Research on EEG Neurofeedback for Anxiety

Society for Neuronal Regulation
9th Annual Conference
Monterey, CA 27-30 October 2001

Anterior Alpha Asymmetry in Anxiety and Depression

Robert Lawson and Eugenia Bodenhamer-Davis
University of North Texas, Denton, TX

Many studies have found that people with depression have less activity in the left frontal region than healthy normals. An EEG correlate of this relative reduced activity is left side alpha asymmetry. It is not known, however, if there are frontal alpha asymmetry variations between individuals with depression only compared to persons with anxious depression.

This exploratory study compared QEEG records of frontal alpha of a group of seven depressed persons with those of a group of nine anxious depressed. Based on clinical observations, it was hypothesized that the depressed group would have more left side alpha asymmetry than the anxious depressed group.

Membership in the depressed group was defined as having an MMPI2 Scale 2 Depression score => 60 and a Scale 7 Psychasthenia score =< 60. The anxious depressed group membership consisted of individuals whose MMPI2 Depression scores were => 60 and Psychasthenia scores >60. Asymmetry was calculated with the following equation: (F3-F4)/(F3+F4) (magnitude, not power, voltage was used). EEG data were recorded on a Lexicor NRS 24 with a linked ear reference and remontaged to a Cz reference.

Results of comparing the QEEG records of the two groups of individuals revealed that 85% of the depressed only group and 55% of the anxious depressed group had left side asymmetry. The depressed only mean asymmetry = -.043; the anxious depressed mean asymmetry = -.0052. The depressed only group had significantly more left side asymmetry (p=. 023). A surprising finding was that the anxious depressed group had higher alpha on both sides than did the depressed only group (p=. 029). The mean frontal alpha, (F3 alpha + F4 alpha)/2, of the depressed only group was 11.8, and 18.0 for the anxious depressed group.

This increased alpha suggests that, relative to the depressed only group, persons who are anxious depressed have reduced frontal activation in both hemispheres.

Roshi Compared with the Rosenfeld Depression Protocol: A Case Report

D. Corydon Hammond, Ph.D., University of Utah, School of Medicine

The Rosenfeld depression protocol for modifying frontal alpha asymmetry is soundly based theoretically on substantial research that finds an alpha asymmetry in depression. Preliminary results on its use are quite encouraging. There are, however, other ways of correcting such a frontal alpha asymmetry and treating depression through neurofeedback. A case report will detail the treatment with neurofeedback of a severely depressed man. His alpha asymmetry was congruent with experimental evidence as measured with the Rosenfeld protocol.

In his initial three sessions following informed consent and using the Rosenfeld protocol, he was, however, unable to demonstrate any improvements during the course of the session. In fact, the asymmetry increased in two of the three sessions. Subsequently, we switched treatment and began using the Roshi, a neurofeedback system utilizing photic stimulation which can be set so that the frequency of photic stimulation varies depending on the patient's existing dominant brainwave.

Electrodes continued to be placed at F3 and F4, and we were reinforcing 15-18 Hz with the patient simultaneously seeking to inhibit alpha and theta frequencies. For the initial 8-10 minutes of a session, we used "complex-adaptive" light stimulation. This consisted of feeding back to the left eye (through varying the frequencies of photic stimulation) information from the F3 (left) electrode site, and vice versa.

The theory is that each eye sees separate EEG information which causes the brain to seek to correct for the apparent error, increasing brain flexibility. It is usually found that the result is a flattening of the spectrum. Following this brief stimulation, the photic stimulation was set to vary so as to pulse on the peak frequency within the 15-18 Hz range, while the patient continued to seek to inhibit slow frequencies and to enhance this beta activity.

Results were rapid and fairly dramatic. After 5 sessions the patient was reporting that he didn't really feel depressed any longer. He received subsequent reinforcing sessions, periodically having him simultaneously hooked to a Lexicor unit, measuring alpha asymmetry using the Rosenfeld protocol, without receiving any feedback linked to the Rosenfeld protocol.

These periodic EEG follow-up evaluations revealed positive physiologic changes in baseline and training measures of alpha asymmetry references to Cz. Psychometric evaluation with 5 measures likewise confirmed the effectiveness of treatment. It thus appears that other neurofeedback protocols are capable to effectively treating depression.

QEEG Relationships With the MMPI-2 Depression Scale and Subscales

Roger deBeus, M.S. University of North Texas, Neurotherapy Lab

Introduction: Depression is a heterogeneous disorder. One demonstration of this is the different classes of antidepressants used to treat it, many with different modes of action. However, most published literature examining QEEG and depression have used measurements with only a single composite score (e.g., the Beck Depression Inventory, Hamilton Depression Rating Scale), and combined genders.

One benefit of utilizing the MMPI-2 is that it considers several factors contributing to depression symptomology. For example, the Harris and Lingoes (1955) subscales include Subjective Depression (D1), Psychomotor Retardation (D2), Physical Malfunctioning (D3), Mental Dullness (D4), and Brooding (D5). Examining these scales in relationship to QEEG measures will help generate a better understanding of this multifaceted disorder.

The hypothesis for this study was that males and females would have different QEEG profiles in relation to their depression scores.

Methods: Participants were part of the patient pool who presented to the UNT Neurotherapy Lab for treatment between 1996 and 1999. As part of the intake procedure, each client participated in a QEEG and completed an MMPI-2. QEEG was collected on the Cadwell Spectrum 32 digital EEG machine. One of the QEEG conditions collected was eyes-closed, which was then analyzed through the NYU database.

Resulting QEEG parameters included absolute power, relative power, asymmetry, and coherence referenced to the ears. Bandwidths examined were Delta (1.5-3.5 Hz), Theta (3.5-7.5 Hz), Alpha (7.5-12.5 Hz), and Beta (12.5-25.0 Hz). The MMPI-2 scales considered for this study included clinical scale Depression, content scale Depression, D1 (Subjective Depression), D2 (Psychomotor Retardation), D3 (Physical Malfunctioning), D4 (Mental Dullness), and D5 (Brooding).

Twenty males and 20 females were analyzed separately by gender. Nonparametric correlations were performed between QEEG parameters and MMPI-2 depression scales.

Findings: The analyses identified distinctive patterns for each QEEG parameter and depression scale for both genders. Overall, the females showed almost four times more significant correlations than the males. For QEEG parameters, the significant findings for females showed more correlations in relative power in the Theta and Alpha bands, and coherence in the Alpha and Delta bands.

Results with the males showed more significant associations with asymmetry involving the Delta, Theta, and Alpha bands. For the depression scales, the females showed the most significant correlations with D4, D5, D1, Depression, and content Depression in descending order. The males showed the most significant correlations with subscales D1, D5, and D3.

More specifically, QEEG correlates with the clinical Depression scale for the female group showed absolute power in frontal Alpha, relative power in frontal Theta and Alpha, and frontal, temporal, and occipital Beta, asymmetry in occipital Beta, coherence in temporal and parietal-occipital Delta, temporal Theta, frontal and frontal-occipital Alpha, and central-parietal Beta. The males showed asymmetry in frontal Theta and temporal Alpha.

Correlations with the content Depression scale for the females showed relative Theta over most of the head; asymmetry in temporal Delta, temporal Alpha, and frontal, central, and temporal Beta; coherence in frontal-temporal Delta, and parietal-occipital and frontal-occipital Alpha. The male group showed relative power in frontal Alpha, and asymmetry in parietal Delta and frontal Theta.

QEEG correlates with the Subjective Depression (D1) subscale for the females showed absolute power in parietal and occipital Theta; relative Theta over most of the head; asymmetry in temporal Alpha and frontal Beta; coherence with temporal and frontal-occipital Delta, frontal-occipital Alpha, and central-parietal Beta. The male group showed correlations with absolute power in parietal and temporal Beta, relative frontal Beta, and asymmetry in frontal Delta, Theta, and Alpha.

Correlations with the Psychomotor Retardation (D2) subscale for the female group showed coherence in frontal Delta and frontal-occipital Alpha. The males showed absolute parietal Delta, asymmetry in temporal Alpha and frontal Beta, and coherence in central Beta.

The Physical Malfunctioning (D3) subscale for the females was associated with relative temporal and occipital Beta; and coherence in frontal, temporal, and parietal-occipital Delta, and frontal-occipital Alpha. The male group showed relative frontal Alpha and Beta, asymmetry in frontal Delta, and coherence in frontal Alpha and Beta.

The Mental Dullness (D4) aspect of depression for the females was correlated with absolute frontal and central Alpha; relative Theta and Alpha over the entire head; asymmetry in frontal Beta; coherence in frontal-temporal Delta, and frontal and frontal-occipital Alpha. The male group showed associations with absolute parietal Delta, and relative frontal Alpha.

The Brooding (D5) component of depression for the female group was related with absolute occipital Theta; relative Theta over most of the head; asymmetry in temporal Delta, frontal, temporal, and occipital Alpha, and frontal and central Beta; coherence in frontal-temporal and frontal-occipital Delta, temporal Theta, frontal-occipital and temporal Alpha, and central-parietal and temporal Beta. The male group showed absolute parietal and temporal Beta, and asymmetry in central, parietal, and temporal Delta, and frontal Theta and Alpha.

Discussion: The results showed (1) unique relationships between QEEG parameters and the depression scales of the MMPI-2, and (2) male and female differences. Each cluster of depression questions elicited a different QEEG profile. Examining subscales or items can aid in the interpretation of diverse depression symptomology and QEEG correlates. Furthermore, this suggests that treatment options are also heterogeneous and that careful assessment should be done to determine which factors contribute most to a client's depression.

Therefore, appropriate electrode sites and modalities can be selected for EEG biofeedback treatment. Another finding was that the males and females had different QEEG profiles for each depression scale although the MMPI-2 profiles were relatively homogeneous for both groups. One interpretation may be that male and female brains experience depression in a different way.

The presence of four times more correlations of the female group compared with males corresponds to the gender effect of two to three times more depression for women found in the general population. Although these findings are preliminary, they support the heterogeneity and complexity of depression.

For additional information, call the
Attention & Achievement Center at 925-280-9100

Disclaimer: The information presented here is for educational purposes only. It is not intended to replace the expert and professional advice of your physician, psychologist, or therapist. Always seek help from qualified professionals in the field of your interest. Our treatments are considered complimentary or alternative to traditional pharmacology and are not licensed or endorsed by the State of California, nor are we licensed healing arts practitioners by the State.