The good news is that an avalanche of scientific discoveries are pouring out the brains of clinicians who treat serious psychiatric / psychological disorders. One of those disorders is called Attention Deficit / Hyperactivity Disorder (ADHD) with increased precalency including ADHD ADHD. The primary treatment for ADHD is medication, especially psychostimulants. However, around 20% of all children with ADHD do not respond to psychostimulants .. Also many children and adults have serious side effects from stimulant medications.
A non-medication treatment with no side effects is EEG biofeedback that reports improved clinical outcome after about 40 to 80 sessions for about 80% of ADHD patients. More good news is that because of advances in modern neuroscience it is possible to achieve clinical improvement in fewer than 20 sessions. For instance, real-time fMRI biofeedback can achieve modification of blood flow in specific brain regions in one 20 minute session (eg, anterior cingulate gyrus ,.
The reason that a single 20 minute session achieves clinical change is because of the ANATOMICAL specificity of the fMRI method where a part of the brain linked to the patient's symptoms is targeted for biofeedback or neurofeedback (NFB). In contrast, surface EEG is diffuse and less specific, for example, the EEG recorded from a single scalp location (eg, Cz) senses sources from widespread regions of the brain and is a mixture of many different frequencies. Surface Z scores improve specificity by isolating dysregulated locations and rhythms, especially when using the Laplacian transform and Low Resolution Electromagnetic Tomography (LORETA) Z scores (ie, real-time or "live" comparisons to a normative database). For example, LORETA Z score biofeedback also often produces results in one 20 minute session because EEG source localization has accuracies similar to fMRI of about 1 cm to 3 cm and there is more more specific than is surface EEG.
Brain Networks and ADHD
Modern neuroscience using different measurement modalities, including functional MRI (fMRI), PET, EEG / MEG, Diffusion Tensor Imaging (DTI), neuropsychological, genetics, and neurochemical studies, has implicated the attention network, default mode network (DMN) and fronto- striatal network abnormalities as contributing to ADHD. Functional imaging studies on ADHD, in particular, have increased almost exponentially over the past decade. A current PubMed search combining the terms 'ADHD and imaging' returned 1,259 results, with more than 100 papers published in the last year alone. Given this vast body of research on neuroimaging of ADHD, this paper can not be comprehensive, nor will it critique individual articles. Instead, it will focus on new advances in understanding of brain networks and connections between nodes of networks within the ADHD literature. It will then seek to place this new information in the context of recent advances in EEG biofeedback in which dysregulation in nodes and connections in relevant brain systems can be made the target of real-time biofeedback with the goal of reinforcing stability and increased information processing in these networks.
Attention Network
Attention is a coercive aspect of consciousness and involves selecting a small subset of stimuli from a larger universe for entry into consciousness and the creation of new memories and new actions. Neuroimaging research on attention has demonstrated three networks: 1 – alerting responses, 2 – orientation responses and 3, executive attention. The alerting network is based in the brainstem with emphasis on the locus coeruleus, cerebellum, basal ganglia and regions of the frontal and parietal cortex. The orienting network consists of a dorsal and ventral branch where the dorsal network consists of the frontal eye fields (FEF) and the intraparietal sulcus and superior parietal lobe (IPS & SPL). The ventral attention network consiststs of parts of the temporal-parietal junction (TPJ) and the ventral frontal cortex (VFC). The executive attention network networks of the cingulate cortex-operational frontal cortex and the fronto-parietal system.
Real-time LORETA Z score EEG biofeedback is used to target current source density in the attention networks depending on the indicators of dysregulation based on age matched Z scores. The goal is to reinforce stability within the various aspects of the human attention network depending on the patient's symptoms. Figure 2 is a Talairach atlas representation of the Brodmann areas (nodes) and connections between Brodmann areas of the dorsal attention network that is used in LORETA Z score NFB.
Default Mode Network
When one is at rest and not engaged in a task and instead ruminating about the past and future (eg, self-narrative) then it is during these reflective moments that there are changes in the synaptic synchronization of millions of neurons connected in a network called the 'Default Mode Network' (DMN). The DMN is suppressed when the attention network is active or anti-correlated with the attention network and is promulgated primarily by the posterior cingulate gyrus, hippocampus, medial frontal lobes, temporal lobes and parietal lobes with approximately five times the number of synaptic connections than any other cortical network. Reciprocal inhibitory and excitatory dynamics occurring in the sub-second time domain that only qEEG is capable of detection in contrast to eye-ball visual examination of EEG traces.
The default mode network is what one's brain does when not engaged in specific tasks. It is the busy or active part of our brains when we are mentally passive. The "DMN is seen to collectively compose an integrated system for autobiographical, self-monitoring and social cognitive functions." It has also been considered as responsible for REST (rapid episodic spontaneous thinking). That is, the spontaneous mind wandering and internal self-talk and thinking that one engages in when not working on a specific task or, when completing a task that is so automatized (eg, driving a car) that our mind starts to wander and generate spontaneous thoughts. It is likely that people differ in the amount of spontaneous mind wandering (which can be both positive creative thinking or distracting thoughts), with some having a very unquiet mind that is hard to turn off, while others can turn off the inner thought generation and self-talk and display tremendously self-focus or controlled attention to perform a cognitively or motorically demanding task. The anterior insula and anterior cingulate gyrus are referred to as the salience network that is involved in switching the DMN and Executive Attention network on and off in a reciprocal manner.
In evaluating and refining attention deficit disorder the link between the attention network (AN) and the default mode network (DMN) is important. For example, imagine a child sitting in a class room attending to a lecture when suddenly a thought from the past intrudes into consciousness, it could be something a parent said or someone else in the past said and there is an attention shift in mood and conscious content away from the lecture to a self-narrative mediated by the DMN. After a few moments one shifts attention back to the lecture but now some of the lecture is missed and it may have been an important part of the lecture. Excessive intrusion of self- narrative memories and thoughts is an example of why both the AN & DMN as well as the insular cortex involved in switching between the two systems may need to be included in a neurofeedback protocol. It is important for the clinician to ask questions about the nature of the problems the client has with sustaining or shifting attention to better predict which aspects of the attention dynamic may be dysregulated in order to select an optimal protocol for LORETA Z score neurofeedback.
Standard surface EEG measures of connectivity such as coherence, phase differences, co-modulation and phase reset have also been applied to Brodmann areas and 3-dimensional networks using Low Resolution Electromagnetic Tomography (LORETA). As mentioned previously, direct measurement of network connectivity has provided increased specificity and improved clinical output in fewer sessions using 3-dimensional LORETA Z score NFB. Recently, the ability to use fundamental network metrics such as phase lock duration and phase shift duration between Brodmann areas complying the attention and default networks has been successfully used for NFB. Figure 6 is an example of phase shift duration between some of the Brodmann areas that represent the default mode network and emphasize the increased temporal and anatomical specificity when using 3-dimensional EEG source analyzes like LORETA.
EEG phase lock duration is correlated approximately 0.86 with EEG coherence, however, the 3-dimensional measurement of phase lock duration between Brodmann areas produces more discrete and precise ranges of phase lock than is observed at the scalp surface. At the level of 3-dimensional imaging there is increased temporal discreteness and temporal precision of phase locking observed with LORETA analyzes of time series from different Brodmann areas that combine the DMN. Also there is a systematic relationship between the duration of phase shift and phase lock and the distance between Brodmann areas as nodes of networks. A frequently heard view is that neurons that are wired together fire together and dysregulation in networks is indicated by either too much or too little functional connectivity involved in neural resource recruitment and resource allocation for brief periods of time. Optimal information processing in networks can now be restructured using NFB at spatial and temporal resolutions that are much greater than in studies that use only one or two scalp electrodes.
Conclusions
The goal of this brief review was to introduce the reader to new advances in EEG network analyzes as applied to EEG biofeedback of the attention and default networks in patients with ADHD. Improvement clinical outcome in fewer sessions can be achieved by thorough clinical assessment that links symptoms to networks of the brain related to the symptoms followed by reinforcement of stability and increased information processing between nodes of the relevant dysregulated networks. The nearly 60 years of science in support of the mechanisms of operant conditioning and the application of operant conditioning to treatment ADHD have reached a level of concern where clinicians can use affordable EEG technology to treat patients and achieve clinical clinical outcomes in fewer sessions than the case in the past. It is important to continue to use modern neuroscience of brain networks for both assessment and treatment while at the same time adhering to the fundamental principles of operant conditioning.