Advances in Assessment and Treatment of ADHD Using Network Analyses
The purpose of this review is to discuss how new advances in neuroimaging and functional network analyses are applied to electroencephalography (EEG) biofeedback or neurofeedback. Clinical efficacy of one or a few scalp EEG recordings used in the treatment of attention-deficit hyperactivity disorder (ADHD) has been repeatedly demonstrated over the past 34 years. However, a problem is that improved clinical outcome often requires 40 to 80 sessions, which is expensive and difficult for patient compliance. This review cites the scientific literature of direct measures of the nodes and connections between nodes in the attention and default mode networks that are correlated with ADHD using functional magnetic resonance imaging, positron emission tomography, and EEG inverse solutions such as low-resolution electromagnetic tomography. Three-dimensional EEG biofeedback that targets dysregulation in Brodmann areas of the attention and default networks provides increased specificity and can result in improved clinical outcome in fewer sessions.
Top row is conventional or standard electroencephalography (EEG) biofeedback in which different units of measurement are used in an EEG analysis (e.g., uV for amplitude, theta/beta ratios, relative power 0% to 100%, coherence 0 to 1, phase in degrees or radians, etc.), and the clinician must “guess” at a threshold for a particular electrode location and frequency and age to reinforce or inhibit a give measure. The bottom row is Z score biofeedback, in which different metrics are represented by a single and common metric (i.e., the metric of a Z score) and the guess work is removed because all measures are reinforced to move Z scores toward Z = 0, which is the approximate center of an average healthy brain state based on a reference age-matched normative database in real time. The goal is to reinforce increased stability in brain networks linked to symptoms.
Brodmann areas (nodes) and connections between Brodmann areas that constitute the dorsal attention network. The Brodmann areas are located in a three-dimensional rendered magnetic resonance image based on the Talairach atlas coordinates that correspond to the dorsal attention network. The network nodes and connections between nodes are colored when a subject fails to reach a Z score threshold during real-time low-resolution electromagnetic tomography Z score neurofeedback (from the Neuroguide [version 2.7.9] BrainSurfer neurofeedback program).
Brodmann areas (nodes) and connections between Brodmann areas that constitute the ventral attention network. The Brodmann areas are located in a three-dimensional rendered magnetic resonance image based on the Talairach atlas coordinates that correspond to the ventral attention network. The network nodes and connections between nodes are colored when a subject fails to reach a Z score threshold during real-time low-resolution electromagnetic tomography Z score neurofeedback (from the Neuroguide [version 2.7.9] BrainSurfer neurofeedback program).
Brodmann areas (nodes) and connections between Brodmann areas that constitute the executive attention network. The Brodmann areas are located in a three-dimensional rendered magnetic resonance image based on the Talairach atlas coordinates that correspond to the executive attention network. The network nodes and connections between nodes are colored when a subject fails to reach a Z score threshold during real-time low-resolution electromagnetic tomography Z score neurofeedback (from the Neuroguide [version 2.7.9] BrainSurfer neurofeedback program).
Brodmann areas (nodes) and connections between Brodmann areas that constitute the default mode network (DMN). The Brodmann areas are located in a three-dimensional rendered magnetic resonance image based on the Talairach atlas coordinates that correspond to the DMN. The network nodes and connections between nodes are colored when a subject fails to reach a Z score threshold during real-time low-resolution electromagnetic tomography Z score neurofeedback (from the Neuroguide [version 2.7.9] BrainSurfer neurofeedback program).
Phase shift durations between Brodmann areas that comprise the default mode network in the x, y, z low-resolution electromagnetic tomography time series directions and the resultant vector in the lower right. The x-axis is the phase shift duration in milliseconds, and the y-axis is the percentage of subjects who exhibited a given phase shift duration for different Brodmann area pairs. The solid line is the eyes-closed condition, and the dashed line is the eyes-open condition. All of the subjects (N = 70) are represented within each curve. For example, 100% of the subjects exhibited a phase shift duration between 18 ms and 35 ms for Brodmann areas 8 and 9 (upper left panel) and similarly for each Brodmann area pair. From Thatcher (2012).
Phase lock durations between Brodmann areas that comprise the default mode network in the x, y, z low-resolution electromagnetic tomography time series directions and the resultant vector in the lower right. The x-axis is the phase shift duration in milliseconds, and the y-axis is the percentage of subjects who exhibited a given phase shift duration for different Brodmann area pairs. The solid line is the eyes-closed condition, and the dashed line is the eyes-open condition. All of the subjects (N = 70) are represented within each curve. For example, 100% of the subjects exhibited a phase shift duration between 250 ms and 500 ms for Brodmann areas 8 and 10 (upper left panel) and similarly for each Brodmann area pair. From Thatcher (2012).
The x-axis is the Euclidean distance between the center voxels of Brodmann areas that comprise the default mode network. The y-axis is the phase shift duration (top) and phase lock duration (bottom). The left row is the left hemisphere Brodmann areas, and the right row is the right hemisphere Brodmann areas. The red line is the fit of an exponential equation T = b1 +
, where T = duration time (ms), d = distance between Brodmann areas (mm), and b1, b2, and b3 are coefficients. R = regression correlation, and p = statistical probability. Phase shift and phase lock are inversely related where Brodmann areas with short phase shift duration exhibit long phase lock durations while Brodmann areas with short phase lock durations exhibit long phase shift durations. From Thatcher (2012).
Robert Thatcher
Contributor Notes