Abstract
Keywords
Children with ADHD suffer from symptoms of inattention, hyperactive and impulsive behavior, or a combination of these two symptom domains (American Psychiatric Association, 2000). Inherent in this description are difficulties in executive function (EF), particularly self-regulation difficulties and uninhibited behavior (Shimoni, Engel-yeger, & Tirosh, 2012).
Many studies have suggested that the core symptoms of ADHD are relevant to the failure of EF. Six domains of EF have been reported thus far, namely, inhibition, working memory, sentence memory, planning, fluency, and shifting (Pennington & Ozonoff, 1996). Among them, a defect in inhibition has been observed in some studies in both children and adults with ADHD (Barkley, 1997; Yasumura et al., 2014). ADHD includes a deficit in behavioral inhibition. A theoretical model is created that connects inhibition to four executive neuropsychological functions that appear to depend on it for their effective execution: (a) working memory, (b) self-regulation, (c) internalization of speech, and (d) reconstitution (behavioral analysis and synthesis; Barkley, 1997).
Inhibitory control is a complex concept that can be defined as the ability to prevent prepotent actions. It represents a loose collection of cognitive processes that are grouped together to facilitate behavioral and cognitive control by suppressing nonproductive behaviors or cognitive processes (Roberts, Fillmore, & Milich, 2011). Prepotent response inhibition, one component of inhibitory control, is the ability to suppress dominant, automatic, or prepotent responses deliberately (Ikeda, Hirata, Okuzumi, & Kokubun, 2010). This element plays a critical role in EF that controls and regulates thoughts and actions. Deficits in prepotent response inhibition have also been observed in population with developmental disorders such as ADHD (Barkley, 1997; Geburek, Rist, Gediga, Stroux, & Pedersen, 2013; Jonkman, van Melis, Kemner, & Markus, 2007).
It is noteworthy that response interference is another component of this concept (van‘t Ent et al., 2009). Response interference refers to the finding that performance deteriorates when a dominant response has to be suppressed to give the alternate response, relative to the condition in which the dominant response and the activated response are the same (Stins, Tinca Polderman, Boomsma, & degeus, 2007).
It is also worth mentioning that deficits in inhibitory control have been related with problems in behavioral regulation, in impulsive violence, as well as a range of clinical disorders. Researchers have suggested that the most substantial neuropsychological findings in children and adults with ADHD are deficits in motor response inhibition and cognitive switching (Boonstra et al., 2005; Rubia et al., 2007; Willcutt et al., 2005).
Neuroimaging studies demonstrate that various areas, including the Frontal Eye Fields (FEF), the pre-Supplementary Motor Area (pre-SMA), the Right Lateral Prefrontal Cortex (rLPFC), and most significantly the Inferior Frontal Gyrus (IFG), play a critical role in the inhibition motor responses and inhibitory controls, thus become active during performance of both Go/No-Go and Stop-signal tasks. Individuals with ADHD show reduced activation, compared with the control group, and anatomical correlations in the rLPFC regions when performing tasks (Depue, Burgess, Willcutt, Ruzic, & Banich, 2010). The role of these areas in inhibitory control has been investigated while using an inhibitory control task and both transcranial magnetic stimulation (TMS) and transcranial direct current stimulation (tDCS; Juan & Muggleton, 2012). The response interference component of inhibitory control is measured by two executive functioning paradigms in color-word Stroop and a Flanker task. It has been demonstrated that high attention problem scores show a relative decrease in activation to response interference in dorsolateral prefrontal cortex (DLPFC) during the Stroop task and parietal and temporal brain regions during the Flanker task (van‘t Ent et al., 2009).
Among various measures of prepotent response inhibition, the most classic is the Stroop color-word test. The Go/No-Go and the stop signal tasks both involve response inhibition as well. Nigg (1999) mentioned that individuals with deficits in inhibitory control should be examined by the Go/No-Go tasks, while Gomez (2003) suggested Stop-Signal paradigms. Whereas a motor response has to be executed or inhibited in the Go/No-Go task, the motor response to a go-stimulus has to be retracted in the stop signal task. Accordingly, both tasks require inhibition on a response (motor) level. In contrast, there is growing evidence that the Stroop task involves inhibition and interference control on the task (cognitive) level (Monsell, Taylor, & Murphy, 2001). Therefore, Stroop and Go/No-Go tasks have been chosen to measure two different dimensions of one cognitive phenomenon, that is, inhibitory control.
Based on the above, it is now hypothesized that the children and adults with ADHD have a deficit in inhibitory control, and the activated areas in the disorder are also recognized. Previous studies have shown that tDCS and TMS can be effective on inhibitory tasks (Depue et al., 2010; Juan & Muggleton, 2012).
tDCS is a safe, cheap, and noninvasive brain stimulation technique that modulates cortical excitability via applying a weak electrical current through the scalp; this procedure has been successfully used as a tool to improve behavioral inhibition (Beeli, Casutt, Baumgartner, & Jancke, 2008; Hsu et al., 2011; Jacobson, Javitt, & Lavidor, 2011). The therapeutic utility of these interventions is currently under investigation for several refractory neuropsychiatric diseases with promising results (Demirtas-Tatlided, Vahabzadeh-Hagh, & Pascual-Leone, 2013). Brain excitability occurs via the application of low-amplitude (0.5-2 mA) direct current through scalp electrodes. This current, through its effects on resting membrane potentials, can lead to increased or decreased neuronal excitability depending on the polarity and spatial arrangement of the electrodes. Cortical excitability is increased under the tDCS anode and decreased under the cathode (Nitsche et al., 2003). During the previous decade, tDCS has also been garnered increasing interest for its application in the treatment of clinical disorders such as depression (Loo et al., 2012), tinnitus (Fregni et al., 2006), stroke (Webster, Celnik, & Cohen, 2006), and Alzheimer’s disease (Boggio et al., 2012). It has been also suggested that repetitive transcranial magnetic stimulation (rTMS) administration over the DLPFC can improve inhibitory control, impulsivity, and decision making (Li et al., 2013).
The modulation of inhibitory control with direct current stimulation on the superior medial frontal cortex has been investigated, and the obtained results have shown that anodal tDCS improved efficiency of inhibitory control. Conversely, cathodal tDCS showed a tendency toward impaired inhibitory control (Hsu et al., 2011). Modulating behavioral inhibition by tDCS combined with cognitive training has been studied. While half of the participants received anodal tDCS (1.5 mA, 15 min) over rIFG, the rest received training but not active stimulation. Findings demonstrated that the brain stimulation accompanying training protocols could enhance their performance (Ditye, Jacobson, Walsh, & Lavidor, 2012). In short, in many studies, it is shown that right DLPFC is involved in behavioral control and dysregulation of this area during application of inhibitory control tasks in ADHD population is investigated. Using stimulation methods have improved inhibition control in this population (Fisher, Aharon-Peretz, & Prett, 2011; Geburek et al., 2013; Woltering, Liu, Rokeach, & Tannock, 2013).
The present study uses two inhibitory control tasks: We chose the Stroop task to measure interference inhibition and the Go/No-Go task to measure prepotent inhibition.
Our research suggests that a large number of brain regions show differences in activity between ADHD adults and the control group. Some of these regions are conceptualized as being sources of attentional control: DLPFC, Anterior Cingulate, Posterior Parietal Cortex, and right Inferior Frontal Cortex; supporting the idea of Banich et al. (2009) and Valera et al. (2005) that these regions exerted control less effectively in individuals with ADHD than members of the control group. It is also noteworthy that targeting a cortical region using TMS/tES can also affect the corresponding region in the other hemisphere via transcallosal connections (Kobayashi & Pascual-Leone, 2003). For example, it has been suggested that HF-rTMS application over right DLPFC decreases left DLPFC activation via transcallosal connections (Mishra, Nizamie, Das, & Praharaj, 2010). To the best of our knowledge, the effect of tDCS application over left DLPFC on inhibitory control in ADHD symptoms has not been investigated so far. In this study, we modulated the activity of left DLPFC to remediate inhibitory control through a possible effect on the corresponding region in the right hemisphere. Our hypothesis is that cathodal stimulation over the left DLPFC can improve the inhibitory control.
Method
Participants
Our population was 350 high school students; out of them, 243 students (age M = 16.40, SD = 1.09) were tested by Conner’s Adult ADHD Rating Scale (CAARS-S:S), which contains a list concerning current symptoms of ADHD. Twenty-three students who had a score 1.5 standard deviation higher than the mean (t > 65) were selected as people with ADHD symptoms. To ensure the existence of the symptoms, the Wender Utah Rating that shows the ADHD symptoms during childhood was used, and results of this questionnaire confirmed symptoms of ADHD in childhood of these students. The Symptom Check List (SCL-25) that measures general psychological health was applied as well, and one of the participants showed anxiety symptoms and therefore was excluded. All participants were right handed and were assessed by the Edinburgh Handedness Inventory. They reported being free from present or past history of neurological or psychiatric disorders, epilepsy, seizures, and head injury or loss of consciousness except one participant who had a past history of seizure. He, together with another student who attended only two sessions and did not complete the experiment, was omitted from the study. Because participants were under legal age, before starting the experiment, their headmaster was instructed about application of the experiment. He then signed the form of consent. The students were free to participate or to withdraw from experiment at any phase. As reported before, one of them participated just for two sessions and then left the experiment. Eventually, we examined 20 high school students who had ADHD symptoms. They were equally distributed by grade and were all aged between 15 and 17 (six 15 years old, six 16 years old, and eight 17 years old).
Cognitive Tasks
The Go/No-Go task
Go/No-Go test is used to measure a participant’s capacity for sustained attention and response control. On a Go/No-Go test, it is required that a participant performs an action on a given certain stimulus (e.g., press a button—Go), and inhibits oneself from that certain action under a different set of stimuli (e.g., not press that same button—No-Go). In this study, a plane appeared to the four directions up, down, left, and right on a laptop’s 14 inch screen and participants were instructed to press the button aligned with the plane (the Go condition), but they had to withhold pressing any button when the sound “Beep” was heard (the No-Go condition). Participants were asked to focus on a cross on the monitor in front of them and to press a response button “as quickly and accurately as possible” to all Go stimuli. This task consisted of 50 stimuli that required response execution on 75% of trials, and the inhibition of a response on the remaining 25% (Figure 1). The planes are static and black on a white screen, the size of plane is 7 × 7 cm. The dependent variables in this study were the proportions of correct responses for the Go and No-Go categories and reaction time of Go. In this task, the correct response to No-Go stimuli assesses the prepotent response inhibition in a way that participants had to refrain from pressing the arrow button when they heard beep with delay. This task has been a modified version of the revised stop signal task proposed by Carter, Farrow, Silberstein, Stough, and Pipingas (2003).

The Go/No-Go task procedure.
The Stroop task
There are different standards of color-word Stroop task in which participants need to conduct the task via a key press of the color (red, blue, green, or yellow) when words were presented. The participants were given four buttons marked with four colors and were instructed to select, as fast as possible, a color among the four colors when presented with the word. There were three types of trials: neutral trials in which the word was written with the black ink (e.g., “red” in black ink) and participant should press the red button that marked on it; during the congruent trial, the word matched the ink color (e.g., “blue” in blue ink); and in the incongruent trials, the word conflicted with the ink color (e.g., “red” in yellow ink) and participants should consider the color of the ink and inhibit the word. The size of the visual stimuli are 7 (length) × 5(width) cm and the presentation duration is until response. There were 25 stimuli for each stage. The proportion of false answer and reaction time of third stage as the main stage that shows interference stimuli was calculated to assess the interference inhibition.
tDCS Methodology
This study consisted of three phases: anodal, cathodal, and sham stimulation. After 8 min of stimulation, the tasks were started and their performance took approximately 7 min. Each experimental session started with Go/No-Go and continued by the Stroop Test. There was a 72-hr interval between sessions. The order of the experimental sessions was fully randomized across participants and sessions. The participants were unaware of the kind of stimulation. At the end of each session, participants responded to a brief questionnaire checking if the participants felt any tDCS side effects that may include headache, vertigo, tingling, itching, dizzying, drowsiness, and nausea. The conditions of each session are shown in Figure 2.

Conditions of each session.
Active electrode was placed on left DLPFC (F3) and return electrode was placed over right supraorbital. tDCS device in use was “Activadose Inotophoresis” manufactured by Activa Tek, battery-driven with a 9-volt battery as its source. The current intensity of 1.5 mA was applied through a pair of saline-soaked sponge electrodes with size of 35 cm2 (7 × 5) for 15 min (with 15-s ramp up and 15-s ramp down). All participants received anodal, cathodal, and sham tDCS in three sessions with at least 72-hr washout period between sessions. The polarity of tDCS was randomized and counterbalanced across participants. As mentioned earlier, we used three stimulation conditions: (a) Actual anodal current tDCS was applied on the left DLPFC (F3) according to the 10-20 EEG electrode systems and the cathode electrode was placed on the right supraorbital (Fp2); (b) actual cathodal current tDCS was placed on left DLPFC according to the 10-20 EEG electrode systems and anode electrode was applied on the right supraorbital (Fp2); (c) sham condition served as a method of control that was performed. During sham condition, current was ramped up for 15 s to generate the same sensation as active condition for the participant and then turned off (Palm et al., 2013).
Statistical Analysis
For the statistical analysis, the data were analyzed using SPSS 18 software. To test the effects of tDCS on the inhibitory control, repeated measures ANOVA was performed for the factor of condition: (three conditions: anodal stimulation on left DLPFC/cathodal stimulation on left DLPFC/the sham condition).
Method
Mauchly’s test was used to evaluate the sphericity of the data before implementing repeated measure ANOVA. Sphericity of hypothesis was violated regarding to Mauchly’s test result, therefore, the greenhouse test result showed that there was significant differences within subject effect. Repeated measures ANOVA was performed to test (a) the effects of tDCS on inhibition accuracy and reaction time in Stroop task to assess the interference inhibition and (b) the effects of tDCS on performance and inhibition accuracy in Go/No-Go task to evaluate prepotent response inhibition. The hypothesis is that cathodal stimulation on left DLPFC can remediate the inhibitory control in ADHD symptoms compared with anode and sham conditions.
Results
First of all, the mean and standard deviation of accuracy and reaction time for each measure (interference inhibition and prepotent response inhibition) were calculated in three conditions for data analysis (anode, cathode, and sham; Table 1).
Means and SDs of Accuracy and Reaction Time in Stroop and Go/No-Go.
The results of repeated measure ANOVA in three conditions are shown separately in interference inhibition which is measured by Stroop task (Table 2) and prepotent response inhibition which is measured by Go-No-Go task (Table 3).
Repeated Measure ANOVA for the Effect of Conditions on Response Inhibition Measured by the Stroop Task.
Repeated Measure ANOVA for the Effect of Conditions on Prepotent Response Inhibition Measured by the Go-No-Go Task.
Repeated measures ANOVA showed no significant difference between three conditions in accuracy (p = .10) and reaction time (p = .31) in interference inhibition measured by the Stroop task. tDCS stimulation had no significant effect on modulating interference inhibition.
Repeated measure ANOVA also indicated that there was a significant difference between conditions in terms of accuracy of No-Go and in accuracy of Go. The Fisher’s least significant difference (LSD) post hoc test showed there are significant differences in accuracy of No-Go in prepotent response inhibition between cathode, anode, and sham conditions in such a way that accuracy of No-Go increased in cathode condition compared with anode (p = .03) and sham (p = .01; 99.58 vs. 96.23 and 95.82). tDCS stimulation can remediate the prepotent response inhibition while cathodal stimulation is on left DLPFC. The LSD also showed the increase in accuracy of Go responses in anode condition comparing with the cathode (p = .03; 98.81 vs. 97.26). All results are shown in Figure 3.

Difference between three conditions in Stroop and Go-No-Go tasks.
Discussion
This study examined the effect of tDCS over left DLPFC on modulation of inhibitory control in adults with ADHD symptoms. The main aim of this study was to determine whether stimulation of left DLPFC, as a cortical structure involved in EF, can remediate the inhibitory control via transcallosal connections. Stroop and Go/No-Go tasks were utilized to evaluate inhibitory control by assessing participants’ ability of interference inhibition and prepotent response inhibition, respectively.
One observation was increase in the proportion of correct No-Go responses by cathodal stimulation. Cathodal stimulation increased participants’ ability to inhibit themselves and not to press the button when they were exposed to No-Go stimulus. Previous studies have shown right DLPFC activity reduction during inhibitory control tasks. This reduction in activity is involved in the impairment of cognitive executive and behavioral control in ADHD population (Banich et al., 2009; Ernst et al., 2003; Moser, Cuini, Weber, & Schroeter, 2009; Rubia et al., 2009; Wild-wall, Oades, Schmidt-Wessels, Christiansen, & Falkenstein, 2009; Wolf et al., 2009; Woltering et al., 2013). It is also demonstrated that the anodal stimulation over right DLPFC can improve inhibitory control (Beeli et al., 2008; Hsu et al., 2011; Jacobson et al., 2011). On the other hand, similar effect of excitatory and inhibitory stimulations on right and left DLPFC areas has been reported in some existing brain stimulation studies and explained based on the inhibitory link between these two regions via transcallosal connections (Kobayashi & Pascual-Leone, 2003; Mishra et al., 2010). The mechanism of interhemispheric inhibition in the human motor cortex is investigated. Selection of stimulation polarity based on the inhibitory connection between regions in two hemispheres has also been performed when using tDCS in motor rehabilitation after stroke (Daskalakis, Christensen, Fitzgerald, Roshan, & Chen, 2002; Schlaug, Renga, & Nair, 2008). This observation is in line with our hypothesis that cathodal stimulation of left DLPFC can improve the inhibition. Accordingly, it is expected that anodal stimulation of left DLPFC would impair participants’ inhibitory control. It has been also taken into account that when participants switch between tasks or different phases of one task, they need to inhibit the current task/phase and engage in the new one. Inhibition of the disengaged task/phase remains active for a period of time and increases the probability of erroneous response in the consecutive task/phase (Dreher & Berman, 2002). Impairment in inhibition ability resulted from anodal stimulation on left DLPFC might increase the probability of unsuccessful disengagement and facilitate Go responses in all trials. This might explain the observation that anodal stimulation increased the accuracy of Go responses in Go/No-Go task.
The result of Stroop task showed no difference between anodal, cathodal, and sham conditions in interference inhibition in accuracy and reaction time. On one hand, previous studies indicated that the ADHD group showed greater color interference than the tDCS group in Stroop task (Yasumura et al., 2014). The Stroop interference is strongly correlated with ADHD symptoms in typically developing children and the symptoms are measured through behavioral rating scales of ADHD that are based on teacher’s report (Ikeda et al., 2010). On the other hand, previous studies have shown that tDCS effects are task dependent; that is, tDCS can modulate performance in difficult cognitive tasks but not simpler ones (Kwon, Kang, Son, & Lee, 2015; Pope, Brentone, & Miall, 2015). Therefore, the absence of tDCS effect on this task can be associated with its simplicity. In contrast, Go-No-Go task places higher load on response inhibition processes compared with other inhibitory tasks (Rubia et al., 2001) and seems to be an appropriate choice for tDCS studies.
This study recommends application of anodal and cathodal stimulation over both left and right DLPFC in a way that cathode is over the left and anode is over the right part. It also suggests that in addition to stimulations and sham conditions, we can have another condition without using device and compare it with other stimulation conditions.
The present study faced some limitations. The first was the low localization capacity of the tDCS device; it caused the area near the target area also to be affected. Furthermore, the results in this study were obtained from only one session per experimental condition in a small group of participants; it would be better to use multiple sessions in a larger one.
Footnotes
Acknowledgements
We appreciate Professor Vincent Walsh and Professor Paul Fitzgerald for reading the early manuscript and providing us with their helpful comments. We also appreciate Dr. Fateme Yavari who provided insight that greatly assisted the research and Dr. Ruhollah Mansouri Sepher who helped us in statistical analysis. We also thank the anonymous reviewers for their helpful comments. We gratefully acknowledge the incentive support of the Cognitive Sciences and Technologies Council.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
