects of single-session stimulation.
The effect of tDCS may be task specific due to the requirement for plasticity induction in the targeted neural pathway [20, 21]. Thus, a transcranial electrical stimulation workshop [20] recommended tDCS to be performed during task execution to enhance modulation effects, yet preliminary studies reported inconclusive results [22,23,24,25,26,27]. The inconclusive results might be related to the fact that existing studies mainly targeted the DLPFC, which had shown mixed cognitive improvement effects. Given that the PPC is also related to working memory performance, it is worthwhile to investigate the potential utility of the PPC as an alternative stimulation target when combining tDCS with concurrent working memory task to enhance working memory in schizophrenia.
The elicitation of tDCS effects necessitates the induction of endogenous N-methyl-d-aspartate receptor (NMDAR) [28]. The NMDAR is also pivotal for cortical activation and synaptic plasticity [29, 30]. Mismatch negativity (MMN) acts as a neurophysiological index for automatic sensory deviation detection and is sensitive to NMDAR agonists [31]. MMN is an effective biomarker of NMDAR deficits in schizophrenia [32, 33], and an indicator of theta band features [34]. Among patients with schizophrenia, there are notable reductions in MMN amplitude, theta band power, and theta band intertrial coherence [35, 36]. Research suggests that MMN can potentially be modulated by targeting specific brain regions with tDCS [37,38,39,40,41]. Additionally, D-serine acts as an endogenous co-agonist of NMDAR [29] and is synthesized in both brain and peripheral tissues [42, 43]. Because D-serine can traverse the blood-brain barrier, its serum levels correlate with its concentration in the brain in animal studies [44]. Hence, both MMN and serum D-serine level can serve as surrogate biomarkers for NMDAR functioning.
Overall, no study has compared the efficacy of tDCS targeting the DLPFC or the PPC during concurrent cognitive task for addressing working memory deficits in schizophrenia. We hypothesized that tDCS targeting the PPC would be more effective than targeting the DLPFC during concurrent working memory task for working memory in schizophrenia. To test this hypothesis, we conducted a randomized controlled clinical trial and utilized biomarkers of the NMDAR (i.e., MMN and D-serine) to attempt to disentangle the intrinsic mechanisms for the therapeutic efficacy.
Methods and materials
Study overview
This single-center, three-arm, double-blind (participant and assessor blind) clinical trial was conducted at the Beijing Anding Hospital affiliated with Capital Medical University in China, from September 2021 to May 2023. This trial was registered in the Chinese Clinical Trial Registry (ChiCTR2000038961), and the protocol was approved by the Ethics Committee of the Beijing Anding Hospital. Participants signed informed consent forms after being fully informed of the trial.
Participants
Participants were outpatients recruited via advertisements, community and outpatient physician referrals. The inclusion criteria included a diagnosis of schizophrenia based on DSM-5 criteria using the Mini International Neuropsychiatric Interview (MINI 7.0.2), aged 18–50 years, right-handed, at least 8 years of education, intelligence quotient (IQ) > 70, and a spatial span test T-score of 30–50. Participants were clinically stable [45], with Positive and Negative Syndrome Scale (PANSS) scores not exceeding 5 on any of the following items: P1 delusion, P3 hallucination, P5 exaggeration, P6 suspicion/victimization, and G9 abnormal thinking content; scored no more than 4 on P2 concept disorder; stayed on current antipsychotic and adjunctive medications for a minimum of 6 weeks, with a fixed antipsychotic dose for at least 2 weeks. Throughout the study, the type and dose of antipsychotic medications remained unchanged. The rationale behind selecting participants with a spatial span test score of 30–50 was based on a computerized cognitive remediation therapy study [46, 47] involving 311 schizophrenic patients, which demonstrated that participants with baseline cognitive impairment of 1–2 standard deviations reaped the most cognitive benefits. The exclusion criteria were detailed in the Supplementary Methods. The study also included healthy controls who were matched with the patient group in terms of sex, age, and educational level. Eligibility criteria for the controls included an age range of 18–50 years, right-handedness, and a minimum education duration of 8 years; an IQ above 70 was required. Additionally, the controls must have had no diagnosed psychiatric disorders, neurological conditions, or other severe physical illnesses, and no family history of psychiatric disorders.
Design
A balanced 1:1:1 allocation was accomplished through block randomization, with allocation concealment implemented using the envelope method. This step resulted in participants being randomly allocated to one of the three groups: an active DLPFC group, an active PPC group, and a sham tDCS group. In the sham tDCS condition, electrode placement was designed to mirror that of active PPC group in half of the subjects and identical to that of active DLPFC group in the other half, consistent with the design in research on healthy individuals [16]. Following the protocol by Valiengo et al. [48], 2 tDCS sessions were conducted daily, with a gap of at least 3 h, over a 5-day period (10 sessions). In all three groups, participants underwent a 20-min visuospatial N-back task [49] during each session. Assessments were conducted by one psychiatrist and two psychologists blind to the patients’ group assignment at baseline, week 1 (the 2nd to 3rd day following the last intervention) and week 2 (the 8th to 9th day following the last intervention). Additionally, healthy controls were included in this study primarily to facilitate the description of impairments and improvements in spatial span test scores across the three patient groups.
Intervention
The high-definition tDCS was administered using a 4 × 1 wire adaptor (Equalizer Box, NeuroConn, Germany) connected to a constant current stimulator (DC-Stimulator Plus, NeuroConn, Germany). Because high-definition tDCS offers a higher spatial focality [20] and more robust and durable modulation of cortical plasticity [50] than conventional tDCS, it is more suitable for investigations on the differential effects of stimulating different brain regions. Considering the abnormalities in the left frontal-parietal cortex during verbal and visuospatial working memory tasks in schizophrenia [9, 11, 51], and the predominant selection of the left hemisphere as the anodal stimulation target in tDCS studies on schizophrenia [13], this study chose the left hemisphere for stimulation targets. The cap based on the EEG 10-5 system was used for localization. For the left DLPFC target, F3 served as the central electrode, with F7, Fz, Fp1, FC3 functioning as surrounding electrodes. For the left PPC target, P3 was chosen as the central electrode, while P7, Pz, O1, CP3 were selected as surrounding electrodes. FC3 and CP3 were chosen to minimize possible effects of active DLPFC on the PPC and active PPC on the DLPFC, respectively. Electrical field simulations were performed using ROAST 3.0 [52] (Supplementary Fig. 2). For the two active stimulation groups, the current intensity was 2 mA for 20 min, with a 40-s ramp-up and ramp-down period. Sham tDCS mirrored this setup, but the real stimulation duration was limited to 40 s.
Outcomes and assessments
The spatial span test is the sole task for working memory in the Chinese version of the MATRICS Consensus Cognitive Battery (MCCB) [53] which is widely recognized as the gold standard for cognitive assessments in clinical trials for schizophrenia [45]. Thus, the primary outcome was changes in spatial span test scores at week 1 from baseline. The secondary outcomes included changes in spatial span test scores at week 2, changes in scores of the color delay-estimation task, digit sequencing task [54], Stroop task, MCCB scales, PANSS [53] scales, Calgary Depression Scale for Schizophrenia (CDSS), MMN and tDCS adverse effects. The standardized T scores were used for the MCCB. d-serine level changes were considered an exploratory outcome.
To evaluate the capacity and precision of visual working memory simultaneously, we implemented the color delay-estimation task as described by Zhao et al. [55]. To assess attention control, we used the Stroop with adaptive response deadline adapted by Draheim et al. [56, 57]. Adverse effects of tDCS were recorded at the end of each treatment session and in the follow-up assessments at week 1 and week 2 using the tDCS Adverse Effects Questionnaire [58]. To evaluate the integrity of the blinding, participants were requested to speculate on whether they had received real or sham stimulation upon completing the week 2 assessment. Details of these tasks are provided in Supplementary Methods.
Biomarkers
We employed an Oddball paradigm, comprised of 90% standard stimuli and 10% duration-deviant stimuli, to elicit the MMN. For each participant, the MMN amplitude was the peak negative amplitude between 140 and 240 ms from the waveform. The waveform was derived from subtracting the average waveform of the standard stimuli from that of the deviant stimuli. MMN theta power and theta intertrial coherence were defined as the peak mean power and intertrial coherence within the theta band (4–7 Hz) during the 140–240 ms interval following the deviant stimulus. Peripheral serum concentrations of D-serine were obtained using the enzyme-linked immunosorbent assay technique. Details are provided in Supplementary Methods.
Statistical analysis
In a study examining the impact of single-session tDCS on working memory in healthy young individuals, the PPC stimulation group showed an effect size of 0.7 compared to the DLPFC stimulation group [16]. Given that our study focused on schizophrenia patients with working memory deficits and employed multi-session tDCS with concurrent cognitive performance, we estimated a larger differential effect size between the DLPFC and PPC stimulation groups, setting it at 1.0. Test power was set at 80%, with a two-tailed α level of 5%. This yielded 17 patients for each of the two groups. Accounting for a potential dropout rate of 15%, we required 20 participants in each group, with additional 20 patients included in the sham stimulation group as a control.
In the intention-to-treat sample, statistical analysis was conducted using SPSS 20.0 (SPSS Inc., Chicago, IL, USA). Histograms and Shapiro-Wilk tests were used to assess the normality of data distribution. For the data that were not normally distributed, logarithmic transformations or non-parametric tests were applied to address the issue. First, when baseline characteristics, chlorpromazine equivalents of antipsychotic medications [59] and adverse effects were compared across the three groups, chi-square tests or Fisher’s exact tests were employed for categorical variables, whereas one-way analysis of variance and Kruskal-Wallis tests were utilized for continuous variables. Second, Chi-square tests or Fisher’s exact tests were applied to compare the number of participants within each group who guessed they were assigned to the real stimulation group, as an assessment of blinding integrity.
Third, the primary efficacy analysis was conducted with linear mixed models to compare changes in spatial span test scores over time (from week 1 to baseline) among the three groups. To delve deeper into the changes of the primary outcome, we have also compared the changes at week 2 from baseline and examined the changes across all three time points among the three groups. The secondary efficacy analyses encompassed comparisons of changes in cognitive performance, clinical symptoms, and MMN across the three groups over specified timepoints (baseline, week 1, and week 2). Additionally, an exploratory analysis was conducted on d-serine levels. Healthy controls underwent the MCCB assessment to characterize baseline impairments in various cognitive domains within the patient groups, as well as to quantify the degree of improvement in spatial span test scores post-intervention. The analysis was performed using independent sample t-tests.
Last, given that the effects of interventions on biomarkers might be immediate, Spearman correlation analysis was used to examine the relationship between changes in MMN and D-serine levels at week 1 from baseline and changes in spatial span test scores at week 1 and 2 from baseline within each group. False discovery rate (FDR) corrections were applied across the groups. The goal was to elucidate possible underlying mechanisms of therapeutic efficacy.
A p-value < 0.05 (two-sided) was deemed statistically significant. Informed by Wobrock et al. [60], effect sizes were calculated based on changes in scores from the baseline for each group using an online tool [61]. Cohen’s d values indicate small (0.2–0.5), medium (0.5–0.8), and large (≥0.8) effect sizes [62].
Results
Participants
Of the 60 participants, 54 (90%) completed the assessment at week 1, and 52 (87%) completed the week 2 assessment (Supplementary Fig. 1). No significant differences were observed across the groups in baseline characteristics except for speed of processing domain in MCCB (F2, 57 = 3.65, p = 0.03) (Table 1). Thirty-five healthy controls were included in the study. Compared with these controls, the patient group exhibited significantly lower IQ and MCCB scores across all domains at baseline (p < 0.05) (Supplementary Table 1).
Primary outcome: spatial span test scores
A significant time-by-group interaction was found in spatial span test scores from week 1 to baseline across the three groups (F2, 51 = 4.66, p = 0.01) (Table 2). Specifically, the active PPC group demonstrated significant improvement in the spatial span test scores than the active DLPFC group at week 1 (F1, 34 = 7.95, p = 0.008, d = 0.94) (Table 2) (Fig. 1A) (Supplementary Figure 3). However, no significant time-by-group interaction was found in spatial span test scores over three time points (baseline, week 1, week 2) across the three groups (Table 3).
At week 1 and 2, the spatial span test scores for both the active DLPFC group and the sham stimulation group remained significantly lower than those of the healthy controls (p < 0.05). However, the scores of the active PPC group approached those of the healthy controls, with no significant differences between the two groups (Fig. 2).
Secondary outcome: cognition
There was a significant time-by-group interaction in color delay-estimation task capacity over time (baseline, week 1, week 2) across the three groups (F2, 52 = 4.04, p = 0.02) (Table 3). Specifically, compared to the active DLPFC group, the active PPC group showed significantly greater augmentation in color delay-estimation task capacity at week 1 (F1, 33 = 4.68, p = 0.04, d = 0.80), and this effect persisted at week 2 (F1, 31 = 5.91, p = .02, d = 0.93). Color delay-estimation task capacity at week 2 was also higher in the active PPC group vs the sham stimulation group (F1, 37 = 5.80, p = .02, d = −0.82). There were no significant differences in the changes in color delay-estimation task capacity between the active DLPFC group and the sham stimulation group (Table 2 and Fig. 1C, D). No time-by-group interactions were found for any other cognitive measures (Tables 2, 3 and Fig. 1B).
Secondary outcome: clinical symptoms
There was a significant time-by-group interaction in negative symptoms over time (baseline, week 1, week 2) across the three groups (F2, 49 = 3.19, p = 0.0498) (Table 3). Specifically, there was significant reduction in negative symptoms at week 2 in the active DLPFC group compared to the active PPC group (F1, 31 = 4.80, p = .04, d = 0.80). The sham stimulation group also exhibited greater improvement in negative symptoms at week 2 compared to the active PPC group (F1, 32 = 7.82, p = 0.009, d = −0.93). There were no significant differences in the changes in negative symptoms between the active DLPFC group and the sham stimulation group (Supplementary Table 2).
The cognitive and clinical outcomes remained consistent with the above results, after accounting for sex, clozapine use, electroconvulsive therapy history, duration of illness, speed of processing and MCCB neurocognitive composite score as covariates in the linear mixed models.
Biomarker: MMN
The active PPC group demonstrated improvement (more negative) in MMN amplitude at week 1 compared to both the active DLPFC group (p = 0.04, d = −0.75) and the sham stimulation group (p = 0.006, d = 0.93). No significant differences were observed in MMN amplitude changes between the active DLPFC group and the sham stimulation group (Supplementary Table 3, Fig. 1E and Supplementary Fig. 4).
Exploratory outcome: d-serine
There were no significant time-by-group interactions in changes of d-serine levels over time across the three groups (Table 3) (Supplementary Table 3).
Correlation between Change in Mismatch Negativity, d-serine and Change in Working Memory
In the active DLPFC group, no significant correlations were observed between the changes in MMN indices and D-serine levels at week 1 from baseline and the changes in spatial span test scores at week 1 and 2 from baseline (p > 0.05 after FDR correction) (Supplementary Fig. 5A, Supplementary Table 5).
Conversely, in the active PPC group, the change in MMN theta intertrial coherence at week 1 positively linked to the changes in the spatial span test scores at week 2 (r = 0.89, p < 0.001 after FDR correction). Within this group, there were no significant associations between the changes in other MMN indices or d-serine levels and the changes in the spatial span test scores (Supplementary Fig. 5B, Supplementary Table 5).
In the sham stimulation group, no significant correlations were found between the changes in MMN indices and d-serine levels at week 1 from baseline and the changes in spatial span test scores at week 1 and 2 from baseline (p > 0.05 after FDR correction) (Supplementary Fig. 5C, Supplementary Table 5).
In the total sample at baseline, neither d-serine level nor MMN measures were correlated with the spatial span test score or delay-estimation task capacity.
Adverse effects and integrity of blinding
Adverse effects of tDCS after the 10th treatment session are detailed in Supplementary Table 4. The active DLPFC group reported higher levels of scalp pain, tingling, and burning sensations compared to the sham stimulation group (p = 0.03, 0.01, and <0.001, respectively). Similarly, the active PPC group exhibited a trend towards significantly higher levels of tingling, itching, and burning sensations relative to the sham stimulation group (p = 0.06, 0.05, and 0.07, respectively). No severe adverse events occurred.
The number of participants who guessed they were assigned to the real stimulation group was 11 (69%), 14 (82%), and 14 (74%) in the active DLPFC group, active PPC group, and sham stimulation group, respectively. The distribution across the groups demonstrated no significant variance (χ22 = 0.90, p = 0.67).
Discussion
Our study demonstrated that tDCS targeting the PPC exhibited an enhancement in working memory compared to tDCS targeting the DLPFC during concurrent cognitive task, reaching a large effect size. After the intervention, the working memory levels of the active PPC group approached those of the healthy controls. Our study also found that in the active PPC group, there was a notable augmentation in MMN amplitude, and improvements in MMN theta band measures linked to enhancements in working memory. Conversely, in both the active DLPFC group and the sham tDCS group, there were no associations between changes in working memory and changes in MMN measures or d-serine levels.
The superiority of tDCS targeting the PPC in our study suggests that the PPC may be a better neuromodulation target for enhancing working memory in schizophrenia and that it also has a causal role in working memory capacity in schizophrenia. Wang et al. [16] similarly found that in healthy individuals, tDCS targeting the PPC enhanced visual working memory capacity, but did not affect precision, and tDCS targeting the DLPFC did not change working memory capacity or precision. Their further analysis highlighted the consistency and stability of the enhancement in visual working memory capacity through tDCS applied to the PPC across individuals. Conversely, they found that the effects of tDCS targeting the DLPFC varied significantly across individuals, which our study also observed. There were no significant differences in the primary outcome between the active PPC group and the sham stimulation group in our study, which may be attributed to the low sensitivity of the spatial span test scores to the intervention. Our findings suggest that delay-estimation task capacity is more sensitive to intervention effects and might be more appropriate as a primary outcome measure. Further validation in future research is essential.
The associations between changes in working memory and changes in MMN in our study imply that enhancements in NMDAR functioning may help bolster working memory when tDCS targets the PPC during concurrent cognitive task, but seem to be non-beneficial for working memory when tDCS targets the DLPFC with concurrent cognitive task. Therefore, direct stimulation via tDCS may not necessarily improve DLPFC functioning for working memory. Moreover, a large multi-site functional magnetic resonance imaging study [63] found that as memory load increased from low to moderate, the change of DLPFC activation was higher in schizophrenic patients compared to healthy controls. This finding suggests that working memory deficits are linked with DLPFC inefficiency instead of hyper- or hypo-frontality in schizophrenia. These findings might explain results from our study and negative cognitive performance in the active DLPFC group in a large-scale tDCS trial conducted by Bulubas et al. [64]. Additionally, the active PPC group demonstrated significant cognitive effects at both week 1 and week 2 compared to the active DLPFC group, yet only at week 2 relative to the sham group. This finding might also be attributed to the complex relationship between working memory and the DLPFC dysfunction in schizophrenia.
We found that tDCS targeting the PPC with concurrent cognitive performance improved MMN amplitude in schizophrenia. Previous research has indicated that the primary benefits of high-order cognitive training in schizophrenia manifest in neurocognition and real-world functioning, with limited impact on MMN [65]. Sehatpour et al. [31] observed that augmenting training with d-serine could elevate MMN amplitude in schizophrenia, displaying a dose-dependent effect. Our finding offers initial evidence for the efficacy of enhancing cognitive training with tDCS on MMN amplitude in schizophrenia and underscores a target-dependent effect, which bears significant clinical implications.
The three groups did not differ significantly regarding improvement in other domains of neurocognition in our study. Tang et al. [66] similarly demonstrated that transcranial magnetic stimulation with individualized targets based on the parietal-hippocampal functional connection improved the visuospatial learning but did not improve other cognitive functions in individuals at clinical high risk for psychosis, as well as schizophrenic patients. tDCS increases the likelihood of neuronal activation [6], thus tDCS may preferentially regulate the brain networks activated in the task, as per the network activity-dependent model of transcranial electrical stimulation [67]. Our study demonstrated that combining tDCS with concurrent visual working memory task could improve visual working memory but not other cognitive functions, lending support to the network activity-dependent model. Likewise, Wang et al. [21] found that tDCS improved motor learning in mice only when applied during movement. This improvement of motor learning by tDCS was task-specific, and tDCS served to strengthen the neural substrates underpinning the given task during intervention. These findings suggest that combining non-specific modulation of tDCS with task-specific activation of brain networks may allow for a more precise modulation of specific brain networks, much akin to the therapeutic effect of individualized TMS based on functional connectivity.
The active DLPFC group demonstrated an advantage in ameliorating negative symptoms relative to the active PPC group in our study. However, negative symptoms were also significantly reduced in the sham group vs the active PPC group. A meta-analysis has found that cognitive training can alleviate negative symptoms [68], which might account for the observed negative symptom improvements in the group receiving sham tDCS along with concurrent cognitive tasks. Additionally, the participants in our study exhibited relatively mild negative symptoms at baseline. This factor, especially in a study with a limited sample size, could accentuate the variability in individual responses to treatment, thereby increasing the likelihood of false-positive findings. Meanwhile, the initial mildness of the negative symptoms suggests that any observed changes might hold limited clinical relevance. Nevertheless, the existence of other unidentified mechanisms influencing these results cannot be ruled out. For instance, it remains to be explored whether the PPC stimulation might disrupt the effect of cognitive training on negative symptoms. Therefore, the effects of high-definition tDCS targeting the PPC on negative symptoms warrant further investigation. Additionally, there were no severe adverse events in our study, indicating the safety and tolerability of high-definition tDCS.
Limitations
The limitations of our study should be noted. First, despite the great challenges in recruiting clinically stable schizophrenia patients for clinical trials, our sample size was limited. Thus, our study should be considered exploratory, primarily serving as a proof-of-concept investigation. The findings need external validation with larger samples to confirm their applicability and reliability. Second, the duration of the follow-up was short. Future research with a longer follow-up period would be valuable. Furthermore, given the economic, portable, user-friendly, and safe advantages of tDCS and cognitive tasks, it is also worthwhile to investigate whether long-term use of home-based tDCS with concurrent cognitive performance can effectively sustain therapeutic effects over time. Last, it should be acknowledged that the significant practice effects might be an influential factor in the absence of notable differences between groups in the MCCB composite scores and some subdomains at week 1 and week 2. This aspect necessitates further investigation.
Conclusions
In conclusion, our study observed that tDCS targeting the PPC as opposed to the DLPFC with simultaneous working memory task improved working memory in schizophrenia. The enhancement in working memory may be associated with an upregulation of NMDAR functioning. However, no significant differences were observed in the primary outcome between the active PPC group and the sham stimulation group. This suggests that claims of the superiority of tDCS targeting the PPC in enhancing working memory warrant further investigation and validation.
Data availability
The data that support the findings of this study are available on request from the corresponding author (CW).
References
Baddeley A. Working memory. Science. 1992;255:556–59.
Galderisi S, Rossi A, Rocca P, Bertolino A, Mucci A, Bucci P, et al. The influence of illness-related variables, personal resources and context-related factors on real-life functioning of people with schizophrenia. World Psychiatry. 2014;13:275–87.
Goldberg TE, Goldman RS, Burdick KE, Malhotra AK, Lencz T, Patel RC, et al. Cognitive improvement after treatment with second-generation antipsychotic medications in first-episode schizophrenia: is it a practice effect? Arch Gen Psychiatry. 2007;64:1115–22.
Keefe RS, Bilder RM, Davis SM, Harvey PD, Palmer BW, Gold JM, et al. Neurocognitive effects of antipsychotic medications in patients with chronic schizophrenia in the CATIE trial. Arch Gen Psychiatry. 2007;64:633–47.
Sinkeviciute I, Begemann M, Prikken M, Oranje B, Johnsen E, Lei WU, et al. Efficacy of different types of cognitive enhancers for patients with schizophrenia: a meta-analysis. NPJ Schizophr. 2018;4:22.
Philip NS, Nelson BG, Frohlich F, Lim KO, Widge AS, Carpenter LL. Low-intensity transcranial current stimulation in psychiatry. Am J Psychiatry. 2017;174:628–39.
Das S, Holland P, Frens MA, Donchin O. Impact of transcranial direct current stimulation (tDCS) on neuronal functions. Front Neurosci. 2016;10:550.
Mencarelli L, Romanella SM, Di Lorenzo G, Demchenko I, Bhat V, Rossi S, et al. Neural correlates of N-back task performance and proposal for corresponding neuromodulation targets in psychiatric and neurodevelopmental disorders. Psychiatry Clin Neurosci. 2022;76:512–24.
Hahn B, Robinson BM, Leonard CJ, Luck SJ, Gold JM. Posterior parietal cortex dysfunction is central to working memory storage and broad cognitive deficits in schizophrenia. J Neurosci. 2018;38:8378–87.
Eriksson J, Vogel EK, Lansner A, Bergstrom F, Nyberg L. Neurocognitive architecture of working memory. Neuron. 2015;88:33–46.
Wu D, Jiang T. Schizophrenia-related abnormalities in the triple network: a meta-analysis of working memory studies. Brain Imaging Behav. 2020;14:971–80.
Smucny J, Dienel SJ, Lewis DA, Carter CS. Mechanisms underlying dorsolateral prefrontal cortex contributions to cognitive dysfunction in schizophrenia. Neuropsychopharmacol. 2022;47:292–308.
Hyde J, Carr H, Kelley N, Seneviratne R, Reed C, Parlatini V, et al. Efficacy of neurostimulation across mental disorders: systematic review and meta-analysis of 208 randomized controlled trials. Mol Psychiatry. 2022;27:2709–19.
Sloan NP, Byrne LK, Enticott PG, Lum J. Non-invasive brain stimulation does not improve working memory in schizophrenia: A meta-analysis of randomised controlled trials. Neuropsychol Rev. 2021;31:115–38.
Narita Z, Stickley A, DeVylder J, Yokoi Y, Inagawa T, Yamada Y, et al. Effect of multi-session prefrontal transcranial direct current stimulation on cognition in schizophrenia: a systematic review and meta-analysis. Schizophr Res. 2020;216:367–73.
Wang S, Itthipuripat S, Ku Y. Electrical stimulation over human posterior parietal cortex selectively enhances the capacity of visual short-term memory. J Neurosci. 2019;39:528–36.
Martin DM, Rushby JA, De Blasio FM, Wearne T, Osborne-Crowley K, Francis H, et al. The effect of tDCS electrode montage on attention and working memory. Neuropsychologia. 2023;179:108462.
Nikolin S, Lauf S, Loo CK, Martin D. Effects of high-definition transcranial direct current stimulation (HD-tDCS) of the intraparietal sulcus and dorsolateral prefrontal cortex on working memory and divided attention. Front Integr Neurosci. 2018;12:64.
Li S, Cai Y, Liu J, Li D, Feng Z, Chen C, et al. Dissociated roles of the parietal and frontal cortices in the scope and control of attention during visual working memory. Neuroimage. 2017;149:210–19.
Bikson M, Brunoni AR, Charvet LE, Clark VP, Cohen LG, Deng ZD, et al. Rigor and reproducibility in research with transcranial electrical stimulation: an NIMH-sponsored workshop. Brain Stimul. 2018;11:465–80.
Wang Y, Wang J, Zhang QF, Xiao KW, Wang L, Yu QP, et al. Neural mechanism underlying task-specific enhancement of motor learning by concurrent transcranial direct current stimulation. Neurosci Bull. 2023;39:69–82.
Burton CZ, Garnett EO, Capellari E, Chang SE, Tso IF, Hampstead BM, et al. Combined cognitive training and transcranial direct current stimulation in neuropsychiatric disorders: a systematic review and meta-analysis. Biol Psychiatry Cogn Neurosci Neuroimaging. 2023;8:151–61.
Lo K, Hopman HJ, Chan SC, Chau W, Cheng P, Cheung KY, et al. Concurrent anodal transcranial direct current stimulation (tDCS) with cognitive training to improve cognition in schizophrenia. Schizophr Res. 2022;241:184–86.
Weickert TW, Salimuddin H, Lenroot RK, Bruggemann J, Loo C, Vercammen A, et al. Preliminary findings of four-week, task-based anodal prefrontal cortex transcranial direct current stimulation transferring to other cognitive improvements in schizophrenia. Psychiatry Res. 2019;280:112487.
Nienow TM, MacDonald AR, Lim KO. TDCS produces incremental gain when combined with working memory training in patients with schizophrenia: a proof of concept pilot study. Schizophr Res. 2016;172:218–19.
Orlov ND, Tracy DK, Joyce D, Patel S, Rodzinka-Pasko J, Dolan H, et al. Stimulating cognition in schizophrenia: a controlled pilot study of the effects of prefrontal transcranial direct current stimulation upon memory and learning. Brain Stimul. 2017;10:560–66.
Schilling TM, Andelfinger V, Bossert M, König M, Wolfenson D, Lang S, et al. No clinically relevant effects of 12 sessions of 2 mA of anodal transcranial direct current stimulation over the left DLPFC in combination with concurrent cognitive training compared to cognitive training only on executive functions in patients with schizophrenia - a randomized controlled trial. Schizophr Res. 2022;248:287–89.
Kronberg G, Bridi M, Abel T, Bikson M, Parra LC. Direct current stimulation modulates LTP and LTD: activity dependence and dendritic effects. Brain Stimul. 2017;10:51–58.
Balu DT, Coyle JT. The NMDA receptor ‘glycine modulatory site’ in schizophrenia: D-serine, glycine, and beyond. Curr Opin Pharmacol. 2015;20:109–15.
Kantrowitz JT, Epstein ML, Beggel O, Rohrig S, Lehrfeld JM, Revheim N, et al. Neurophysiological mechanisms of cortical plasticity impairments in schizophrenia and modulation by the NMDA receptor agonist D-serine. Brain. 2016;139:3281–95.
Sehatpour P, Iosifescu DV, De Baun HM, Shope C, Mayer MR, Gangwisch J, et al. Dose-dependent augmentation of neuroplasticity-based auditory learning in schizophrenia: A double-blind, placebo-controlled, randomized, target engagement clinical trial of the NMDA glutamate receptor agonist d-serine. Biol Psychiatry. 2023;94:164–73.
Greenwood LM, Leung S, Michie PT, Green A, Nathan PJ, Fitzgerald P, et al. The effects of glycine on auditory mismatch negativity in schizophrenia. Schizophr Res. 2018;191:61–69.
Rosburg T, Kreitschmann-Andermahr I. The effects of ketamine on the mismatch negativity (MMN) in humans - a meta-analysis. Clin Neurophysiol. 2016;127:1387–94.
Javitt DC, Lee M, Kantrowitz JT, Martinez A. Mismatch negativity as a biomarker of theta band oscillatory dysfunction in schizophrenia. Schizophr Res. 2018;191:51–60.
Lee M, Sehatpour P, Hoptman MJ, Lakatos P, Dias EC, Kantrowitz JT, et al. Neural mechanisms of mismatch negativity dysfunction in schizophrenia. Mol Psychiatry. 2017;22:1585–93.
Erickson MA, Ruffle A, Gold JM. A meta-analysis of mismatch negativity in schizophrenia: from clinical risk to disease specificity and progression. Biol Psychiatry. 2016;79:980–87.
Wang X, Guo Y, Zhang Y, Li J, Gao Z, Li Y, et al. Combined behavioral and mismatch negativity evidence for the effects of long-lasting high-definition tDCS in disorders of consciousness: a pilot study. Front Neurosci. 2020;14:381.
Weigl M, Mecklinger A, Rosburg T. Transcranial direct current stimulation over the left dorsolateral prefrontal cortex modulates auditory mismatch negativity. Clin Neurophysiol. 2016;127:2263–72.
Impey D, de la Salle S, Knott V. Assessment of anodal and cathodal transcranial direct current stimulation (tDCS) on MMN-indexed auditory sensory processing. Brain Cogn. 2016;105:46–54.
Impey D, Knott V. Effect of transcranial direct current stimulation (tDCS) on MMN-indexed auditory discrimination: a pilot study. J Neural Transm. 2015;122:1175–85.
Chen JC, Hämmerer D, Strigaro G, Liou LM, Tsai CH, Rothwell JC, et al. Domain-specific suppression of auditory mismatch negativity with transcranial direct current stimulation. Clin Neurophysiol. 2014;125:585–92.
Wolosker H, Blackshaw S, Snyder SH. Serine racemase: a glial enzyme synthesizing D-serine to regulate glutamate-N-methyl-D-aspartate neurotransmission. Proc Natl Acad Sci USA. 1999;96:13409–14.
Montesinos GC, Mani AR. The role of D-serine in peripheral tissues. Eur J Pharmacol. 2016;780:216–23.
Hashimoto A. Effect of the intracerebroventricular and systemic administration of L-serine on the concentrations of D- and L-serine in several brain areas and periphery of rat. Brain Res. 2002;955:214–20.
Buchanan RW, Keefe RS, Umbricht D, Green MF, Laughren T, Marder SR. The FDA-NIMH-MATRICS guidelines for clinical trial design of cognitive-enhancing drugs: what do we know 5 years later? Schizophr Bull. 2011;37:1209–17.
Zhang, Fan J, Zhu H, Wang Yunhui X, Tan Yunlong, Yang Fude, et al. The clinical effect of computerized cognitive remediation therapy in schizophrenic patients with different levels of cognitive impairment. Chin J Psychiatry. 2020;53:328–34.
Tan S, Zhu X, Fan H, Tan Y, Yang F, Wang Z, et al. Who will benefit from computerized cognitive remediation therapy? Evidence from a multisite randomized controlled study in schizophrenia. Psychol Med. 2020;50:1633–43.
Valiengo L, Goerigk S, Gordon PC, Padberg F, Serpa MH, Koebe S, et al. Efficacy and safety of transcranial direct current stimulation for treating negative symptoms in schizophrenia: a randomized clinical trial. JAMA Psychiat. 2020;77:121–29.
Working Memory and Plasticity Lab. https://wmp.education.uci.edu/software/. Accessed 10 July 2023.
Kuo HI, Bikson M, Datta A, Minhas P, Paulus W, Kuo MF, et al. Comparing cortical plasticity induced by conventional and high-definition 4 × 1 ring tDCS: a neurophysiological study. Brain Stimul. 2013;6:644–48.
Lee J, Folley BS, Gore J, Park S. Origins of spatial working memory deficits in schizophrenia: an event-related fMRI and near-infrared spectroscopy study. PLoS ONE. 2008;3:e1760.
Huang Y, Datta A, Bikson M, Parra LC. Realistic volumetric-approach to simulate transcranial electric stimulation-ROAST-a fully automated open-source pipeline. J Neural Eng. 2019;16:56006.
Shi C, Kang L, Yao S, Ma Y, Li T, Liang Y, et al. The MATRICS consensus cognitive battery (MCCB): co-norming and standardization in China. Schizophr Res. 2015;169:109–15.
Keefe RS, Goldberg TE, Harvey PD, Gold JM, Poe MP, Coughenour L. The brief assessment of cognition in schizophrenia: reliability, sensitivity, and comparison with a standard neurocognitive battery. Schizophr Res. 2004;68:283–97.
Zhao YJ, Ma T, Zhang L, Ran X, Zhang RY, Ku Y. Atypically larger variability of resource allocation accounts for visual working memory deficits in schizophrenia. Plos Comput Biol. 2021;17:e1009544.
Draheim C, Tshukara JS, Engle RW. Replication and extension of the toolbox approach to measuring attention control. Behav Res Methods. 2024;56:2135-57.
Draheim C, Tsukahara JS, Martin JD, Mashburn CA, Engle RW. A toolbox approach to improving the measurement of attention control. J Exp Psychol Gen. 2021;150:242–75.
Brunoni AR, Amadera J, Berbel B, Volz MS, Rizzerio BG, Fregni F. A systematic review on reporting and assessment of adverse effects associated with transcranial direct current stimulation. Int J Neuropsychopharmacol. 2011;14:1133–45.
Leucht S, Samara M, Heres S, Davis JM. Dose equivalents for antipsychotic drugs: the DDD method. Schizophr Bull. 2016;42 Suppl 1:S90–94.
Wobrock T, Guse B, Cordes J, Wolwer W, Winterer G, Gaebel W, et al. Left prefrontal high-frequency repetitive transcranial magnetic stimulation for the treatment of schizophrenia with predominant negative symptoms: a sham-controlled, randomized multicenter trial. Biol Psychiatry. 2015;77:979–88.
Lenhard W, Lenhard A. Computation of effect sizes. 2016. https://www.psychometrica.de/effect_size.html. Psychometrica.
Cohen J. Statistical power analysis for the behavioral sciences. New York: Academic Press; 2013.
Potkin SG, Turner JA, Brown GG, McCarthy G, Greve DN, Glover GH, et al. Working memory and DLPFC inefficiency in schizophrenia: the FBIRN study. Schizophr Bull. 2009;35:19–31.
Bulubas L, Goerigk S, Gomes JS, Brem AK, Carvalho JB, Pinto BS, et al. Cognitive outcomes after tDCS in schizophrenia patients with prominent negative symptoms: Results from the placebo-controlled STARTS trial. Schizophr Res. 2021;235:44–51.
Best MW, Milanovic M, Iftene F, Bowie CR. A randomized controlled trial of executive functioning training compared with perceptual training for schizophrenia spectrum disorders: effects on neurophysiology, neurocognition, and functioning. Am J Psychiatry. 2019;176:297–306.
Tang Y, Xu L, Zhu T, Cui H, Qian Z, Kong G, et al. Visuospatial learning selectively enhanced by personalized transcranial magnetic stimulation over parieto-hippocampal network among patients at clinical high-risk for psychosis. Schizophr Bull. 2023;49:923–32.
Fertonani A, Miniussi C. Transcranial electrical stimulation: what we know and do not know about mechanisms. Neuroscientist. 2017;23:109–23.
Cella M, Preti A, Edwards C, Dow T, Wykes T. Cognitive remediation for negative symptoms of schizophrenia: a network meta-analysis. Clin Psychol Rev. 2017;52:43–51.
Funding
This study was supported by the National Natural Science Foundation of China (81971250 [to CW], 81971287 [to XL], 82171501 [to FZ]), and the Beijing Hospitals Authority Clinical Medicine Development of Special Funding Support (ZLRK202335 [to QB]).
Author information
Authors and Affiliations
Contributions
Concept and design: WH, LZ, ZW, CW. Acquisition, analysis, or interpretation of data: WH, JL, QW, HL, XQ, YD, LZ, CW. Drafting of the manuscript: WH. Critical revision of the manuscript for important intellectual content: FZ, JL, FD, QB, AL, ZC, XL, CW. Statistical analysis: WH, FZ. Obtained funding: FZ, XL, CW, QB. Administrative, technical, or material support: FD, QB, FZ, AL, ZW, XL, CW. Supervision: CW.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Hou, W., Zhou, F., Wang, Q. et al. Effect of transcranial direct current stimulation with concurrent cognitive performance targeting posterior parietal cortex vs prefrontal cortex on working memory in schizophrenia: a randomized clinical trial. Transl Psychiatry 14, 279 (2024). https://doi.org/10.1038/s41398-024-02994-w
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41398-024-02994-w
- Springer Nature Limited