Biofeedback and Self-Regulation, Vol. 1, No.

1, 1976

Electromyographic Biofeedback for Relief

of Tension in the Facial and Throat Muscles
of a W o o d w i n d Musician 1
John R. Levee 2
Veterans Administration Hospital, Sepulveda, California, and
Department of Psychology, UCLA

Michael J. Cohen and William H. Rickles

Veterans Administration Hospital, Sepulveda, California, and
Department of Psychiatry, UCLA School of Medicine

Electromyographic (EMG) biofeedback, for the relaxation of specific

throat and facial muscles, was given to a woodwind musician. The patient
had a nineteen-year history of tics and high levels of tension in his throat
and facial muscles. Eventually these problems progressed to a point that
interfered with his ability to perform as a professional woodwind musician.
Following detoxification from alcohol and Dexamyl, and after a period of
psychotherapy, EMG biofeedback relaxation training was started for the
muscles specifically showing chronically high tension levels. The EMG
training consisted of four phases designed to help the patient progressively
lower tension and generalize these newly learned techniques to his professional life. He had a total of twenty treatments of approximately 45 minutes
each. This procedure resulted in dramatic reductions in tension levels of the
specific throat and facial muscles along with increased proficiency as a
musician and in psychological functioning.
The technique called biofeedback refers to methods for providing external
and augmented information about ongoing internal, physiological activity.
The feedback information is usually presented in the form of a meter reading, a tone that varies in pitch, a light that changes in intensity, or some
'Appreciation is expressed to Dr. Wayne Hanson for technical assistance and to Dr. E. Ruth
Mezquita for medical supervision of the patient.
2Requests for reprints should be addressed to Dr. John R. Levee, Psychological Services,
Veterans Administration Hospital, Sepulveda, California 91343.
113
1976 Plenum Publishing Corporation, 227 West 17th Street, New Y o r k , N.Y. 10011. No
part of this publication may be reproduced, stored in a retrieval system, or transmitted, in
any form or by any means, electronic, mechanical, photocopying, microfilming, recording,
or otherwise, without written permission of the publisher.

114

Levee, Cohen, and Riekles

combination of the above methods. To be effective, the changes in these

feedback signals should be directly and linearly related to fluctuations in the
physiological system being monitored. The individual receiving feedback
can then use this information to control voluntarily the functioning of this
physiological system.
Biofeedback of electromyographic (EMG) activity has been of substantial interest and demonstrated efficacy in treating a wide variety of disorders. Electromyography refers to the recording of electrical signals produced by contracting bundles of muscle fibers, and is directly related to tension levels of the muscle (for an excellent review of EMG applications see
Basmajian, 1972). The uses to which EMG biofeedback have been put include generalized relaxation for chronically anxious people (Raskin, Johnson, & Rondestvedt, 1973), reduction of muscle tension in patients with
neck injuries (Jacobs & Felton, 1969), the control and alleviation of tension
headache (Budzynski, Stoyva, Adler, & Mullaney, 1973), muscle retraining
in hemiplegics (Johnson & Gerton, 1973), subvocalization during silent
reading (Hardyck & Petrinovich, 1969), and spasmodic torticollis (Brudny,
Grynbaum, & Korein, 1974).
Generally, the situations in which EMG biofeedback has been of demonstrated therapeutic value involved the improper functioning of specific
muscles. Current procedures for generalized relaxation have not worked
well (see Raskin et al., 1973), but the reasons for this are most likely procedural. Basmajian (1972) further suggested that EMG feedback could be
used with normally functioning muscles as an aid to learning a skill.
Basmajian has studied the lip and cheek muscles of wind-instrument musicians of varying degrees of proficiency. A definite relationship was established between the skill of the players and specific electromyographic patterns.
The case herein reported involves a wind-instrument musician who
over the past nineteen years experienced increasing tightness in the muscles
of the lips, cheek, and throat. At the time EMG biofeedback training was
started the condition had progressed to the extent that the patient had difficulty playing at a professional level, and work as a musician was becoming
infrequent.

METHOD

Case History
The patient is a 52-year-old male musician who plays flute, clarinet,
and saxophone. He was hospitalized on a neuropsychiatric ward of a

EMG Biofeedback of Facial and Throat Muscles

115

Veterans Administration Hospital two years ago. His complaints, at that

time, were (a) a tightness of his jaw and a constriction of his throat which
were interfering with his profession, (b) alcoholism, and (c) DexamyP
dependence. Early in his career the patient was a successful musician with a
number of substantial contracts at several movie studios.
In 1955, during a party at his home, he noticed the first tic in his lower
lip. By the end of the year it had radiated to his mouth and throat,
tightening the muscles there and in the face. At this point he started psychoanalytically oriented psychotherapy which he continued for about two years
with no significant relief of symptoms.
In 1965, he was hospitalized for a suspected myocardial infarction,
with a final diagnosis of esophageal spasm. At about this time his drinking
became increasingly habitual. His speech began to deteriorate and his job
performance degenerated to the point that he played second-chair parts,
and only on occasional calls. In 1971, he was admitted to Scripps Clinic in
La Jolla for diagnosis of this throat and jaw problem. The resulting diagnosis was craft palsy of a psychogenic nature, and hypnotherapy and mood
elevators were prescribed. Results were not significant. Shortly after this, he
was referred to private behavioral psychotherapy to learn Jacobsonian relaxation of affected muscle tension and was given cassette tapes for home
implementation of training. He reported no relief from Jacobsonian therapy, although he admitted he did little home practice. Just prior to Veterans
Administration hospitalization, he was drinking a pint to a quart of vodka
daily, mostly in the evenings. On occasions he became violent and abusive
to his wife. During the day he took 10-15 Dexamyl tablets in an attempt to
continue performing as a musician. Financial difficulties became extreme
and his wife threatened divorce, motivating him to enter the Veterans
Administration Hospital.
Status at Time o f Admission
Upon admission, the patient was diagnosed as an alcoholic with habituation to Dexamyl. He had no thought or perceptual disorders. He was
able to abstract well and to give information adequately. He had little or no
insight into his problems, and although admitting to heavy drinking, denied
that it affected his musical performance, and denied any marital conflict.
He was very slow and careful in forming words and phrasing responses,
alluding constantly to restrictions of his musculature. He was referred to the
Neurological Clinic with negative findings.
The patient was referred also to the Speech and Audiology Clinic with
the following results: He showed mild intention tremor of the fingers and
~Dextroamphetamine sulfate and amobarbital,

116

Levee~ Cohen, and Rickles

thumb. He possessed normal hearing sensitivity bilaterally to pure-tone

audiometric examination and had no receptive language disturbance. His
speech intelligibility was rated as good, although he showed dysfluency in
mild clonic blocks, with some consonant repetitions. The patient had appropriate vocal pitch and intensity and had good vocal quality on sustained
vowels. On oral examination he showed a slight protrusion and excessive
internal motion of the mandible when speaking. He had mild tongue tremors, particularly while sustaining the high back-tongue v o w e l s , / i / a n d / u / .
In addition, he had extreme hypertension of the neck and facial muscles
while speaking. His tension increased with prolonged speaking and was
more pronounced in stressful, anxiety-provoking situations, particularly
when the patient spoke to his wife or gave music lessons. He reported pain
in the masseter muscles, often when anticipating speaking situations. The
same symptoms occurred during individual psychotherapy and intermittently while the patient played one of his wind instruments. The clinical impression of speech mechanics, in general, was of physical and functional integrity, but with underlying emotional disturbance. Overall recommendations were that treatment should be aimed at reducing tension in target muscles of the face and throat and that psychotherapy should be addressed to
anxiety-producing conflicts. Direct speech therapy was not recommended.

Method of Treatment
The patient was detoxified from alcohol and Dexamyl, placed on 300
mg of Quaalude (later changed to 75 mg Sinequan, three times a day), and
at his request placed on 250 mg Antabuse (subsequently reduced to 125 mg).
He became responsive to treatment on the hospital ward, attending individual and group psychotherapy on a regular weekly basis. His individual
psychotherapy was client-centered/behavioristically oriented. Therapy was
directed to helping the patient recognize and express underlying feelings of
inadequacy and hostility which related to his wife, brothers, and father, all
of whom had been accomplished musicians. Vocational reconstruction was
also started.
The patient attended psychoanalytically oriented group psycnomerapy on a weekly basis for ten weeks, but then he discontinued, complaining
that he was constantly criticized by other group members and that he found
it difficult to respond. He was hospitalized for 5 months and then discharged to continue on outpatient psychotherapy one hour a week. During
the transition from inpatient to outpatient status the patient took on parttime jobs playing in bands and also teaching students to play woodwind instruments. A year after discharge he was playing full time in a well-known
band and carrying a teaching roster of 25-30 students.

EMG Biofeedback of Facial and Throat Muscles

117

With continuing demands for his playing, and some anticipation o f

playing first-chair positions, the patient developed complaints of increasing
constriction in his mouth and throat muscles. It was at this point in therapy
that he was referred for E M G relaxation training of target muscles.

Biofeedback Training
The patient was placed on an EMG feedback unit for training of muscle relaxation.' Feedback was a visual display of seven light-emitting diodes
placed in a vertical column. The gain of the feedback unit could be varied to
reflect 2.5, 5, 10, 20, or 30/aV of peak-to-peak muscle tension per light
when activated from skin-surface electrodes (disk cups). For example, if the
unit was set for a gain setting of 10/aV a n d four lights were on (lighting was
sequential from bottom to top), 4 0 - 5 0 / a V of muscle tension was present.
The high-pass and low-pass filters were set at 1.0 and 1200 Hz respectively,
and the internal noise of the E M G unit varied between 2 and 5/aV, depending on input frequency. The training consisted of four phases.
First Phase (Three Sessions). Electrodes were placed on the frontalis
muscle (referent), the bridge of the nose (ground), and alternately each
session on the right and left orbicularis oris about midway below and between the quadratus labii superior, infraorbital heads, and zygomaticus
muscles. Three 40-minute training sessions were run at a gain setting of 10
/aV/light. The patient was instructed to turn off as many lights as possible,
and once he was comfortable with the E M G unit's performance, he was left
by himself for feedback training. By the end of the first training session, he
was able to reduce a 70/aV display (all lights) to a fluctuation between 10
and 20 gV. This level of performance was maintained over the following
two sessions. The patient reported a comfortable easing of tension in the
orbicularis oris and a shift of target tension to the throat. Table I outlines
beginning and ending tension levels for the various phases of the training.
Second Phase (Three Sessions). To deal with target tension in the
larynx, electrodes were shifted to (a) the infrahyoid muscle area alongside
the trachea and larynx, and (b) posterior to the chin on the genioglossus
muscle close to the geniohyoideus. The ground electrode was placed alternately on the right or the left mandible bone as the infrahyoid electrode was
alternated from session to session. The initial training session in this second
phase was run for 40 minutes, resulting in a final tension level of 40/aV. At
the start of the second session, the patient reported much throat tension. He
had spent almost the entire Easter weekend at work in a band playing
'We thank Dr. Barbara B. Brown for use of the EMG feedback unit designed and built in her
laboratory.

118

Levee, Cohen, and Rickles

Table I. Beginning and Final Tension Levels over
the Several Sessions o f Training
Session

Start ~ V )

Phase I

1
2
3

70
70
70

Phase II

1
2
3

100
100
35

Phase III

1
2 (flute
intro.)
4

35
210

Phase IV

1
5
10

210
35
35
35

Finish (~V)
10-20
15
15
40
100
25
5
10-5
10-5
10
10
5

several instruments, and throughout the 40-minute session was unable to

reduce his tension below 100/aV. In the third session, the patient was expressing some discouragement about previous training sessions, so the gain
setting was reduced to facilitate his ability to turn out lights. He became
more relaxed in this session, mimicked playing his flute as he experienced
more control of the lights, and ended with a 25 ~V reading.
Third Phase (Four Sessions). The patient was moved into a singlewalled acoustically controlled IAC 403 series chamber for the subsequent
four sessions. Electrode placement was the same as for the second phase.
The patient was able to adapt to a 10-/aV to 5-/aV level of relaxation in about
10-15 minutes. At this point of relaxation, he was instructed to play his
flute and to randomly play scales or familiar pieces for brief periods and
then relax between playing by returning EMG readings as close as possible
to the preplaying relaxation level of 10/aV to 5/aV. In the first session, he
was able to return immediately to preplaying levels. In the following three
sessions, he was able to maintain a 10-/aV to 5-/aV level of relaxation before
playing of his flute and to return to this level within seconds after intermittent playing.
The patient reported not only continued reduction of muscle tension
in mouth and throat areas during on-the-job performance, but experienced
a new and increased sense of control and refinement, which he felt was influenced by use of the acoustical chamber. The patient was urged to continue relaxation in a similar manner during practice sessions at home and in
daily performance at work.
Fourth Phase (Ten Sessions). With a decrease of symptoms, the patient continued on the same regimen without the acoustical chamber for ten

EMG Biofeedback o f Facial and Throat Muscles

119

more sessions. He was able to maintain a 10-~V relaxation level for the first
six sessions, which he then reduced to a continuous 5 ~V for the remaining
four sessions.
Follow-up. Once-a-month follow-up sessions have continued for 6
months, with the patient able to maintain the 10-~V to 5-~V level of relaxation.
In his work performance, from the fourth phase onward, he has been
able to advance from third- to first-chair positions on three instruments,
and during the follow-up period has been performing solo parts on demand
of his conductor, with neither a return of discomfort in target areas nor a
reported shift of tension to any new areas.

DISCUSSION
Over a number of years the patient has undergone treatment in a
number of psychotherapeutic modalities. All these treatments left him
essentially unchanged in one of the areas most important to his functioning
and identity as a person: performing well as a musician. Specific muscles of
the patient continued to deteriorate as did the patient's ability to play his
instruments. Biofeedback EMG training, started after physiological integrity was restored through detoxification," and after psychotherapy
allowed him to deal with the specific symptom, was directed at relaxing
target muscles. With EMG training, the patient has learned to control target
muscles and function well again as a musician. Why was biofeedback training successful after other modalities were not? A comprehensive answer to
this question is not possible. However, an attractive formulation would not
necessarily ascribe success or failure to any one of the multiple therapeutic
interventions.
Each therapeutic modality could be viewed as directed at different behavioral, conceptual, psychophysiological, or psychodynamic levels of
functioning. With resolution of difficulties at each of the levels (e.g., detoxification, vocational reconstruction), biofeedback could be seen as a
common final pathway. Through the various therapeutic interventions, the
initial symptom was divested of primary and secondary gains. The patient
could then approach the muscle constriction more simply, as a learning task
attended by only the usual facilitating anxieties associated with a learning
situation. As of this writing there has been no return of symptoms, nor have
there been any signs of symptom substitution or breakdown of adaptive
psychological defenses. Indeed, his present level of professional function
continues to enhance his self-esteem and to enrich lais marital and family
relationships.

120

Levee, Cohen, and Rickles

The progressive development o f this case involved (a) initial b r e a k d o w n

into a psychogenic tic, (b) m o r e widespread b r e a k d o w n o f coping functions, (c) restoration o f function via multiple p s y c h o t h e r a p e u t i c m o d a l ities, and finally (d) a direct a p p r o a c h to the initial s y m p t o m via biofeedback. This evolutionary progression o f b o t h b r e a k d o w n a n d cure suggests
that biofeedback holds m u c h promise as an adjunct to the psychotherapeutic treatment o f psychophysiological disorders involving the muscles.
F r o m this point o f view, the use o f biofeedback treatment alone or
introducing biofeedback training at an earlier stage o f this patient's breakd o w n and treatment p r o b a b l y w o u l d not have been as useful. The timing
and adjunctive therapeutic requirements o f b i o f e e d b a c k t h e r a p y o f these
and related psychophysiological disorders are issues for further research.

REFERENCES

Basmajian, J. V. Electromyography comes of age. Science, 1972, 176, 603-609.

Brudny, J., Grynbaum, B. B., & Korein, J. Spasmodic torticollis: Treatment by feedback display of the EMG. Archives of PhysicalMedicine and Rehabilitation, 1974, 55, 403-408.
Budzynski, T. H., Stoyva, J. M., Adler, C. S., & Mullaney, D. J. EMG biofeedback and
tension headache: A controlled outcome study. Psychosomatic Medicine, 1973, 35,
484-496.
Hardyck, C. D., & Petrinovich, L. F. Treatment of subvocal speech during reading. Journal of
Reading, 1969, 1, 1-11.
Jacobs, A., & Felton, C. S. Visual feedback of myoelectric output to facilitate muscle relaxation in normal persons and patients with neck injuries, Archives of Physical Medicine
and Rehabilitation, 1969, 50, 34-39.
Johnson, H. E., & Gerton, W. H. Muscle re-education in hemiplegia by use of electromyographic device. Archives of Physical Medicine and Rehabilitation, 1973, 54, 320-325.
Raskin, M., Johnson, G., & Rondestvedt, J. W. Chronic anxiety treated by feedback-induced
muscle relaxation. Archives of General Psychiatry, 1973, 28, 263-267.
(Original received December 15, 1974)