Recently, I was asked to contribute to a chapter on EEG Neurofeedback for a text on Psychiatry and Alternative Medicine to be published by the American Psychiatric Association Press. Since I had promised in my last blog to write more about EEG feedback, I have decided to post the chapter contribution here.
NEUROFEEDBACK IN A PSYCHIATRIC PRACTICE
Thomas M Brod, MD
Los Angeles, California
When I was first told about biofeedback by a nurse-practitioner my wife and I had met on a vacation in 1979, my mind was closed. I internally dismissed it as a cultural bloom, unreliable and over-hyped related to the “Test Your Biorhythm” devices then appearing in the foyers of coffee shops. That introductory conversation went on for several hours and when we stopped I was unconvinced that it could cure all manner of ills, but understood that machine-facilitated “relaxation training” was legitimate.
I understood that a mindful person could achieve a calm state to reverse stress, and that biofeedback techniques related to muscle relaxation, hand-warming and electrodermal response (GSR) sped up the acquisition of the ability of a person to control their physiology. I began to absorb the stress literature like a sponge (ah youth!) and actually realized that biofeedback really could help all manner of ills.
Within a year, I had opened a small multi-modality extension of my practice and called it “Self-Regulation Medical Group.” Another 18 months later I organized a major conference on Stress & Relaxation at UCLA at which many of these concepts were introduced to the local medical community. And less than ten years after that, I brought neurofeedback into my practice. Brainwave biofeedback as a field has continued to develop and change, as has my practice, ever since.
Brainwave biofeedback (“neurofeedback”) is self-regulation with a totally different paradigm than the relaxation-training model of peripheral biofeedback. Now we no longer speak of the “self” getting information about itself and learning to “let go” of intention (and stress). Now we speak of the brain—an information processing device of unimaginable complexity—getting information about itself…and changing itself with the acquisition of that information.
By shifting the explanatory metaphor from “Self” to “brain”, the ground has shifted. Consciousness is no longer a primary focus (although there are protocols for enhancing meditative practice just as there are for other forms of peak performance). Brainwave biofeedback is a treatment that, while depending on a sentient person being present, is typically passive for that person; the training happens to the person without consciously learning to change[1]
My current psychiatric practice stands on three legs, psychotherapy, medication and biofeedback. The first two are familiar to all; it’s biofeedback and the resultant need to prescribe less medication that often needs to be explained.
In the first decade of using neurofeedback, I trimmed back using peripheral feedback such as temperature, muscle tension, and electrodermal (GSR) feedback. Brainwave biofeedback is superior, offering treatment for the same wide range of stress-related difficulties even as it helps sharpen mental acuity; meditation skills are facilitated with alpha/theta feedback even as that submodality is helping people process traumatic events. However, I have in fact adopted a valuable form of peripheral feedback, heart-rate variability (HRV) into my practice because it provides a simple and reliable method for people to have something valuable to actively practice both while they are (passively) receiving neurofeedback and at home using a simple smart-phone app.
In my office, we use several different systems to give patients brainwave feedback. Each works, but each works differently, selecting a different aspect of brainwave monitoring to signal the brain about itself. And by in large each is mystifying in how its information leads to the alteration in mental functioning we see. As with many medications, we know fragments of mechanisms without a complete understanding of mechanisms of action. One of the things we know is that different proprietary neurofeedback systems offer feedback through different aspects of the EEG signal.
In the paragraphs that follow, I will be schematic, emphasizing differences in clinical applications for various systems. In truth, my clinical choices among the systems are intuitive and empirical but not based on head-to-head comparison studies, which are extremely rare. We psychiatrists are used to differentiating treatments based on small differences and personal clinical experience. Consider the differences in SSRIs and how you choose among them.
The earliest type developed, still in active use, separated out certain frequency bands for inhibition (e.g., 2-7 Hz) and other bands (e.g., 10-14 Hz and 14-18 Hz) to be rewarded. This is traditional neurofeedback, which tends to be site-specific. Wired sensors—often scarily introduced to patients as “electrodes”—are pasted to specific places on the scalp with the expectation that feedback of the selected cortical brainwave activity under the sensor will shift that brainwave activity. Feedback is typically from a screen in front of the patient displaying a representation of frequency and amplitude shifts hundreds of times a second; nowadays the display is often from any commercial DVD that distorts and clarifies according to undesired and desired brainwave activity.
These “traditional” neurofeedback systems are, of course, evolving with technology’s capability. Most systems now use multi-channel signal analysis. Some crunch massive databases to give real-time z-score comparisons to desired norms.
In my own practice, I do not use the complex z-score training, preferring feedback systems that are far simpler to administer and—as far as I have seen—equally effective. I do use a four-channel system with site-specific feedback to enhance communication across specific areas of the brain; the application is especially valuable for autistic spectrum and learning disorders. Incidentally, the latter applications are the only ones in which I have found consistent value in using a quantitative EEG (QEEG) to guide treatment protocols.
The field of EEG biofeedback is mature enough to have developed some surprising and highly efficient ways to feed information about the EEG signal back to the brain. Among our systems, we have one that analyzes tiny shifts in dominant frequency about 100 times a second. Another uses complex wavelet signal analysis to anticipate state shifts and “warn” the brain almost instantly. A third system, sometimes called Infra-Low Frequency (ILF), gives feedback on a very slow and long component of the EEG signal in the range of 0.0001 to 0.1 cycles per second (Hertz)—compare that to the alpha signal which is 9-10 cycles per second! We believe that ILF feedback is affecting key neural network processing through its effects on glial cells. Each of these systems is clinically effective for the wide range of indications neurofeedback has been discovered to help. As Siegfried Othmer has pointed out, the brain has an extraordinary innate capacity to discern its own activity when “shown” many different aspects of the brainwave signal.
Three brief clinical vignettes of hundreds possible follow. I have chosen these cases because in recent months I have been able to collect subjective symptom tracking data through an individualized and automated secure on-line data collection system.
A 48 year old woman consulted me after she had heard a voice saying that she must suffer again. She recognized an incipient return of recurrent “psychotic depression” and insisted that she really did not want to take drugs again if at all possible, and I agreed to see if that were possible. She had two weekly sessions of psychotherapy with me and concurrently began neurofeedback sessions. In seven weeks (10 neurofeedback sessions), her symptom ratings shifted dramatically. On ten symptoms she had chosen here severity fell from an average of 8 to 4.7 (her depression ranking went from 9 to 4; her anxiety and sleep measures dropped from 8-4; at this time, however, her lack of pleasure and low self-esteem only shifted from 8 to 5 and 6 respectively). At this writing she is still getting weekly neurofeedback.
An eight year-old girl with mild Pervasive Developmental Disorder has been in twice-weekly neurotherapy (only) for behavioral and academic difficulties. Her mother listed the top ten targets for treatment. In 15 sessions over 11 weeks, her mother has noted a drop in “Inflexibility” from 7 to 4; in “Agitation” from 8 to 4; in “Anger” from 7 to 4; and a lowering of oppositional or defiant behavior from 5 to 2. We have noted, however, that her academic performance has not improved and are shifting our protocol from 2-channel ILF to 4-channel coherence training for learning difficulties.
A hyperactive and anxious boy of ten with ADHD has had 16 sessions over 9 weeks. According to his mother’s ratings, “Inflexibility” has dropped from 8 to 3; “Hyperactivity” has dropped from 6 to 2; “Rages” have gone from 6 to 3; “Impulsivity” has fallen from 8 to 3. Six measures of attention have dropped (“Difficulty Following Directions”, “Difficulty Organizing”, “Not Listening”, “Poor Concentration”, “Sustained Attention”, and “Unmotivated.”), from an average of 5.83 to an average of to 1.83.
There are no “typical” patients and no “typical” success with neurofeedback. These cases are “ordinary” in the sense that we often, but not always, have success comparable to these cases.
Follow-up hint: One site that is particularly rich in neurofeedback resources is www.EEGInfo.com. See the pull-down menus for “What is Neurofeedback” and “Research”.
[1] At a deeper level, one could accurately say that old-fashioned peripheral biofeedback (the original relaxation training) is also about getting conscious intention out of the way. The key difference really is that the original notions of biofeedback were based on training of consciousness to move away from intention, and modern neurofeedback is more intentionally passive and the training is not mediated by consciousness (technically, by mixing both operant and classical conditioning and adding an exercise model of re-functioning plasticity).