Pone 0215095

RESEARCH ARTICLE
Transcranial direct current stimulation in

attention-deficit hyperactivity disorder: A
meta-analysis of neuropsychological deficits
Mohammad Ali Salehinejad ID1,2☯*, Miles Wischnewski3☯, Vahid Nejati4,5, Carmelo
M. Vicario2,6, Michael A. Nitsche2,7
1 Ruhr-University Bochum, International Graduate School of Neuroscience, Bochum, Germany, 2 Leibniz
Research Centre for Working Environment and Human Factors, Department of Psychology and
Neurosciences, Dortmund, Germany, 3 Donders Institute for Brain, Cognition, and Behaviour, Radboud
University Nijmegen, The Netherlands, 4 Department of Psychology, Shahid Beheshti University, Tehran,
a1111111111 Iran, 5 Department of Psychology, University of Regensburg, Regensburg, Germany, 6 University of
a1111111111 Messina, Department of Scienze Cognitive della Formazione e degli Studi Culturali, Messina, Italy,
a1111111111 7 University Medical Hospital Bergmannsheil, Department of Neurology, Bochum, Germany
a1111111111
a1111111111 ☯ These authors contributed equally to this work.
* salehinejad@ifado.de
Abstract
OPEN ACCESS
Transcranial direct current stimulation (tDCS) is a promising method for altering cortical
Citation: Salehinejad MA, Wischnewski M, Nejati V,
Vicario CM, Nitsche MA (2019) Transcranial direct excitability with clinical implications in neuropsychiatric diseases. Its application in neurode-
current stimulation in attention-deficit hyperactivity velopmental disorders especially attention-deficit hyperactivity disorder (ADHD), is in early
disorder: A meta-analysis of neuropsychological stage and promising but its effectiveness has not been systematically examined yet. We
deficits. PLoS ONE 14(4): e0215095. https://doi.
conducted a meta-analysis on the effectiveness of tDCS on the most studied neuropsycho-
org/10.1371/journal.pone.0215095
logical symptoms of ADHD, which is the first reported meta-analysis of tDCS studies on
Editor: Alessio Avenanti, University of Bologna,
ADHD. Data from 10 randomized controlled studies (including 11 separate experiments) tar-
ITALY
geting inhibitory control, and/or working memory (WM) in ADHD were included. Results
Received: February 4, 2019
show that overall tDCS significantly improved inhibitory control. Sub-analyses further show
Accepted: March 26, 2019 that dorsolateral prefrontal cortex (dlPFC) (but not right inferior frontal gyrus) tDCS and
Published: April 12, 2019 anodal (but not cathodal) tDCS significantly improved inhibitory control with a small effect
Copyright: © 2019 Salehinejad et al. This is an size. Anodal dlPFC-tDCS had the largest significant effect on inhibitory control with a small-
open access article distributed under the terms of to-medium effect size. Additionally, a significant improving effect of tDCS on inhibitory con-
the Creative Commons Attribution License, which trol accuracy (but not response time) and WM speed (but not accuracy) were found. Overall,
permits unrestricted use, distribution, and
this meta-analysis supports a beneficial effect of tDCS on inhibitory control and WM in
reproduction in any medium, provided the original
author and source are credited. ADHD with a small-to-medium effect size. TDCS seems to be a promising method for
improving neuropsychological and cognitive deficits in ADHD. However, there might be a
Data Availability Statement: All relevant data are
within the manuscript and its Supporting dissociation between neuropsychological deficits and clinical symptoms of ADHD and there-
Information files. fore, the significance of this meta-analysis for clinical purposes is limited. Future studies
Funding: Publication of this article was funded by should systematically evaluate the role of inter-individual factors (i.e., ADHD subtype, types
the Open Access Fund of the Leibniz Association. of the deficit) and stimulation parameters (i.e., site, polarity, intensity, duration, repetition
MAS receives support from the National
rate) on tDCS efficacy in ADHD population and examine whether benefits are long-term.
Organization of Science, Research & Technology,
Deputy of Scholarship and Students Affairs, Iran,
grant number: 95000171. The funder had no role
in the study design, data collection, and analysis,
PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 1 / 26

tDCS in ADHD: A meta-analysis
decision to publish or preparation of the Introduction

manuscript.
Attention-deficit hyperactivity disorder (ADHD) is the most common neurodevelopmental
Competing interests: MAN is a member of the
disorder with an estimate of 11% prevalence in school-age children [1]. ADHD is primarily
Scientific Advisory Board of Neuroelectrics. All
other authors declare no competing interests. This
characterized by symptoms of inattention, hyperactivity, impulsivity [2] and various cognitive
does not alter our adherence to PLOS ONE policies dysfunctions [3] that often persist into adolescence and adulthood. Apart from symptoms of
on sharing data and materials. hyperactivity and inattentiveness, a wide range of cognitive deficits is observed in individuals
with ADHD such as problems in attention, inhibitory control working memory (WM), plan-
ning, problem-solving and executive functions [4–7]. Executive dysfunctions, especially inhib-
itory control and working memory, are pervasive and influential in ADHD pathophysiology to
the extent that ADHD was labeled a disorder of cognitive control [7, 8] but, cognitive deficits
in ADHD are heterogeneous [9–11].
A precise description of ADHD neuropathology is difficult to delineate due to its neuropsy-
chological heterogeneity [12, 13], and substantial overlap between ADHD and typically devel-
oping children [14]. However, based on neuroimaging and neuropsychological findings
relatively distinct brain regions can be identified to account for ADHD hallmark symptoms.
Poor inhibitory control resulting from executive dysfunctions (i.e., inhibition-based model)
and impulse-control deficits leading to hyperactivity (i.e., motivational dysfunction model) are
two influential theories for neural foundations of ADHD [15, 16]. These major symptoms
have been linked to frontostriatal circuitry and prefrontal cortex (PFC) abnormalities and can
be diverged in the “prefrontal hypothesis of ADHD” [17]. According to this hypothesis, pre-
frontal regions including the dorsolateral PFC (dlPFC), orbitofrontal cortex (OFC) and infe-
rior frontal gyrus (IFG) have altered activity in ADHD. Specifically, hypoactivity of the
bilateral dlPFC, right inferior frontal gyrus (rIFG) and OFC, as well as smaller prefrontal vol-
umes, are associated with behavioral deficits in ADHD [18–22]. Additionally, an adaptive
increase of posterior parietal activity accompanies the respective hypoactivity of frontostriatal
regions during performance of executive tasks [19]. These findings suggest that ADHD symp-
toms/causes stem from disturbances in large-scale brain networks [23, 24] and structural,
functional and neurochemical abnormalities in cortical and subcortical structures [13, 17, 25],
which should be considered in therapeutic interventions.
Pharmacological and non-pharmacological interventions have been among the most com-
monly used and recommended treatment options in ADHD. Pharmacological interventions
can broadly be divided into stimulant and non-stimulant medications [26, 27]. Non-pharma-
cological or psychological interventions can be categorized into behavioral interventions, cog-
nitive training and neurofeedback training [27, 28]. Stimulant medications have been so far
the most effective psychopharmacological treatment and adherence to treatment is reported
relatively high. Although current treatment options especially pharmacological interventions
have been relatively effective and successful, the efficacy of these treatment options has been
inconsistent in ADHD [27]. For example, the efficacy of pharmacological treatments is not
long-lasting enough [29] and suffers from the partial response or non-response [30], adverse
effects and relatively poor adherence [31]. On the other hand, neurofeedback training is found
to have limited effects on ADHD when rated by blind evaluators [32]. Due to the heterogeneity
of ADHD pathophysiology, symptoms and treatment response, there might be a need for new
treatment or complementary treatment options, especially regarding interventions based on
recent findings of the neuropathology of ADHD.
The significant role of brain abnormalities in ADHD symptomology encouraged research-
ers to probe new treatment options that target ADHD symptoms by modulating, altering and
remediating deficient neurocognitive brain functions. Transcranial direct current stimulation
(tDCS) is a non-invasive, painless, and well-tolerated brain stimulation technique that has

been gaining increased popularity in human neuroscience research in both healthy and clinical
populations during last decade [33–35]. It induces a weak direct current applied to the scalp
which modulates cortical excitability by shifting resting membrane potential, [36] and can
induce both, acute and neuroplastic alterations of cortical excitability at the macroscopic level
[37]. In the motor cortex, it has been suggested that anodal stimulation increases cortical excit-
ability while cathodal stimulation decreases it [38] although such polarity-dependent effect is
not always linear and could be affected by external and individual difference factors especially
stimulation parameters such as current density, polarity, stimulation duration and/ or geomet-
rical montage of electrodes [39, 40].
TDCS has been applied for modulating a variety of cognitive functions [41–44] or improv-
ing symptoms/deficits in healthy and clinical populations [45] (for an overview see [46, 47]).
Its application in neurodevelopmental populations, especially ADHD, has gained attention in
recent years [48, 49] while the efficacy of the method in ADHD is still unclear and needs fur-
ther research. Different stimulation parameters and montages have been used in previous
studies and different symptoms and deficits have been targeted by tDCS which does not evalu-
ate the potential of this method in ADHD. Moreover, most clinical tDCS studies so far have
focused on adult populations and there is a great need for more research into their therapeutic
uses in pediatric patients [49, 50]. Considering the fact that pediatric brains still have to go
through various stages of neurodevelopment and have accelerated neuroplasticity compared
to adults [51], tDCS could be considered a potential therapy for some pediatric disorders, par-
ticularly when there are no other safe viable alternatives [52]. That being said, knowledge
about tDCS effects in ADHD, is still relatively limited and warrants further investigation espe-
cially through systematic review or meta-analytic studies.
To date, meta-analysis or specific systematic review is available for the effects of tDCS on
ADHD. The present meta-analysis, therefore, aims to investigate the effectiveness of tDCS on
neuropsychological deficits in ADHD. Moreover, we did further sub-analyses on the available
studies in order to (1) evaluate overall and deficit-specific efficacy of tDCS in ADHD, (2) iden-
tify the brain regions that are prominently involved in ADHD pathophysiology tDCS response,
and (3) examine stimulation parameters that are important for tDCS efficacy in ADHD. Since
inhibitory control and WM deficiencies are among the major neuropsychological deficits of
ADHD, [53] and mediated by the genetic risk for ADHD, [54], we performed a meta-analysis
on the effects of tDCS on inhibitory control and WM performance in respective patients.
Method
Our meta-analysis follows the guidelines of the Preferred Reporting Items for Systematic
reviews and Meta-Analyses (PRISMA) [55] which consists of a checklist intended to facilitate
preparation and reporting review/meta-analysis studies by identifying, selecting, and critically
appraising relevant research, and collecting and analyzing data from the included studies. A
brief version of the PRISMA checklist is available as supporting information (See S1 Table).
The following steps are based on the PRISMA protocol.
Eligibility criteria
To ensure a high level of methodological accuracy, only peer-reviewed published studies were
included in our analysis. Only studies on humans using tDCS in experimental designs with the
control condition (i.e., sham stimulation and/or baseline control) with reported full-text in
English were retained. Studies were included if (1) they had a randomized placebo- (sham50)
controlled or baseline-controlled design, (2) measured at least inhibitory control, and/or work-
ing memory performance, (3) reported data (text, figures, tables, appendices) which contained

effect size values or mean and standard deviation values in order to calculate effect sizes. Par-
ticipants in the studies were required to be between 3 to 18 years of age (for childhood and
adolescence ADHD) and 18 to 65 years of age (for adult ADHD) and to have a clinical diagno-
sis of ADHD (any subtype) based on the Diagnostic and Statistical Manual of Mental Disor-
ders (DSM) or meet accepted cut-offs on validated ADHD symptom rating scales. Studies
which included patients under medication for ADHD treatment in either the control or the
active arm were not excluded.
Search strategy and study selection

A comprehensive literature search was conducted, including the PubMed database, Web of
Science, Medline, and Scopus with the final search updated on January 2019. Search terms
included attention-deficit hyperactivity disorder OR ADHD OR attention disorders AND
transcranial direct current stimulation OR tDCS OR transcranial electrical stimulation OR tES
(see Fig 1). Furthermore, a manual search was carried out over the reference sections of
retrieved studies and review articles. After removal of duplicates, the full text was assessed and
studies that were eligible according to the above-mentioned inclusion criteria were selected.
The final search identified a total of 241 articles with 1 additional article identified by reference
lists and recent reviews. Records were reduced to 28 articles after removing duplications. A
content expert examined each title and abstract for eligibility. 18 articles were excluded for not
meeting the inclusion criteria. Thus 10 articles remained for full-text assessment and data
extraction. 8 of these studies investigated inhibitory control and 5 studies explored tDCS
effects on WM performance. Articles were screened and selected independently by the first
and second authors. Respective study characteristics are summarized in Tables 1 and 2.
Outcome variables
We limited our focus on neuropsychological variables that were targeted in most ADHD stud-
ies using tDCS. Based on the most commonly available data, inhibitory control and WM were
identified as the variables for which the data were extracted. Inhibitory control and WM are
central components of ADHD executive dysfunctions, as proposed by causal models of
ADHD, [14, 53, 56] and encompass the most consistently impaired domains in ADHD [8].
Moreover, inhibitory control and WM are major neuropsychological deficits from the neuro-
scientific perspective, including the prefrontal-striatal model of ADHD [57]. For the meta-
analysis on inhibitory control, the following tasks were included: (1) Go/No-Go, (2) Stop Sig-
nal Task (SST), 3) Flanker, 4) Stroop, 5) Continuous Performance Test (CPT), and 6) Neuro-
psychological Development Assessment (NEPSY II). In the Go/No-Go task, inhibitory process
is reflected by the ability to inhibit a motor action in the case of specific number of “No-Go”
trial [58]. Similarly, in the SST, participants should stop responding to “Go” trials after the
stop signal is presented which shows the ability to inhibitory control. In the Stroop task, inhib-
itory control is reflected by the ability to suppress the meaning of a written word and focus on
the color [59]. For the Flanker task response, inhibition is measured by identifying the target
defined by its location while ignoring one or more distracting items that flank the targets in
the same or opposite direction [60]. In the CPT, which is primarily a test of attention, the
errors of commission and omission reflect inhibitory control ability [61]. Finally, in NEPSY-II
inhibitory control is measured by looking at a series of shapes or arrows and naming the shape
or direction or an alternative response, depending on the color or shape of the arrow [62]. For
the meta-analysis on WM, the following tasks were included: 1) N-back, 2) Digit Span, and 3)
Corsi block-tapping test. In the N-back task, subjects need to identify a stimulus that repeats
the one presented “n” items before its onset. In the digit span task, subjects are read or shown

Fig 1. (A) Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram of selection
of studies, (B) Bias assessment in individual studies. tDCS = Transcranial Direct Current Stimulation;
ADHD = attention-deficit hyperactivity disorder; IC = Inhibitory Control; WM = Working Memory.
https://doi.org/10.1371/journal.pone.0215095.g001
a list of digits and asked to recall them in order. In the Corbi Cubes test, the subject repeats
sequences of touches in different cubes (either forward or backward) [63].
Risk of assessment bias

Quality assessment of the included studies was performed using the Cochrane Collaboration’s
tool for risk of bias in randomized trials [64]. For each study, the authors judged the risk of
selection bias, performance bias, detection bias, attrition bias, reporting bias, and other biases.
Risk of bias was categorized as low, high or uncertain. The final rating was established through
consensus with the involvement of the senior author. The assessment of each study and the
percentage of bias are presented in Fig 1B.
Data extraction and statistical analysis

Hedges’ g was used as a measure of effect size. This effect size reflects the tDCS-induced differ-
ence compared to 1) sham, 2) baseline or 3) sham and baseline (i.e. posttest–pretest tDCS vs

Table 1. Characteristic of studies included in meta-analysis for the effecs of tDCS on inhibitory control.
# Authors N Mean age tDCS montage Intensity Duration Polarity On-/off- Control Task Outcome Hedges’
(target/ line g
reference)
1 Allenby et al 37 37.17 (range F3/Fp2 (25 2 mA 3 days x Anodal Offline Baseline CPT False positive errors 0.42
(2018) 18–56) cm2 both) 20 min + sham
True positive errors -0.06
Response time -0.11
F3/Fp2 (25 2 mA 3 days x Anodal Offline Baseline SST Reaction time -0.18
cm2 both) 20 min + sham
2 Bandeira et al 9 11.1 ± 2.8 F3/Fp2 (35 2 mA 5 days x Anodal Offline Baseline NEPSY II Total errors 0.12
(2016) cm2 both) 30 min
Completion time 0.54
3 Breitling et al 21 14.33 (range F8/mastoid (35 1 mA 20 min Anodal Online Sham Flanker Omission errors -0.11
(2016) 13–17) cm2 both) task
Comission errors 0.46
Reaction time -0.14
Reaction time 0.13
variability
F8/mastoid (35 1 mA 20 min Cathodal Online Sham Flanker Omission errors -0.60
cm2 both) task
Reaction time 0.13
Reaction time -0.02
variability
4 Cosmo et al 30 31.8 ± 11.6 F3/F4 (35 cm2 1 mA 20 min Anodal Offline Sham Go/No-go Correct responses -0.39
(2015) both) task
(letters)
Omission errors 1.17
Comission errors -0.15
Go/No-go Correct responses 1.09
task
(fruits)
Omission errors 0.44
5 Munz et al 14 12.3 ± 1.4 F3+F4/both 0–0.25 mA 5 x 5 min Anodal Offline Sham Go/No-go Reaction time 0.88
(2015) mastoids (0.5 (oscillatory) task
cm2 all)
Reaction time 0.83
variability
6 Nejati et al 15 10 ± 2.2 F3/F4 (25 cm2 1 mA 15 min Anodal Offline Sham Go/No-go Go accuracy 0.14
(2017) both) task
experiment 1
No-go accuracy 0.56
Reaction time -0.41
Stroop Accuracy 1.11
task
Reaction time 0.98
7 Nejati et al 10 9 ± 1.8 F3/Fp2 (25 1 mA 15 min Anodal Offline Sham Go/No-go Go accuracy 0.41
(2017) cm2 both) task
experiment 2
No-go accuracy 0.57
Reaction time -0.20
(Continued )

Table 1. (Continued)
(target/ line g
reference)
F3/Fp2 (25 1 mA 15 min Cathodal Offline Sham Go/No-go Go Accuracy 0.41
cm2 both) task
No-go accuracy 1.04
Reaction time -0.20
8 Soltaninejad 20 Range 15– F3/Fp2 (35 1.5 mA 8 min Anodal Offline Sham Go/No-go Go accuracy -0.05
et al (2015) 17 cm2 both) task
No-go accuracy 0.03
Reaction time 0.23
task
Reaction time 0.23
F3/Fp2 (35 1.5 mA 8 min Cathodal Offline Sham Go/No-go Go accuracy -0.54
cm2 both) task
No-go accuracy 0.73
Reaction time -0.02
task
Reaction time 0.11
Reaction time 0.02
9 Sotnikova et al 13 14.33 ± 1.3 F3 (13 cm2)/ 1 mA 30 min Anodal Online Sham Go/No-go accuracy (hits -1.03
(2017) Cz (35 cm2) +correct rejections/
total number of
stimuli)
Reaction time 0.16
Reaction time -0.14
variability
tDCS = transcranial direct current stimulation; F3 = left dlPFC; F4 = right dlPFC; F8 = inferior frontal gyrus; Fp1 = left supraorbital area; Fp2 = right supraorbital area;
online = task performance during tDCS; offline = task performance after tDCS; CPT = Conners Continuous Performance Task; SST = Stop Signal Task (SST).
https://doi.org/10.1371/journal.pone.0215095.t001
posttest–pretest sham). When the effect size Cohen’s d was reported, the effect sizes were
extracted directly from the selected articles. Otherwise, d was calculated using means and
pooled standard error of mean, which were gathered from the results section, figures or tables.
From these effect sizes, Hedges’ g values were calculated to correct for effect inflation due to
small sample sizes [65]. In this meta-analysis, positive values reflect a tDCS-induced increase
of inhibitory control or WM performance, whereas negative values indicate a tDCS-induced
decrease of inhibitory control or WM performance compared to sham/baseline values. A
weighted average was calculated from these Hedges’ g values to determine the cumulative
effect size (Ē) and 95% confidence intervals. Subsequently, the Z-statistic and p-value were cal-
culated to investigate whether Ē differed significantly from zero.
For the main investigation on the effects of tDCS on inhibitory control in ADHD patients,
three meta-analytic steps were performed. First, an unsigned analysis, in which the general
effect of frontal tDCS, independent of polarity was analyzed. This analysis included 46
extracted effect size values. We further explored the effect of electrode montage by comparing
tDCS setups that primarily targeted the dlPFC (38 extracted effect size values) or the rIFG (8
extracted effect size values). Second, a signed analysis was performed in which the effects of
anodal (34 extracted effect size values) and cathodal tDCS (12 extracted effect size values) were

Table 2. Characteristic of studies included in the meta-analysis for the effects of tDCS on working memory.
(target/ line g
reference)
1 Bandeira et al 9 11.1 ± 2.8 F3/Fp2 (35 cm2 2 mA 5 days x 30 Anodal Offline Baseline Digit span Amount -0.87
(2016) both) min forward
Digit span Amount -0.40
backward
Corsi cube Amount -0.45
forward
Corsi cube Amount 0.08
backward
2 Nejati et al 15 10 ± 2.2 F3/F4 (25 cm2 1 mA 15 min Anodal Offline Sham 1-back task Accuracy 0.15
(2017) both)
experiment 1
Reaction time 1.70
3 Nejati et al 10 9 ± 1.8 F3/Fp2 (25 cm2 1 mA 15 min Anodal Offline Sham 1-back task Accuracy 1.14
(2017) both)
experiment 2
Reaction time 0.82
F3/Fp2 (25 cm2 1 mA 15 min Cathodal Offline Sham 1-back task Accuracy 0.53
both)
Reaction time 0.52
4 Prehn- 12 12.1 (range F3+F4/both 0–0.25 mA 5 x 5 min Anodal Offline Baseline Digit span Amount -0.61
Kristensen et al 10–14) mastoids (0.5 (oscillatory) + sham
(2014) cm2 all)
5 Soff et al (2017) 15 14.2 ± 1.2 F3 (3.14 cm2)/ 1 mA 5 days x 20 Anodal Offline Baseline QB (1-back) QB score (errors 0.50
Cz (12.5 cm2) min + sham task and reaction
time)
6 Sotnikova et al 13 14.33 ± 1.3 F3 (13 cm2)/ Cz 1 mA 30 min Anodal Online Sham 1-back task Accuracy -0.99
(2017) (35 cm2)
Reaction time -0.05
Reaction time 0.18
variability
2-back task Accuracy -1.14
Reaction time 0.65
Reaction time 1.06
variability
tDCS = transcranial direct current stimulation; F3 = left dlPFC; F4 = right dlPFC; Fp2 = right supraorbital area; online = task performance during tDCS; offline = task
performance after tDCS; QbTest = Quantified Behavior Test.
determined separately. This analysis was then followed by an exploration of tDCS montage.
Finally, since the majority of outcome measurements investigated both, accuracy (i.e., amount
of correct responses or error rate) and speed (i.e. reaction time or reaction time variability) on
inhibitory control tasks, separate analyses were performed for these variables (27 and 19
extracted effect size values respectively). For the WM analyses, not enough observations could
be gathered for cathodal tDCS stimulation (n = 2) and therefore no separate analyses were
performed.
For all analyses, the sample distribution was checked for normality by the Kolgomorov-
Smirnov test. Additionally, total heterogeneity (Qtotal) was tested against the χ2 distribution
with n-1 degrees of freedom to determine whether the variance of effect sizes was greater than
to be expected from sampling error [65]. To further investigate possible effects of publication

bias, the fail-safe number, based on the Rosenthal method (α < 0.05), was calculated, which
gives an indication of the amount of null findings that are needed to render the cumulative
effect non-significant [66]. Data were analyzed using MetaWin 2.1[67] and IBM SPSS 22.0. All
statistical tests were tested against a significance level of α � 0.05 (two-tailed).
Results
Overview
In total, 11 separate experiments from 10 studies published from 2014—April 2018 were
included in this meta-analysis. In 8 of 10 studies (or 9 of 11 experiments), effects of tDCS on
inhibitory control was investigated [6, 68–74] and in 5 of 10 studies (or 6 of 11 experiments),
tDCS effects on WM were explored [6, 69, 74–76]. In what follows, we describe the effect size
obtained from studies to examine the efficacy of tDCS in ADHD major neuropsychological
symptoms (Please see Fig 2). Brief details of each study are also summarized in Tables 1 and 2.
Risk of bias
The authors’ judgment on the risk of bias is displayed in Fig 1B. Overall, the risk of bias was
low. However, uncertainty about the risk does exist, at least to a certain extent, for selection
bias, performance bias, and detection bias. Especially, for determination of the latter, informa-
tion was missing in the majority of studies. A selective reporting bias was observed for the
study of Munz et al. [72] since accuracy data for the experimental conditions are not reported
in the manuscript, nor in any appendices. In all studies, blinding of participants was reported,
but blinding of experimenters was not reported in two studies [70, 73]. We need to note that
true blinding procedure of participants requires debriefing at the end of treatment about the
active or sham mode of stimulation confirmed by statistical analysis. This was reported or con-
firmed only in three included studies implicating that blinding of participants should be inter-
preted with caution. All studies used randomization to allocate participants to different
experimental conditions.
Effects of tDCS on inhibitory control in ADHD patients

A significant cumulative effect size (Ē) of 0.197 (Z = 2.76, p = 0.006) was observed for a general
tDCS effect on inhibitory control, taking polarity not into account Kolmogorov-Smirnov’s test
of normality showed that the distribution of the effect sizes was not significantly different from
a normal distribution (lower bound p = 0.20) and total heterogeneity of the effect sizes was not
significant (Qtotal = 45.21, p = 0.463). The fail-safe number indicated that 221 unpublished
null-findings would be required to render the effect non-significant. Exploration of montage
showed that only dlPFC stimulation (l-dlPFC and bilateral) (Ē = 0.243, Z = 2.92, p = 0.004),
but not rIFG stimulation (Ē = 0.005, Z = 0.04, p = 0.971) yielded a significant increase of accu-
racy rates in inhibitory control task performance.
Subsequently, polarity-dependent effects were investigated. Studies using anodal tDCS
showed a significant Ē of 0.232 (Z = 2.72, p = 0.007), with a fail-safe number of 171 showing
that anodal tDCS significantly improved inhibitory control. This sample was distributed nor-
mally (lower bound p = .20) and showed no significant heterogeneity (Qtotal = 33.05,
p = 0.465). As for the stimulation polarity-independent analysis, this effect was driven by stud-
ies using a left and bilateral dlPFC montage (Ē = 0.255, Z = 2.66, p = 0.008), whereas the rIFG
montage did not yield a significant effect (Ē = 0.084, Z = 0.33, p = 0.745). In contrast to anodal
tDCS, cathodal tDCS did not show a significant overall effect (Ē = 0.089, Z = 0.63, p = 0.530).

Fig 2. Meta-analysis and forest plot results including Hedges’ g and 95% confidence interval and Cumulative effect size of tDCS on
inhibitory control (top) and working memory (down).
https://doi.org/10.1371/journal.pone.0215095.g002

Table 3. Meta-analysis results for the effects of tDCS on inhibitory control in ADHD patients.
Cumulative effect size Normality Heterogeneity
Analysis N Ē 95% CI Z p-v0alue Fail-safe number KS test p-value Qtotal p-value
Polarity-independent
All studies 46 0.197 0.057–0.336 2.758 0.0058 221 0.088 LB 0.200 45.21 0.463
dlPFC only 38 0.243 0.080–0.406 2.922 0.0035 97 0.077 LB 0.200 36.87 0.475
rIFG only 8 0.005 -0.261–0.271 0.037 0.9705 0 0.195 LB 0.200 6.411 0.493
Polarity-dependent
Anodal tDCS 34 0.232 0.065–0.400 2.723 0.0065 171 0.095 LB 0.200 33.05 0.465
dlPFC only 30 0.255 0.067–0.443 2.658 0.0079 56 0.085 LB 0.200 28.98 0.466
1
rIFG only 4 0.084 -0.422–0.589 0.325 0.7452 0 2.25 0.523
Cathodal tDCS 12 0.089 -0.189–0.367 0.628 0.5300 0 0.133 LB 0.200 11.27 0.421
dlPFC only 8 0.194 -0.212–0.600 0.937 0.3488 0 0.114 LB 0.200 7.06 0.423
1
rIFG only 4 -0.075 -0.635–0.486 -0.263 0.7926 0 3.01 0.390
Speed vs Accuracy
Accuracy 27 0.250 0.042–0.459 2.356 0.0185 36 0.096 LB 0.200 26.30 0.447
Speed 19 0.093 -0.075–0.261 1.085 0.2779 0 0.208 0.031 18.08 0.451
tDCS = Transcranial Direct Current Stimulation; dlPFC = dorsolateral prefrontal cortex; LB = lower bound; rIFG = right inferior frontal gyrus; Ē = cumulative effect
size; CI = Confidence interval; KS = Kolmogorov-Smirnov’s test of normality; Qtotal = total heterogeneity represented by Cohen’s Q; Significant results are highlighted
in bold. dlPFC refers to either left dlPFC or bilateral dlPFC (for detailed information refer to Tables 1 and 2 under tDCS montage column). 1KS test could not be
performed because of too small sample size
Finally, an analysis was performed separating outcomes measures that focused on accuracy
or amount of errors compared to the speed of response. The results showed that tDCS signifi-
cantly increased accurate responses in inhibitory control tasks (Ē = 0.250, Z = 2.36, p = 0.019).
No significant cumulative effect was found for speed (Ē = 0.093, Z = 1.09, p = 0.278). For this
last analysis, a deviation from normality was observed (p = 0.031). Results are summarized in
Table 3.
Effects of tDCS on working memory in ADHD patients

No significant cumulative effect was observed for tDCS on working memory, without taking
polarity into account (Ē = 0.160, Z = 0.81, p = 0.418). Also, no effect of tDCS was observed
when only studies with an anodal montage were included (Ē = 0.116, Z = 0.53, p = 0.600).
However, when separating outcomes for accuracy and speed, a significant effect of tDCS on
speed was observed. TDCS led to a faster response time (Ē = 0.681, Z = 2.43, p = 0.015), with a
fail-safe number of 17. The sample was normally distributed (lower bound p = .20) and no sig-
nificant heterogeneity was seen (Qtotal = 5.88, p = 0.437). These results should be interpreted
with caution, given the low sample size (N = 7). Moreover, results showed that tDCS did have
no significant effect on accuracy of working memory task performance (Ē = -0.187, Z = -0.76,
p = 0.446). Results are shown in Table 4.
Discussion
ADHD is a major neurodevelopmental disorder with remarkable heterogeneity in symptoms,
etiologies, and treatment response. Recent studies have emphasized on executive dysfunctions
and brain functional abnormalities underlying ADHD pathophysiology [14, 20, 54, 57] sug-
gesting that modulation of cortical activity in the involved brain regions via tDCS might

Table 4. Meta-analysis results for the effects of tDCS on working memory in ADHD patients.
Cumulative effect size Normality Heterogeneity
Analysis N Ē 95% CI Z p-value Fail-safe number KS test p-value Qtotal p-value
Polarity independent
All studies 18 0.160 -0.227–0.547 0.811 0.4176 0 0.111 > 0.200 17.36 0.430
Polarity-dependent
Anodal tDCS 16 0.116 -0.317–0.549 0.525 0.5996 0 0.107 > 0.200 15.46 0.419
Speed vs Accuracy
Accuracy 11 -0.187 -0.668–0.295 -0.762 0.4461 0 0.163 > 0.200 10.09 0.433
Speed 7 0.681 0.131–1.231 2.427 0.0152 17 0.130 > 0.200 5.88 0.437
tDCS = Transcranial Direct Current Stimulation; Ē = cumulative effect size; CI = Confidence interval; KS = Kolmogorov-Smirnov’s test of normality; Qtotal = total
heterogeneity represented by Cohen’s Q; Significant results are highlighted in bold
ameliorate neuropsychological symptoms [77]. Exploring the impact of tDCS especially on

ADHD neuropsychological deficits has significantly increased in the last few years. However,
no meta-analysis about tDCS efficacy in ADHD is available. We conducted a meta-analysis of
randomized, placebo-or-baseline-controlled trials of tDCS application in ADHD, which is the
first meta-analysis for ADHD patients.
Overall, results of this meta-analysis show that tDCS improved inhibitory control and WM
in ADHD. Further sub-analyses yielded the following findings: (1) tDCS has an overall signifi-
cant cumulative effect on inhibitory control in ADHD with a small effect size, (2) when the tar-
geted brain region is taken into account, only tDCS over the dlPFC had a significant effect on
inhibitory control (small-to-medium effect size), but not tDCS over the rIFG, (3) when stimu-
lation polarity was taken into account, only anodal, but not cathodal tDCS had a significant
effect on inhibitory control, (4) when both polarity and targeted region are taken into account,
only anodal tDCS of the dlPFC had a significant effect on inhibitory control with a small-to-
medium effect size, (5) and when analyzing inhibitory control outcomes separately, tDCS had
a significant cumulative effect on accuracy, but not speed (i.e., reaction time). With regard to
WM (6), tDCS had an overall significant effect only on performance speed but not on accu-
racy. In what follows we discuss tDCS effects on these two major target deficits, the stimulation
parameters that could yield to larger effects, and finally clinical and methodological implica-
tions for future studies.
tDCS effects on inhibitory control in ADHD

The significant overall effect of tDCS on inhibitory control in ADHD is in line with a previous
meta-analysis about the effects of this technique on a variety of cognitive functions (e.g., WM,
executive functions, language) [78–80] and neuropsychiatric populations [80–82]. Moreover,
the small effects size of the overall effect of tDCS found in our meta-analysis is consistent with
small-to-medium reported effect sizes in previous meta-analyses about the effectiveness of
tDCS [78, 82–84]. However, the effects of tDCS on inhibitory control were associated with
larger effect size when target area (i.e., electrode positioning) and stimulation polarity (anodal
vs cathodal) were taken into account.
Anodal dlPFC tDCS had the largest effect size (small-to-medium effect) on inhibitory con-
trol in ADHD populations, whereas anodal rIFG tDCS had no significant effect. For stimula-
tion over the dlPFC, both left and right dlPFC were targeted in different studies but the left
dlPFC was almost always stimulated with the target electrode. First of all, this is in line with

previous findings about significant involvement of the dlPFC in ADHD symptoms, especially
executive dysfunctions [16, 18]. Inhibitory control is a major component of dlPFC-supported
executive functioning [85, 86] which is impaired in ADHD. The dlPFC is hypoactive in
ADHD, which is associated with insufficient suppression of dlPFC signaling not only in the
resting state, but also during the performance of cognitive tasks [20, 87]. Recent meta-analyses
of neuroimaging studies about functional abnormalities in ADHD populations showed a bilat-
eral hypoactivity of the dlPFC [88] and reduced left medial frontal cortex activation [89] in
ADHD populations during inhibitory control, WM and attention task performance. This
hypoactivity of the dlPFC is assumed to underlie the attentional deficit, impaired inhibitory
control and executive dysfunctions in ADHD and therefore, the pathophysiological rationale
for the therapeutic application of tDCS is to increase dlPFC activation with anodal stimulation.
It further supports the “cognitive dysfunction or inhibition-based model” of ADHD which
suggests that inhibition-based executive deficits are a core deficit in ADHD [56, 90].
Secondly, it highlights the importance of stimulation site for the tDCS effects on inhibitory
control. Our findings showed that dlPFC tDCS and not rIFG tDCS is effective. Moreover, the
target area of the reference electrode impacts on tDCS effects on inhibitory control. TDCS
effects on inhibitory control were more effective when two specific regions were stimulated by
the target and reference electrodes. The left dlPFC-right OFC montage was the most effective
montage and all of the four experiments which used this protocol reported improved perfor-
mance. In contrast, the left dlPFC-right dlPFC tDCS montage, which was used in two studies,
did not improve any inhibitory control outcome measure. In two studies that targeted only
one prefrontal region (i.e., left dlPFC-Cz tDCS, left/right dlPFC- mastoid tDCS), inhibitory
control was also improved (see Table 5). Thirdly and in addition to the stimulation site, the
results of this meta-analysis demonstrate the importance of stimulation polarity in the tDCS
effects on inhibitory processes in ADHD. Anodal tDCS, especially over dlPFC yielded the larg-
est effect size for improving inhibitory control in ADHD. Even the two studies that reported
significant effects of cathodal tDCS on response inhibition, argued that cathodal tDCS over the
left dlPFC was probably effective in improving inhibitory control in ADHD by increasing
activity of the right dlPFC through the inhibitory link between contralateral dlPFC regions via
the transcallosal connections. That being said, further systematic investigations are still
required to more robustly specify the contribution of potential stimulation sites and polarity
on response inhibition in ADHD populations.
Additionally, the findings suggest that heterogeneity of ADHD subtypes and brain abnor-
malities should not be neglected in interpreting the tDCS effects. This heterogeneity could
explain why rIFG tDCS was not as effective as dlPFC tDCS in the present analysis. For exam-
ple, both, rIFG and left/right dlPFC abnormalities are shown to be associated with inhibitory
control deficits in ADHD [13, 18, 53]. However, a differential involvement of the dorsal/ven-
tral divisions within the lateral PFC is suggested in ADHD pathophysiology. Dorsal regions
(i.e., dlPFC) are suggested to support inhibitory control of cognitive processes, while more
ventral regions (e.g., rIFG) support motor domains of inhibitory control [18, 91]. This shows
that ADHD subtypes (i.e., inattention type vs hyperactive type) are important for variations in
tDCS effects and implicates that the rIFG tDCS might be well-suited for those groups of
patients with more hyperactive than inattentive symptoms. Moreover, dlPFC has a global con-
tribution to executive functioning and response inhibition regardless of modality [42]. This
global contribution of the dlPFC in all aspects of response inhibition might explain the signifi-
cant implication of the dlPFC, but not rIFG in ADHD inhibitory processes. Furthermore,
ADHD is a disorder with disturbances in large-scale inter-related brain networks [23, 24]. The
dlPFC is a crucial part of the prefrontal network and compared to other prefrontal regions, has
many connections with other cortical/subcortical areas [92, 93]. It is possible that dlPFC

Table 5. Characteristics of the stimulation site and polarity in the studies investigated tDCS effects on inhibitory control and working memory.
Inhibitory control
# Authors Stimulation target (site) polarity Performance improvement
1 Breitling et al (2016) right IFG—left mastoid Anodal No
2 Soltaninejad et al (2015) Left dlPFC—right supraorbital area (OFC) Anodal No
Left dlPFC—right supraorbital area (OFC) Cathodal Yes
3 Allenby et al (2018) Left dlPFC—right supraorbital area (OFC) Anodal Yes
4 Bandeira et al (2016) Left dlPFC—right supraorbital area (OFC) Anodal Yes
5 Nejati et al (2017) Exp 2 Left dlPFC—right supraorbital area (OFC) Anodal No
Left dlPFC—right supraorbital area (OFC) Cathodal Yes
6 Nejati et al (2017) Exp 1 Left dlPFC—right dlPFC Anodal No
7 Cosmo et al (2015) Left dlPFC—right dlPFC Anodal No
8 Sotnikova et al (2017) Left dlPFC—vertex Anodal Yes
9 Munz et al (2015) Left dlPFC–mastoid Anodal Yes (only RT)
right dlPFC—mastoid Anodal Yes (only RT)
Working memory
1 Soff et al. (2017) Left dlPFC—vertex Anodal Yes (both accuracy and RT)
2 Sotnikova et al (2017) Left dlPFC—vertex Anodal Yes (both accuracy and RT)
3 Nejati et al (2017) Exp 1 Left dlPFC—right dlPFC Anodal Yes (only RT)
4 Nejati et al (2017) Exp 2 Left dlPFC—right supraorbital area (OFC) Anodal Yes (both accuracy and RT)
Left dlPFC—right supraorbital area (OFC) Cathodal No
5 Bandeira et al (2016) Left dlPFC—right supraorbital area (OFC) Anodal No
6 Prehn-Kristensen et al (2014) Left dlPFC / right dlPFC–mastoid Anodal Yes
tDCS = transcranial direct current stimulation; dlPFC = dorsolateral prefrontal cortex; IFG = inferior frontal gyrus; OFC = orbitofrontal cortex.
stimulation effectively hits the entire prefrontal network including the dlPFC-rIFG sub-net-
work, whereas, rIFG stimulation has more restricted effects. Such a global contribution of the
dlPFC to other prefrontal region functions has been shown for various executive functions
[42] which might be true for ADHD too.
tDCS and WM performance in ADHD

Our meta-analysis showed that tDCS does not have an overall significant effect on WM perfor-
mance accuracy in ADHD, although it led to significantly faster response time. To be more
specific, it was shown that anodal left dlPFC tDCS, but not cathodal stimulation improved
response time during WM performance in ADHD. This finding should be discussed consider-
ing physiological processes underlying inhibitory control and WM, task characteristics, and
results of previous meta-analyses about tDCS effects on WM. First of all, this result shows a
dissociation of tDCS effects on inhibitory control and WM. Such a dissociation has also been
shown for neuropsychological deficits in ADHD [53]. Inhibitory control is an executive con-
trol function [42] and requires executive attention to inhibit a prepotent response and moni-
toring, whereas memory tasks can deploy attentional resources [53]. Different effects of tDCS
on inhibitory control and WM thus could be attributed to different cognitive processes under-
lying WM and inhibitory control as well as target area-specific effects of tDCS, which need to
be studied in future systematically. Characteristics of the respective WM tasks can also affect
the results. For example, performance on the n-back task can still be faster while accuracy can-
not due to ceiling effects in case of near-optimal accuracy already without intervention.

Another factor that could be a source of variance of the tDCS effects on WM in the included
studies is the difference of stimulation sites which is an important feature of tDCS efficacy.
The stimulation montages of the studies on WM included anodal left dlPFC-right OFC tDCS
(two experiments), left dlPFC-right dlPFC tDCS (two experiments), and left dlPFC-vertex
tDCS (two experiments). Different montages used in these studies can explain the observed
effects and may explain why some studies demonstrated no effects on WM accuracy. In four of
the experiments, the reference electrode was positioned over regions that are involved in cog-
nitive performance (i.e., right OFC and right dlPFC). In this case, it might be speculated that
the cathodal return electrode compromises performance, and thus partially antagonizes anodal
tDCS effects over the target region. The other two experiments used unilateral protocols that
targeted the mere region of interest without positioning the cathodal return electrode over a
potentially involved region and interestingly, in these studies both WM accuracy and RT
improved, and prolonged effects outlasting the stimulation were also reported. Therefore, an
electrode positioning which avoids such antagonistic effects might be advantageous.
Recent meta-analyses and review articles about the effects of tDCS on cognitive functioning
showed mixed results regarding memory tasks. For example, a recent systematic review of 188
trials reported significant effects of anodal tDCS on RT only for executive functioning tasks,
but not memory tasks in both, healthy and neuropsychiatric populations [80]. Similar to our
finding, another meta-analysis of 12 tDCS WM trails found that tDCS significantly improved
RT, but not accuracy [78]. A more recent meta-analysis about the effects of tDCS on WM
(measured with n-back and digit span tasks) found a significant effect of anodal tDCS on accu-
racy and RT for offline, but not online tasks in healthy subjects; in neuropsychiatric popula-
tions, tDCS had a significant effect on WM accuracy (both offline and online) but not RT [84].
Finally, a recent meta-analysis of 61 studies investigated the effect of tDCS on cognitive tasks
in relation to task and stimulation parameters. They found that in contrast to offline-task per-
formance, online-task performance resulted in increased accuracy in clinical populations [94].
Interestingly, in 5 out of 6 studies included in our meta-analysis for WM, offline task perfor-
mance was measured in ADHD individuals. Thus, the non-significant effect of tDCS on WM
accuracy we found could be due to offline task performance in the majority of studies, which
might be less efficient. Nevertheless, the results of the effect of tDCS on WM in ADHD should
be treated with caution due to the limited number of qualified studies that did not allow us to
disentangle polarity- and area-dependent effect of tDCS.
Clinical and methodological implications

This meta-analysis found that modulation of the activity of brain areas involved in the patho-
physiology of ADHD with tDCS has a significant effect on response inhibition and WM, two
major executive dysfunctions in ADHD [14, 53, 54]. This suggests that tDCS over the dlPFC
might be suited as a potential therapeutic approach in ADHD, similar to other neuropsychiat-
ric syndromes which are responsive to this method [45]. Increasing evidence from clinical and
cognitive neuroscience describes ADHD as a disorder with functional/structural abnormali-
ties, especially of the frontostriatal circuitry and PFC. Accordingly, modulation of the involved
brain areas with non-invasive brain stimulation is a promising approach for improving neuro-
psychological symptoms in ADHD patients. The efficacy of this method might, however,
depend on ADHD subtypes/symptoms and stimulation parameters, which need to be investi-
gated systematically in future studies.
The findings of this meta-analysis have some important clinical and methodological impli-
cations to be considered for future tDCS studies in ADHD. The results of the included studies
showed beneficial effects of tDCS in ADHD, and at the same time indicate that effects of tDCS

on ADHD deficits (namely inhibitory control and WM) are strongly dependent on 1) individ-
ual and inter-individual factors (e.g., type of symptoms/deficits) and 2) also stimulation
parameters (i.e., site or cortical target, polarity, intensity, duration, and repetition rate). For
example, heterogeneous ADHD subtypes in the analyzed studies (i.e., inattentive or hyperac-
tive) might partially explain the non-significant effect of anodal rIFG tDCS on response inhibi-
tion, since dlPFC activity contributes to response inhibition in all ADHD subtypes (especially
individuals with inattentive symptoms), while the rIFG is assumed to be specifically involved
in ADHD individuals with more hyperactivity subtypes. Therefore, the inter-individual vari-
ability is an important factor which generally affects the efficacy of tDCS [95] and might be
specifically relevant for the treatment of ADHD patients, that vary in subtypes. Furthermore,
findings of those studies that investigated tDCS effects on other ADHD deficits showed that
for those EF domains that involve motivational and emotional processing, tDCS over both,
prefrontal and frontopolar areas is more effective compared to dlPFC-only tDCS. For example,
the Wisconsin Card Sorting Test (WCST) is a well-documented measure of EF and primarily
measures cognitive flexibility. This task involves aspects of both, hot (e.g., task switching, disin-
hibition, WM) and cold EFs (i.e., executive control, inhibition) and results showed that only
when both, dlPFC and OFC were stimulated, cognitive flexibility improved. Therefore, the
common stimulation protocols in ADHD that primarily target dorsolateral frontostriatal net-
works may not be ideal for all executive dysfunction in ADHD, especially hot executive dys-
functions. This highlights the necessity of symptom-driven protocols implicating that for each
specific impairment (e.g., response inhibition, WM, selective attention, interference control,
cognitive flexibility, etc.) a specific stimulation montage that targets the most relevant cortical
region would be most appropriate.
Another important implication regarding the clinical efficacy of tDCS in ADHD concerns
stimulation parameters (i.e., intensity, duration, repetition rate, polarity, and site). The stimu-
lation intensities applied in the included studies ranged from 1 mA (seven experiments) to 2
mA (two experiments) and two experiments applied tDCS with 1.5 mA intensity. Findings
from other clinical populations showed that higher intensities of stimulation can result in
more prominent symptom improvement, for example in tinnitus [96], or cognitive
impairment in Parkinson’s disease [97]. This is in further accordance with physiological find-
ings of tDCS about larger tDCS after-effects in motor cortex plasticity as a result of tDCS with
higher intensities [98]. This was supported by the findings from two studies conducted on
adult ADHD. The study with 2 mA intensity reported significant improvement in inhibitory
control performance [68] while the study with 1 mA intensity reported no significant effects
on Go/No-Go task performance [71] despite the fact both studies targeted left dlPFC with
anodal tDCS. Stimulation intensity has, therefore, an impact on the effectiveness of tDCS,
which is not necessarily linear [99] and should be considered in clinical applications.
With regard to stimulation duration, the length of stimulation duration varied from 8 min
to 30 min and the longest duration of stimulation reported was five consecutive days of
30-min stimulation (Tables 1 & 2). The effects of different stimulation durations were not
addressed systematically in ADHD studies so far. Longer stimulation duration, similar to stim-
ulation intensity, could be an important factor to improve the efficacy of tDCS effects, includ-
ing clinical effects. Prolongation of stimulation duration for increasing efficacy of the
intervention works similarly well as enhancing stimulation intensity. One advantage of
enhancing stimulation duration compared to intensity might be that this does not increase the
probability of side effects like itching and tingling [33, 100], which might be specifically rele-
vant for studies in children. Findings from other neuropsychiatric disorders show that
repeated sessions of tDCS are more effective in reducing symptoms [33]. All studies included
in this meta-analysis except one [69] examined effects of tDCS after one single session. Higher

clinical efficacy might be achieved with a longer duration of stimulation, and repetitive inter-
ventions, as shown for example in non-invasive brain stimulation studies in depression [101–
103].
Stimulation polarity is another important parameter that determines the efficacy of tDCS.
The physiological mechanism of tDCS effects, depending on the polarity, is the induction of
LTP- and LTD-like plasticity [36, 104]. Anodal tDCS induces LTP-like plasticity, based on sub-
threshold depolarization effects on membrane potentials, and respective enhancement of spon-
taneous neuronal activity, while cathodal tDCS has antagonistic effects [105]. In accordance
with the pathophysiological foundation of ADHD that includes underactivation of the lateral
and inferior PFC, anodal tDCS-generated excitability enhancement is conceptually a more
promising approach in ADHD. The findings of this meta-analysis are in accordance with this
assumption and showed that anodal, but not cathodal tDCS improved inhibitory control and
WM in ADHD which could be attributed to the stimulation-induced compensation for regional
cortical hypoactivity as well as alteration of functional cortical network connectivity following
anodal tDCS that improves cognitive performance [106]. Nevertheless, this does not mean that
cathodal tDCS is not of clinical interest in ADHD. Functional abnormalities in ADHD are not
limited to hyperactivation of specific regions but also hyper-sensitiveness of other regions, espe-
cially those involved in motivational and emotional processing [11]. Moreover, findings from
cathodal stimulation of the motor cortex showed that motor cortex excitability alterations
induced by tDCS are intensity-dependent and nonlinear [99] which is not well studied for non-
motor functions. It might be that cathodal tDCS over regions pathologically hyperactive in the
disease might also be beneficial, which has not been systematically explored so far.
The cortical target area or stimulation site, another important parameter, contributes to the
clinical efficacy of tDCS in ADHD. The contribution of this parameter was specifically clear in
the studies which investigated tDCS effects on ADHD WM deficits. Only when the potentially
involved cortical region (left dlPFC in WM) was stimulated with the target electrode, both
WM accuracy and RT significantly improved and the effects were long-lasting. In these cases,
the reference electrode was placed over a non-contributing area (i.e., vertex). In contrast, when
the reference electrode targeted homologous regions of the contralateral hemisphere (i.e., right
dlPFC or right supraorbital) the effects were not present or reduced to RT improvement only.
A comparable pattern but in a different way was observed for tDCS effects on inhibitory con-
trol. Here, tDCS effects were more effective when two specific potentially involved regions
were stimulated by the target and reference electrodes. The left dlPFC-right OFC montage was
the most effective montage and all of the four experiments which used this protocol reported
improved performance. In contrast, the left dlPFC-right dlPFC tDCS montage, which was
used in two studies, did not improve any inhibitory control outcome measure. This demon-
strates that the target area or areas significantly contribute to the tDCS effectiveness in ADHD.
To conclude, results from this meta-analysis support the cognitive benefits of tDCS in
ADHD. However, for these cognitive benefits to be clinically useful, effects need to be sus-
tained for longer durations. All of the studies included in this meta-analysis applied tDCS in
an acute and short-term mode with no follow-up examination which does not allow to infer
long-term efficacy of tDCS in improving ADHD neuropsychological deficits. Similarly, mea-
suring inhibitory control and WM was done either during stimulation or right/shortly after
intervention which hinders any robust conclusions about long-term effects of tDCS. One spe-
cific important note here is the dissociation between neuropsychological deficits and clinical
symptoms of ADHD, which means improvement in inhibitory control and WM after or dur-
ing tDCS does not say too much about the improvement of clinical symptoms. Moreover,
applying statistically-proven blinding procedures, which was not reported in most of the
included studies in this meta-analysis, is important for making firm conclusions about the

clinical efficacy of tDCS. Therefore, the results of this meta-analysis should be interpreted with
caution, especially when it comes to the clinical efficacy of tDCS in ADHD. Nevertheless, we
need to note that the studies included in this meta-analysis are actually the first tDCS studies
on ADHD that primarily aimed to investigate whether tDCS has any beneficial effects on
ADHD symptoms or not. Despite short-term tDCS-induced cognitive benefits in ADHD, the
clinical efficacy of this technique needs further investigation in follow-up designs and in com-
parison to other interventions. Moreover, it requires stimulation protocols optimization using
the information about the individual neural activity associated with deficits and task execution
(individual level) and general stimulation parameters [49].
Safety of tDCS
There are unique issues concerning the safety, applicability, and ethics of tDCS application in
pediatric populations which is mainly due to limited available data from children compared to
the adult population [52]. However, the general safety of tDCS with standard protocols has
been proven by a large body of evidence in recent years and a recent review concluded that
application of conventional tDCS in human trials has not yet produced any reports of a Serious
Adverse Effect or irreversible injury [107]. In pediatric populations, no severe adverse events
have been reported, and even in children with epilepsy, seizures do not seem to worsen with
tDCS [52]. The most frequently reported side-effects within studies included in that systematic
review were headache, itchiness, and redness at the site of the stimulation. Similarly, no signifi-
cant side-effects were reported in the studies included in our meta-analysis. This is important
due to some concerns about the application of tDCS in children. One specific concern is about
current intensity due to children’s thinner skulls and the smaller distance between scalp and
brain. Computational models of current flow within the brain suggest that about 50% of the cur-
rent strength applied in adults result in respective effects in children [52] implicating that 0.5
mA applied in children results in similar physiological effects as 1 mA in adults [108]. From the
studies included in this meta-analysis, eight studies applied a current intensity of 1 mA or
lower, two studies applied 1.5 mA and one study applied 2 mA, and no significant adverse effect
was reported (See Tables 1 & 2). However, the caveat still stands that studies included in this
meta-analysis were not designed as safety studies and mostly included single session tDCS inter-
ventions. Therefore, systematic safety monitoring, especially for the clinical application of
tDCS, is recommended. Moreover, it is important to keep in mind that the developing brain
has “sensitive” or “critical” periods where the effects of interventions affecting the brain could
be stronger than usual. This suggests that the risk to induce maladaptive neural plasticity due to
tDCS might be high which necessitates the priority of dose-finding studies and longitudinal
monitoring of tDCS-induced neuroplasticity in pediatric ADHD population [49].
Limitations and future directions

Despite promising results and novelty, this meta-analysis has some limitations to be consid-
ered. First, despite that our meta-analysis found a significant effect of tDCS on neuropsycho-
logical symptoms (namely inhibitory control and WM) in ADHD populations, the efficacy of
this method on other executive dysfunctions, and clinical symptoms cannot be directly derived
from these results. Inhibitory control and WM were chosen because they are two common def-
icits examined in most ADHD-tDCS studies which allowed us to run a meta-analytic study.
Secondly, although we showed that anodal tDCS over the dlPFC has significant effects on IC
in ADHD, no effects were seen for cathodal tDCS and rIFG tDCS. However, given the small
number of studies investigating cathodal/rIFG tDCS, these results have to be interpreted with
caution. With increasing interest in the application of tDCS in ADHD, future research might

be able to disentangle subtle protocol differences for ADHD subtypes. Thirdly, adaptation of
stimulation protocols based on symptom subtypes, ADHD subgroups, specific cognitive defi-
cits, and neuroanatomical differences is lacking in currently available studies and therefore, it
is difficult to rate the potential of tDCS in ADHD.
Future tDCS studies should systematically investigate the following lines of research, to
reveal the real clinical potential of tDCS for ADHD treatment. First, preliminary data suggest
beneficial effects of tDCS on other cognitive dysfunctions in ADHD including cognitive flexi-
bility, problem-solving, selective attention, and hyperactivity symptoms involved in ADHS [6,
70, 109]. Second, based on the findings of this meta-analysis, future tDCS studies on ADHD
populations, with a statistically-proven blinding procedure (especially double-blinded trials),
are recommended to target the stimulation area based on the symptoms subtypes. The above-
mentioned stimulation parameters are other important factors that might affect the efficacy of
the intervention and should be explored systematically. Lastly, future studies, especially those
with clinical implications, require examining the long-term efficacy of tDCS in ADHD. With
the expected increasing number of tDCS studies in ADHD populations, future meta-analytic
studies might be able to deliver a more realistic picture about the effectiveness of tDCS in dif-
ferent subtypes of ADHD. It is lastly of note that other modalities of electrical stimulation such
as transcranial alternating current stimulation (tACS), which has been shown to have neuro-
plastic effects beyond its impact on brain oscillations [110], or transcranial random noise stim-
ulation (tRNS) can be a potentially interesting avenue for future research in ADHD, which is
associated with abnormal brain oscillations [24].
Conclusion
The findings of this meta-analysis of tDCS interventions in ADHD suggest an improvement of
neuropsychological deficits (i.e., inhibitory control and WM) by tDCS. Stimulation polarity
and target area are relevant for the efficacy of tDCS in ADHD. Anodal dlPFC tDCS had a sig-
nificantly superior effect on inhibitory control compared to cathodal/sham stimulation and
anodal rIFG tDCS. TDCS significantly increased response accuracy of inhibitory control per-
formance and decreased response time in WM tasks. Although our findings suggest improving
effects of tDCS in ADHD neuropsychological deficits, the clinical utility of tDCS cannot be
firmly rated with the currently available findings. Application of this method as a therapeutic
intervention will require optimizing stimulation protocols based on general stimulation
parameters and individual and inter-individual factors for improvement of clinical efficacy,
exploration of clinical symptoms in addition to surrogate parameters, and achievement of sus-
tained clinical benefits by tDCS over longer durations of time. Thus, future research is needed
to more thoroughly explore and refine optimal stimulation parameters required for tDCS-
based cognitive improvement and implementing robust experimental designs in different
ADHD subtypes. Broadly speaking, the potential for tDCS as a non-invasive brain stimulation
technique to safely improve neuroplasticity and treat neurological and neurodevelopmental
disorders is encouraging. Future studies utilizing tDCS will further increase our understanding
of neural networks and how to treat their pathological states in ADHD and other neurodeve-
lopmental disorders including autism and learning disabilities.
Supporting information
S1 Table. The Preferred Reporting Items for Systematic reviews and Meta-Analyses
(PRISMA) 2009 checklist.
(DOC)

Acknowledgments
Publication of this article was funded by the Open Access Fund of the Leibniz Association.
Author Contributions
Conceptualization: Mohammad Ali Salehinejad.
Formal analysis: Miles Wischnewski.
Funding acquisition: Michael A. Nitsche.
Methodology: Mohammad Ali Salehinejad, Miles Wischnewski.
Supervision: Vahid Nejati, Carmelo M. Vicario, Michael A. Nitsche.
Visualization: Miles Wischnewski.
Writing – original draft: Mohammad Ali Salehinejad, Miles Wischnewski.
Writing – review & editing: Mohammad Ali Salehinejad, Michael A. Nitsche.
References
1. Visser SN, Danielson ML, Bitsko RH, Holbrook JR, Kogan MD, Ghandour RM, et al. Trends in the Par-
ent-Report of Health Care Provider-Diagnosed and Medicated Attention-Deficit/Hyperactivity Disor-
der: United States, 2003–2011. Journal of the American Academy of Child & Adolescent Psychiatry.
2014; 53(1):34–46.e2. https://doi.org/10.1016/j.jaac.2013.09.001 PMID: 24342384
2. American Psychiatric Association. Diagnostic and statistical manual of mental disorders (DSM-5):
American Psychiatric Pub; 2013.
3. Castellanos FX, Sonuga-Barke EJS, Milham MP, Tannock R. Characterizing cognition in ADHD:
beyond executive dysfunction. Trends in Cognitive Sciences. 2006; 10(3):117–23. https://doi.org/10.
1016/j.tics.2006.01.011 PMID: 16460990
4. Hudec KL, Alderson RM, Patros CHG, Lea SE, Tarle SJ, Kasper LJ. Hyperactivity in boys with atten-
tion-deficit/hyperactivity disorder (ADHD): The role of executive and non-executive functions.
Research in Developmental Disabilities. 2015;45–46:103–9.http://dx.doi.org/10.1016/j.ridd.2015.07.
012.
5. Cortese S, Ferrin M, Brandeis D, Buitelaar J, Daley D, Dittmann RW, et al. Cognitive Training for Atten-
tion-Deficit/Hyperactivity Disorder: Meta-Analysis of Clinical and Neuropsychological Outcomes From
Randomized Controlled Trials. Journal of the American Academy of Child & Adolescent Psychiatry.
2015; 54(3):164–74.https://doi.org/10.1016/j.jaac.2014.12.010.
6. Nejati V, Salehinejad MA, Nitsche MA, Najian A, Javadi A-H. Transcranial Direct Current Stimulation
Improves Executive Dysfunctions in ADHD: Implications for Inhibitory Control, Interference Control,
Working Memory, and Cognitive Flexibility. Journal of Attention Disorders. 2017;0
(0):1087054717730611. https://doi.org/10.1177/1087054717730611 PMID: 28938852.
7. Kasper LJ, Alderson RM, Hudec KL. Moderators of working memory deficits in children with attention-
deficit/hyperactivity disorder (ADHD): A meta-analytic review. Clinical Psychology Review. 2012; 32
(7):605–17. https://doi.org/10.1016/j.cpr.2012.07.001 PMID: 22917740
8. Willcutt EG, Doyle AE, Nigg JT, Faraone SV, Pennington BF. Validity of the Executive Function Theory
of Attention-Deficit/Hyperactivity Disorder: A Meta-Analytic Review. Biological Psychiatry. 2005; 57
(11):1336–46. https://doi.org/10.1016/j.biopsych.2005.02.006 PMID: 15950006
9. Roy A, Oldehinkel AJ, Hartman CA. Cognitive Functioning in Adolescents with Self-Reported ADHD
and Depression: Results from a Population-Based Study. Journal of Abnormal Child Psychology.
2017; 45(1):69–81. https://doi.org/10.1007/s10802-016-0160-x PMID: 27138748
10. Roberts M, Ashinoff BK, Castellanos FX, Carrasco M. When attention is intact in adults with ADHD.
Psychonomic Bulletin & Review. 2017. https://doi.org/10.3758/s13423-017-1407-4 PMID: 29181782
11. Rubia K. Cognitive Neuroscience of Attention Deficit Hyperactivity Disorder (ADHD) and Its Clinical
Translation. Frontiers in Human Neuroscience. 2018; 12(100). https://doi.org/10.3389/fnhum.2018.
00100 PMID: 29651240
12. Fair DA, Bathula D, Nikolas MA, Nigg JT. Distinct neuropsychological subgroups in typically develop-
ing youth inform heterogeneity in children with ADHD. Proceedings of the National Academy of Sci-
ences. 2012; 109(17):6769–74.

13. Samea F, Soluki S, Nejati V, Zarei M, Cortese S, Eickhoff SB, et al. Brain alterations in children/ado-
lescents with ADHD revisited: A neuroimaging meta-analysis of 96 structural and functional studies.
Neuroscience & Biobehavioral Reviews. 2019; 100:1–8.https://doi.org/10.1016/j.neubiorev.2019.02.
011.
14. Nigg JT. Neuropsychologic Theory and Findings in Attention-Deficit/Hyperactivity Disorder: The State
of the Field and Salient Challenges for the Coming Decade. Biological Psychiatry. 2005; 57(11):1424–
35. https://doi.org/10.1016/j.biopsych.2004.11.011 PMID: 15950017
15. Rubio B, Boes AD, Laganiere S, Rotenberg A, Jeurissen D, Pascual-Leone A. Noninvasive Brain Stim-
ulation in Pediatric Attention-Deficit Hyperactivity Disorder (ADHD):A Review. Journal of Child Neurol-
ogy. 2016; 31(6):784–96. https://doi.org/10.1177/0883073815615672 PMID: 26661481.
16. Zhu Y, Jiang X, Ji W. The Mechanism of Cortico-Striato-Thalamo-Cortical Neurocircuitry in Response
Inhibition and Emotional Responding in Attention Deficit Hyperactivity Disorder with Comorbid Disrup-
tive Behavior Disorder. Neurosci Bull. 2018. https://doi.org/10.1007/s12264-018-0214-x PMID:
29508250
17. Seidman LJ, Valera EM, Makris N. Structural Brain Imaging of Attention-Deficit/Hyperactivity Disorder.
Biological Psychiatry. 2005; 57(11):1263–72. https://doi.org/10.1016/j.biopsych.2004.11.019 PMID:
15949998
18. Depue BE, Burgess GC, Willcutt EG, Ruzic L, Banich MT. Inhibitory control of memory retrieval and
motor processing associated with the right lateral prefrontal cortex: Evidence from deficits in individu-
als with ADHD. Neuropsychologia. 2010; 48(13):3909–17. https://doi.org/10.1016/j.neuropsychologia.
2010.09.013 PMID: 20863843
19. Belen R, Aaron DB, Simon L, Alexander R, Danique J, Alvaro P-L. Noninvasive Brain Stimulation in
Pediatric Attention-Deficit Hyperactivity Disorder (ADHD): A Review. Journal of Child Neurology.
2015; 31(6):784–96. https://doi.org/10.1177/0883073815615672 PMID: 26661481
20. Langleben DD, Austin G, Krikorian G, Ridlehuber HW, Goris ML, Strauss HW. Interhemispheric asym-
metry of regional cerebral blood flow in prepubescent boys with attention deficit hyperactivity disorder.
Nuclear Medicine Communications. 2001; 22(12):1333–40. 00006231-200112000-00009. PMID:
11711904
21. Christakou A, Murphy CM, Chantiluke K, Cubillo AI, Smith AB, Giampietro V, et al. Disorder-specific
functional abnormalities during sustained attention in youth with Attention Deficit Hyperactivity Disor-
der (ADHD) and with Autism. Mol Psychiatry. 2012; 18:236. https://doi.org/10.1038/mp.2011.185
https://www.nature.com/articles/mp2011185#supplementary-information. PMID: 22290121
22. Bush G, Valera EM, Seidman LJ. Functional Neuroimaging of Attention-Deficit/Hyperactivity Disorder:
A Review and Suggested Future Directions. Biological Psychiatry. 2005; 57(11):1273–84. https://doi.
org/10.1016/j.biopsych.2005.01.034 PMID: 15949999
23. Cai W, Chen T, Szegletes L, Supekar K, Menon V. Aberrant Time-Varying Cross-Network Interactions
in Children With Attention-Deficit/Hyperactivity Disorder and the Relation to Attention Deficits. Biologi-
cal Psychiatry: Cognitive Neuroscience and Neuroimaging. 2018; 3(3):263–73.https://doi.org/10.
1016/j.bpsc.2017.10.005.
24. Marcos-Vidal L, Martı́nez-Garcı́a M, Pretus C, Garcia-Garcia D, Martı́nez K, Janssen J, et al. Local
functional connectivity suggests functional immaturity in children with attention-deficit/hyperactivity
disorder. Hum Brain Mapp. 2018:n/a-n/a. https://doi.org/10.1002/hbm.24013 PMID: 29473262
25. Hoogman M, Bralten J, Hibar DP, Mennes M, Zwiers MP, Schweren LSJ, et al. Subcortical brain vol-
ume differences in participants with attention deficit hyperactivity disorder in children and adults: a
cross-sectional mega-analysis. The Lancet Psychiatry. 2017; 4(4):310–9. https://doi.org/10.1016/
S2215-0366(17)30049-4 PMID: 28219628
26. Sibley MH, Kuriyan AB, Evans SW, Waxmonsky JG, Smith BH. Pharmacological and psychosocial
treatments for adolescents with ADHD: An updated systematic review of the literature. Clinical Psy-
chology Review. 2014; 34(3):218–32. https://doi.org/10.1016/j.cpr.2014.02.001 PMID: 24632046
27. Catalá-López F, Hutton B, Núñez-Beltrán A, Mayhew AD, Page MJ, Ridao M, et al. The pharmacologi-
cal and non-pharmacological treatment of attention deficit hyperactivity disorder in children and ado-
lescents: protocol for a systematic review and network meta-analysis of randomized controlled trials.
Systematic Reviews. 2015; 4(1):19. https://doi.org/10.1186/s13643-015-0005-7 PMID: 25875125
28. Cortese S, Ferrin M, Brandeis D, Holtmann M, Aggensteiner P, Daley D, et al. Neurofeedback for
Attention-Deficit/Hyperactivity Disorder: Meta-Analysis of Clinical and Neuropsychological Outcomes
From Randomized Controlled Trials. Journal of the American Academy of Child & Adolescent Psychia-
try. 2016; 55(6):444–55.https://doi.org/10.1016/j.jaac.2016.03.007.
29. Molina BSG, Hinshaw SP, Swanson JM, Arnold LE, Vitiello B, Jensen PS, et al. The MTA at 8 Years:
Prospective Follow-up of Children Treated for Combined-Type ADHD in a Multisite Study. Journal of

the American Academy of Child & Adolescent Psychiatry. 2009; 48(5):484–500.https://doi.org/10.

1097/CHI.0b013e31819c23d0.
30. Faraone SV, Biederman J, Spencer TJ, Aleardi M. Comparing the Efficacy of Medications for ADHD
Using Meta-analysis. Medscape General Medicine. 2006; 8(4):4–. PMC1868385. PMID: 17415287
31. Hsu S, Piedra AN, Froehlich TE, Brinkman WB, Epstein J. 6.18 Caregiver-Reported Side Effects of
Methylphenidate and Comorbid Oppositional Defiant Disorder Diagnosis in Children With Attention-
Deficit/Hyperactivity Disorder (ADHD). Journal of the American Academy of Child & Adolescent Psy-
chiatry. 2017; 56(10, Supplement):S282.https://doi.org/10.1016/j.jaac.2017.09.363.
32. Daley D, van der Oord S, Ferrin M, Danckaerts M, Doepfner M, Cortese S, et al. Behavioral Interven-
tions in Attention-Deficit/Hyperactivity Disorder: A Meta-Analysis of Randomized Controlled Trials
Across Multiple Outcome Domains. Journal of the American Academy of Child & Adolescent Psychia-
try. 2014; 53(8):835–47.e5.https://doi.org/10.1016/j.jaac.2014.05.013.
33. Lefaucheur J-P, Antal A, Ayache SS, Benninger DH, Brunelin J, Cogiamanian F, et al. Evidence-
based guidelines on the therapeutic use of transcranial direct current stimulation (tDCS). Clinical
Neurophysiology. 2017; 128(1):56–92. https://doi.org/10.1016/j.clinph.2016.10.087 PMID: 27866120
34. Vicario CM, Salehinejad MA, Felmingham K, Martino G, Nitsche MA. A systematic review on the thera-
peutic effectiveness of non-invasive brain stimulation for the treatment of anxiety disorders. Neurosci-
ence & Biobehavioral Reviews. 2019; 96:219–31.https://doi.org/10.1016/j.neubiorev.2018.12.012.
35. Harty S, Brem A-K, Cohen Kadosh R. Neurocognitive Effects of tDCS in the Healthy Brain. In: Brunoni
A, Nitsche M, Loo C, editors. Transcranial Direct Current Stimulation in Neuropsychiatric Disorders:
Clinical Principles and Management. Cham: Springer International Publishing; 2016. p. 103–41.
36. Nitsche MA, Paulus W. Excitability changes induced in the human motor cortex by weak transcranial
direct current stimulation. The Journal of Physiology. 2000; 527(3):633–9. https://doi.org/10.1111/j.
1469-7793.2000.t01-1-00633.x PMID: 10990547
37. Kuo M-F, Polanı́a R, Nitsche M. Physiology of Transcranial Direct and Alternating Current Stimulation.
In: Brunoni A, Nitsche M, Loo C, editors. Transcranial Direct Current Stimulation in Neuropsychiatric
Disorders: Clinical Principles and Management. Cham: Springer International Publishing; 2016. p.
29–46.
38. Nitsche Boggio, Fregni Pascual-Leone. Treatment of depression with transcranial direct current stimu-
lation (tDCS): a review. Experimental neurology. 2009; 219(1):14–9. https://doi.org/10.1016/j.
expneurol.2009.03.038 PMID: 19348793
39. Nitsche MA, Cohen LG, Wassermann EM, Priori A, Lang N, Antal A, et al. Transcranial direct current
stimulation: State of the art 2008. Brain Stimulation. 2008; 1(3):206–23. https://doi.org/10.1016/j.brs.
2008.06.004 PMID: 20633386
40. Woods AJ, Martin DM. Clinical Research and Methodological Aspects for tDCS Research. In: Brunoni
A, Nitsche M, Loo C, editors. Transcranial Direct Current Stimulation in Neuropsychiatric Disorders:
Clinical Principles and Management. Cham: Springer International Publishing; 2016. p. 393–404.
41. Ghanavati E, Salehinejad MA, Nejati V, Nitsche MA. Differential role of prefrontal, temporal and parie-
tal cortices in verbal and figural fluency: Implications for the supramodal contribution of executive func-
tions. Sci Rep. 2019; 9(1):3700. https://doi.org/10.1038/s41598-019-40273-7 PMID: 30842493
42. Nejati V, Salehinejad MA, Nitsche MA. Interaction of the Left Dorsolateral Prefrontal Cortex (l-DLPFC)
and Right Orbitofrontal Cortex (OFC) in Hot and Cold Executive Functions: Evidence from Transcra-
nial Direct Current Stimulation (tDCS). Neuroscience. 2018; 369(Supplement C):109–23.https://doi.
org/10.1016/j.neuroscience.2017.10.042.
43. Kuo M-F, Nitsche MA. Effects of Transcranial Electrical Stimulation on Cognition. Clin EEG Neurosci.
2012; 43(3):192–9. https://doi.org/10.1177/1550059412444975 PMID: 22956647.
44. Ghanavati E, Nejati V, Salehinejad MA. Transcranial Direct Current Stimulation over the Posterior
Parietal Cortex (PPC) Enhances Figural Fluency: Implications for Creative Cognition. Journal of Cog-
nitive Enhancement. 2018; 2(1):88–96. https://doi.org/10.1007/s41465-017-0059-7
45. Lefaucheur J-P. A comprehensive database of published tDCS clinical trials (2005–2016). Neurophy-
siologie Clinique/Clinical Neurophysiology. 2016; 46(6):319–98. https://doi.org/10.1016/j.neucli.2016.
10.002 PMID: 27865707
46. Filmer HL, Dux PE, Mattingley JB. Applications of transcranial direct current stimulation for under-
standing brain function. Trends in Neurosciences. 2014; 37(12):742–53. https://doi.org/10.1016/j.tins.
2014.08.003 PMID: 25189102
47. Shah-Basak PP, Hamilton RH, Nitsche MA, Woods AJ. Transcranial Direct Current Stimulation in
Cognitive Neuroscience. In: Knotkova H, Nitsche MA, Bikson M, Woods AJ, editors. Practical Guide to
Transcranial Direct Current Stimulation: Principles, Procedures and Applications. Cham: Springer
International Publishing; 2019. p. 597–625.

48. Vicario C, Nitsche M. Transcranial direct current stimulation: a remediation tool for the treatment of
childhood congenital dyslexia? Frontiers in Human Neuroscience. 2013; 7(139). https://doi.org/10.
3389/fnhum.2013.00139 PMID: 23626530
49. Vicario CM, Nitsche MA. Chapter 9—tDCS in Pediatric Neuropsychiatric Disorders. In: Oberman LM,
Enticott PG, editors. Neurotechnology and Brain Stimulation in Pediatric Psychiatric and Neurodeve-
lopmental Disorders: Academic Press; 2019. p. 217–35.
50. Ciechanski P, Kirton A. Chapter 5—Transcranial Direct-Current Stimulation (tDCS): Principles and
Emerging Applications in Children. Pediatric Brain Stimulation. Oxford: Academic Press; 2016. p.
85–115.
51. Stortelder F, Ploegmakers-Burg M. Adolescence and the Reorganization of Infant Development: A
Neuro-Psychoanalytic Model. The Journal of the American Academy of Psychoanalysis and Dynamic
Psychiatry. 2010; 38(3):503–31. https://doi.org/10.1521/jaap.2010.38.3.503 PMID: 20849240
52. El-Hagrassy MM, Jones F, Rosa G, Fregni F. CNS Non-invasive Brain Stimulation. In: Gilleran JP,
Alpert SA, editors. Adult and Pediatric Neuromodulation. Cham: Springer International Publishing;
2018. p. 151–84.
53. Sergeant JA, Geurts H, Huijbregts S, Scheres A, Oosterlaan J. The top and the bottom of ADHD: a
neuropsychological perspective. Neuroscience & Biobehavioral Reviews. 2003; 27(7):583–92.https://
doi.org/10.1016/j.neubiorev.2003.08.004.
54. Nigg JT, Gustafsson HC, Karalunas SL, Ryabinin P, McWeeney SK, Faraone SV, et al. Working Mem-
ory and Vigilance as Multivariate Endophenotypes Related to Common Genetic Risk for Attention-Def-
icit/Hyperactivity Disorder. Journal of the American Academy of Child & Adolescent Psychiatry. 2018;
57(3):175–82.https://doi.org/10.1016/j.jaac.2017.12.013.
55. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-
analyses: The PRISMA statement. International Journal of Surgery. 2010; 8(5):336–41. https://doi.
org/10.1016/j.ijsu.2010.02.007 PMID: 20171303
56. Sonuga-Barke EJS. Causal Models of Attention-Deficit/Hyperactivity Disorder: From Common Simple
Deficits to Multiple Developmental Pathways. Biological Psychiatry. 2005; 57(11):1231–8. https://doi.
org/10.1016/j.biopsych.2004.09.008 PMID: 15949993
57. Castellanos FX, Proal E. Large-scale brain systems in ADHD: beyond the prefrontal–striatal model.
Trends in Cognitive Sciences. 2012; 16(1):17–26. https://doi.org/10.1016/j.tics.2011.11.007 PMID:
22169776
58. Georgiou G, Essau CA. Go/No-Go Task. In: Goldstein S, Naglieri JA, editors. Encyclopedia of Child
Behavior and Development. Boston, MA: Springer US; 2011. p. 705–6.
59. Stuss DT, Floden D, Alexander MP, Levine B, Katz D. Stroop performance in focal lesion patients: dis-
sociation of processes and frontal lobe lesion location. Neuropsychologia. 2001; 39(8):771–86.https://
doi.org/10.1016/S0028-3932(01)00013-6. PMID: 11369401
60. Mullane JC, Corkum PV, Klein RM, McLaughlin E. Interference Control in Children with and without
ADHD: A Systematic Review of Flanker and Simon Task Performance. Child Neuropsychology. 2009;
15(4):321–42. https://doi.org/10.1080/09297040802348028 PMID: 18850349
61. Conners CK, Staff M, Connelly V, Campbell S, MacLean M, Barnes J. Conners’ continuous perfor-
mance Test II (CPT II v. 5). Multi-Health Syst Inc. 2000; 29:175–96.
62. Brooks BL, Sherman EMS, Strauss E. NEPSY-II: A Developmental Neuropsychological Assessment,
Second Edition. Child Neuropsychology. 2009; 16(1):80–101. https://doi.org/10.1080/
09297040903146966
63. Kessels RPC, van Zandvoort MJE, Postma A, Kappelle LJ, de Haan EHF. The Corsi Block-Tapping
Task: Standardization and Normative Data. Appl Neuropsychol. 2000; 7(4):252–8. https://doi.org/10.
1207/S15324826AN0704_8 PMID: 11296689
64. Higgins JPT, Altman DG, Gøtzsche PC, Jüni P, Moher D, Oxman AD, et al. The Cochrane Collabora-
tion’s tool for assessing risk of bias in randomised trials. BMJ. 2011; 343. https://doi.org/10.1136/bmj.
d5928 PMID: 22008217
65. Larry H, Ingram O. Statistical Methods for Meta Analysis. Orlando, FL: Academic Press; 1985.
66. Rosenthal R. The file drawer problem and tolerance for null results. Psychological Bulletin. 1979; 86
(3):638–41. https://doi.org/10.1037/0033-2909.86.3.638
67. Rosenberg MS, Adams DC, Gurevitch J. MetaWin: statistical software for meta-analysis. Sunderland,
MA: Sinauer Associates Sunderland, MA; 2000.
68. Allenby C, Falcone M, Bernardo L, Wileyto P, Rostain A, Ramsay JR, et al. Transcranial Direct Current
Brain Stimulation Decreases Impulsivity in ADHD. Brain Stimulation. 2018.https://doi.org/10.1016/j.
brs.2018.04.016.

69. Bandeira ID, Guimarães RSQ, Jagersbacher JG, Barretto TL, Jesus-Silva JRd, Santos SN, et al.
Transcranial Direct Current Stimulation in Children and Adolescents With Attention-Deficit/Hyperactiv-
ity Disorder (ADHD):A Pilot Study. Journal of Child Neurology. 2016; 31(7):918–24. https://doi.org/10.
1177/0883073816630083 PMID: 26879095.
70. Breitling C, Zaehle T, Dannhauer M, Bonath B, Tegelbeckers J, Flechtner H-H, et al. Improving Inter-
ference Control in ADHD Patients with Transcranial Direct Current Stimulation (tDCS). Front Cell Neu-
rosci. 2016; 10:72. https://doi.org/10.3389/fncel.2016.00072 PMC4834583. PMID: 27147964
71. Cosmo C, Baptista AF, de Araújo AN, do Rosário RS, Miranda JGV, Montoya P, et al. A Randomized,
Double-Blind, Sham-Controlled Trial of Transcranial Direct Current Stimulation in Attention-Deficit/
Hyperactivity Disorder. PLOS ONE. 2015; 10(8):e0135371. https://doi.org/10.1371/journal.pone.
0135371 PMID: 26267861
72. Munz MT, Prehn-Kristensen A, Thielking F, Mölle M, Göder R, Baving L. Slow oscillating transcranial
direct current stimulation during non-rapid eye movement sleep improves behavioral inhibition in atten-
tion-deficit/hyperactivity disorder. Front Cell Neurosci. 2015; 9:307. https://doi.org/10.3389/fncel.2015.
00307 PMC4531340. PMID: 26321911
73. Soltaninejad Z, Nejati V, Ekhtiari H. Effect of Anodal and Cathodal Transcranial Direct Current Stimula-
tion on DLPFC on Modulation of Inhibitory Control in ADHD. Journal of Attention Disorders. 2015.
https://doi.org/10.1177/1087054715618792 PMID: 26689935
74. Sotnikova A, Soff C, Tagliazucchi E, Becker K, Siniatchkin M. Transcranial Direct Current Stimulation
Modulates Neuronal Networks in Attention Deficit Hyperactivity Disorder. Brain Topogr. 2017; 30
(5):656–72. https://doi.org/10.1007/s10548-017-0552-4 PMID: 28213645
75. Prehn-Kristensen A, Munz M, Göder R, Wilhelm I, Korr K, Vahl W, et al. Transcranial Oscillatory Direct
Current Stimulation During Sleep Improves Declarative Memory Consolidation in Children With Atten-
tion-deficit/hyperactivity Disorder to a Level Comparable to Healthy Controls. Brain Stimulation. 2014;
7(6):793–9. https://doi.org/10.1016/j.brs.2014.07.036 PMID: 25153776
76. Soff C, Sotnikova A, Christiansen H, Becker K, Siniatchkin M. Transcranial direct current stimulation
improves clinical symptoms in adolescents with attention deficit hyperactivity disorder. J Neural
Transm. 2017; 124(1):133–44. https://doi.org/10.1007/s00702-016-1646-y PMID: 27853926
77. Cosmo C, Ferreira C, Miranda JG, do Rosário R, Baptista A, Montoya P, et al. Spreading effect of
tDCS in individuals with attention-deficit/hyperactivity disorder as shown by functional cortical net-
works: a randomized, double-blind, sham-controlled trial. Frontiers in Psychiatry. 2015; 6(111). https://
doi.org/10.3389/fpsyt.2015.00111 PMID: 26300790
78. Brunoni AR, Vanderhasselt M-A. Working memory improvement with non-invasive brain stimulation of
the dorsolateral prefrontal cortex: A systematic review and meta-analysis. Brain and Cognition. 2014;
86:1–9. https://doi.org/10.1016/j.bandc.2014.01.008 PMID: 24514153
79. Price AR, McAdams H, Grossman M, Hamilton RH. A Meta-analysis of Transcranial Direct Current
Stimulation Studies Examining the Reliability of Effects on Language Measures. Brain Stimulation.
2015; 8(6):1093–100. https://doi.org/10.1016/j.brs.2015.06.013 PMID: 26210573
80. Dedoncker J, Brunoni AR, Baeken C, Vanderhasselt M-A. The effect of the interval-between-sessions
on prefrontal transcranial direct current stimulation (tDCS) on cognitive outcomes: a systematic review
and meta-analysis. J Neural Transm. 2016; 123(10):1159–72. https://doi.org/10.1007/s00702-016-
1558-x PMID: 27145765
81. Kalu UG, Sexton CE, Loo CK, Ebmeier KP. Transcranial direct current stimulation in the treatment of
major depression: a meta-analysis. Psychological Medicine. 2012; 42(9):1791–800. Epub 2012/01/12.
https://doi.org/10.1017/S0033291711003059 PMID: 22236735
82. Bastani A, Jaberzadeh S. Does anodal transcranial direct current stimulation enhance excitability of
the motor cortex and motor function in healthy individuals and subjects with stroke: A systematic
review and meta-analysis. Clinical Neurophysiology. 2012; 123(4):644–57. https://doi.org/10.1016/j.
clinph.2011.08.029 PMID: 21978654
83. Shiozawa P, Fregni F, Benseñor IM, Lotufo PA, Berlim MT, Daskalakis JZ, et al. Transcranial direct
current stimulation for major depression: an updated systematic review and meta-analysis. Interna-
tional Journal of Neuropsychopharmacology. 2014; 17(9):1443–52. https://doi.org/10.1017/
S1461145714000418 PMID: 24713139
84. Hill AT, Fitzgerald PB, Hoy KE. Effects of Anodal Transcranial Direct Current Stimulation on Working
Memory: A Systematic Review and Meta-Analysis of Findings From Healthy and Neuropsychiatric
Populations. Brain Stimulation. 2016; 9(2):197–208. https://doi.org/10.1016/j.brs.2015.10.006 PMID:
26597929
85. Diamond A. Executive Functions. Annual Review of Psychology. 2013; 64:135–68. https://doi.org/10.
1146/annurev-psych-113011-143750 PMC4084861. PMID: 23020641

86. Elliott R. Executive functions and their disordersImaging in clinical neuroscience. British Medical Bulle-
tin. 2003; 65(1):49–59. https://doi.org/10.1093/bmb/65.1.49
87. Hoekzema E, Carmona S, Ramos-Quiroga JA, Fernández VR, Bosch R, Soliva JC, et al. An indepen-
dent components and functional connectivity analysis of resting state fMRI data points to neural net-
work dysregulation in adult ADHD. Hum Brain Mapp. 2014; 35(4):1261–72. https://doi.org/10.1002/
hbm.22250 PMID: 23417778
88. Lei D, Du M, Wu M, Chen T, Huang X, Du X, et al. Functional MRI reveals different response inhibition
between adults and children with ADHD. Neuropsychology. 2015; 29(6):874–81. https://doi.org/10.
1037/neu0000200 PMID: 25938917
89. McCarthy H, Skokauskas N, Frodl T. Identifying a consistent pattern of neural function in attention defi-
cit hyperactivity disorder: a meta-analysis. Psychological Medicine. 2014; 44(4):869–80. Epub 2013/
05/13. https://doi.org/10.1017/S0033291713001037 PMID: 23663382
90. Cepeda NJ, Cepeda ML, Kramer AF. Task Switching and Attention Deficit Hyperactivity Disorder.
Journal of Abnormal Child Psychology. 2000; 28(3):213–26. https://doi.org/10.1023/
a:1005143419092 PMID: 10885680
91. Aron AR, Poldrack RA. Cortical and Subcortical Contributions to Stop Signal Response Inhibition:
Role of the Subthalamic Nucleus. The Journal of Neuroscience. 2006; 26(9):2424. https://doi.org/10.
1523/JNEUROSCI.4682-05.2006 PMID: 16510720
92. Miller EK, Cohen JD. An Integrative Theory of Prefrontal Cortex Function. Annual Review of Neurosci-
ence. 2001; 24(1):167–202. https://doi.org/10.1146/annurev.neuro.24.1.167 PMID: 11283309.
93. Mario Vicario C, Martino D. The neurophysiology of magnitude: One example of extraction analogies.
Cognitive Neuroscience. 2010; 1(2):144–5. https://doi.org/10.1080/17588921003763969 PMID:
24168285
94. Dedoncker J, Brunoni AR, Baeken C, Vanderhasselt M-A. A Systematic Review and Meta-Analysis of
the Effects of Transcranial Direct Current Stimulation (tDCS) Over the Dorsolateral Prefrontal Cortex
in Healthy and Neuropsychiatric Samples: Influence of Stimulation Parameters. Brain Stimulation.
95. Datta A. Inter-Individual Variation during Transcranial Direct Current Stimulation and Normalization of
Dose Using MRI-Derived Computational Models. Frontiers in Psychiatry. 2012; 3(91). https://doi.org/
10.3389/fpsyt.2012.00091 PMID: 23097644
96. Shekhawat GS, Sundram F, Bikson M, Truong D, De Ridder D, Stinear CM, et al. Intensity, Duration,
and Location of High-Definition Transcranial Direct Current Stimulation for Tinnitus Relief. Neuroreha-
bil Neural Repair. 2016; 30(4):349–59. https://doi.org/10.1177/1545968315595286 PMID: 26180052.
97. Boggio PS, Ferrucci R, Rigonatti SP, Covre P, Nitsche M, Pascual-Leone A, et al. Effects of transcra-
nial direct current stimulation on working memory in patients with Parkinson’s disease. J Neurol Sci.
2006; 249(1):31–8. https://doi.org/10.1016/j.jns.2006.05.062 PMID: 16843494
98. Jamil A, Batsikadze G, Kuo H-I, Labruna L, Hasan A, Paulus W, et al. Systematic evaluation of the
impact of stimulation intensity on neuroplastic after-effects induced by transcranial direct current stim-
ulation. The Journal of Physiology. 2017; 595(4):1273–88. https://doi.org/10.1113/JP272738 PMID:
27723104
99. Batsikadze G, Moliadze V, Paulus W, Kuo MF, Nitsche MA. Partially non-linear stimulation intensity-
dependent effects of direct current stimulation on motor cortex excitability in humans. The Journal of
Physiology. 2013; 591(7):1987–2000. https://doi.org/10.1113/jphysiol.2012.249730 PMID: 23339180
100. Brunoni AR, Nitsche MA, Bolognini N, Bikson M, Wagner T, Merabet L, et al. Clinical research with
transcranial direct current stimulation (tDCS): Challenges and future directions. Brain Stimulation.
101. Salehinejad MA, Ghanavai E, Rostami R, Nejati V. Cognitive control dysfunction in emotion dysregula-
tion and psychopathology of major depression (MD): Evidence from transcranial brain stimulation of
the dorsolateral prefrontal cortex (DLPFC). Journal of Affective Disorders. 2017; 210:241–8. https://
doi.org/10.1016/j.jad.2016.12.036 PMID: 28064113
102. Meron D, Hedger N, Garner M, Baldwin DS. Transcranial direct current stimulation (tDCS) in the treat-
ment of depression: Systematic review and meta-analysis of efficacy and tolerability. Neuroscience &
Biobehavioral Reviews. 2015; 57:46–62.https://doi.org/10.1016/j.neubiorev.2015.07.012.
103. Nejati V, Salehinejad MA, Shahidi N, Abedin A. Psychological intervention combined with direct electri-
cal brain stimulation (PIN-CODES) for treating major depression: A pre-test, post-test, follow-up pilot
study. Neurology, Psychiatry and Brain Research. 2017; 25:15–23.https://doi.org/10.1016/j.npbr.
2017.05.003.
104. Nitsche MA, Nitsche MS, Klein CC, Tergau F, Rothwell JC, Paulus W. Level of action of cathodal DC
polarisation induced inhibition of the human motor cortex. Clinical Neurophysiology. 2003; 114
(4):600–4.https://doi.org/10.1016/S1388-2457(02)00412-1. PMID: 12686268

105. Monte-Silva K, Kuo M-F, Hessenthaler S, Fresnoza S, Liebetanz D, Paulus W, et al. Induction of Late
LTP-Like Plasticity in the Human Motor Cortex by Repeated Non-Invasive Brain Stimulation. Brain
Stimulation. 2013; 6(3):424–32. https://doi.org/10.1016/j.brs.2012.04.011 PMID: 22695026
106. Polanı́a R, Nitsche MA, Paulus W. Modulating functional connectivity patterns and topological func-
tional organization of the human brain with transcranial direct current stimulation. Hum Brain Mapp.
2011; 32(8):1236–49. https://doi.org/10.1002/hbm.21104 PMID: 20607750
107. Bikson M, Grossman P, Thomas C, Zannou AL, Jiang J, Adnan T, et al. Safety of Transcranial Direct
Current Stimulation: Evidence Based Update 2016. Brain Stimulation. 2016; 9(5):641–61. https://doi.
org/10.1016/j.brs.2016.06.004 PMID: 27372845
108. Moliadze V, Andreas S, Lyzhko E, Schmanke T, Gurashvili T, Freitag CM, et al. Ten minutes of 1mA
transcranial direct current stimulation was well tolerated by children and adolescents: Self-reports and
resting state EEG analysis. Brain Research Bulletin. 2015; 119:25–33. https://doi.org/10.1016/j.
brainresbull.2015.09.011 PMID: 26449209
109. Cachoeira CT, Leffa DT, Mittelstadt SD, Mendes LST, Brunoni AR, Pinto JV, et al. Positive effects of
transcranial direct current stimulation in adult patients with attention-deficit/hyperactivity disorder A
pilot randomized controlled study. Psychiatry Research. 2017; 247:28–32. https://doi.org/10.1016/j.
psychres.2016.11.009 PMID: 27863315
110. Wischnewski M, Engelhardt M, Salehinejad MA, Schutter DJLG, Kuo MF, Nitsche MA. NMDA Recep-
tor-Mediated Motor Cortex Plasticity After 20 Hz Transcranial Alternating Current Stimulation. Cere-
bral Cortex. 2018:bhy160–bhy. https://doi.org/10.1093/cercor/bhy160 PMID: 29992259

Pone 0215095

Uploaded by

Copyright:

Available Formats

Pone 0215095

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Pone 0215095

Uploaded by

Copyright:

Available Formats

RESEARCH ARTICLE

Transcranial direct current stimulation in

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 1 / 26

decision to publish or preparation of the Introduction

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 2 / 26

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 3 / 26

Search strategy and study selection

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 4 / 26

Risk of assessment bias

Data extraction and statistical analysis

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 5 / 26

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 6 / 26

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 7 / 26

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 8 / 26

Effects of tDCS on inhibitory control in ADHD patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 9 / 26

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 10 / 26

Effects of tDCS on working memory in ADHD patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 11 / 26

ameliorate neuropsychological symptoms [77]. Exploring the impact of tDCS especially on

tDCS effects on inhibitory control in ADHD

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 12 / 26

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 13 / 26

tDCS and WM performance in ADHD

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 14 / 26

Clinical and methodological implications

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 15 / 26

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 16 / 26

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 17 / 26

Limitations and future directions

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 18 / 26

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 19 / 26

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 20 / 26

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 21 / 26

the American Academy of Child & Adolescent Psychiatry. 2009; 48(5):484–500.https://doi.org/10.

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 22 / 26

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 23 / 26

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 24 / 26

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 25 / 26

PLOS ONE | https://doi.org/10.1371/journal.pone.0215095 April 12, 2019 26 / 26

You might also like