Damian Sendler: It’s been 45 years since EEG-neurofeedback has been studied, and the most recent meta-analyses of RCTs only show small/medium effects in comparison to non-active controls. Functional magnetic resonance imaging and near-infrared spectroscopy were used in three small studies to pilot neurofeedback of frontal activations in ADHD, but the results showed no benefit over the control groups. Repetitive transcranial magnetic and direct current stimulation (rTMS/tDCS) has been used in the treatment of ADHD. Most studies have found that rTMS has little effect on improving cognitive function or symptoms. Only two out of three tDCS studies that focused on the dorsolateral prefrontal cortex showed clinical improvements in meta-analyses of the results. In one RCT, trigeminal nerve stimulation was found to have a moderate effect on ADHD symptoms. Modern neurotherapeutics are appealing because of their relative safety and neuroplastic potential. It’s important that they’re rigorously tested for their clinical and cognitive efficacy in a variety of settings, as well as for their potential for individualized treatment.
Damian Jacob Sendler: Symptoms of attention deficit/hyperactivity disorder (ADHD) include persistent and impairing symptoms of age-inappropriate inattention and/or hyperactivity/impulsivity (DSM-5) According to Thomas et al. (2015), 7 percent of children worldwide are affected by the disorder. Comorbidities and poor academic and social outcomes persist in a significant number of cases into adulthood (Thomas, et al., 2015) [2].
Dr. Sendler: Deficits in higher-level cognitive functions, known as “executive functions,” are mediated by late developing fronto-striato-parietal and fronto-cerebellar networks in ADHD patients (Rubia, 2013) [3]. “Cool” EF, such as motor response inhibition, working memory, sustained attention, response variability, and cognitive switching (Pievsky & McGrath, 2018; Rubia, 2011; Willcutt et al., 2008), as well as in temporal processing (in particular in time discrimination and estimation tasks) [4,5,6] (Noreika et al., 2013; Rubia et al., 2009) [7,8]. Temporal discounting and gambling tasks have both revealed impairment in the so-called “hot” EF functions of motivation control and reward-related decision-making. However, evidence for a hot EF deficit has been less consistent than evidence for a cool EF deficit [5,8,9] (Noreika et al., 2013; Plichta & Scheres, 2014; Willcutt et al., 2008), in accordance with diagnostic criteria [5,8]. There is more evidence of cognitive deficits in children with ADHD than in adolescents or adults with ADHD [6,10]. (Groen et al., 2013; Pievsky & McGrath, 2018). More than 30% of patients with cognitive impairments show no EF impairments (Nigg et al. 2005; Roberts et al. 2017) [11,12], further illustrating the wide range of cognitive impairments.
Psychostimulant medication that enhances catecholamines in the brain, reaching an effect size of 0.8, is the most effective treatment for ADHD, with about 70% of patients responding [13]. Cortese and colleagues (2018). According to fMRI studies, stimulant medication increases activation and interconnectivity in the inferior frontal and striatal regions and decreases activation in the default mode network in these regions [14]. Cognitive improvements may be attributable to both of these interventions (Rubia et al., 2014) [15,16]. (Coghill et al., 2014; Pievsky & McGrath, 2018).. ( Nuclear transporter/receptor blockers Atomoxetine and Guanfacine are used as a secondary treatment because they increase brain catecholamines by 0.56 and 0.67, respectively. Cortese and colleagues (2018). It’s controversial because of the potential for abuse and diversion that stimulant prescriptions have increased dramatically in recent decades worldwide. Aside from these common side effects, stimulants can also cause disturbances in sleep and appetite, as well as irritability and nausea/vomiting. Other common side effects include labile moods, headaches, and a decrease in growth (Cortese et al., 2018). In addition, only half of patients can tolerate it, some comorbid conditions (such as cardiovascular malfunctions and sleep disorders) necessitate caution, and adherence can be poor, especially in adolescents. As a matter of fact, long-term efficacy has not been shown in meta-analyses, nor in observational or epidemiological studies [13,17]. A disagreement exists (Coghill, 2019) [18] between the two studies.
Neurotherapeutics have a distinct advantage over stimulant medications because they can directly target the brain function deficits that have been identified in ADHD, unlike stimulant medications, which were originally developed for other medical conditions, such as bronchodilation, headache, and blood pressure [19–20] (Connolly et al., 2015). Many studies on the differences in brain function between ADHD patients and healthy controls using electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) have been conducted since the 1970s (e.g. Satterfield, 1973; Satterfield et al., 1973) [21,22] and over the past 2.5 decades that have provided us with neurofunctional biomarkers that could be targeted with neurotherapeutics such as neurofeedback or non-invasive brain stimulation techniques.
Increases in slower oscillations, such as delta, theta or alpha, but also faster beta frequencies bands during resting conditions were most relevant to ADHD electrophysiology findings in ADHD (Loo & Makeig, 2012)[23]. Slower frequencies in particular decrease in frequency as we grow older, so the oscillatory profile shows problems with maturation and arousal. As the debate over whether or not higher frontocentral TBR is associated with ADHD heats up [24], a finding that is becoming increasingly contentious. Resting EEG markers have been linked to reduced attention, hypoarousal, and maturational lag in previous studies (Snyder et al., 2015), suggesting an association between ADHD and these resting EEG markers is solid. In spite of maturational effects (Buyck & Wiersema, 2014, 2015; Liechti et al., 2013), scientific efforts to replicate this hypothesis have failed to show a consistent increase in TBR in ADHD [25,26,27,28], and a relationship between TBR and arousal has been questioned [29]. Clarke and colleagues (2019). For example, a meta-analysis found that a lower effect size in ADHD may be due to a decrease in sleep duration, which may be linked to an increase in the number of people with ADHD who have a higher level of TBR than healthy people [30]. (Arns et al., 2012). The heterogeneity of ADHD has been shown to be a possible explanation for the inconsistent results. In fact, only three of the five EEG clusters showed increased TBR [31] in patients with ADHD (Clarke et al., 2011), with 60% of children with ADHD showing increased theta activity. More recent research shows that 35% of people with ADHD have high levels of TBR [32]. [33] A study by Bussalb et al. (2019). Disparities in concentration, cognitive effort, activation and drowsiness are likely to confound TBR as a biomarker for ADHD [33] (Drechsler et al., 2020), consistent with findings that theta activity increases only after longer EEG recordings are made in ADHD Zhang and colleagues (2019). The current evidence in the field makes it premature to make definitive statements about the usefulness of the TBR ratio as a diagnostic test for ADHD [35], according to a recent review on ADHD resting EEG power research. Clarke and colleagues (2020) Deviating TBR has been taken into account in recent EEG-NF studies, which propose a cut-off (i.e., >4.1) and apply TBR-NF only to the subgroup with high TBR ratios [36,37].
Event-related potentials, on the other hand, appear to be more consistent than the controversial research on electrophysiological oscillations (ERPs). Task-locked ERPs are defined as reflecting cognitive, sensory and motor brain responses. ADHD was found to have abnormalities in stimulus discrimination, resource allocation, inhibition, preparation, error detection, and conflict processing via different ERP components [38–39]. (Barry et al., 2003, Johnstone et al., 2013). Although these changes do not appear to be ADHD-specific, the limited utility they provide as diagnostic biomarkers is concerning [23]. (Loo & Makeig, 2012). Meta-analysis [40] found moderate to large effects for specific ERPs associated with late cognitive processing related to attentional preparation and resource allocation like P300 and contingent negative variation (CNV), but the results were characterized by substantial heterogeneity and modest effect sizes that limit their use in clinical applications. Kaiser et al. (2020) conducted the current meta-analysis [40]. Moreover, it is imperative that these components be studied in a systematic manner, as most studies used a variety of tests and measures, making reliable interpretation of classification accuracy and effect size particularly difficult [41]. (Gamma & Kara, 2020).
In the last two decades of MRI research, there has been consistent evidence of ADHD-related deficits in brain structure and function. A neurodevelopmental disorder, like ADHD, is now recognized. Several structural volumetric studies in ADHD have shown reduced gray matter in subcortical regions, including the basal ganglia and insula [42,43,44,45] (Hoogman et al., 2017; Lukito et al., 2020; Nakao et al., 2011; Norman et al., 2016), as well as limbic areas such as the amygdala and hippocampus [42] (Hoogman et al., 2017). Hoogman and colleagues (2019); Lukito and colleagues (2020); Norman and colleagues (2016; Lukito et al., 2020); Norman et al., 2016). The peak of cortical thickness and surface area maturation in frontal, temporal, and parietal regions has been delayed (Shaw et al., 2007; Shaw et al., 2012) [47,48]. Fronto-striatal, fronto-cerebellar and interhemispheric white matter tracts have also been found to be impaired in the disorder (Aoki et al., 2018; Chen et al., 2016) [49,50], as well as long-distance tracts such as fronto-occipital tracts. (See Rubia, 2018 for a review) [51].
Neurotherapeutics have been developed to target several of the neurofunctional biomarkers identified by fMRI studies in children with ADHD. This disorder has been linked to a wide range of dysfunction in the lateral inferior and dorsolateral prefrontal cortex as well as the basal ganglia, medial frontal, cingulate, and orbital frontals and the dissociated fronto-parietal fronto-parietal, fronto-limbic & fronto-cerebellar networks they form a part of (Rubia, 2018) [51].
ADHD is the subject of a number of fMRI meta-analyses, the majority of which include cool EF studies. Several frontostriatal, fronto-parietal, and fronto-cerebellar brain regions in ADHD show cognitive domain-dissociated brain dysfunctions. Among our three meta-analyses, which included the most patients, we found that people with ADHD have consistently lower activation in key areas of cognitive control than healthy controls, including the right inferior prefrontal cortex (IFC)/anterior insula, the supplementary motor area (SMA), the anterior cingulate cortex (ACC), and the striatal region (SRC). This finding has been replicated in other studies. (Hart and colleagues, 2013; Lukito and colleagues, 2020; Norman and colleagues, 2016) A number of smaller meta-analyses focusing on inhibition tasks also found DLPFC underactivation in addition to Cortese et al(2012), .’s Lei et al(2015), .’s and McCarthy et al(2014) .’s findings [53,54,55]. (Cortese, Lei, and McCarthy, 2014; Lei, Lei, and McCarthy, 2015) The right hemispheric dorsal attention network, comprised of the right DLPFC, right inferior parietal cortex, and caudal portions of both the basal ganglia and the thalamus, showed reduced activation in 171 ADHD patients compared to 178 healthy controls in our meta-analysis of a wide range of fMRI studies on attention tasks, including selective, divided, and sustained attention, as well as alerting and mental rotation. There was an increase in activation in the right cerebellum and left cuneus in ADHD patients to compensate for the decreased activation of frontal DLPFC-parieto-cerebellar attention network in the frontal part of the network. [52] In a study by Hart et al. During attention tasks, the right anterior cingulate was found to be significantly less active in a meta-analysis of 11 fMRI datasets [55]. Cortese and colleagues (2012). More than 150 patients with attention deficit hyperactivity disorder (ADHD) were found to have lower activation in the left inferior parietal lobe, left IFC, and right lateral cerebellum compared to 145 healthy controls, according to an fMRI meta-analysis of 11 studies on timing functions. [56] In all critical regions mediating timing functions (Hart et al., 2012), According to Wiener and coworkers, (2010). While some large studies and meta-analyses have found right and left IFC underactivation [55,58], this meta-analysis found that 111 ADHD patients relative to 113 control subjects had reduced activation in bilateral middle and superior PFC, as well as the left MFC/ACC [53] (McCarthy et al., 2014). (Cortese et al., 2012; van Ewijk et al., 2015). In two large comparative meta-analyses, the right IFC dysfunction during cognitive control tasks was found to be disorder-specific to ADHD relative to OCD and ASD [44,45]. (Lukito et al., 2020; Norman et al., 2016). There are deficits in different fronto-striato-cerebellar networks in ADHD patients that have been found in the IFC/ACC/SMA fronto-striato-thalamic (inhibition), right DLPFC fronto-striato-thalamo-parietal (attention), bilateral DLPFC, IFC, and MFC/ACC (working memory), and left IFC-parieto-cerebellar regions (timing). RUBIA (Rubia, 2018).
Children with ADHD have shown reduced activation in the ventromedial prefrontal cortex (vmPFC), orbitofrontal cortex (OFC), and striato-limbic regions during tasks of “hot” EF, such as reward-related decision making or temporal discounting, in addition to deficits in several of these lateral fronto-striato-parietal and fronto-cerebellar regions that mediate so-called “cool” EF. Findings on deficits have been more inconsistent (Plichta & Scheres, 2014; Rubia, 2018), but they are still present in many studies
The dorsal and ventral attention and cognitive control networks, in particular, show evidence of decreased inter-regional functional connectivity during cognitive tasks and while resting (Rubia, 2018; Sripada et al., 2014; Sripada et al., 2014). [51,59,60].
Nonetheless, it has been found that not only task-positive regions, but also areas of the default mode network (DMN), which includes the ventromedial frontal cortex, posterior cingulate, precuneus, and inferior parietal and temporal regions, are abnormal in function in ADHD [61]. (Raichle, 2015). Many of the studies and meta-analyses discussed above show that individuals with ADHD have more active DMN regions, including those in the rostromedial prefrontal cortex (Hart et al., 2013), the posterior cingulate and the precuneus (Christakou et al., 2013), as well as increased activation in these areas during interference inhibition and other cognitive control (Hart etal., 2013). (Hart et al., 2012). For ADHD patients, the results show that they have a hard time controlling their interoceptive attention orientation and mind-wandering, which intrudes into their already weak exteroceptive attention processes, likely resulting in an increased level of distraction and impulsivity. Attention-demanding higher-level cognitive control tasks may be impaired in ADHD because of an immature pattern of poor activation of task-relevant and age-correlated task-positive brain activation networks and decreased deactivation of the DMN (Rubia, 2018) [51].
Neurotherapeutics could potentially target the most consistently dysfunctional regions, such as the right IFC, right DLPFC, ACC, right inferior parietal lobe, or the basal ganglia. IFC and DLPFC have already been used in fMRI/NIRS-neurofeedback or brain stimulation therapies to target these regions of the brain for neuromodulation. As an additional benefit of using fMRI-NF, it is possible to target the entire ADHD network, including the dorsal and ventral attention and cognitive control systems (Sripada et al., 2014). fMRI-NF has the potential to be a useful neurotherapeutic tool for the future if it can reduce DMN levels. We have shown in ADHD patients after fMRI-NF of the right IFC (Rubia et al., 2019) [65] that upregulation of the IFC/DLPFC with brain stimulation or neurofeedback can indirectly downregulate areas of the DMN because of evidence for an anti-correlation between the IFC/DLPFC and the DMN [59] (Sripada et al., 2014).
As neuroimaging research has progressed, it has discovered that the brain is highly plastic, particularly during childhood and adolescence [66,67], but also in middle and older adulthood [68,69] (Jancke, 2009; Rapoport & Gogtay, 2008). (Draganski et al., 2004; Draganski & May, 2008). Juggling, for example, can change the structure of the brain after just a few weeks or months of practice [68,69] (Draganski et al., 2004; Draganski & May, 2008), studying for an exam [70] (Draganski et al., 2006), or learning to meditate [71] (Dodich et al., 2019). Neuromodulation treatments like non-invasive brain stimulation or neurofeedback are now more appealing in clinical settings because of these new understandings of the brain’s neuroplastic potential [51,72]. (Rubia, 2018, Ashkan et al., 2013). Because children and adolescents have more rapid neural plasticity than adults do after brain stimulation, it’s especially important in the early stages of disorders in young people [73] (Anderson et al., 2011); evidence shows that this is most effective in children and adolescents) (Brunoni et al., 2012).
Using EEG and fMRI studies to identify ADHD biomarkers has made it possible to target these biomarkers with neurotherapeutics over the past few decades. Treatments aimed at correcting ADHD’s underlying neurobiological abnormalities may be promising, given the strong evidence for electrophysiological and neuroimaging functional deficits in the disorder. For over 45 years, EEG-NF has been used to study ADHD, but the results have been mixed. It’s too early in the game for fMRI or NIRS-neurofeedback to provide a clear picture of its potential efficacy. Over the past decade, the number of non-invasive brain stimulation studies has increased exponentially. However, the number of studies has been relatively small and the study designs have been extremely diverse. As a result, findings on how to improve cognition have been inconsistent, and there has been very little evidence on how to improve clinical behavior. Here, we’ll take a look at some of the ways neuromodulation has been used to treat ADHD.
Damian Jacob Markiewicz Sendler: With the help of real-time audio or visual feedback of their brain activation, participants in an operant conditioning procedure known as neurofeedback (NF) learn to voluntarily self-regulate specific regions or networks of their brains. This can be gamified for children in an appealing way. ADHD has been the most successfully treated with NF using electrophysiological neurofeedback because it is characterized by a lack of self-control [75] (Schachar et al., 1993). (EEG-NF).
By the year of publication, effect sizes (ES) in meta-analyses of EEG neurofeedback studies were calculated for their impact on global ADHD symptoms. MPROX: ratings by parents/proximal raters; PBLIND: ratings by presumably blinded raters. * Studies that followed a predetermined procedure.
Randomized controlled trials (RCTs) are considered the gold standard in clinical research while non-randomized studies are considered a weak experimental design [89] (Norris & Atkins, 2005) in the first meta-analysis [78]. Here is a meta-analysis [90] that addresses this issue by including only RCTs and blinding criteria for the clinical outcome, such as ADHD core symptoms (Sonuga-Barke et al 2013). The term “probably blinded” raters was coined by these authors to describe the type of assessment performed by teachers who aren’t aware of the patient’s treatment assignment. It was found that the clinical effect was reduced to a trend level for the likely blinded raters (such as parents) when these two new requisites were met, but remained significant for unblinded raters (such as parents) with medium effect sizes. These new findings led to a change in the recommendation to use EEG-NF in the treatment of ADHD.
In 2014, Micoulaud-Franchi et al. [82] (Micoulaud-Franchi et al., 2014) updated Sonuga-meta-analysis Barke’s from 2013 [84] (Sonuga-Barke et al., 2013), including the subdomains of the core ADHD symptoms, i.e. inattention, hyperactivity, and impulsivity. Only the inattention subdomain showed a significant effect when the core symptom domains were evaluated separately by the probably blinded raters.
An updated version of Sonuga-analysis Barke’s was released by the same group two years later, increasing the analysis from 8 to 13 RCTs with parent-ratings and 4 to 8 RCTs with probably blinded ratings [80] (Cortese et al., 2016) for the European ADHD guidelines group. A small effect size was observed for parents’ ratings, but a large effect size was observed for all of the possibly blinded ratings except inattention. In Micoulaud-Franchi [82] (Micoulaud-Franchi et al., 2014), the discrepancy in the blinded findings in the subdomain of inattention appears due to the selection of different blinded outcomes in the same studies.
Using only three studies that used standard protocols for EEG-NF [76] (Arns et al., 2014), the effects on ADHD symptoms were also significant for probably blinded rater, however, subsequent large-scale standard NF trials [36,91] (i.e., Arnold and colleagues in 2020; Strehl and colleagues in 2017) could not substantiate this. It is noteworthy that Bussalb et al. [32] in their meta-analysis [Bussalb et al., 2019] systematically evaluated additional factors that could affect the efficacy of the NF. In the end, they found that the severity of NF, rather than the length of treatment, was linked to greater efficacy, and that teachers were less sensitive to the symptoms of their patients. They also recommended that NF be tested using placebo-controlled interventions.
EEG-NF efficacy in ADHD has improved, as can be seen from this, in terms of the quality and certainty of the consideration and evaluation. Since there are so many different training modalities available, neurofeedback should be considered an umbrella term (such as Coherence training, asymmetry feedback, etc). Efforts should be made to standardize this issue, which is critical. A few large studies have been published recently, but so far, the standard protocols have met these requirements.
Additionally, the selection of an adequate control group was addressed in the most recent comprehensive meta-analysis [83] (Riesco-Matas et al., 2021), which compared EEG-NF with non-active control groups (waiting-list controls, treatment as usual) and active control groups. [83] EEG-NF was found to be more effective in assessing inattention by blinded raters than non-active control groups. This finding is similar to that of Micoulaud-Franchi et al. [82]. (Micoulaud-Franchi et al., 2014). As an active control condition, like medication, EEG-NF was no longer superior to EEG-NF. Studies of neurofeedback and other neurotherapies, such as acupuncture, benefit greatly from taking into account active elements in control conditions, as well as grading these active elements consistently. There has been a recent consensus statement on evidence-based ADHD treatments that excluded studies and meta analyses with non-active and heterogeneous controls such as waiting control or treatment as usual (Faraone et al., 2021). However, blinded raters may be able to detect some genuine NF-effects in real-world settings that may be overlooked by this method.
The patient’s preferences and the cost-benefit analysis are equally important considerations. Pharmacotherapy, as previously discussed, has a number of drawbacks, including side effects and inconsistencies in long-term outcomes. For six months after treatment, a recent meta-analysis examined whether EEG-NF had any long-lasting effects compared to non-active conditions, and found that it had small to medium positive effects when compared with non-active conditions and comparable positive effects with active conditions, such as pharmacotherapy [81]. (Van Doren et al., 2019). Thus, EEG-NF appears to have a delayed beneficial effect, as for example in a study where the superiority of stimulants over NF observed at treatment end [93] (Geladé et al., 2016) was no longer significant at the six-month follow-up, and ADHD core symptoms compared to a semi-active control condition (physical exercise) were similar at treatment end but were reduced with NF relative to the exercise control condition at follow-up [93] (Geladé Contrary findings from the largest study to date, which assessed longer-term EEG-NF effects, showed that although the improvement of ADHD core symptoms in comparison to the baseline remained large and stable after treatment at six months follow up [94] (Aggensteiner et al., 2019), it was no longer superior to a semi-active condition, suggesting considerable unspecific long-term effects.
Damien Sendler: There is a lot of debate about the specificity of EEG-efficacy. NF’s During the past decade, researchers have been attempting to separate the true effects of neuromodulation from the non-specific effects. A study by Strehl et al. [91] examined this issue by comparing an EEG-NF group with an EMG-BF semiactive control group and controlling for unspecific effects such as the high-tech training setting, interaction, learning, time, motivation and effort. The results showed that EEG-NF had a clinical advantage one month after treatment ended. Keeping an eye on these variables is critical, as the clinical effects of this type of time-consuming training could otherwise be attributed to non-specific psychosocial [95] factors. It is possible that treatments in high-tech settings, such as those used by Thibault and colleagues (Wood & Kober, 2018) or by Thibault et al. (2016) or Thibault et al. (2018) may have stronger placebo effects [96,97,98,99,100] (Thibault and colleagues (2016) or Thibault & Raz (2016) or Thibault and Raz (2018)) When conducting intervention research, the use of a sham-feedback condition is often considered the gold standard.
When comparing TBR-NF with a double-blind placebo group in the Collaborative Neurofeedback Group 36 study, the researchers selected only participants with an elevated TBR for the double-blind sham-NF placebo group, allowing for individualisation. After 13 months of follow-up, both groups showed large uncontrolled clinical effects and a reduced need for medication, but the EEG-NF group failed to demonstrate clinical superiority for EEG-NF despite more TBR learning (67 percent in the NF versus 59 percent in the sham group) [36]. Large nonspecific clinical effects have been observed in both groups, but the exact mechanism by which they occur is still unknown.
Damian Sendler
There should be a direct correlation between neuromodulation improvement and clinical improvement because the primary goal of neuromodulation is to help patients self-regulate their trained parameters. [101] (Zuberer et al., 2015) and complicated by delayed effects, as discussed above, or the indirect effects of effort and skill acquisition (Zuberer et al., 2015) (Gevensleben, Albrecht, et al., 2014). Even so, less than 70 percent of those treated with NF improved their ability to self-regulate [94]. A “dose-response” relationship between learned regulation and clinical improvement is only seen in about half of the studies (Aggensteiner et al., 2019) [103] Remarkably, (Drechsler and co-authors, 2007). After SCP-NF, three studies found a significant link between ADHD core symptoms and improved brain self-regulation [94,103,104]. (Aggensteiner et al., 2019; Drechsler et al., 2007; Strehl et al., 2006). Even though the results of some recent frequency-band NF studies were expected to show a link between self-regulation and symptom reduction following the end of treatment [36,105] (Arnold et al., 2020; Janssen et al., 2016), they were unexpectedly positive in the semi-active control group (Aggensteiner et al., 2019). Intriguingly, significant association was found at 6-month follow-up in the study by Arnold et al., 2020, suggesting a possible delayed effect. In order to distinguish between effects that are specific and those that are not, studies examining the links between brain activity and behavior are required. The effects of self-regulation have not yet been studied in depth enough to draw a firm conclusion.
Predicting who will respond to EEG-NF is an important consideration in the treatment process. Clinical responses to theta-modulating neurofeedback were predicted by increased theta activity, and stronger oscillatory parietal alpha activity and stronger task-related preparatory SCPs explained nearly 30% of the clinical outcome variance after SCP-NF [106,107]. (Gevensleben, Kleemeyer, et al., 2014; Gevensleben, Moll, et al., 2014; Wangler et al., 2011). However, these intriguing findings need to be replicated by other researchers.
Future research should investigate the specificity of self-regulation and the mechanisms that underlie individual clinical effects, taking into account reduced medication use and long-term improvement in ecological settings. If NF is tailored to the needs of each individual, it is not yet clear whether limiting TBR training to those with elevated TBR improves outcomes.
While many studies have examined the efficacy of EEG-NF in the treatment of ADHD over the past 45 years, there is still debate about whether or not “blinded” raters were used in some of these studies [32,80]. (Bussalb et al., 2019; Cortese et al., 2016). Furthermore, more research is needed to determine the precise effects of EEG-NF and the link between NF self-regulation and clinical improvement. To improve EEG-NF self-regulation and improvement over time, future studies should consider increased artefacts and altered reward learning in ADHD [118] (for example, Aase & Sagvolden, 2005) and further systematically investigate why some participants show low regulation performance in the current studies.
It’s still in its infancy when it comes to fMRI-NF and NIRS-NF research. Small proof-of-concept studies with fMRI- and NIRS-NF have yielded promising results. However, larger, double-blind, placebo-controlled randomised controlled trials are needed to further evaluate the potential efficacy of fMRI or NIRS-NF in ADHD treatment. – A lot is still unknown about the optimal protocol for rtfMRI neurofeedback or NIRS neurofeedback, such as the optimal target region for neurofeedback, the number and duration of neurofeedback sessions, whether self-regulation in specific regions can reach saturation or plateau, or how and which interindividual differences affect learning of brain self-regulation after how many sessions. It’s not clear how the transfer affects patient behavior in the clinic. Other questions remain unanswered, such as the best methods for reinforcing learning or developing cognitive skills in children using fMRI or NIRS-NF. It’s also not clear how regional fMRI-NF affects non-targeted cognitive functions and regions that aren’t self-regulated in neurofeedback studies. One must consider the potential costs of downregulating neighboring, interconnected, or contralateral brain regions as a result of self-regulation training in one brain region. When we studied the brain activity of adolescents with ADHD, we found that the active rIFC group had less parahippocampal control region activation than the control group, suggesting that the self-regulation of a particular region leads to the downregulation of other regions in the brain (Alegria et al., 2017).
It’s one of the most intriguing findings from the existing NF studies that there is evidence for longer-term delayed consolidation effects, which appear to be more pronounced at follow-up than at post-NF treatment assessment points [76–111]. However, one recent study showed no superiority over a semi-active control group at six months follow-up [94]. In this paper by Agensteiner et al. These effects of delayed consolidation support the idea that brain self-regulation via NF affects neuroplasticity and may therefore have unique long-term efficacy.advantages. In contrast to stimulants, which have no effect on neuroplasticity and may even lose their effectiveness over time [13,120], this would be a clear advantage. There is a possibility of brain adaptation [121] (Cortese et al., 2018) For more information, see Fusar-Poli et al. Neurofeedback has been shown to alter the structure and function of the brain in humans, including changes in cortical excitability and white matter tracts [122]. (Sitaram et al., 2017). However, it is unknown whether these modifications will remain stable over time. Neurofeedback therapies are likely to be popular due to their longer-lasting neuroplastic effects and apparent lack of side effects.
Damian Jacob Sendler
Only in the last decade have non-invasive brain stimulation therapies, such as rTMS and tDCS, been used to treat ADHD. When used in conjunction with the long-term potentiation (LTP) mechanisms that are mediated by GABA and glutamate, these stimulation techniques have the potential to have lasting effects on neurons. It was found that (Demirtas-Tatlidede et al., 2013). The effects of stimulation on cognition can last up to a year in healthy people and in patient groups [124,125]. Research has shown that (Ruf et al., 2017; Katz et al., 2017). When it comes to treating ADHD, positron emission tomography (PET) studies show that anodal frontal transcranial direct current stimulation (tDCS) can release neurotransmitters such as dopamine Better attention (Borwick et al., 2020) was also associated with these interventions (129). The effects on noradrenaline have been shown to be indirect (Fukai et al., 2019) [130,131]. (Adelhöfer et al., 2019). Animal and human studies have shown that rTMS over prefrontal regions alters neurotransmitter systems, including serotonin, dopamine release and metabolite levels in the striatum, as well as striatal glutamate release and concentration [132,133]. (Moretti et al., 2020; Poh et al., 2019). Furthermore, studies have shown that the synergistic effects of functional targeting (Cramer et al., 2011; Kuo & Nitsche, 2012; Ziemann & Siebner, 2008) make cognitive training and stimulation more effective together [132,133,134,135,136].
However, there is disagreement over whether or not blinded raters were used in many meta-analyses of EEG-NF randomized controlled trials, which consistently show small to medium effect sizes for symptom improvements. Cortese et al. (2016) found a significant increase in the number of patients with chronic obstructive pulmonary disease (COPD). The specificity of EEG-effects NF’s as well as its long-term efficacy need to be studied in greater detail. Precision medicine relies heavily on the study of factors that can predict individual responses.
It has only recently been piloted in ADHD studies using higher spatially resolved neuroimaging techniques such as NIRS and fMRI to show feasibility. But there are some promising results that need to be tested in a larger sample size. Smaller and more rigorous RCTs are needed to determine if neurofeedback training using NIRS or fMRI neurofeedback can be used to treat some individuals with ADHD. NIRS and fMRI neurofeedback protocols need to be thoroughly tested because optimal settings are unknown for either technology. Neurofeedback modalities have not tested for potential negative effects on non-regulated brain regions, but this knowledge is necessary for ethical reasons.
More than a dozen small studies of non-invasive brain stimulation for children and adults with attention deficit hyperactivity disorder (ADHD) have been conducted in recent years, most of them using either TMS or tDCS in either one or five sessions, focusing on the DLPFC or IFC. There hasn’t been much progress made with TMS studies. Studies of tDCS’ effects on the DLPFC show only small improvements in cognition, according to meta-analyses (Salehinejad et al., 2019; Westwood et al., 2021). Only three studies examined whether tRNS could help with clinical problems like inattention, and the results were mixed. TDCS has been shown to improve clinical and cognitive functions, but further studies are needed to determine how it affects non-targeted cognitive or behavioral functions.
Knowing how to best stimulate different patient populations is also needed, just like for fMRIs and NIRS-NFs (i.e., stimulation site, intensity, duration, frequency, electrode size, inter-electrode distance, etc.). ADHD patients may benefit more from brain stimulation and cognitive training than from brain stimulation alone. Cognitive training tasks that target ADHD-related functions must be developed in conjunction with brain stimulation techniques in order to be effective. There are few risks associated with using transcranial direct current stimulation (tDCS) or transcranial direct current stimulation (tRNS) to treat mental health issues that first manifest in childhood. Kruse and Kadosh (2013) This promise, however, needs to be thoroughly tested in large RCTs of various protocols in order to be taken seriously. As a result, there is a need to thoroughly investigate the potential costs of localized brain stimulation on non-targeted functions before it is applied to patients. Proof-of-concept studies using tRNS and TNS have shown promising results in improving ADHD symptoms, but further research is needed.
Also, the financial efficiency of various neurotherapies must be considered. As compared to tDCS or other brain stimulation devices (such as tACS, tRNS), TMS devices cost more than USD 50,000, which is significantly more expensive. In addition, more office space is needed for rTMS. The pay-per-use business model employed by many rTMS manufacturers is also very inefficient. It is more expensive to administer rTMS because a clinician/technician must receive a greater amount of training for rTMS than for tDCS. Even some TMS devices may require an MRI scan to locate them. As an alternative, because tDCS is compact, portable, and easy to use, you can buy one commercially and start using it right away, without the need for an outside therapy provider. When it comes to treating pain, cost-effectiveness analyses show that tDCS is more cost-effective than rTMS, with lower costs and greater efficacy [200] even when both techniques are used by professionals (Zaghi et al., 2018). In light of the fact that tDCS shows comparable, relatively small effects in improving ADHD symptoms or cognition, it may be more cost-effective than rTMS treatment. Because tDCS has fewer side effects and is easier to use, it is a better option for treating young children than other forms of electrical stimulation. In comparison to rTMS and tDCS, TNS appears to be the most cost-effective non-invasive brain stimulation treatment because of its higher efficacy in improving ADHD symptoms and the fact that it can be purchased commercially and used at home during sleep without the need for a therapist at a relatively low cost of about USD 1000 per device with additional electrode costs.
fMRI-neurofeedback devices are far more expensive than EEG-NF or NIRS-NF in terms of up-front costs, ranging from USD800–1000 for an hourly MRI scan to several million dollars for the device itself. EEG-NF certified equipment costs less than US$10,000. Evidence suggests that fMRI-NF learning is possible in fewer sessions than EEG-NF [96]. (Thibault et al., 2016). Full treatment costs for 30–40 sessions have been estimated at USD4000 to 6000, similar to pharmacotherapy over 5–10 years [201], for ADHD using EEG-NF, which typically requires 25-40 sessions of 45-60 minutes each (Garcia Pimenta et al., 2021). Compared to fMRI-NF, which requires fewer sessions, this means that the administering therapist will incur higher costs. To compensate for the higher costs of fMRI-NF, the lower session and therapist time costs compared to EEG-NF could be offset by qualified behavioural psychotherapeutic support during training, which may be critical for the clinical effects of NF to be maintained and transferred to the next patient. There are currently only a few centers equipped with the necessary hardware and software to conduct fMRI-NF or fNIRS-NF studies. Many private and some clinical centers around the world now offer EEG-NF, which has fewer exclusion criteria and is more widely available. Conclusions from cognitive neuroscience on ADHD have opened up translational neuroscience studies in an effort to use neurofunctional biomarkers as treatment targets for neurotherapeutics. For ADHD, neurotherapeutics appear to be a viable alternative to medication because of their lack of side effects and the potential for long-term neuroplastic effects, which medication cannot provide. Although neurotherapy has been proven effective in the short- and long-term, further testing is needed to determine the optimal “dose” (i.e., the optimal target site; the intensity of stimulation; frequency of stimulation/neurofeedback sessions), as well as the potential for individualised treatment based on clinical or cognitive ADHD subtypes, as well as the potential costs that accompany the benefits. Neurofeedback, brain stimulation, and medication are all likely to help ADHD patients in different clinical or cognitive subgroups, and establishing this knowledge is critical to the benefit of individual patients.