295

Treatment of chronic fatigue with

neurofeedback and self-hypnosis
D. Corydon Hammond
Department of Physical Medicine & Rehabilitation,
University of Utah School of Medicine, Salt Lake City,
UT, USA
A 21 year old patient reported a relatively rapid onset of serious chronic fatigue syndrome (CFS), with her worst symptoms being cognitive impairments. Congruent with research
on rapid onset CFS, she had no psychiatric history and specialized testing did not suggest that somatization was likely.
Neuroimaging and EEG research has documented brain dysfunction in cases of CFS. Therefore, a quantitative EEG was
done, comparing her to a normative data base. This revealed
excessive left frontal theta brainwave activity in an area previously implicated in SPECT research. Therefore, a novel
treatment approach was utilized consisting of a combination
of EEG neurofeedback and self-hypnosis training, both of
which seemed very beneficial. She experienced considerable
improvement in fatigue, vigor, and confusion as measured
pre-post with the Profile of Mood States and through collaborative interviews with both parents. Most of the changes
were maintained at 5, 7, and 9 month follow-up testing.
Keywords: Chronic fatigue, EEG biofeedback, neurofeedback, hypnosis

1. Introduction
Chronic Fatigue Syndrome (CFS) is characterized by
fatigue so severe that 43% of patients are unable to attend school or work [18] and as many of 85% of patients
report cognitive impairments [12,16]. Cognitive impairments usually include concentration/attention and
memory [5,7,8,17,21,23,25]. No consistent pathogenic
mechanism has been identified as a cause, and, thus,
its existence is often regarded as controversial and it
Address for correspondence:

D. Corydon Hammond, Ph.D., University Medical Center, PM&R, 50 No. Medical Dr., Salt Lake City,
UT 84132, USA. Fax: +1 801 585 5757; E-mail: D.C.Hammond@
m.cc.utah.edu.
NeuroRehabilitation 16 (2001) 295300
ISSN 1053-8135 / $8.00 2001, IOS Press. All rights reserved

is often considered nothing more than somatization.

Duffy [9], a pediatric neurologist at Harvard, has indicated that CFS may be associated with several possible etiologies, including viral and bacterial infections,
surgery, mild head injury, and toxic exposure. Thus, it
appears to be associated with a heterogeneous group of
factors [14].
CFS patients are often diagnosed as having somatization disorder (SD) when, in fact, very few CFS patients
seem to fit the strict definition for SD [15], and most
CFS patients endorse SD symptoms only after the onset
of CFS. Johnsons and DeLucas [7,15] research suggests that psychiatric factors alone cannot sufficiently
explain the broad range of somatic symptoms reported
by CFS patients but labeling of a condition is not
trivial because it can affect treatment services [15,
p. 56].
DeLuca et al. [6] found that CFS patients who have
a gradual onset of the illness tend to be more likely
to have a comorbid psychiatric diagnosis. This finding is supported by research by Mawle et al. [19] who
found CFS patients with a gradual onset had significantly more major life events occurring in the year prior
to onset compared to CFS patients with a rapid onset.
Similarly, patients with multiple chemical sensitivities
who do not have a clear date of onset are more likely to
have a psychiatric diagnosis (62%) than those with an
identifiable onset (26%) [10]. The Johnson and DeLuca
group seems to believe that patients with a more rapid
onset of symptoms are more likely to be associated
with a disease entity, such as a virus (e.g., Epstein-Barr
virus, enteroviruses, a putative human T lymphotropic
virus type III) that may eventually be isolated.
Some studies have suggested that an encephalitic
process may be involved with CFS [2,13,20,24]. For
example, Schwartz et al. [24] found that CFS patients
had significantly more defects (81%) throughout the
cortex on SPECT scans than did normal subjects (21%).
The most common sites of defects in CFS patients,
and the only ones that were significantly different from
normal control subjects, were in the lateral frontal cortex, lateral temporal cortex, and basal ganglia. The

296

D.C. Hammond / Treatment of chronic fatigue with neurofeedback and self-hypnosis

CFS patients had more than 10 times the number of

defects as control subjects. These authors concluded
that complaints of afflicted patients, particularly those
involving the CNS, can be misdiagnosed or even considered by some to be factitious. The finding of abnormal neuroimaging studies in the vast majority (94%) of
patients with CFS indicates, however, that this condition is associated with physiologic changes that can be
observed objectively (p. 939). They cite that CFS may
be due to a virus or to vascular problems. The frontal
and temporal lobes were also implicated by Ichise et
al. [13] and Goldstein et al. [11]. Ray et al. [21] found
hypoperfusion in the basal ganglia and Costa et al. [3,
4] in the brainstem.
Unpublished data on QEEGs on Duffys [9] patients
have found that many of them have high amplitude
sharp EEG alpha, scattered sharp waves, and a lack
of any tendency to fall asleepunusual in a syndrome
associated with fatigue . Duffy expresses the belief
that CFS appears to involve the brain and result in
changes in its electrical activity. The fact that QEEG
findings were reasonably consistent across most CFS
patients suggests that brain involvement may be much
more common than thought. In 28 females with CFS,
Billiot et al. [1] also found increased 57 Hz theta at
Cz compared to age matched controls.
DeLuca et al. [7,8] have found that CFS patients with
a sudden onset experience more difficulties with cognitive impairment (e.g., verbal memory, attention, and
information processing) than patients with a gradual
onset, than CFS patients with psychiatric comorbidity,
and than controls. The findings by DeLuca et al. [7,
8] have important implications for treatment. CFS
patients with gradual onset and a psychiatric history
may well benefit more from psychotherapeutic and psychotropic medication management. On the other hand,
patients with a more rapid onset and without a history
of psychiatric comorbidity may benefit more from education about symptom management and from cognitive
rehabilitation, including EEG biofeedback (neurofeedback).
Neurofeedback is considered a controversial alternative medicine practice by many professionals because of the limited amount of controlled research
that exists thus far. However, Stermans [26] review
of neurofeedback research with uncontrolled epilepsy
concludes that neurofeedback treatment of uncontrollable epilepsy is not experimental. The peer-reviewed
studies include randomized, controlled trials, including placebo controls. Sterman documents that both
immediate and sustained functional changes that are

consistent with an elevation of seizure thresholds occur

in brainwave patterns following neurofeedback training. In addition, the research finds that 82% of subjects across studies experienced a significant reduction
of seizures. We thus know that neurofeedback can produce physiological changes in brain activity. Moderately good research also exists on neurofeedback applications to ADD/ADHD, but only two case reports have
ever been published concerning the use of neurofeedback with chronic fatigue. Thus, its application in this
area must be regarded as experimental.

2. Case background
The patient was a twenty-one year old, single female who experienced a relatively sudden onset of CFS
and who had no prior history of psychiatric comorbidity. Three years prior to seeing me, she had contracted
mononucleosis while in her senior year of high school.
She was an A student and graduated with honors, despite being unable to attend school for a while. The
following year she was diagnosed with CFS by an internist. She appeared to be a very intelligent young
woman, perhaps with a type-A personality.
The hardest symptoms for the patient were the cognitive impairments. She described problems with concentration and memory which had kept her from attending university. Her mother indicated that not being able to think is her biggest symptom. On a rating scale she indicated that since contracting CFS she
started projects but tended not to finish them, was easily
distracted, did not get much done, and had a short attention span. She also experienced sensitivity to noise,
and light sensitivity. For 24 hours each day she had
enough energy to do some things (e.g., go to the store),
but then she would get a little drifty mentally. Someone else would usually need to drive here anywhere that
she wanted to go because she felt so easily distracted.
As part of a thorough evaluation, the patient was
administered the Profile of Mood States, and a battery
of tests associated with the high risk model of threat
perception and evaluation of somatization [27]. She
scored in the average range (6) on the Harvard Group
Scale of Hypnotic Susceptibility, not displaying either
the high or low hypnotizability that appears to predispose someone to somatization. She also scored low
on other predisposing factors such as catastrophizing
(measured by the Dysfunctional Cognitions Inventory),
and negative affectivity (measured by the Eysenck Personality Inventory). She did appear prone to repres-

D.C. Hammond / Treatment of chronic fatigue with neurofeedback and self-hypnosis

sion, scoring at the 99th percentile on the L scale of the

Eysenck, but her score was only slightly elevated on
another measure of repression, the Marlowe-Crowne.
Thus, she did not seem to display much in the way of
predisposing factors for somatization.
On the Schedule of Recent Experiences she did display enough life changes in the previous 3 years to
trigger potential somatic reactions, but on the Hassles
Scale, she did not display an unusual number of hassles or irritants in her life. Mitigating and buffering
against the influences of changes in her life, she appeared to have an above average support system, with
which she was quite satisfied. It appeared on the Coping Responses Inventory that she was below average in
positive coping skills, and prone to using more avoidance coping skills. Some psychophysiological stress
profiling was done which revealed high electrodermal
response to stress, but a difficulty with the computer
prevented the results from being preserved.
Digital EEG was collected with recording electrodes
placed over the 19 standard regions defined by the International 10/20 System, referenced to linked ears.
All electrode impedance levels were below 3000 ohms,
with less than 1000 ohms difference between all sites.
A bipolar recording channel was used to monitor eye
movement artifact. The QEEG data were compared
via a Z-score transformation to a database of normal
subjects, the Hudspeth NeuroRep Analysis and Report
System with the revised Thatcher Lifespan EEG Reference Database. The results did not reveal significant problems with coherence, phase, or asymmetries.
However, it revealed a mild left frontal theta excess
(Fp1, F3, F7, Fz), which was maximal at F3 where it
was 1.06 Z-scores in excess. A mild alpha excess was
also found posteriorly at occipital sites and to a lesser
degree at parietal and temporal sites. The spectral topography revealed that the problem with excess theta
anteriorly was most prominent at 6 and 7 Hz in the
central and left central frontal areas.

3. Treatment
Since the patient appeared relatively normal psychologically, a treatment strategy consisting of stress management and neurofeedback was proposed to the patient and her family. They responded very positively
and we engaged in an informed consent process.
A strategy was selected to decrease the left frontal
theta excess. She was treated with three sessions a week
using a Lexicor NRS-24 neurofeedback unit utilizing a

297

protocol reinforcing an increase in low beta (1215 Hz)

while inhibiting theta at F3. Within only a few sessions,
symptomatic improvement was occurring. In fact, five
days after her first session, her parents reported that
she seemed more perky and energetic for 23 days.
By session five, she reported engaging in more activities. Two efforts in sessions six and seven to change
the training protocol to increase 15-18 Hz activity at
F3 resulted in unpleasant responses, and we returned to
reinforcing low beta. Reports from the patient of improvement continued, and were independently verified
on a regular basis from her mother and father.
In the fourteenth session of neurofeedback, we realized that we had not yet begun stress management training. I was impressed that the patient frequently had
cold hands, although this had not seemed to be the case
during psychophysiological stress profiling. Therefore,
she was asked to engage in self-monitoring, taking her
hand temperature every 23 hours throughout the day
with a thermistor that was provided. She was also given
a self-hypnosis tape and encouraged to use it on a daily
basis. The self-monitoring revealed wide fluctuations
in hand temperature during the day, generally ranging
from a high of about 92 F to lows in the mid 70s. Subsequently, she began using a homework form, recording
her subjective units of distress and hand temperatures
before and after self-hypnosis practice. Soon another
self-hypnosis tape was made for her. Over about a
three-month period, significant changes occurred in her
hand temperature. Instead of reaching the mid to high
80s after practicing self-hypnosis, she began reaching
temperatures of 9395 degrees. Instead of her beginning temperatures being in the high 70s to low 80s,
her hand temperatures began to generally be in the high
80s to low 90s before she started self-hypnosis. This
seemed to be an important component of her treatment
and came to be a valued skill.
After 15 sessions of neurofeedback and considerable
improvement, the patient was provided with a training
session using a neurofeedback unit called the Roshi.
It is a neurofeedback system that utilizes photic stimulation that can be set so that the frequency of photic
stimulation varies depending on the patients existing
dominant brainwave. Another unique feature of the
Roshi is that it also trains simultaneously at two electrode sites. The Roshi system samples at 128 samples/second and amplifies the raw EEG at a gain of
82000 in each channel. The filter bandwidth is on the
order of 2 to 42 Hz, at 3 dB points. These signals
are, then, subjected to a Fast Fourier Transform (FFT)
that is performed at the 128 samples/second data rate.

298

D.C. Hammond / Treatment of chronic fatigue with neurofeedback and self-hypnosis

Table 1
Pre-post treatment and follow-up percentile scores on the profile of mood
states
Fatigue
Vigor
Confusion

Pre-Tx
95.5%
33%
81.6%

Post-Tx
31%
84%
2.3%

Fig. 1. Percentile scores on the POMs fatigue scales.

Although the use of a cosine-tapered window is implemented, it can be optionally selected by the user, at the
output of the FFT, not on the raw EEG input. These
techniques bolster the power of Roshis unique audiovisual stimulation (AVS) system. These Fourier RMS
magnitudes are, then, subjected the standard 1 to 4 Hz
(Delta), 4 to 7 Hz (Theta), 8 to 13 (Alpha), 12 to 15 Hz
(SMR or low beta) and 15 to 20 Hz (Beta) frequencies. An added uniqueness of the Roshi system is in its
use of interhemispheric Fourier products in its training
modalities, with very low feedback latency. The A/D
conversion rate is 128 samples per second. Still another
beneficial advantage of the Roshi is that it also trains
simultaneously at two referential electrode sites.
We used a program called SMR Max. This program
provides photic stimulation, moment to moment, that is
focused on the patients peak frequency in the 1215 Hz
range, reinforcing this low beta band while simultaneously inhibiting the ranges from 4 to 12 Hz. We placed
electrodes at F3 and F4. Her immediate post-session
response was: The other sessions have been good,
but this was terrific! Due to a technical problem, it
was not possible for her to have another session on the
Roshi for 10 days. A few subsequent sessions with
traditional neurofeedback received continued positive
responses, but she still wanted to return to training on
the Roshi at the first opportunity. In her next session on

5 Months
34.5%
46%
3.6%

7 Months
65.5%
69.2%
4.5%

9 Months
58%
84%
6.7%

the Roshi, she reported a parallel experience to her first

trial on Roshi, and her strong subjective response was
that treatment with Roshi was clearly more powerful.
Subsequently, all training took place on the Roshi.
In part because of considerable progress, and in part
due to financial considerations, the patient decided to
stop weekly appointments at the end of 40 sessions.
Instead of seeing her 23 times weekly, sessions 3640
were scheduled 710 days apart. In anticipation of a
cessation of regular sessions, it was recommended that
maintenance may be facilitated by using a light/sound
(L/S) machine. She purchased a Photosonix Muse
Sharp unit with peripheral vision glasses that had
green LED lights around the outside, with clear plastic in the middle. This was suggested to permit her to
engage in a favorite activity, reading, while utilizing
photic and auditory stimulation. It was suggested that
she use 14 Hz stimulation, the middle of the range that
we had been reinforcing, twice daily for 15 minutes.
She began to do this after the 37th neurofeedback session. She immediately reported that her hands were
warmer following 15 minutes use of the L/S machine.
She reported generally feeling energized for 34 hours
following a L/S session. However, on occasions when
she felt unusually fatigued, a L/S session resulted in
feeling exhausted and having a headache afterwards.
After 40 sessions of neurofeedback, we agreed to
have once monthly reinforcement sessions. These sessions also provided opportunities to check her selfhypnosis homework forms. The Profile of Mood States
was used to provide an additional objective evaluation
of her progress, and one of her parents accompanied
her to each session, allowing external validation about
progress. It was learned that emphasizing balance in
her lifestyle was very important. Occasionally, enjoying how well she was feeling, she would overdo it.
After doing a great deal for one or several days, she
usually paid a price in feeling more fatigued for several
days afterwards. Operating on the assumption that a
virus may be involved, we emphasized that she might
well continue to have a vulnerability that would necessitate her being very attuned to herself, and keeping
balance in her lifestyle in other words, by not overdoing it. Self-hypnotic practice with a tape daily, and

D.C. Hammond / Treatment of chronic fatigue with neurofeedback and self-hypnosis

299

Fig. 2. Percentile scores on POMs vigor and confusion scales.

for 3-4 minutes several times daily, seemed important

in managing stress and maintaining balance. On one
occasion, a cold set her back for over a week. In another
case, she got the flu and it took her a week to bounce
back. But, in general, as the months went by, she continued to maintain her improvements and, at times, felt
that she continued to progress. Of particular delight to
her parents was the fact that her vocabulary and sense
of humor returned. Table 1 summarizes her improvements on the Profile of Mood States scales for Vigor, Fatigue and Confusion from pre-neurofeedback, to posttreatment (conclusion of 40 neurofeedback sessions),
and in follow-up evaluations at 5, 7, and 9 months.
Figures 1 and 2 show her profiles on the POMS taken
at these periods. At all testing times, she displayed
below average levels of tension, depression, and anger.
Therefore, these scores are not shown in summary tables and figures. At the 9 month follow-up she remains
stable. Her activity level has improved substantially,
and she is beginning to take her first college class.

4. Conclusion
Given the preliminary QEEG data of Duffy [9] and
Billiot et al. [1] demonstrating excess slow brainwave
activity, the neuroimaging research documenting hy-

poperfusion, and the research demonstrating cognitive

impairments in CFS, neurofeedback appears to hold
definite promise as one component in the multidimensional treatment of CFS. This particular case is of interest not only because of her objectively documented
and externally verified improvements, but also because
of the use of photic stimulation.
The author has had considerable experience in using traditional neurofeedback equipment with patients
and then having them work on the Roshi, which trains
on two electrode sites simultaneously and uses photic
stimulation to assist in guiding the brain toward the
desired goal. The subjective experience of the large
proportion of these patients is very similar to the response of this patient. The Roshi feels more powerful.
This has especially seemed to be the case in patients
with fatigue and head injuries. This patients indication
that training with a traditional neurofeedback unit was
helpful, but that training with the Roshi was terrific,
and her desire to only continue training on the Roshi, is
not atypical. Perhaps such reactions occur because the
photic stimulation provides extra assistance in guiding
the brain to the desire state, facilitating learning, and/or,
perhaps such enthusiastic reactions are associated with
the photic stimulation encouraging increased vascular
flow to hypoperfused areas of the brain. Further research should explore these issues and the potential

300

D.C. Hammond / Treatment of chronic fatigue with neurofeedback and self-hypnosis

of neurofeedback to assist in the rehabilitation of CFS

patients.

[14]

[15]

References
[1]

[2]

[3]
[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

K.M. Billiot, T.H. Budzynski and F. Andrasik, EEG patterns

and chronic fatigue syndrome, Journal of Neurotherapy 2(2)
(1997), 2030.
D.P.R. Buchwald, D.I. Cheney and B. Peterson et al., A chronic
illness characterized by fatigue, neurologic and immunologic
disorders and active herpes virus type 6 infection, Annals of
Internal Medicine 116 (1992), 103113.
D.C. Costa, J. Brostoff, V. Douli and P.J. Ell, Postviral fatigue
syndrome, British Medical Journal 304 (1992), 1993.
D.C. Costa, C. Tannock and J. Brostoff, Brainstem perfusion
is impaired in chronic fatigue syndrome, Quarterly Journal of
Medicine 88 (1995), 767773.
J. DeLuca, S.K. Johnson, D. Beldowicz and B.H. Natelson, Neuropsychological impairments in chronic fatigue syndrome, multiple sclerosis, and depression, Journal of Neurology & Neurosurgical Psychiatry 58 (1995), 3843.
J. DeLuca, S.K. Johnson, S.P. Ellis and B.H. Natelson, Sudden
vs gradual onset of chronic fatigue syndrome differentiates
individuals on cognitive and psychiatric measures, Journal of
Psychiatric Research 31(1) (1997), 8390.
J. DeLuca, S.K. Johnson, S.P. Ellis and B.H. Natelson, Cognitive functioning is impaired in patients with chronic fatigue
syndrome devoid of psychiatric disease, Journal of Neurology,
Neurosurgery, & Psychiatry 62 (1997), 151155.
J. DeLuca, S.K. Johnson and B.H. Natelson, Information processing in chronic fatigue syndrome and multiple sclerosis,
Archives of Neurology 50 (1993), 301304.
F.H. Duffy, QEEG in chronic fatigue syndrome. Unpublished
material published on F.H. Duffys web site associated with the
Clinical Neurophysiology Laboratory, Boston Childrens Hospital, 1999, http://fhdno2.tch.harvard.edu/ww/qeeg/qeegcfs.
htm1.
N. Fiedler, H.M. Kipen, J. DeLuca, K. Kelly-McNeil and
B.H. Natelson, A controlled comparison of multiple chemical sensitivities and chronic fatigue syndrome, Psychosomatic
Medicine 58 (1996), 3849.
J.A. Goldstein, I. Mena, E. Jouanne and I. Lesser, The assessment of vascular abnormalities in late life chronic fatigue
syndrome by brain SPECT: Comparison with late life major
depressive disorder, Journal of Chronic Fatigue Syndrome 1
(1995), 5579.
J. Grafman, Neuropsychological features of chronic fatigue
syndrome, in: Chronic fatigue syndrome, S.E. Straus, ed.,
Marcel Kekker, New York, NY, 1994, pp. 181284.
M. Ichese, I.E. Salit and S.E. Abbey et al., Assessment of
regional cerebral perfusion by Tc-HMPAO SPECT in chronic
fatigue syndrome, Nuclear Medicine Communications 13
(1992), 767772.

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

L.A. Jason, L. Wagner, R. Taylor, M.T. Ropacki, J. Shlaes, J.

Ferrari and S.P. Slavich, Chronic fatigue syndrome: A new
challenge for health care professionals, Journal of Community
Psychology 23 (1995), 143164.
S.K. Johnson, J. DeLuca and B.H. Natelson, Assessing somatization disorder in the chronic fatigue syndrome, Psychosomatic Medicine 58 (1996), 5057.
A.L. Komaroff, Clinical presentation of chronic fatigue syndrome, in: Chronic fatigue syndrome, G.R. Bock and J. Whelen, ed., Ciba Foundation Symposium 173 (1993), 4361.
L.B.B. Krupp, M. Sliwinski, D.M. Masur, F. Friedberg and
P.K. Coyle, Cognitive functioning and depression in patients
with chronic fatigue syndrome and multiple sclerosis, Archives
of Neurology 51 (1994), 705710.
A.R. Lloyd, I. Hickie and C.R. Boughton et al., Prevalence of
chronic fatigue syndrome in an Australian population, Medical
Journal of Australia 153 (1990), 522528.
A.C. Mawle, L. Steele, M. Reyes, J.G. Dobbins and W.C.
Reeves, A case-control study of immunologic parameters in
chronic fatigue syndrome (CFS) patients. Paper presented at
the first annual meeting of the American Chronic Fatigue Syndrome, October, 1994, Ft. Lauderdale, FL. Cited in J. DeLuca
et al., 1997.
B.H. Natelson, J.M. Cohen, I. Brassloff and H.J., A controlled
study of brain magnetic resonance imaging in patients with fatiguing illnesses, Journal of Neurological Science 120 (1993),
213217.
C. Ray, L. Phillips and R.C. Weir, Quality of attention n
chronic fatigue syndrome. Subjective reports of everyday attention and cognitive difficulty, and performance on tasks of
focused attention, British Journal of Clinical Psychology 32
(1993), 357364.
M. Riccio, C. Thompson, B. Wilson, D.J. Morgan and A.F.
Larient, Neuropsychological and psychiatric abnormalities
in myalgic encephalomyelitis: A preliminary report, British
Journal of Clinical Psychology 31 (1992), 111120.
C.A. Sandman, J.L. Barton, K. Nackoul, J. Goldstein and F.
Fidler, Memory deficits associated with chronic fatigue immune dysfunction syndrome, Biological Psychiatry 33 (1993),
618623.
R.B. Schwartz, B.M. Garada, A.L. Komaroff, H.M. Tice, M.
Gleit, F.A. Jolesz and B.L. Holman, Detection of intracranial abnormalities in patients with Chronic Fatigue Syndrome:
Comparison of MRI imaging and SPECT, American Journal
of Radiology 162 (1994), 935941.
A.P. Smith, P.O. Behan and W. Bell, Behavioral problems
associated with chronic fatigue syndrome, British Journal of
Psychology 84 (1993), 411423.
M.B. Sterman, Basic concepts and clinical findings in the
treatment of seizure disorders with EEG operant conditioning,
Clinical Electroencephalography 31 (2000), 4555.
I. Wickramasekera, Somatization: Concepts, data, and predictions from the high risk model of threat perception, Journal of
Nervous & Mental Disease 183(1) (1995), 1523.