< -2312: *uéIH-Ls-
u?

 

J: ‘ 0.: . m.-.7~"4-~.—K.L.w_1.u w- )klto‘II .
359‘? h. ;IE§W:E:1 .314“ ._ ~..~ ﬁhw‘”
‘Ciu‘: )V‘S‘lei“ {L n... éﬁis: a 4%.- g 1. ‘
u.
- 11-." -

.151“ "Int .63:cm my

@13: . “it.
51:53.? .ﬁ:"-" wsy‘.
‘11 ~3$~3n

. . . .1. 1.;
.C“: 11%;?“ ‘f’i‘x ‘14:.ng
€533“: 33:121.; ‘5“; -' 3“? “ maid.“ -
(:31..le "" “ft-A3311. 1 ‘1. 1:32... '
"11.1.71. 13. 45$: ‘L‘L‘i‘ ‘ .;- is}; ‘N 533%.?“ "1::
3}“ 3n 11.2“!) in. ‘31:.{LLC‘1 ‘1‘: *1
‘1 “11“«51‘11‘1 ” ‘=“:‘~““ﬂ
in”. a... 1:111. '1;- 141:3“.‘1
....t. 151.-.)“. ﬁr.»
L‘L L «NU 1 ﬁg“. 5‘24“.-
gm 1 “ 1' "1‘ I I “$51211. . é:;“"“
.. T..‘v.."7 .- - >1“ 3' " ... =
iii: ”‘35“ “‘g %‘&111.‘““‘ 0:1. {L z‘k‘ﬁ‘ 'L 3915M its; HWﬁX-«Ei
“is {It ‘
. “1.11."...3
i win‘ 1V‘1Q1”“ ‘LZM ”‘33“. ﬁlm‘wwl {RE‘U‘LQ'HE ‘12:...3“
fldnv‘ ’1'“ .. 3- . ‘.““L~.‘1“t‘~“"11""»» “1.3%.; "1:554? 36‘5“-
Wq! INN.“ ‘ E1. Lt? 0‘8 $“‘1,L11.“:E;£~ _€r-‘L-.«... C“.
$31": ‘3‘?“ ”$121 52%.?“ 1“?“ .3311 1:41; ‘1‘“1‘1‘14'45‘ .‘n 1'1.“ r‘ 11‘1‘3‘31. '“
1113.}: 37:24“ {(‘Lu 12.35% 1.2111} 111551;; 1 “‘J'RLL ‘L ‘1‘.
”glam“ 1111‘ IE?“ 93an {5111‘ ~31“ “‘15 3" 11-31%
{23.51711}. 13:3“: . ‘11.”? 113:1, “515.1
1.. $1.11 :1. ‘1" '

Kai-hi,

‘ 113113;, 1}“ ‘L‘ {fa
.' M1 ' - ‘x “‘1‘ 3,“: - ‘7
"It.“ ”‘51:“ 1.11:1. 1% ‘ “1%“ ‘13:.‘1‘E‘m 1“} ~"u
2:1. 1. “7%:1‘11‘1‘1‘Latw " 1‘1‘1‘3‘4‘1L“
511‘,“ W?“ "‘ 31:.
11.1.51. ‘5‘“? .r "“15: 3%3113311‘1“

 

‘1.“- :1: .3;ng

~13... ‘21; 5‘15.“ 1:
111511.13 1“

‘1.
"”3.
“$5 ‘1:
3‘51?

{.31}: ‘ .. ”‘1 1 51' 1 . . ,
v ‘31 M31” “EVE“ 4:3 ‘3‘ u. ‘ 1 " “ﬁg ’
.v N . «1k ‘3‘. . ‘13-» ~ 71.1"».
“51.41.313.113; 1.111315133223311 . 31:15. , 111113» . “*"ayzwlu
' 1* 11:72....‘311? 132'." ‘ ‘11!" ‘13.:4.‘ . ‘1. km ”52.2%
g" ‘1“. $5M“: 15.31.13}; “R11.” .' .. . . e ”a. ,5 .1 .g‘it‘é mﬂ“
’3: (kn‘h-m‘ixu _.. “4‘ '1'.» "u 3
11,1 "v‘néﬁm‘ 1.1 ‘1‘:
:' 111;; I‘ni‘i ”551:13‘1 ’ 3’ 1513‘ “ﬂ “ $1.} .
~1-- - 3‘21‘1317‘.‘ . . . 1‘
, ‘iﬂ 513.133.9311.: a1; , 5:35.15. . _ .__ ‘ . ,1, . § .3 '
;,. . 3L1... I‘LL“ .4 .1.‘ r}. "ER “y""‘"..~r’ ;- 51;} ' ' .
“’11.. ‘3 1.155111 -1... “~21 - .-. * .1 “1 ~ “‘7‘“ 1: 531%“
=5..." 15.5% 1%. 3.3.1. 3‘13... 1.1% a”. .
{“333}: {'9ng i . ..
. '- 53;“;

‘ ‘f" "It: 1““; a
$1115 1" .1“ it“. "1 ‘1'"
. - .12131,’ . . ‘ . I. .h ‘1 I 7' ' 5‘" L‘ .‘1.’
“Ea 1.“ “E1“ ’1‘“ “1%: ‘ R 11:13:: ﬁ‘zgf’t . . 1:. ‘1 - Rh: ‘1‘? gﬁﬁi‘: ﬁg: ‘9.“
1‘9“» "3:11:13“? ‘3" 31".‘1. ‘i‘. '11 n. h * . 1': l- 3'" £15421“
1‘ ‘ 'h‘w' “ ‘ﬁk‘éﬁ‘hﬁ‘gﬁ: * 1."? 1"" '5‘“ I‘ék‘ﬁaﬁri‘ .
$9M " ‘2

‘ I
11:1 3‘5“

{$3.4 351113
2“ ‘13:“ “ $5131 .13“? ‘3“
u a???“
1‘ “‘11" .195: 11%;} 4.1% 1H"“‘“1‘1§'

1&1“ 19,31 32..“ 1‘4““ 35a. . .. m 1mm.
11%;? £535 .. ‘1‘ f . {1 471b, ".15“ 11.

i '1‘“ ‘1': u“? ‘C
‘l‘i‘ 1% $11“ 12.. «‘1‘ 1:2 ’0“3“-3..~icz“f€%“ 4w 1
1; ~- . if“

V.”
LE‘Mfw-‘Lk‘ﬁ‘ L ‘

L ‘1. ‘ - 54‘ ..
I "l. rwg‘gém‘l; 9‘- Ski} 7...:3“:15‘%:‘:L§V “L 1-1.3
‘ '1??in . .K“ . ‘T 1“}:1.‘ .. 33...} '.: ..‘.
5‘3 1.“. 3 . .7. ..

 

 

 

 

 

 

TH ESlS

\\\\\\\\\\\\\\\\\\\\\\\\\\ \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\l 1* tJ JJ 1

3 10458 9118 J . p, ..

This is to certify that the

thesis entitled

EXPLORATION OF A COMBINED PROGRAM OF ELECTROMYOGRAPHIC
BIOFEEDBACK AND PROGRESSIVE RELAXATION AS A
TREATMENT APPROACH WITH SCHIZOPHRENICS

presented by

Carol Keating

 

has been accepted towards fulﬁllment
of the requirements for

Ph.D. Counseling and

degree in _
Educat1onal Psychology

 

 

u/Jv— Mm?

Major professor

Date May 1981

 

0-7639

 

 

 

MSU

LIBRARIES
n

 

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

 

 

 

 

 

 

 

EXPLORATION OF A COMBINED PROGRAM OF ELECTROMYOGRAPHIC
BIOFEEDBACK AND PROGRESSIVE RELAXATION AS A
TREATMENT APPROACH WITH SCHIZOPHRENICS

By

Carol Keating

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Counseling and Educational Psychology

1981

ABSTRACT

EXPLORATION OF A COMBINED PROGRAM 0F ELECTROMYOGRAPHIC
BIOFEEDBACK AND PROGRESSIVE RELAXATION AS A
TREATMENT APPROACH WITH SCHIZOPHRENICS

By

Carol Keating

Persons diagnosed as schizophrenic have been shown to be aroused
on skeletal-muscle measures. Researchers have suggested that such
physiologic arousal is reflective of responses occurring in other
systems where cognition and emotional behaviors are represented. It
has also been hypothesized that schizophrenics are in a state of
chronic physiological disequilibrium which impacts on their ability
to modulate or change levels of responsivity to correspond to varia-
tions in internal and external events. The physiological arousal,
then, has been viewed as an important contributing factor to the emo-
tional and cognitive instability of the schizophrenic.

Electromyographic biofeedback relaxation training has been used
with highly anxious individuals who, like schizophrenics, experience
muscular tension. The technique focuses on muscular relaxation, which
in turn has been demonstrated to effect consequent improvements in
emotional functioning. The purpose of the present study was to
develop and examine the effectiveness of a combined treatment program
of electromyographic biofeedback and modified progressive relaxation

exercises in alleviating select symptoms with schizophrenics.

Carol Keating

One male and three females were accepted for the study based on
three criteria which included (a) diagnosis of schizophrenia without
paranoid ideation, (b) functional stability, and (c) experiences of
physical tension. All subjects were required to have the consent of
their psychiatrists in order to participate. Each subject had a his-
tory of at least three psychiatric hospitalizations and was being
maintained on an antipsychotic medication.

A single case study was used with an A-B time series research
design. Observations were made for each subject across 22 sessions
which included a pre- and post-treatment interview, 7 baseline and l3
EMG biofeedback relaxation training sessions. The continuous dependent
variables which were compared across baseline and treatment sessions
included: EMG values during biofeedback training and during cogni-
tive test performance, scores on each of five cognitive tests known as
the Repetitive Psychometric Measures, and scores on three self-rating
scales of tension. The Taylor Manifest Anxiety Scale was administered
pre- and post—treatment.

The split-middle method of analysis was used to assess linear
trends in the data, and binomial tests were applied to determine the
significance of differences for the continuous measures.

The major results of the study were:

l. All subjects significantly decreased EMG values during bio-
feedback training compared to baseline resting levels.

2. Subject D significantly decreased EMG values during cogni-

tive test performance following treatment intervention.

Carol Keating

3. Subject A significantly increased scores on three out of four
cognitive tests and Subject D significantly increased scores on one
out of five cognitive tests following treatment intervention.

4. Subjects B and C significantly decreased ratings of general
levels of tension following treatment intervention.

5. All subjects significantly increased ratings of their ability
to modify levels of tension following treatment intervention.

6. Subjects did not improve in their ability to detect changes
in levels of tension following treatment intervention.

7. Only Subject D did not decrease scores on the Taylor Mani-
fest Anxiety Scale at the end of treatment.

The treatment was found to be clinically significant in view of
the changes which occurred in physiological, cognitive, and emotional
behaviors. The treatment process was particularly beneficial in
helping individuals identify and modify ideation which inhibited suc-
cessful relaxation. Further studies are needed which test the clini-
cal significance of the treatment with a wider variety of subjects,
specifically those who are psychologically less stable. In addition,
it would be important to establish the long—term benefits of continued

practice with these techniques.

@ Copyright by
CAROL KEATING
198T

 

To my mother, Maude Pelak, for her unconditional support and
much-appreciated wisdom.

To my father, Walter Pelak, who would have been proud and whom
I would like to have shared this experience with.

To my son, John Keating, for the joy he brings and the tolerance

he's shown.

.” A; _.-.—-—/

ACKNOWLEDGMENTS

I would like to acknowledge the following people for their
contributions.

My doctoral committee for guidance and expertise: Dr. William
Hinds, committee chairman; Dr. William Farquhar; Dr. Norman Kagan;
and Dr. Robert Winborn.

My special friend, Este Wolkinson, who showed caring and sensi-
tivity during a particularly difficult time.

My close friend, Lionel Rosen, who helped lay the foundations

and continued to encourage my efforts over the years.

 

TABLE OF CONTENTS

LIST OF TABLES ........................
LIST OF FIGURES ........................

Chapter
I. THE PROBLEM ......................

Introduction ....................
Need ........................
Purpose .......................
General Hypotheses .................
EMG Levels ....................
Cognitive Tests ..................
Self-Rating Scales ................
Definition of Terms .................
Overview ......................

II. RELATED RESEARCH ...................

Relationship of Skeletal-Muscle System to
Other Variables ..................
The Relationship of Skeletal-Muscle Response
Tendencies to Other Variables in Schizophrenia
Theories of Psychophysiological Arousal in

Schizophrenia ...................
Electromyographic Biofeedback as a Method of
Relaxation Training ................
Progressive Relaxation ...............
The Effects of Neuroleptics on Skeletal-Muscle
Measurement and Cognition .............
Summary and Implications of Related Research . . . .
III. DESIGN OF THE STUDY ..................
Sample .......................
Instrumentation ...................
Self-Report Rating Scales .............
Biofeedback Equipment ...............
Repetitive Psychometric Measures .........

Taylor Manifest Anxiety Scale (TMAS) .......

iv

Pag
vi

vii

90')?

heeeee e ww

 

Research Design ...................
Methodology .....................
Pre- and Post-Treatment Interviews ........
Baseline .....................
Treatment .....................
Hypotheses .....................
Differences in EMG Levels .............
Differences in Test Scores of Cognitive Abilities .
Differences in Self-Ratings of Tension ......
Analysis ......................
Split-Middle Method ................
Data Graphs ....................
Summary .......................

IV. ANALYSIS OF RESULTS ..................

Hypotheses I and II: Differences in EMG Levels . . .
Hypotheses III, IV, V, VI, and VII: Differences in
Scores on Repetitive Psychometric Measures
Hypotheses VIII, IX, and X: Differences in Scores
on Self-Rating Scales of Tension .........
Clinical Observations ................
End of Session Resting EMGs ............
Final Four Biofeedback Training Sessions .....
EMG Changes During the Pre- and Post-Treatment
Interviews ...................
Taylor Manifest Anxiety Scale ...........
Progressive Relaxation ..............
Subjects' Experiences During Biofeedback Training .
Summary .......................

V. SUMMARY AND CONCLUSIONS ................

Summary .......................
Conclusions .....................
Electromyographic Levels .............
Repetitive Psychometric Measures .........
Self-Report Rating Scales .............
Clinical Observations ...............
Limitations .....................
Implications for Future Research ..........

APPENDICES ..........................
A. SELF-REPORT RATING SCALES ...............
B. REPETITIVE PSYCHOMETRIC MEASURES ...........

”Mm”.- —.-

 

Page

C. TAYLOR MANIFEST ANXIETY SCALE ............. 132
D. SUBJECT'S CONSENT FORM FOR ELECTROMYOGRAPHIC
BIOFEEDBACK STUDY .................. 136
E. PERSONAL DATA INVENTORY ................ I39
F. PHYSICIAN'S CONSENT FORM FOR ELECTROMYOGRAPHIC
BIOFEEDBACK STUDY .................. 141
REFERENCES .......................... I43

 

Table

3.1
4.1

 

LIST OF TABLES

Treatment Schedule and Phases ..............

Percentage of Change for Median EMG Levels From 20
Minutes Rest During Baseline to 30 Minutes of
Biofeedback Training During Treatment .........

Binomial Tests of Phase Changes for Research Hypotheses
Involving EMG Levels .................

Percentage of Change for Median EMG Levels From
Baseline to Treatment During Cognitive Tests .....

Median Scores for Repetitive Psychometric Measures
Tests During Baseline and Treatment Phases ......

Binomial Tests of Phase Changes for Research
Hypotheses Involving Repetitive Psychometric Measures .

Binomial Tests of Phase Changes for Research
Hypotheses Involving Self-Report Rating Scales

Binomial Tests of Phase Changes for End of Session
lO-Minute Resting EMGs ................

Median EMG Values for Subjects During Pre- and
Post-Treatment Interviews ...............

Percentage of Change for Subjects TMAS Test Scores

Page

50

61

62

64

68

69

77

81

84
85

 

Figure
2.l
2.2

8
9
IO

b#4>h-¥>-b-J>J>4>J>

.ll

 

LIST OF FIGURES

Pathways From Skeletal Muscle to Neocortex ......

Nonmonotonic Relationship Between Arousal and
Performance .....................

Monotonic Relationship Between Attention and
Performance .....................

Psychophysiological Feedback Loop ...........

Daily EMG Levels for 20 Min. Rest to 30 Min.
Biofeedback Training ................

Daily EMG Levels During Cognitive Tests ........
Daily Scores on Flexibility of Closure ........
Daily Scores on Number Facility ............
Daily Scores on Perceptual Speed ...........
Daily Scores on Speed of Closure ...........
Daily Scores on Visualization .............
Daily Scores on Self-Rating Scale #l .........
Daily Scores on Self-Rating Scale #2 .........
Daily Scores on Self-Rating Scale #3 .........
Daily EMG Levels for l0 Min. at End of Session

viii

Page

27

27
31

6O
63
65
67
71
73
74
76
78
80
82

 

 

 

CHAPTER I

THE PROBLEM

Introduction

 

Schizophrenia is a complex mental illness which can best be
understood from a multidimensional perspective. Research and theory
which emphasize genetics, predisposition and learning, environment,
family relationships, and errors in metabolism are illustrations of
the many ways in which this mental illness has been studied (Freedman,
Sadock, & Kaplan, l976). Likewise, therapeutic interventions such
as drug therapy and various forms of psychotherapy reflect treatment
approaches which derive from the various hypotheses put forth con-
cerning the etiology of schizophrenia.

One approach to studying schizophrenia lies in the field of
psychophysiology. Researchers in this field such as Malmo, Whatmore
and Kohli, and Goldstein and the theories of Mednick (l958), Broen
(1968), and Tecce and Cole (l972) focus on characteristics of physio-
logical arousal and behavioral response tendencies to stress in
schizophrenics.

Psychophysiological research has shown that schizophrenics are
abnormally aroused on skeletal-muscle measures and that this arousal
is manifested during rest and in response to stressors. Theories
have examined this arousal and explore the possible explanations for

the individual's apparent inability to modulate response tendencies.

l

 

 

 

 

They attempt also to explain how arousal in the skeletal-muscle
system is linked to the symptomatology of delusions, emotional blunt-
ing, hallucinations, and disordered thought processes observed in
schizophrenia. This particular aspect of research forms the basis
for the present study, which proposes a treatment program aimed at
teaching persons diagnosed as schizophrenic to more effectively modu-
late responses in the skeletal-muscle system.

Research in psychophysiology indicates that abnormal physio—
logical arousal is a contributing factor in behavioral responses and
in the maintenance or production of symptoms in schizophrenics. For
example, Malmo and Shagass (l949) examined the relationship of
anxiety to physiological arousal. They found that the closest cor-
respondence between degree of anxiety and degree of physiologic dis-
turbance was reflected in measurements of striate muscle activity.

On these measures, schizophrenics most resembled a comparison group
of nonschizophrenics whose predominant symptom was anxiety. Malmo,
Shagass, and Davis (l95l) and Malmo, Shagass, and Smith (l95l) also
observed that during cognitive and performance tasks schizophrenics
evidenced significantly greater levels of muscular activity than non-
schizophrenic controls. They hypothesized that their experimental
findings reflected the types of reactions one might observe in every-
day stress situations among psychiatric patients.

The relationship between skeletal-muscle activity and degree
of psychological disturbance was the focus of another study by
Goldstein (l965). She compared schizophrenics with other groups of

psychiatric inpatients and normal controls on a number of physiological

 

variables which included measures of muscle tension. Schizophrenics
showed the highest degree of physiologic disturbance reflected in
the skeletal muscles, and they also manifested the highest degree of
psychologic disturbance.

Additional studies focused on the direct relationship of
skeletal-muscle activity to schizophrenic symptomatology and behav-
ior. Whatmore and Ellis (l958) and Whatmore (l966) concluded that
hyperactivity of the motor system of schizophrenics was intimately
tied with states of emotion and cognition. They observed that
schizophrenics in remission exhibited labile skeletal—muscle measures,
with tension levels varying greatly from day to day. An increase in
lability was observed to precede acute schizophrenic episodes, while
a reduction in skeletal-muscle tension was followed by clinical
improvement.

Fenz and Felner (1970) hypothesized that the hyperactivity noted
in the motor systems of schizophrenics was indicative of a common
basic deficit in these individuals in their inability to modulate or
change levels of responsivity to correspond to variations in internal
and external stimulation. They felt that the ineffective modulation
was characteristic of responses occurring in the cognitive and
behavioral as well as in the physiological domains. Others, including
Whatmore and Malmo, have incorporated this notion into their theories
of mechanisms involved in schizophrenic symptomatology.

Schizophrenics were compared with highly anxious subjects in
several of the studies cited. The excessive muscle tension in

anxious persons is believed to be a manifestation of emotions which

 

 

evoke defensive behavior, particularly the activation of muscles in
preparation to defend against threats to one's well-being (Brown,
l978). Others, including Whatmore and Kohli, theorized that the
excessive muscle tension in schizophrenics is also part of an emo-
tional response. While the total symptom picture of these two popu-
lations is quite dissimilar, skeletal-muscle activity as well as
degree of anxiety which is experienced often appears similar between
these two groups.

Electromyographic (EMG) biofeedback relaxation training is a
form of treatment which has been utilized with anxious individuals
who manifest abnormal levels of skeletal-muscle tension. Most EMG
biofeedback programs combine this technique with relaxation exer-
cises for home practice (Brown, l978). The focus of EMG biofeedback
and relaxation exercises is the modification of a physiological
response in the form of decreased muscle tension. The rationale is
that the newly acquired response pattern, decreased muscle tension,
is antagonistic to that observed in stress reactions or anxiety
states (Schwartz & Beatty, 1977). Generalizing to other systems,
Gaardner and Montgomery (1977) believed that an alteration in one
response system, in this case a physiological one, can interact and
impact on the autonomic and central nervous systems as well where
cognitive, behavioral, and emotional reactions are also represented.
They used the term "psychophysiological feedback“ to illustrate this
interaction between systems, and they viewed the skeletal-muscle
system as a main vehicle for communication between the brain and the

environment.

 

Studies of therapeutic outcomes of EMG biofeedback relaxation'
training provide illustrations of the interactive effects of
decreased skeletal-muscle arousal. For example, Raskin and her col-
leagues (1973) found that chronically anxious individuals could
better manage situational anxiety in their daily lives following
training. Townsend, House, and Addario (1975) also treated chronic-
ally anxious patients and found changes in mood disturbance and in
trait and state anxiety measures. In a more recent study, Quieting
Response Training, which combines various relaxation techniques includ-
ing EMG biofeedback and Progressive Relaxation, was used with psychi-
atric inpatients (Ford, Strong, Stroebel, & Szarek, 1981). A
significant finding was that these techniques could be successfully
used with various psychiatric populations, regardless of diagnosis.
Until recently, the use of this type of treatment with psychiatric
inpatients was not recommended because of the possibility of increas-
ing pathological symptoms.

This study explores the possible benefits of using EMG biofeed-
back training with persons diagnosed as schizophrenic. The founda-
tions for this work are based on aspects of research and theory in
the psychophysiology of schizophrenia and work which has been per-
formed in the area of biofeedback training with patient populations
who share some of the same characteristics in terms of anxiety and
skeletal-muscle responses.

Specifically, a treatment program was developed which combines
EMG biofeedback training with home exercises in the form of modified

Progressive Relaxation. To date, few researchers have explored the

 

use of these techniques with such a population. The aim of the pro-
gram was to observe first the changes which occurred in skeletal-
muscle tension levels and to assess consequent changes in formal
cognitive functions and subjective feelings of tension and anxiety.
Measures were selected which would sample behaviors in the physiologi-

cal, cognitive, and behavioral/emotional domains.

Need

Evidence indicates that muscle tension, primarily the ability
to effectively modulate arousal response tendencies, is a signifi-
cant component of the total symptom picture of schizophrenics. The
significance lies in the belief that physiological tension does not
exist independently of emotional and cognitive responses. To date,
antipsychotic medications have produced the most profound changes in
schizophrenic symptomatology with one of the benefits being a
decrease in general levels of arousal (Freedman, Sadock, & Kaplan,
1976). Tecce and Cole (1972) found, however, that this only per—
tained to certain autonomic measures and not to somatic or skeletal-
muscle arousal. The question arises as to the significance of this
finding. It seems highly probable that even though ”arousal” is
decreased with medications and some symptoms are alleviated, that the
schizophrenic individual remains quite tense and that this is part of
an emotional or anxiety response.

If, even with highly effective medications, the schizophrenic
continues to experience emotional or physical difficulties, then it

is apparent that the need exists for additional therapeutic

——_—7J

 

interventions. This may, in part, be provided by the more conven-
tional modes of psychotherapy, but may be addressed in yet another
way.

The notion of teaching schizophrenics to become cognizant of
their physiological reactions as they relate to other behaviors,
such as emotion and thinking, has only begun to be explored. While
the therapeutic effectiveness of antipsychotic medications cannot be
disputed, they do not enhance the individual's self—awareness, nor
do they teach the patient how to deal with stressful events in their
daily lives. This researcher believes that therapeutic interven-
tions which are based on existing data regarding the nature of schizo-
prenia should be explored. The need is not to dispute the effec-
tiveness of contemporary treatments already in use. Rather, the
need lies in exploring additional interventions to be used in con-

junction with accepted practices.

Purpose

There is considerable support for the theory that the abnormally
high skeletal-muscle tension of schizophrenics is indicative of the
difficulties experienced by them in dealing with internal and exter-
nal stressors. Research findings suggest that a relationship exists
between these motor irregularities and in the production and main-
tenance of symptoms of schizophrenics. While these issues have
already been touched upon, a more extensive review will follow.

Other persons diagnosed as being chronically anxious and who

also manifest unusually high skeletal—muscle tension have been helped

 

 

to alleviate their feelings of tension through EMG biofeedback relaxa—
tion training. While these two populations are different in many
ways, certain similarities do exist, specifically in anxiety and
tension represented in the skeletal-muscle system. It has been
recently suggested that similar techniques can be used successfully
with various psychiatric populations. Therefore, it was the purpose
of this investigation to explore the effectiveness of EMG biofeedback
techniques in alleviating select symptoms in persons diagnosed as

schizophrenic.

General Hypotheses

Hypotheses for the study were divided into three parts, each
part pertaining to a different type of dependent variable. The
variables were: (a) electromyographic levels, (b) scores on five

cognitive tests, and (c) scores on three self-rating scales.

EMG Levels
1. The EMG levels (measured in microvolts) will be lower
following the point of biofeedback training than prior
to treatment.
2. The EMG levels (measured in microvolts) will be lower
during the administration of cognitive tests following

the point of biofeedback training than prior to treatment.

Cognitive Tests
3. Scores on each of five cognitive tests will increase fol-
lowing the point of biofeedback training than prior to

treatment.

 

 

 

Self-Rating Scales

4. Scores on general level of tension will decrease following
the point of biofeedback training than prior to treatment.

5. Scores of the ability to modify levels of tension will
increase following the point of biofeedback training than
prior to treatment.

6. Scores of the ability to detect change in levels of tension
will increase following the point of biofeedback training

than prior to treatment.

Definition of Terms
The following definitions are listed to clarify terms used in
this study:

Peripheral nervous system. The peripheral nervous system con-

 

sists of nerves entering and leaving the brain and spinal cord and

is subdivided into two other systems: (a) The Somatic Nervous System
or skeletal-muscle system, consists of motor nerves emerging from

the spinal cord in the ventral roots and sensory nerves entering

the dorsal roots. These receptors carry information to nerve
impulses which innervate the Central Nervous System. These sensory
and motor systems are integrated at several points in the nervous
system, including the spinal cord, cerebellum, and cerebral cortex.
(b) The Autonomic Nervous System consists of sensory and motor

nerves serving the heart, glands, and smooth muscle of the viscera.
These structures maintain homeostasis ofinternal mechanisms (Tuttle &

Schottelius, 1969).

 

 

lO

Antipsychotic medications. These drugs are known as major
tranquilizers or neuroleptics and are often the drugs of choice used
in the treatment of schizophrenia. Subclassifications of these
drugs are penothiazine derivatives, thioxanthene derivatives, and
butyrophenone derivatives (Freedman, Sadock, & Kaplan, 1976).

Schizophrenia.’ Schizophrenia is a mental disorder of psychotic

 

level characterized by disturbances in thinking, mood, and behavior.
The causes of schizophrenia remain unknown. The disorder consists
of the following subclassifications: hebephrenic, catatonic, para-
noid, schizoaffective, childhood, residual, latent, chronic undif-
ferentiated, and acute schizophrenia (Freedman, Sadock, & Kaplan,
1976). This study used the Taylor and Abrams (1975) modification of
the Feighner et al. (1972) criteria for schizophrenia. The follow-
ing are suggested criteria for use in research:

1. Duration of episode greater than six months.

2. Clear consciousness.

3. Presence of either delusions, hallucinations, or formal
thought disorder. Thought disorders include the following:
(a) verbigeration (meaningless repetition of words or phrases),
(b) nonsequiturs (illogical thought whereby an inference or conclu-
sion does not follow from established evidence), (c) word approxi-
mation, (d) neologisms (a new word or condensation of several words
which are formed in order to express a complex idea), (e) blocking
(thought deprivation or involuntary cessation of thought or speech),
(f) derailment (many incomprehensible and disconnected utterances).

These six criteria refer to the lack of understandable speech.

 

 

 

 

4. Absence of broad affect.

5. Absence of signs and symptoms sufficient to make a diag-
nosis of affective disease.

6. Absence of alcoholism or drug abuse within one year of
admission.

7. Absence of clinical signs and symptoms of coarse brain dis—
ease or major medical illness known to produce significant behavioral
changes.

Biofeedback. Biofeedback is the use of instrumentation to
reflect physiological processes of which the individual is not nor-
mally aware. A person receives immediate information about a bio—
logical process, such as muscle tension, skin-surface temperature,
brain-wave activity, heart rate, blood pressure, and galvanic skin
response (Fuller, 1977).

Electromyogram (EMG). The EMG is a recording of electrical

 

responses from muscles which detects levels and changes in muscle
contractions (Schwartz & Beatty, 1977).

Electromyographic feedback. This process relays information

 

through visual or auditory amplification of levels and changes in
the contraction of muscles (Schwartz & Beatty, 1977).

Progressive relaxation. Progressive Relaxation is a technique

 

developed by E. Jacobson in 1938 which emphasizes the relaxation of
voluntary skeletal muscles (Benson, 1975). Sets of muscles are
alternately tensed and relaxed. The exercises emphasize discrimina-

tion of tension versus relaxation.

 

12

, Overview

In Chapter II, literature relevant to the theory and research
focusing on skeletal-muscle responses and physiological arousal in
schizophrenics is reviewed. Literature is also reviewed which
relates to studies in the field of electromyographic biofeedback and
Progressive Relaxation techniques. In addition, this chapter con-
tains general reviews of the possible relationship which exists
between the skeletal-muscle systems and other variables and the
effects of antipsychotic medications on skeletal-muscle measures.
In Chapter III the design and analysis are presented, including a
description of the sample, methodology, and instrumentation. The
formal research hypotheses are also presented in this chapter. An
analysis of the data is presented in Chapter IV with an interpreta-
tion of the results for each hypothesis. Included in Chapter V are
a summary of this investigation, a discussion of the findings, and
implications for further research in the area of EMG biofeedback

training for schizophrenics.

 

 

 

CHAPTER II

RELATED RESEARCH

The literature reviewed in this chapter is divided into seven
sections. The first is a review of the hypothesized relationship
between the skeletal-muscle system to other variables. It provides
background information for the next two sections, which are reviews
of research and theory focusing on skeletal-muscle responses and
physiological arousal in schizophrenics. The fourth and fifth sec-
tions pertain to electromyographic biofeedback and Progressive Relaxa-
tion, techniques which have been used to decrease skeletal-muscle
tension in various patient populations. The following section reviews
the effects of neuroleptic medications on skeletal-muscle measures.
(These medications are the preferred drugs used in persons diagnosed
as schizophrenic.) The last section is a summary of the implications

of the research for the present study.

 

Relationship of Skeletal-Muscle System
to Other Variables

In order to understand the significance of some of the studies
presented in this chapter, it is important to understand the hypothe-
sized relationship which exists between the skeletal-muscle system
and other variables involving cognition and behavior. The symptoms

of schizophrenia permeate all three areas of functioning.

l3

 

 

14

Figure 2.1 is a schematic representation of brain structures
which serve as a relay system for excitatory and inhibitory impulses
going to and coming from other parts of the body, including the
skeletal-musculature. In essence, the process which will be described
is an elaborate feedback loop where impulses originating at one site
can impact on and affect other sites.

It is generally agreed that the hypothalamus and portions of
the brain stem form the Reticular Activating System (RAS), believed
to be that part of the brain which mediates inhibitory and excita—
tory impulses. Neuronal pathways form connections with the RAS and
structures such as the limbic system, the cortex, and the skeletal-
musculature. The limbic system is believed to contribute to the
subjective experience of emotions, while the cortex is that part of
the brain where cognition or thinking is represented (Lezak, l976).

Gellhorn (1964) reviewed experimental evidence which examined
the interactive effects between the somatic (skeletal-muscle) system
and structures within the brain. It was observed that an increase
in proprioceptive discharges in the skeletal-musculature was corre-
lated with states of mental alertness and cortical excitation. Con-
versely, a decrease or inhibition of proprioceptive discharges was
correlated with a diminished state of cortical excitation. Utiliz-
ing the figure and descriptions of structures within the brain, it
can be seen how states of excitation or inhibition might also be
occurring simultaneously at other sites.

Whatmore and Kohli (1974) focused on the skeletal-muscle system

as being most important in the maintenance of fear and symptoms of

 

 

15

A.¢mmp .w—sox a wLoEpmcz

 

 

 

 

                 

 

 

 

 

 

 

 

 

 

 

 

 

 

soggy .xmpcooom: op mpomse _mgwﬁmxm Eocw mmeSpmm H—.N mczmwm
«3353
63362505
- 200 35% A L. 23:8
new 82335 .-.....H......\ C2285
E2m>m
9.29:3 830:0: .. .
xmtoo tomcom .m,“ .
can 3829: o... KmanEmooan—E
3.3.2.83321/ 3529:. 3:294
m:Em.a50a>I.\ /m=Em_a.._._.
5.50m
€05 623:3
E2»? 03.5.. \ 3.60 .222
__:m:_0 «22.0 new BEEP;—

xetooooz

 

 

 

 

l6

anxiety. They hypothesized that excessive skeletal-muscle activity

observed in these emotional states resulted in excitatory signaling

to the Reticular Activating System, which in turn affected states of
emotion through effects on the limbic system as well as on ideation

through effects at the cortical level.

Psychophysiological research in the field of schizophrenia
attempts to find physiological correlates of emotion and cognition.
Lang, Rice, and Sternbach (in Greenfield & Sternbach, 1972) emphasized
that an understanding of the interrelation of response systems is
fundamental to the psychology of emotion and to the development of
the effective treatment of emotional disorders. The literature
reviewed in this chapter examines data which integrate physiological
arousal and skeletal-muscle response tendencies in schizophrenics as
they relate to other response systems.

The Relationship of Skeletal-Muscle Response
Tendencies to Other Variables in Schizophrenia

Studies in the field of schizophrenia are often confusing because
of the heterogeneous nature of this diagnostic group. There are
numerous ways of subclassifying schizophrenics to create more homoge-
neous samples for study. Three of the most common classificatory
systems found in the literature include: (a) process/reactive,

(b) chronic/acute, and (c) paranoid/nonparanoid. The criteria for
these groups are based on duration of illness, history of the develop-
ment of the illness, and type and severity of symptomatology. An
attempt will be made to clarify the nature of samples used in the

studies reviewed.

 

17

The results of years of investigation can be summarized as
follows. Among schizophrenics, (a) skeletal—muscle measures differ—
entiate among the subject groups more clearly than other measures,
such as heart rate or brain wave activity; (b) a high degree of con-
sistency exists among individuals in measures of muscular tension
from one stress situation to another; (c) disturbances in striate
muscle activity correspond to degree of anxiety; (d) skeletal-muscle
tension bears a relationship to symptom production; (e) skeletal-
muscle activity is believed to be representative of stress reactions
experienced in everyday life situations; and finally, (f) observed
emotional withdrawal and emotional blunting in schizophrenics are not
indicative of general diminished physiological responsiveness to
stimuli.

Malmo and his co-workers generated considerable interest in the
role of muscle tension in symptom production and its relationship to
emotional states. They looked at changes in physiological variables
during the presentation of experimental stressors in psychiatric
patients. In 1949, Malmo and Shagass studied reactions to thermal
pain stimulation in an attempt to investigate the relationship of
physiologic disturbance to anxiety. They made comparisons among
"early" schizophrenics (mean duration of illness was 1.5 years),
psychiatric patients with anxiety as the predominant symptom, a mixed
group of psychiatric patients, and a nonpatient control group. They
found that the greatest degree of physiological disturbance occurred
among the most anxious and early schizophrenic groups. This occurred

during a seven-minute prestimulus period as well as during pain

 

 

18

stimulation. In addition, the clearest correspondence between degree
of anxiety and degree of physiologic disturbance was fOund in measures
of striate muscle activity which were similar among anxious and
early schizophrenic groups. Overreaction to pain stimulation, as
reflected in the signaling of pain experience with stimuli of low
intensity, occurred more frequently among both groups (anxious and
schizophrenic), with the least amount of discrimination occurring among
early schizophrenics. They concluded that this heightened state of
expectation also appeared to be characteristic of degree of anxiety.
These findings provided the groundwork for the following investigations.
Malmo, Shagass, and Davis (1951) expanded on the 1949 study
with the addition of a group of chronic schizophrenic patients.
(Mean duration of illness was 4.5 years, and all had been committed
to an institution for an average of three years.) Two more stressors,
rapid discrimination and mirror drawing tasks, were presented to the
subjects. In general, they found that muscular tension tended to be
abnormally high in psychiatric patients. During the preparation
(instruction) period of the tests, while marked elevations were
observed on some autonomic measures in all groups, only skeletal-
muscle tension was considerably greater in the patients. These con-
siderations added further weight to the issue that the skeletal-motor
aspect of emotional states is an important one. Muscular tension was
also observed to be a highly consistent measure within individuals from
one: stress situation to another. Finally, levels of muscle tension
under stress were as high in chronic schizophrenics as in other psy-

chiatric patients, with the exception that they were hyporeactive to

 

 

19

pain stimulation. The authors concluded that excessive muscular
tension was probably of considerable importance in the production of
symptoms.

Using data from the Malmo, Shagass, and Davis study, Malmo,
Shagass, and Smith (1951) addressed an additional issue of observed
behavioral withdrawal and affective dulling in chronic schizo-
phrenics. Such symptoms had been tentatively explained by others in
terms of low physiological reactivity. Malmo, Shagass, and Smith
summarized that chronics, while being equally reactive on skeletal-
muscle measures in anticipation of a stressor, were less reactive on
presentation of the stressor. In other words, tension levels remained
high, but showed little fluctuation in reaction to the stressors.
Chronics also executed fewer button presses, an indication of the
experience of pain, than did the other patients. The authors con-
cluded that the withdrawal and affective dulling observed in schizo-
phrenics were the result of a diminished ability to execute purpose-
ful behavior in response to environmental cues, rather than being
due to diminished Central Nervous System activity.

Goldstein (1965) also investigated the relationship between
anxiety, overt emotionality, and physiologic responsiveness. She
compared groups of subjects with diagnoses of psychosis (10 of 15 of
these patients were schizophrenic and none were chronic), neurosis,
and character disorder with normal controls. All subjects were rated
for degree of anxiety on the Taylor Manifest Anxiety Scale and the
Freeman Manifest Anxiety Scale. Results showed that psychotics

exhibited the highest levels of response on skeletal-muscle measures

 

 

20

during the presentation of a stressor (an auditory stimulus), and
they were significantly more reactive, with greater variability from
baseline to stress situations, on skeletal-muscle measures. She
concluded that degree of physiologic disturbance was related to
degree of psychological maladjustment despite observations of dimin-
ished emotional responsivity among psychotics.

Whatmore and Ellis (1958) observed levels of skeletal-muscle
tension in schizophrenics during a 30-minute rest period and found
exaggerated motor activity compared to a group of normal controls.
Although it was not stated, the schizophrenics in this study appeared
to resemble Malmo's early or acute schizophrenics. Whatmore continued
observations of skeletal-muscle activity among various patient popula-
tions and found that schizophrenics who were in remission exhibited
labile EMGs, with day-to—day readings alternating from being markedly
elevated to falling within the range of the nonpatient control sub-
jects (Whatmore, 1966). Important here is that EMGs were abnormally
variable when compared to a group of depressed controls, even though
the patients were in excellent clinical condition. He observed that
an increase in lability and an elevation of baseline measures preceded
acute psychotic episodes, with the episodes occurring at high points
in the EMG curve. Conversely, a sustained reduction in electromyo-
graphic levels was followed by improvement in clinical condition.

His findings, that there was a relationship between manifest symptoma-
tology and skeletal—muscle tension,1ed to the conclusion that there
was an important connection between such tension and the variables

of emotion and cognition. He hypothesized that hyperactivity in the

 

21

skeletal-musculature was characteristic of levels of arousal or
excitation also occurring within the limbic system, Reticular Activat-
ing System, and cortical sites within the brain.

Later studies confirmed that schizophrenics, with labels of
acute and chronic, and process and reactive, were overaroused on
measures of skeletal-muscle tension. This was consistent during
rest, and during the presentation of stressors. This was not true
for paranoid subjects, which is why individuals falling within this
subgroup were excluded from the present study. Fenz and Velner
(1970) concluded that a common basic deficit in schizophrenics lies
in their inability to modulate or change levels of responsiveness to
correspond to variations in internal and external stimulation. In
behavioral terms, the deficit seems to indicate that the schizophrenic
individual is unable to evaluate the emotional demands of a situation
and to respond to it appropriately. Issues of overarousal at the
physiological level and the effects upon behavior and cognition are
central to psychophysiological theories of schizophrenia, which are
reviewed in the next section.

Theories of Psychophysiological Arousal
in Schizophrenia

 

The theories of Mednick, Broen, and Tecce and Cole are reviewed
in this section. They are exemplary of theories which focus on dis-
turbances in psychological functions, such as attention and learning,
and the relationship of these functions to physiological variables.

A common theme of these three theories is that in schizophrenia, a

state of physiological disequilibrium exists, whereby chronic states

 

 

 

22

of arousal impact on the individual's ability to process and filter
information. The behavioral results are seen as disorganization at
the cortical level, producing such symptoms as the inability to focus
thoughts, increased attention to irrelevant stimuli, and delusional
thinking. The theories do not explain etiology, but attempt to
describe psychophysiological interactions.

Mednick (1958) suggested that there must be a predisposition for
schizophrenia, possibly genetic in nature, which results in limita-
tions in terms of the individual's ability to adapt adequately to
events in his environment. The characteristics of this predisposi-
tion involve excessively strong physiological reactions to mild stress
and excessively slow recovery rates of physiological arousal. Mednick
hypothesized that this arousal may have a role in the development of
certain learned defensive cognitive styles which characterize the
thought disorder of schizophrenia. One of the key factors of this
theory is that the individual learns to focus on nonstressful stimuli.

Physiological arousal was viewed by Mednick as a variable which
contributes to the strength of the individual's drives, where increas-
ing arousal is accompanied by increased drive levels. The preschizo-
phrenic, experiencing excessive arousal, would be in a heightened
state of drive according to this theory. Mednick further predicted
that increasing drive states resulted in increased response strength
of any habit tendencies. Consequently, the highly aroused preschizo—

phrenic individual responds to a wider range of stimuli more quickly

and with greater intensity. In terms of the learning theory on which

 

 

 

 

23

Mednick based his theory, conditioned responses are acquired more
rapidly and excessive stimulus generalization occurs.

Following is a brief description of the progression of the
schizophrenic psychosis. The preschizophrenic was described by
Mednick as being an extremely anxious individual, fearful of many
situations. New stimulus events and experiences which appear simi-
lar to those the individual has already learned to fear tend also to
elicit anxiety responses. There appears to be an almost circular
chain of events: increased arousal, increased drive levels, exces-
sive stimulus generalization, and further heightened arousal. The
individual, seeking to reduce levels of anxiety, learns to avoid the
feared events, but since the original fears never extinguish, this
does not represent a successful adaptation. The psychosis develops
out of these learned defensive habits and usually is precipitated by
an event or crisis which interfered with the anxiety—avoidant
responses. This event serves to exacerbate the individual's already
high levels of anxiety. (Anxiety and arousal appear to be used inter-
changeably in Mednick's theory.) The conditions which have been
described continue to escalate: drive levels become unusually high
and generalization to stimuli is increased, whereby an even greater
number of events provoke fear responses. Finally, an upper limit
or "threshold" of psychological and physiological arousal and stimulus
generalization is reached. The individual can no longer control the
number and quality of thoughts. The ability to discriminate relevant
from irrelevant stimuli is diminished and increased generalization

results in remote associations gaining in response strength. Thoughts

 

 

 

 

24

speed, associations are not logical, and the individual attempts to
reach a rational solution to these fearful events in the form of
delusions.

Mednick explained the cognitive process of schizophrenia pri-
marily as a learned problem, and he cited important research in con-
ditioning and generalization which supported such an interpretation.
Broen's theory, which is reviewed next, postulates that set patterns
of physiological responding are causally related to schizophrenic
symptomatology. His idea about the developing cognitive style of
the schizophrenic appears quite similar to Mednick's, even though the
emphasis is on physiological rather than learning factors.

Broen (1968) agreed with Mednick that the schizophrenic is highly
anxiety—prone, with slow rates of recovery from high levels of physio-
logical arousal. He cited research which demonstrated that skeletal-
muscle tension is a consistent and predominant indication of this
arousal. He also hypothesized that changes in cognition were a direct
result of this chronic arousal. Two factors, physiological arousal

and "response strength ceiling, are central to Broen's theory of
schizophrenia.

He explained response strength ceiling in this way. In normals,
when a stimulus evokes more than one response tendency, the approp-
riate response tendency is much stronger than competing response
tendencies and, therefore, will usually occur. These response ten-

dencies are hierarchically ordered in accordance with their approp-

riateness. In schizophrenics, this hierarchy seems to be partially

collapsed, where the strengths of dominant and competing responses

 

 

 

 

 

25

are more nearly equal. The collapse in the hierarchy of responses is
also viewed as being a predispositional factor. The interaction of
arousal and collapsed response strength ceilings results in cogni—
tive disorientation.

The theory states that since the recovery from arousal in
schizophrenics is retarded, new situations tend to have an additive
effect on tension which has not yet abated. As with Mednick, Broen
postulated that physiological arousal increases attentiveness to
remote associations and stimuli. When arousal is initially high and
the strongest response in a response hierarchy is restricted at
ceiling strength, an increase in arousal can only facilitate the
competing responses, and the result is a sensory overload on a cog-
nitive system which is capable of processing a limited amount of
information. This hypothesis parallels the upper threshold of stimu-
lation which Mednick referred to.

Broen cited the research of Malmo, which indicated that as
levels of physiological activation increase in the schizophrenic,
responsivity to stimuli decreases. He hypothesized that a physio-
logical inhibitory set occurs with thoughts directed toward nonstress
ideation. This represents an effort to decrease arousal and restabil—
ize or reach some form of homeostasis.

The result of inhibitory feedback is not always simply to

control the overactive system. . . . Inhibitory processes
stimulated by the activation of a physiological system can
have fairly important effects on other systems. . . . For

example, it seems that a spurt in cardiovascular functioning
activates an inhibitory process that not only provides homeo-
static feedback to the heart, but also tends to counteract
the activation of processes involved with sensory receptivity
(Broen, 1968, p. 216).

 

 

 

26

Broen's statement suggests that tension and physiological arousal
are causally related to behavioral changes in schizophrenics,
specifically, attention to remote stimuli and decreased utilization
of cues in the environment.

Tecce and Cole (in Mostofsky, 1976) presented a theory of
schizophrenia which incorporated other theories and empirical find-
ings already reviewed. The theory attempts to define the relation-
ship which exists between physiological arousal and cognitive behav-
iors of attention and distraction. These two concepts are similar
to the concepts of excessive stimulus generalization of Mednick's
theory and the disturbances of associative processes due to irrele-
vant responses of Broen's theory. Tecce and Cole considered the
relationship between arousal, represented in the Reticular Activating
System, and in changes of skeletal-muscle tension, as being a complex
interaction between states of excitation and inhibition which impact
on cognitive interference. Like Broen and Mednick, they did not
attempt to resolve the cause-effect issue in terms of the etiology
of schizophrenia.

Tecce and Cole hypothesized a two-process theoretical model,
based in part on the earlier works of a theory by Hebb. The two
hypotheses, schematicized in Figures 2.2 and 2.3, are:

1. Performance of schizophrenics is positively and mono-
tonically related to attention.

2. Performance of schizophrenics is nonmonotonically (an
inverted U) related to levels of arousal.

Stated differently, performance increases with increasing attention,

whereas performance increases with increasing arousal only to a

 

 

 

 

27

Performance '

 

Low

 

Low High'
Arousal

Figure 2.2: Nonmonotonic relationship between Arousal
and Performance

High

Performance

 

Low

 

Low High
Attention

Figure 2.3: Monotonic relationship between Attention and
Performance.

 

 

 

28

maximum level. As arousal increases above maximum level, a decre-
ment in performance occurs. This decrement in performance occurs
because excessively high levels of arousal produce behavioral impair-
ment by distraction. Conversely, Tecce and Cole stated that dis-
traction can also produce elevations in arousal.

The interaction between arousal, attention, and distraction
has its basis in the excitatory and inhibitory regulation of the
Reticular Activating System according to this theory. Tecce and Cole
concluded:

The direct exclusion of extraneous stimuli (including irrele-

vant thoughts) by focused and narrowed attention processes

appears to occupy a central role in psychotherapeutic and

drug treatment methods. Transcendental meditation appears

to be an especially effective technique in altering states

of consciousness by producing a decrease in distraction and

a lowering of physiological functions (Mostofsky, 1976, p. 208).

The following sections reviewed in this chapter deal with ele-
ments of biofeedback and progressive relaxation which can also be
instrumental in decreasing levels of physiological arousal.

Electromyographic Biofeedback as a Method
of Relaxation Training

 

 

Biofeedback involves the use of instrumentation to reflect
psychophysiological processes of which the individual is not normally
aware and which may be brought under voluntary control. It is the
integration of bodily functions and cognition, or the mind-body rela-
tionship, which is central to the concepts of biofeedback. A
person is given information about his or her biological condi-
tions such as muscle tension, skin-surface temperature, or brain

wave activity, and the cognitive awareness or understanding of a

 

 

 

 

29

physiological activity enables the individual to alter the physio-
logical response (Fuller, 1977). EMG biofeedback is used in relaxa-
tion training to effect a decrease in muscle—contraction activity.
Electrodes attached on the surface of the skin over a particular
muscle measure electrical discharges from the muscle fibers. The
electrical activity comprises the EMG recording. As the muscle
becomes more tense, indicating increased contraction, the auditory
signal heard by the subject increases in pitch. This feedback pro-
vides immediate information to the subject of the state of muscle
tension (Olton & Noonberg, 1980). As Green and Green (1977) point
out, in learning to voluntarily control normally unconscious processes,
one does not become aware of the biological activity in neuronal path—
ways, but one increases awareness of the mental-emotional factors
which influence physiological states.

Brown (1978) summarized the concepts involved in the biofeed-
back process. First, conditioning results from rewarding biological
activities for performance, where the biofeedback monitor is used as
the reward in shaping the desired activity. Second, EMG biofeedback
is used as a stress-reduction technique, where a developing aware-
ness of the feeling of relaxation occurs, allowing the discrimination
of differences between tension and relaxation. Finally, biofeedback
involves cognitive mechanisms. She emphasized this concept of higher
mental activity as being an important factor involved in the control
of physiological functions.

Brown believed that in complex emotional problems, where muscle

relaxation was achieved, it seemed that the patient's attention

 

 

3O

shifted to insights related to the problem as the individual became
aware of the subjective mechanisms which maintained stress. Stress
is maintained by: (a) excessive rumination and preoccupation which
focus on the stressors, and (b) internal feedback of information
about the status of physiological reactions to stress, where states
of tension are interpreted as emotional stress reactions which
become reinforcing. Brown stated:

Between the information input and the resulting learned

physiological control, extraordinarily complex information

processing must occur, where appropriate information to

the brain is associated, integrated, evaluated, put into

memory, and the product of this brain activity is used to

activate patterns of neural activity that result in dis-

creet, channelled directions to control the selected
physiologic activity (p. 22).

 

Brown believed that in biofeedback, attention is redirected from
fixed, self-reinforcing, circular mental activities by supplying
information about the physiological mechanisms involved in maintain-
ing such mental activity. This idea seems to be compatible with the
theories of arousal in schizophrenia which emphasized that thoughts
of the schizophrenic are redirected to inappropriate ideation in an
effort to reduce stress. Clearly, this type of thought distraction
contributes to the maladaptive behavior of the schizophrenic.
Gaardner and Montgomery (1977) provided a model, illustrated in
Figure 2.4, which describes how EMG biofeedback relates to other
variables. The figure illustrates the interaction between external
stimuli, muscular tension, and cortical activation. They endorsed
what others, including Jacobson (1938) and Benson (1975), have also

emphasized: that anxiety cannot exist in the presence of deep muscle

 

31

 

A.mRm_ .mewaompcoz w Locvcmww Eogmv

 

 

2:235
6:380:25
cc: o_Eo:2:<

 

 

 

 

E236

.aoo_ xownuwww FmomePmexzmocoxm:

"v.m mgzmw:

 

3.33:5.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1 O .I .ll 0.2m
l d% l I | _

_ _

_ Ammocmzoﬁmcoov _ mE2m>w
14% 68602: Ila—TV Bamcow Al EoE:9_>:m
AIM—1 33:60 AI..“I<. 2:339me

— J

_ a.» e: u ,

_ mcozocsu _

_ 39:22 0mm

_ cc: w>_::moo _

_ _

_

.lllllllllL

Emhw>w m=O>mmZ .— m ZmO

 

 

 

 

 

32

relaxation. Using the diagram, minimizing activity of the muscular
system (Channel C) and consequent focusing of conscious awareness
(Channel 0) prohibits the existence of physiological stress. Accord-
ing to this hypothesis, cognitive functions are, in part, dependent
on the state of the muscular and sensory systems.

The effects of muscle relaxation on other bodily systems were
also reviewed by Basmajian (1979). He observed that with the achieve-
ment of states of muscle relaxation, heart rate and cortical arousal
are decreased. He discussed the probability that in learned muscle
relaxation, brainstem inhibitory mechanisms were activated. The rela-
tionship of skeletal-muscle responses to brain structures was dis-
cussed in conjunction with Figure 2.1.

Schwartz and Beatty (1977) emphasized, as well, that EMG bio—
feedback training deals with somatic reactions which are part of a
broader set of issues relating to stress and adaptation to stress.
These stress reactions and emotional processes are the result of an
individual's cognitive appraisals of the significance of an event.

The control of such somatic reactions is seen as an important aspect
of the emotions and self-regulation. They believed that it is the
self-regulation which provides the link between biofeedback and the
emotions.

Clinical studies have shown that EMG relaxation training has
been used successfully for chronic anxiety with beneficial changes
in muscular tension and other anxiety symptoms. Some of these studies
are cited in the following section, which provides the rationale for

combining relaxation techniques. Several other researchers have

 

 

 

33

examined the efficacy of using EMG biofeedback as relaxation training
with various psychiatric populations.

Jessup and Neufeld (in Stoyva, Kamiya, Barber, Miller, & Shapiro,
1979) made comparisons between four relaxation techniques, including
EMG biofeedback, with hospitalized psychiatric patients. Fifteen out
of the 20 subjects had some form of depression as their major diag-
nosis. The diagnoses of the other five were not given, although none
of the subjects were psychotic. Subjects in the EMG biofeedback
group were given brief instructions over an intercom in which they
were asked to use a tone to help them relax. They were given four
sessions on consecutive days of training. The treatment did not sig-
nificantly affect EMG measures, and insignificant changes on a mood
checklist were found from pre to post testings. The authors con—
cluded that there was no support for using EMG biofeedback treatment
for inducing relaxation in the samples studied.

A second study by Ford, Stroebel, Strong, and Szarek (1981) pro-
vided data suggesting that Quieting Response Training, which combines
EMG biofeedback with thermal feedback, deep breathing, progressive
relaxation, and autogenic techniques, is a useful technique for use
with psychiatric inpatients. Although the study assessed a combina-
tion of techniques which were different from those used in the present

study, the findings suggested that patients with various diagnoses

 

 

could benefit from techniques which emphasize self-regulation. Thirty-

seven inpatients participated in the study with diagnoses which
included psychotic, neurotic, personality disorders, and I'special"

disorders. It is probable that some of the 15 psychotics were

 

 

 

 

 

 

34

schizophrenic, even though this was not specified. Outcome measures
which included changes in the frequency, severity, and duration of
symptoms as well as changes in medication indicated that success of
treatment was unrelated to diagnosis or presenting complaint. There
were no differences in the percentage of subjects who experienced
success when comparisons were made between the inpatients and a gen-
eral clinic population. In addition, there were no problems for
patients in terms of increases in pathological symptoms or motivation.

Studies support that the techniques outlined in the present study
would be beneficial for persons diagnosed as schizophrenic. These
individuals are known to be psychotic from time to time and manifest
some symptoms which are similar to individuals who are chronically
anxious.

The idea of self-regulation in biofeedback training seems par-
ticularly appealing in working with schizophrenics, who, because of
the nature of their illness, often become dependent on others to
provide continuous reassurance and support in treatment. To date,
psychotherapy, institutionalization, and antipsychotic medications
have been the means by which this support has been provided (Freedman,

Sadock, & Kaplan, 1976).

Progressive Relaxation

 

Progressive Relaxation is a set of exercises that emphasize the relaxa-
tion of skeletal muscles. The technique seeks to achieve increased
control over the skeletal muscles until the individual is able to

produce very low levels of tension in major muscle groups (Benson,

 

 

 

 

35

1975). The technique was originally developed by E. Jacobson, whose
major thesis was that anxiety and relaxation are mutually exclusive.
Since individuals generally have little awareness of the sensation of
relaxation, the technique requires that the individual tense a set of
muscles as hard as he can until real tension is experienced. The
muscles are then relaxed and the individual tries to focus on the
difference between tension and relaxation. As practice continues,
the individual begins to discriminate more finely the different
degrees of tension and relaxation (Brown, 1978). Jacobson's Progres-
sive Relaxation requires extensive amounts of time for practice.
Therefore, abbreviated instructions such as those provided by
Basmajian (1979) have been recommended for home practice. Modified
instructions were also used in this study.

The primary reason for combining EMG biofeedback with relaxa-
tion techniques is that they are complimentary and result in the
added efficiency and effectiveness of muscle relaxation. The follow-
ing studies indicate why this is the case.

Researchers have demonstrated that while self-report measures of
relaxation may be equal to, or higher, in groups of subjects training
in Progressive Relaxation, EMG measures are significantly lower in
groups training with EMG biofeedback. Canter, Kondo, and Knott (1975)
made comparisons between two groups of chronically anxious patients.
One group received EMG biofeedback training and the other was taught
progressive muscle relaxation. Patients given feedback training
learned to decrease EMG levels to a greater degree than patients who

learned Progressive Relaxation. Results from symptom checklist:s

 

 

 

 

 

36

revealed, however, that the latter group may have obtained greater
relief from anxiety. Reinking and Kohl (1975) also examined the
effectiveness of EMG relaxation training with other relaxation thera-
pies. While all groups in the study reported increased relaxation
on self-report measures, measures of EMG activity showed that the
speed of learning and depth of relaxation were superior with EMG
biofeedback.

While improvements in symptoms of anxiety have been found in
patient populations trained with EMG biofeedback, the question
arises as to the generalization of physiological relaxation to
untrained muscle sites. Alexander (1975) tested this hypothesis in
groups of normal adults trained on the frontalis muscle located on
the forehead. No generalization of relaxation occurred. While reduc—
tions in frontalis EMGs occurred, subjects showed increases for arm
EMGs, while values for a leg muscle remained at low levels throughout
sessions. Alexander also found no reliable correlation between sub-
jective feelings of relaxation and levels of frontalis muscle tension.
His use of normal samples with initially low EMG levels limits these
findings but raises important issues for combining techniques.

Coursey (1975) also investigated the problem of generalization
of relaxation. He concluded that EMG feedback was more effective in
lowering tension in a specific muscle than simple verbal instructions
to relax. The muscle site used for EMG training in this study was
also the frontalis. Results from a questionnaire showed that only
the upper parts of the body (face, forehead, upper arms, and hands)

correlated with the subjective sense of relaxation.

 

 

 

 

37

Several final points have been made which provide information
as to the effectiveness of combining EMG feedback with relaxation
instructions such as Progressive Relaxation. Lader and Matthews
(1971) emphasized that the neuroleptics seem to reduce subjective
feelings of arousal but do not actually affect EMG levels. The
physiological feedback of muscle tension allows one to observe the
unfelt tension. In addition, individuals prominent in the field of
biofeedback techniques, such as Gaardner and Montgomery, and Brown,
emphasized that home practice of relaxation is extremely important
since the major objective of the biofeedback treatment is to teach
the patient voluntary control of tension without the use of instru-
mentation. They generally suggest tape recordings of instructions
for relaxation, which the patient can use in the home between feed-
back sessions.

The Effects of Neuroleptics on Skeletal—
Muscle Measurement and Cognition

 

The widely accepted use of neuroleptics in the treatment of
schizophrenia necessitates a brief review of some of the changes in
symptomatology produced by these drugs.

Spohn, Lacoursiere, Thompson, and Coyne (1977) reported changes
toward normalization in the variables of attention and perception
along with a reduction in the severity of manifest symptomatology
following the administration of a neuroleptic. These changes were
attributed to the reductions in autonomic hyper-arousal in chronic
schizophrenics being treated with neuroleptics. There were no

reported changes observed on cognitive tests, which included a

 

38

shortened version of the Wechsler Adult Intelligence Scale and tests
of memory.

In a review of the effects of drugs on physiological activity,
Tecce and Cole (1972) summarized that the neuroleptics reduced
generally elevated basal levels of autonomic physiological activity
(except for heart rate), as well as reactivity to stimuli. Such
changes were often observed to be accompanied by behavioral improve-
ment. However, they were unable to find consistent evidence for
similar changes in EMG measures. The Spohn et al. study did not
include measures of EMG responses.

Lehman (1975) explained that neuroleptic drugs diminish arousal
of the Central Nervous System without inhibiting the higher cortical
centers at any dose level. He hypothesized that the result of the
administration of the drugs was to effect a decrease of excessive
perceptual input through inhibition at the reticular formation. This
may be one reason why changes in formal cognitive functioning are not
observed to occur with these drugs. Lader and Matthews (1971) noted
the drugs which act on the Central Nervous System may reduce subjec-
tive tension without accompanying reductions in EMG levels.

In some of the studies cited in this chapter, subjects were
receiving doses of a neuroleptic medication. The results indicated
that the ability to modulate skeletal-muscle tension during rest and
in response to stressors continued to be problematic.

Neuroleptics appear to normalize autonomic activity and reac-
tivity, to improve attentional-perceptual deficits, and to reduce

subjective anxiety. It should be expected, then, that in the

 

 

 

39

population under study, one would find a low correlation between
subjective tension and EMG levels. The therapeutic effectiveness of
normalizing EMG levels remains open to question, particularly as it
might affect cognitive performance and the individual's perceived

ability to modify stress reactions.

Summary and Implications of Related Research

 

The review of psychophysiological research in schizophrenia has
shown that schizophrenics are abnormally aroused on skeletal-muscle
measures. It is believed that this tension is representative of
anxiety experienced by the schizophrenic and that a relationship
exists between such tension and the maladaptive behaviors observed in
schizophrenia. Specifically, an increase in tension has been observed
to occur prior to the appearance of symptoms, and a decrease in ten-
sion accompanies clinical improvement. To date, neuroleptic medica-
tions have been one of the most effective means of treatment for the
schizophrenic, resulting in improvements of attention and perception
as well as decreasing autonomic arousal. These medications do not
appear to impact, however, on skeletal-muscle tension, which remains
elevated.

Combined techniques of electromyographic biofeedback and Progres-
sive Relaxation have been effective in reducing skeletal—muscle ten—
sion and in decreasing subjective feelings of tension in various
patient populations, including chronically anxious and psychotic
individuals. In addition, it is emphasized in the literature that

one of the primary aims of these techniques is to help the individual

4o

become more aware of stress reactions represented in the skeletal-
muscle system by (a) focusing on minute-to-minute alterations in
tension and (b) becoming more aware of factors, internal and external,
which serve to maintain or increase tension.

In this study, persons diagnosed as schizophrenic were taught
to alter skeletal-muscle tension, utilizing a combined program of EMG
biofeedback and modified Progressive Relaxation exercises. Changes
in performance on cognitive tasks, subjective feelings of tension,
anxiety, and EMG levels were measured to examine the effectiveness of

the treatment program.

 

CHAPTER III

DESIGN OF THE STUDY

The design of the study involved recording EMG averages and
administering the Repetitive Psychometric Measures and Self—Rating
Scales of tension during Baseline and Treatment phases to subjects
on an individual basis. An additional measure, the Taylor Manifest
Anxiety Scale, was administered pre-Baseline and post-Treatment.

These results were included in the Clinical Observations of Chapter IV
but were not included in hypothesis testing. A description of the
sample, design, methodology, and analysis arepresented in this chap-

ter. A description of the split-middle technique is also included.

Sample

Following is a description of the four subjects selected for the
study:

Subject 1 was a 20—year-old, single male who lived at home with
his divorced mother and younger brother. He graduated from high school
and was enrolled in a vocational rehabilitation program during the
study. He had a history of three psychiatric hospitalizations, each
lasting about four weeks. He was being maintained on 400 mg of
Thorazine per day for six months prior to the study. This was changed
to 50 mg of Loxitane per day by the end of the study. Both medications
are antipsychotic agents.

41

42

Subject 2 was a 37-year-old, married woman with one teenage
daughter. The subject was the elder of two children. At the time
of the study, she was enrolled in graduate-level courses and had
earned one masters-level degree. She had a history of six previous
psychiatric hospitalizations, the first occurring after the birth of
her daughter. Length of stays in the hospital varied in duration,
and none was less than several weeks. Her medications included 3-5 mg
of Prolixin, 900 mg of Lithium, and 100 mg of Seconol per day. Only
the Prolixin is an antipsychotic medication. The Lithium was used in
combination with the Prolixin, and the Seconol is a sedative used for
sleep.

Subject 3 was a 24-year-old woman who was single and lived at
home with her mother and father. She had one older brother who lived
away from home. She completed 11 grades of school and did not have a
high school diploma. She was attending a day program at a community
mental health center and was enrolled in two high school equivalent
classes at the time of the study. Her first psychiatric hospitaliza-
tion occurred when she was 11 years old, and she had subsequently been
admitted to psychiatric hospitals at least eight times. One of the
hospitalizations, at age 16, lasted for three years. Throughout the
study, she was taking 100 mg of Serental per day, which is an anti-
psychotic medication.

Subject 4 was a 28-year-old, divorced woman with one child, whom
she gave up for adoption. She lived with her older brother but fre-
quently spent evenings at her parents' home. At the time of the

study, she had an associate degree and was attending day care at a

43

community mental health center. She was also enrolled in two college
courses at a junior college. She had three psychiatric hospitaliza-
tions with stays from one to two months in duration. At the time of
the study, she had been maintained on 20 mg of Navane for one year.
This antipsychotic medication was discontinued on the ninth training
session of the study.

Initially, the study was presented to psychiatrists in the
East Lansing and Lansing areas as a research project designed to
investigate the effects of an electromyographic biofeedback treatment
program on EMG levels, subjective experiences of tension, and on cog-
nitive performance in persons diagnosed as schizophrenic. A request
was made that psychiatrists recommend clients for the study who they
felt were not classified as paranoid schizophrenic and who (a) fit
the research criteria, (b) were stable enough to participate, and
(c) experienced some difficulty with tension. In turn, clients who
agreed to participate with the consent of their psychiatrist were
asked to contact the experimenter. This arrangement insured that
issues of client/physician confidentiality were honored.

Each potential subject was given an introductory session in which
the purpose and procedures of the study were reviewed. They were
informed that three to four hour—long sessions per week were required,
and they were shown examples of questions and tests which would be
administered during the study. In addition, clients were told that
they would be asked to practice relaxation at home with the use of

tape-recorded instructions provided by the experimenter. This session

44

was also used to demonstrate the biofeedback equipment, with an
explanation of the measurement of muscle tension.

Four out of five individuals interviewed were selected for the
study. One male was excluded because of his evident paranoid delu-
sions, which might have prevented him from understanding and using
the instructions in a straightforward manner without incorporating

them into his delusions.

Instrumentation

Three types of measures were used in order to test the major
hypotheses. They included (a) three Self-Report Rating Scales,
(b) frontalis EMG values, and (c) five cognitive/perceptual measures
known as the Repetitive Psychometric Measures. In addition, the
Taylor Manifest Anxiety Scale was used to compare pre- and post-
measures of anxiety for individuals. Test scores from the TMAS
could not be statistically analyzed because they were not a continuous
measure, which is a requirement of the research design. Results from

this measure were included in the Clinical Observations of Chapter IV.

Self-Report Rating Scales

Three scales were developed by the experimenter in order to pro-
vide quantitative measurements of subjective changes in tension
experienced throughout the treatment program. In addition to the
quantitative data, another purpose for utilizing these measures was
to enhance the subject's awareness of variations in states of tension.

Subjects rated themselves on a scale of 1-5 or l-lO, depending on

45

the scale, for: general level of tension (Scale 1), ability to
decrease tension at will (Scale 2), and awareness of changes in ten-

sion throughout the day (Scale 3). (See Appendix A.)

Biofeedback Eguipment

Information about the feedback myograph and digital integrator
was obtained from the instruction manuals for the Autogen 15006 and
HT—lO by Autogenic Systems, Inc., of Berkeley, California. The feed-
back myograph used was the 1500c model by Autogenics Systems. This
model detects signals as low as 0.1 microvolts (NV) and has six audio
feedback modes. The EMG amplitude is an integral averaged measure
rather than a peak-to-peak measure. The unit is equipped with a
standard bandpass of 100-200 Hz (cycles per second). Visual feed-
back is presented on a logarithmically scaled amplitude meter. There
are three silver/silver chloride surface electrodes provided which
are attached to the skin by adhesive discs. The digital integrator,
the Autogen HT-lO model, is an instrument which provides average EMG
values during data collection. It can be calibrated to average values

from one second to 20 minutes.

Rppetitive Psychometric Measures

 

These five tests by Moran and Mefferd (1959) were developed for
study of physiological and biochemical correlates of human behavior.
They were adapted from tests already known to represent distinct
factors of mental abilities through factor—analytic studies. The

measures may be repeated often because of the 20 alternate, equated

 

46

forms for each test. They are timed, paper-and-pencil tests, requir-
ing a total time of about 15 minutes to administer. (See Appendix B
for examples.) Following is a description of each test.

1. Flexibility of Closure (FC). The task of FC is to retain
the image of a given configuration despite the influence
of other distracting configurations. The subject copies
specified geometric figures onto matrices of dots. The
score is the number of figures correctly copied in 3
minutes.

2. Number Facility (NF). This test requires the addition of
one- or two-digit numbers in sets of three. The score is
the number of correct answers obtained in 3 minutes.

3. Perceptual Speed (PS). This test measures the speed with
which an individual crosses out every digit in rows of ran-
dom digits, which is like the one indicated for that par-
ticular row. The score is the number of digits correctly
marked in 2.5 minutes.

4. Speed of Closure (SC). This test measures the ability to
unify a disparate perceptual field into a single percept.
The task is to circle all four-letter words in rows of
random letters. The score is the number of correctly
circled words in 2.5 minutes.
5. Visualization (V). This test is associated with the ability
to visualize the outcome of objects in space. The subject's
task is to visually follow the path of a line embedded in a
set of tangled lines, from left to right, and to place the
number of the line in the appropriate cell. The score is
the number of cells correctly numbered in 3 minutes.
Standardization of the measures was obtained initially by con-
structing 30 alternate forms of each test through random assignment
of items from a large pool of items. All 30 forms of each test were
administered twice to subjects in random order. Paired scores were
plotted and items were exchanged between forms in which disparate
levels of difficulty were observed. The procedure, using the new
forms, was repeated on new subjects. Each form of the test was

adjusted toward the mean performance. All forms which varied more

47

than 0.5 50 from the subject's own means were discarded. The 20
forms of the test which varied least from the subject's means were
selected.

Intercorrelations between the tests were established by admin-
istering Form 1 to 164 subjects. The correlations ranged from .09 to
.44. Test-retest correlations between Form 1 and Form 2 ranged from
.72 to .94.

In a second study, Moran and Kimble, Jr. (1964) equated the 20
alternate forms for level of difficulty. They administered the forms
to samples of 544 and 508 college students. All alternate forms were
found to be reliably different in difficulty level for all but the
Number Facility Test. Correction factors were provided in order to
decrease this variability.

Moran and Mefferd claimed that the tests were appropriate for
use with psychiatric populations in terms of their ability to sustain
levels of motivation. They found that such individuals were able to

perform satisfactorily for as many as 246 consecutive days.

Taylor Manifest Anxiety Scale (TMAS)

 

This true-false test was devised by Janet Taylor (1953) to measure
manifest symptoms of anxiety. Five clinicians rated 200 items from
the Minnesota Multiphasic Personality Inventory (MMPI) for their
descriptions of manifest anxiety. Sixty-five items were retained
with 80% agreement. These items plus 135 buffer items were given to
a normative sample of 352 students. Several modifications retained

50 of the original 65 items through an analysis of internal consistency.

 

48

The buffer items were changed so that the total test has 225 items.
This latter form is referred to as the Biographical Inventory. Many
studies have been carried out on the 50-item test, which was revised
in terms of wording to increase its applicability to various popu-
lations. This 50-item test (see Appendix C) is known as the At scale
in Dahlstrom et a1. (1975). The following data refer to the At scale,
which was used in this study.

A test of clinical validity reported a correlation of .44
(significant beyond the .01 level) between psychologists' ratings
and scores on the At scale. Mean scores of 233 normals and 51 neu—
rotic patients were reliably different at the .001 level. Other
studies show mixed results with respect to clinical ratings and
scores on the scale. Apparently, the At scale doesn't consistently
agree with observers' ratings of anxiety. It does, however, discrimi-
nate between various psychiatric populations and normals with differ-
ences in scores being statistically significant beyond the .01 level.
The scale has a correlation of .92 with the Pt scale of the MMPI,
which measures such things as anxiety and dread, low self-confidence,

doubts, sensitivity, and moodiness.

Research Design
This study employed a single case study with an A-B time series
research design. The notation for this design according to Kratoch-
will (1978), where repeated interventions are presented in the B

Phase, is: 010 O 0 O O 0 I 0 I O ..I 0 Observations across

2 3 4 5 6 7 1 8 2 9" 9 12'
Baseline and Treatment are represented by 0. I signifies the

 

49

introduction of treatment. The dependent variables measured on a
repeated or continued basis over time included: electromyogram
values (EMGs), three Self-Report Rating Scales, and five cognitive
tests (Repetitive Psychometric Measures). The treatment effects of
electromyographic biofeedback training, combined with relaxation
exercises, were contrasted against the dependent variables across

four individuals at the same relative points in time of the treatment.

Methodology

 

All sessions presented during the study were given on an indi-
vidual basis and averaged 60 minutes for each session. Subjects
participated in four sessions per week for a total of 22 sessions
per subject. With the exception of the first and final sessions (the
Pre- and Post-Treatment Interviews), only the subject and experimenter
were present in the room. Subjects were seated on a comfortable
chair or sofa, depending on preference. Sessions were divided into
the following phases: (a) Pre-Treatment Interview, (b) Baseline,

(c) Treatment, and (d) Post-Treatment Interview. The design of
phases is represented in Table 3.1. All EMG values recorded during
these sessions represented one-minute averages and were taken from
the frontalis muscle. The two active electrodes were placed approxi-
mately one inch above the eyebrows and two inches from the upper mid-
point of the nasal bones (equidistant from the nasion). The ground
electrode was placed between the active electrodes. Note that at

the end of each session an additional 10 minutes of resting EMGs were

recorded. These values were included to determine the nature of

50

.:owmm:omwn $0 memo:

u:=a Low mg: mmgzmmme emcee .mcwpmmp mvmmcpoaxc :w umuzpocw mmocp m:m mmcsmmma mmmghm

 

Acowmmmm Fv

 

 

 

 

A.:we OFV mgim ccoowm A.:PE opv mm_mum _
NN mug: metamam m<zp was: newbmam mewpam-:_mm pcaamwwuwmwwoa
onmmmm uLoomm :mme:JEu< ncoumm :wpmwcwsu<
>H mm<zm
. A.:PE opv A.:PE omv mm—mom Amcowmmmm av
Pm w_ was“ mcwpmmm mcwcwmch mcwpmmuepmm “cospmmcp
onmmmm ULQUmm xomnvwwwowm empmwcw5u< QHHH mm<za
. A.:Fe on mwzm accomm A.:wE omv mmeom Amcowmmmm mv
N_ m was“ mcwummm mz:m mcwcwmch mcwpmmuwpmm “cmEpmweh
onmmmm Ugoomm prmwcwEU< xumncmmbowm memwcwEU< MHHH mm<zm
m
. A.:we OFV moZm neoomm A.:we omv mm_mom Amcowmmmm NV
m N was: m:_pmam azam WGZM mcwpmmm m::pem-cpmm a=w_amem
onmmmm uLOUmm :wpmwcw2u< ecoomm :mpmwcwsu< HH mm<zm
m
. . Acowmmmm Pv
_ A ::E o_v mag: oeouam A ewe opv mmpaam zmp>emecH
2 was: mewpmmm m<ze was: mewpmam m::pem-:_mm peaeJamee-mLa
onmmm neoomx :mpmwcwsu< neoomm :mamw:FEn< H mm<zm

 

 

 

 

 

 

mamas: vcm m—znmgom “cospmmge "_.m mpnwh

 

   

51

change in EMG values following a task. Results were not included in
hypotheses testing but were included in the Clinical Observations

of Chapter IV.

Pre— and Post-Treatment Interviews

 

During the Pre-Treatment Interview, subjects were asked to fill
out a Subject Consent Form (Appendix D) and a Subject Information
Form (Appendix E). Subjects were instructed that they could discon-
tinue participation at any time during the course of the study, and
they were also informed that the treatment administered in the study
did not guarantee any beneficial results. Prior to beginning the
Baseline sessions, the subjects' physicians were also asked to fill
out a Physician's Consent Form (Appendix F).

The format for the Pre- and Post-Interview Sessions continued
in the following manner. Subjects were first asked to respond to
the three Self-Rating Scales. Next, surface electrodes from the
biofeedback equipment were attached, and the subject was asked to
sit quietly for 10 minutes while EMGs were recorded. Following this,
the Taylor Manifest Anxiety Scale was administered by a third person
while EMG averages were recorded by the experimenter. At the end of
the session, the interviewer left the room, and 10 additional minutes
of resting EMG averages were recorded. The information gained from
these two pre- and post-sessions was for purposes of discussion, and

data were not included in statistical tests of significance.

 

52

Baseline

Baselines were established in the first seven sessions following
the Pre-Treatment Interview. Each session began with subjects' rat-
ings on the three Self-Rating Scales, at which point electrodes were
attached and subjects were asked to sit quietly for 20 minutes while
resting EMGs were recorded. Next, the five Repetitive Psychometric
Measures were administered while EMGs were recorded. The final part
of the session consisted of recording 10 additional minutes of resting

EMGS.

Treatment

The Treatment phase consisted of the next 13 sessions. During
the first nine sessions, the format was identical to the Baseline
sessions with the exception of the 20-minute resting EMGs. Instead,
following the administration of the Self-Rating Scales, biofeedback
training was given for 30 minutes while EMGs were recorded. In addi-
tion, the subject was given a tape with relaxation instructions fol-
lowing the format of Progressive Relaxation exercises. They were
instructed to use the tape once each day.

The tenth through thirteenth sessions measured only the Self—
Rating Scales and 30-minute training EMG levels. As nine data points
were sufficient for the data analysis, these last sessions were
included to provide more time to assess whether or not the maximum

level of muscle relaxation was attained for each subject.

53

Hypotheses

 

The following research hypotheses were developed to test for

changes in EMG levels, scores on the Repetitive Psychometric Measures,

and for scores on Self-Rating Scales of tension.

Differences in EMG Levels

I.

II.

Differences in Test Scores

Null Hypothesis:

Alternative
Hypothesis:

Null Hypothesis:

Alternative
Hypothesis:

 

of Cognitive Abilities

 

III.

IV.

Null Hypothesis:

Alternative
Hypothesis:

Null Hypothesis:

There will be no difference in the linear
trend for EMG levels between 20 minutes
of rest during Baseline and 30 minutes of
biofeedback training during Treatment.

There will be a reduction in the linear
trend for EMG levels between 20 minutes
of rest during Baseline and 30 minutes
of biofeedback training following the
point of treatment intervention.

There will be no difference in the linear
trend for EMG levels during cognitive
tests between Baseline and Treatment
phases.

There will be a reduction in the linear
trend for EMG levels during cognitive
tests following the point of treatment
intervention during the Treatment phase.

There will be no difference in the linear
trend for scores on Flexibility of Closure
between Baseline and Treatment phases.

There will be an increase in the linear
trend for scores on Flexibility of Closure
following the point of treatment interven-
tion during the Treatment phase.

There will be no difference in the linear
trend for scores on Number Facility
between Baseline and Treatment phases.

 

Alternative
Hypothesis:

54

There will be an increase in the linear
trend for scores on Number Facility fol-
lowing the treatment intervention during
the Treatment phase.

V. Null Hypothesis: There will be no difference in the linear
trend for scores on Perceptual Speed
between Baseline and Treatment phases.

Alternative There will be an increase in the linear

Hypothesis: trend for scores on Perceptual Speed fol-
lowing the point of treatment intervention
during the Treatment phase.

VI. Null Hypothesis: There will be no difference in the linear
trend for scores on Speed of Closure
between Baseline and Treatment phases.

Alternative There will be an increase in the linear
Hypothesis: trend for scores on Speed of Closure fol-
lowing the point of treatment intervention
during the Treatment phase.
VII. Null Hypothesis: There will be no difference in the linear

Alternative
Hypothesis:

trend for scores on Visualization between
Baseline and Treatment phases.

There will be an increase in the linear
trend for scores on Visualization follow-
ing the point of treatment intervention
during the Treatment phase.

 

Differences in Self-
Ratings of Tension

 

 

There will be no difference in the linear
trend for ratings of general level of ten-
sion on Self-Report Rating Scale 1 between
Baseline and Treatment phases.

VIII. Null Hypothesis:

There will be a decrease in the linear
trend for ratings of general level of ten-
sion on Self-Report Rating Scale 1 follow-
ing the point of treatment intervention
during the Treatment phase.

Alternative
Hypothesis:

 

w , . . M
:1 M¢;,_:, _

 

55

IX. Null Hypothesis: There will be no difference in the linear
trend for ratings of the ability to modify
levels of tension on Self-Report Rating
Scale 2 between Baseline and Treatment

phases.
Alternative There will be an increase in the linear
Hypothesis: trend for ratings of the ability to modify

levels of tension on Self-Report Rating
Scale 2 following the point of treatment
intervention during the Treatment phase.

X. Null Hypothesis: There will be no difference in the linear
trend for ratings of the ability to detect
changes in levels of tension on Self—Report
Rating Scale 3 between Baseline and Treat-
ment phases.

Alternative There will be an increase in the linear

Hypothesis: trend for ratings of the ability to detect
changes in levels of tension on Self-
Report Rating Scale 3 following the point
of treatment intervention during the Treat-
ment phase.

Analysis

The split—middle method of analysis was used in this study to
test for the effects of a combined treatment program of electromyo-
graphic biofeedback and modified Progressive Relaxation exercises.
This analysis was applied to an A-B time series design for single
subjects. The purpose of this analysis was to determine if data
during Phase B were significantly different from the data during
Phase A on measures of: (a) EMG levels, (b) scores on five tests
of cognitive abilities, and (c) scale values for three rating scales
of tension. One-tailed tests of significance were used with an
a = .05 level to determine if there was a significant difference in
the trend of the data curves from A to B. Graphic data curves for

dependent measures were presented for visual inspection of trends.

 

 

 

 

 

56

Split-Middle Method

 

The split-middle technique (White, 1974) is a method of describ-
ing behavior change over time which can be applied to single-subject
and group designs. The primary purpose of the technique is to char-
acterize linear trends in the data and to predict future performance
based on the trend estimation. In addition, a binomial test can be
applied to the projected A_slope in order to assess the statistical
significance of the changes in behavior which occurred during the
intervention, or 8 phase (Hersen & Barlow, 1976).

In applying the split-middle technique, a separate slope or
"celeration" line is estimated from the plotted data for each phase.
Median data points from phases A and B are used to determine these
lines, so that 50% of the data fall above the slope and 50% fall
below. The statistical significance of change across phases is
evaluated once these lines are determined. The null hypothesis upon
which the test is made is that there is no change in performance
across A and 8 phases. In other words, the celeration line of the
A phase should be a valid estimate of the celeration line of the
8 phase. The binomial test determines the probability of attaining
x number of data points on Phase B which fall above or below the pro-

jected A slope.

Data Graphs

 

Thoresen (1974) and White (1972) recommended transforming data
for the split-middle analysis onto a semilog scale. They reported

that this type of scale reduces variability and makes the trend more

 

57

suitable for linear expression. The slope which is generated from
the transformed data is comparable to a least squares regression line
and is less influenced by extreme data points.

The graphs presented in this study followed these recommenda-
tions. The y axes, which represent values of the dependent variables,
were expressed on a log to the base 10 scale. A true value of 10 is
therefore transformed to a value of l on the graphs. The §_axes,

representing days, were not transformed to log units.

Summary

Four subjects from the Lansing and East Lansing areas were
selected for the study. They were selected on the basis of a diag-
nosis of schizophrenia based on defined criteria and with their
physician's consent that they were stable enough to participate. In
addition, all subjects had experienced at least three major psychotic
episodes with hospitalization, and all were being maintained on anti-
psychotic medication. They varied in terms of marital status and
level of education obtained, and all were nonpaid volunteers.

Subjects participated in a combined treatment program of EMG
biofeedback training and Progressive Relaxation while EMG levels, the
Repetitive Psychometric Measures tests, and Self-Rating Scales of
tension were administered on a repeated basis over time. These
measures were representative of changes which might occur in the
physiological, cognitive, and behavioral domains. Data regarding
levels of anxiety were also obtained using the Taylor Manifest

Anxiety Scale.

 

58

An A-B single case time series design was used with the split-
middle method trend analysis. Binomial tests of significance were
applied to test the main hypotheses. Scores on EMG levels, five
psychometric measures, and three Self-Rating scales were the dependent
variables. The purpose of using the split-middle technique with
significance testing was to analyze intraindividual changes occurring
over repeated measures of the dependent variables from Baseline to
Treatment phases. Data from the Taylor Manifest Anxiety Scale were
included for purposes of discussion and were not included in hypothe-
sis testing.

The results of the hypotheses tests and an interpretation of

these results are presented in Chapter IV.

 

CHAPTER IV

ANALYSIS OF RESULTS

The statistical hypotheses and an analysis of the data are pre-
sented in Chapter IV. All hypotheses relate to comparisons made
between events in the seven baseline and first nine treatment sessions
of the research schedule. The hypotheses were tested using the split-
middle method of trend analysis with binomial tests of significance.
One-tailed tests of significance were determined for each dependent
variable for each subject at the a = .05 level. An analysis of the
data from clinical observations is also presented. Following is a
summary of the results.

Hypotheses I and II: Differences
in EMG Levels

 

 

I. Null Hypothesis: There will be no difference in the linear trend
for EMG levels between 20 minutes of rest dur-
ing Baseline and 30 minutes of biofeedback
training during Treatment.

Alternative There will be a reduction in the linear trend

Hypothesis: for EMG levels between 20 minutes of rest during
Baseline and 30 minutes of biofeedback training
during Treatment, following the point of treat-
ment intervention.

The plotted daily EMG levels and celeration lines for 20 minutes
of rest during Baseline sessions and 30 minutes of biofeedback train-
ing during Treatment sessions are presented for all subjects in
Figure 4.1. Median EMG levels obtained during the two phases and

59

 

 

6O

mv

mv

.mcwcwwgu xumnvmmwown .:PE om Op pmw: .:wE ON :0» m—w>mp ozm xpwmo H—.¢

 

 

 

 

 

“553: w>¢D 3:35
mm NN «w 0
v 0 V1 # u 0 $. V Gaol
.6vg.
Lawn.
.6Nw.
e uuahngm //////
iémw.
2958.; w>¢D 2.533
mm NN - o
n 4 n 4 4 u 4 “uo~.-

 

cumﬂo=m

N

 

lieu.

.rwm.

.éwm.

 

Loom.

lWOA-OHJIW JD 001

llDA-OHOIW AD 007

3953.; w>¢D
mv mm NN

mczmw:

—~

m:w_mmmm

 

. b p p
d 4 d

T

 

m scannsm

2258:. wrco
we mm NN

db

 

«a

1P

u uo—.I

.6+_.

60m.

bum.

 

Pémm.

m:__mmom

“ 00—.I

 

.1 b if P bl
1 1 1 ‘1 1

.1
\I
-.I|II
|

_ uuaﬂnam

 

iévg.

.émm.

 

gémo.

lWOA-OHUIH JD 001

lWOA-OHJIW 30 007

 

 

61

the percentage of change across phases are shown in Table 4.1. EMG
levels were decreased from Baseline to Treatment by 43% for Subject 1,
58% for Subject 2, 40% for Subject 3, and 73% for Subject 4. The
results of the statistical analysis were significant for all subjects
at the p_= .002 level. These are shown in Table 4.2. The null
hypothesis for no differences between Baseline and Treatment phases
for EMG levels was rejected in favor of the alternative hypothesis

in four out of four cases. Subjects decreased EMG levels during 30

minutes of biofeedback training compared to resting baseline measures.

Table 4.1: Percentage of Change for Median EMG Levels From 20 Minutes
Rest During Baseline to 30 Minutes of Biofeedback Training
During Treatment

 

 

 

Subject (SSeAiQI (3o hiRItEISEeed- Percent
Resting EMG) back Training) Of Change

1 2.33 Mv 1.32 Mv -43%

2 3.58 Mv 1.51 NV -58%

3 4.20 Mv 2.50 NV -40%

4 5.02 Mv 1.34 NV -73%

 

- indicates a decrease from Baseline to Treatment.

Mv = microvolts.

62

Table 4.2: Binomial Tests of Phase Changes for Research Hypotheses
Involving EMG Levels

 

 

. Baseline- Baseline-
Subject Treatment Treatment
(Hp (H2)
1 .002* .253
2 .002* 1.0
3 .002* 1.0
4 .002* .002*
Across all 4 of 4 1 of 4
subjects significant significant

 

 

*Significant at .05 level.

II. Null Hypothesis: There will be no difference in the linear
trend for EMG levels during cognitive tests
between Baseline and Treatment phases.

Alternative There will be a reduction in the linear trend
Hypothesis: for EMG levels during cognitive tests follow-
ing the point of treatment intervention during
the Treatment phase.
The difference in daily EMG levels during cognitive tests
between Baseline and Treatment phases is shown in the plotted data

in Figure 4.2. The percentage of change from Baseline to Treatment

for median EMG levels obtained during cognitive tests is shown in

Table 4.3. After treatment, EMG levels increased by 26% and 6% for

Subjects 1 and 3 during the administration of cognitive tests. A

decrease of 11% and 12% occurred for Subjects 2 and 4. Binomial test

results (Table 4.2) show that EMG changes from Baseline to Treatment

were not significant for Subjects 1, 2, and 3 and that the decrease

 

 

.mpmmp m>4pwcmoo mc4czu m_w>m— GEM >448:

w>mo 85.58.;
wv mm NN ——

P

“F
~0-
<1-
‘1-

w:_.wmom

h

o

 

(P
d

D
4 d

1

J

éoowﬂ.

 

 

 

63

W>GD 26.5mm: 6:133

mv

.boon.

Ibooov.

 

g.ooovm.

 

P
1

 

N Suannzm

 

.éoowa.

.Boowm.

.6ooov.

 

.6oovm.

lJOA-OUOIW 30 001

IWOA-DUDIW 50 001

mv

mm

“N.¢ mczmw:

w»¢o
Nu

P b

ucwEumwcp

—~

m:__wmom
o

 

 

mv

1!

qt-

 

L
11

b
d

boomu.

Soowm.

6oo94.

 

:

 

 

m,uomnn:m éoovm.

w>¢D 45.58.; 6:138

mm NN 4_ o
4 4 4 4 4. 4 4 4. o

iaoomg.

do
L.ooowm.
.boowv.

P Humﬂaam

Lfoovm.

 

 

lWOA-OUUIW 30 001

lWOA-OHDIW JD 001

64

in EMG levels was significant for Subject 4 at the p_= .002 level.
The null hypothesis was rejected in favor of the alternative in one
out of four cases. Only Subject 4 decreased EMG levels during cog-

nitive test task performance following treatment intervention.

Table 4.3: Percentage of Change for Median EMG Levels From Baseline
to Treatment During Cognitive Tests

 

 

. . Percent of
Subject Baseline Treatment Change

1 1.49 Mv 1.87 Mv +26%

2 1.70 Mv 1.52 Mv -ll%

3 2.91 Mv 3.08 Mv + 6%

4 2.94 Mv 2.60 Mv -12%

 

My = microvolts.
+ indicates an increase from Baseline to Treatment.

- indicates a decrease from Baseline to Treatment.

Hypotheses III, IV, V, VI, and VII:
Differences in Scores on Repetitive
Psychometric Measures

 

 

 

III. Null Hypothesis: There will be no difference in the linear
trend for scores on Flexibility of Closure
between Baseline and Treatment phases.

Alternative There will be an increase in the linear trend

Hypothesis: for scores on Flexibility of Closure following
the point of treatment intervention during the
Treatment phase.

Plotted scores and celeration lines for the Repetitive Psycho-

metric Measures test of Flexibility of Closure are shown in Figure 4.3.

 

 

 

 

.mcsmoro 4o >44_wnwxm—4 :o mwcoom xrwmo 4m.¢ mczm44

 

 

 

 

 

 

 

 

 

 

w>¢o 6:133 m>¢D
m. mm «N 44 o mm NN 44 o
.r 4 4 4 1» .r 4 04.: 4 4 .r 4 O—.I
.Sww. .DmN.
.1
0
1.600. Flu .OOD.
nu
1:
e uponnam no
«J
. 0 . .
45mm do 5mm
3
4 4884234
450m.4 48m;
rug
rho
W>¢D wcw—wmem m>¢o
m4 mm mm 44 mm NN 4. o
411 4 4 4 4 o4.- 4 4 4 4 o4.u
amw. 483.
N sawhnzm MW
bow. no .oom.
V 0
.4 . ..J
. 4%. no
-: - - emm. nu 8mm.
mm 4 Sumﬂnam
bom.4 - éom.4

 

 

 

 

 

38038 90 001

38038 90 001

 

.

.41..
«444.4. ..

 

 

66

The median scores obtained during Baseline and Treatment phases are
given in Table 4.4. There were increases across phases in median
scores for the number of items correctly completed for all subjects.
The increase was 4 points for Subject 1, 3 points for Subject 2,

.5 points for Subject 3, and 1.9 points for Subject 4. Results from
the binomial tests shown in Table 4.5 indicate that overall increases
were not statistically significant at the (x = .05 level. The null
hypothesis for no differences in linear trend between Baseline and
Treatment phases was therefore not rejected in favor of the alterna-
tive in any of the four cases. There were no significant improve-
ments in subjects' ability to correctly copy geometric figures onto

matrices of dots following treatment intervention.

IV. Null Hypothesis: There will be no difference in the linear
trend for scores on Number Facility between
Baseline and Treatment phases.

Alternative There will be an increase in the linear trend

Hypothesis: for scores on Number Facility following the
treatment intervention during the Treatment
phase.

The daily plotted scores for the Repetitive Psychometric Measures
test of Number Facility for subjects are shown in Figure 4.4. Results
from Table 4.4 indicate that Subjects 1, 3, and 4 increased their
median scores for number of items correctly completed from Baseline
to Treatment by 1, 9, and 6 points, respectively. There was a decrease
of 1 point for Subject 2. The statistical analysis (Table 4.5) showed
that the overall increase in scores was significant for Subject 1 at
the p_= .019 level and for Subject 4 at the p_= .002 level. The null

hypothesis for no differences was rejected in two out of four cases,

 

 

 

67

mv

mm

w>¢o

NN

 

.xpw—womd 483532 :0 mmcoom >Fwwo

acceucmch
44

8:448mom

 

db

4:-

fr

 

8 3883333

w>¢o

NN

1;-

up

P b
u d

 

 

“caswcoch
4"

o:__mmm:

Do.—

:ONN.~

Juvv.—

.omm.u

 

imumo.~

 

N 8883333

..

P b -
4 d 4

 

oo._

éNN.~

6vv._

dam.“

 

émm.~

38038 30 001

38038 30 001

43.3 833044

m>¢o
m. an NN

acmepmmcp

~—

mcw—omcm

 

by
d

 

m uommozm \\

w>¢o
8» mm NN

b

[‘1

 

acmEummcp

«a

h
1

d>

m:4_mmom

co.—

.6NN.—

.Evv.~

Liumm._

 

8888.4

 

-
..
‘
db

1

 

4 8884334

<F

 

P
1 [‘1

00.4

.6NN.~

&O+v.4

30mm.—

 

3388.4

38038 30 001

38038 30 001

 

68

.zppsuw4mwu mo

Fo>mp :o4 :wumswum .444804408 umgoom mE¢44 4o Longs: :wwume ms» pcmmmgamg mmcoom

 

 

 

 

 

 

 

m.Nm m.mm m.mm —.Fm m.Nm _.pop o.©m 0.0m m.m~ o.¢_ v
o.©N m.mp m.nm m.FN N.ww m.©m o.wN o.m— w.N m.N m
o.¢N o.mN m.Pm o.mN o.mw m.mm 0.0m o.nm m.m m.© N
m.mN F.NN N.om o.mm N.mm m.¢© o.mN O.VN m.m— m.m —
h m H m H m H m H m
441v Aozv mcamopo Amzv :mmam 43:4 44444884 Amzv mczmopu nomwnzm
:0488N4483m4> 4o ummam Fmspamo:m: :mnszz mo 444—4nwx844

 

mommza acmEHNmLH

use mcwpwmmm m:4::a mpmmh m¢g=mm¢2 owepmsocoxm: m>wpwpmamm :04 mucoom :mwuwz 43.3 84584

69

.48384 33. 48 838844433484

 

 

pcmowwwcmwm ucmow4wcm4m pcmow4wcm4m pcmow4wcm4m pcmow4wcm4m mpomhnzm
3 4o 4 3 40 o 3 4o 4 3 40 N 3 mo o 448 mmoco<
000.4 0.4 coo.4 emoo. ooo.4 w
ooo.— 0.4 mom. mmm. mmm. m
mmm. 0.4 000.4 034. omm. N
amoo. 0.4 amoo. «mpo. OFm. F
4438 4838 4838 4438 4438
“cospmmge acmEpmmcp “cospmmch “cospmmch acmEpmmLH pomwazm
18:448m8m -m:_memm -m:4_wmmm -w:4_mmmm um:4_mmmm

 

m>wpwpmamm m:4>_o>:4

mommgpoazz :ocmwmmm

mmezmwmz oweumEogozm:
:o4 mmmcm;o 888:: 4c mummh _m_so:4m 4m.3 84:84

 

 

70

indicating that two subjects improved in their ability to correctly

add sets of numbers following treatment intervention.

V. Null Hypothesis: There will be no difference in the linear trend
for scores on Perceptual Speed between Baseline
and Treatment phases.

Alternative There will be an increase in the linear trend

Hypothesis: for scores on Perceptual Speed following the
point of treatment intervention during the
Treatment phase.

Figure 4.5 shows the daily plotted scores for the Repetitive
Psychometric Measures test of Perceptual Speed from Baseline to Treat-
ment phases. Changes in the median scores obtained during both phases
shown in Table 4.4 indicate that during Treatment, Subject 1 increased
median scores by 10.3 points, Subject 2 showed an increase of 9.1
points, and Subject 3 increased scores by 11.3 points. There was a
decrease during Treatment in median scores for Subject 4 of 8.2 points.
Results from the binomial tests (Table 4.2) show that the overall score
increase was statistically significant for Subject 1 at the p_= .002
level. The improvements noted in scores for Subjects 2 and 3 were not
significant. Therefore, the null hypothesis for no difference in
linear trend from Baseline to Treatment was rejected in favor of the
alternative in one out of four cases. These results indicate that one

subject improved in the ability to correctly identify given digits

from rows of random digits following treatment intervention.

VI. Null Hypothesis: There will be no difference in the linear trend
for scores on Speed of Closure between Baseline
and Treatment phases.

 

 

.nmmam 4834388483 :0 484084 >448o 4m.3 843m44

 

 

 

 

 

71

 

 

 

 

 

 

 

 

 

 

 

 

W>¢D 4:48.558...— oE4-8448m w>¢D 44448458.:- 44:44:44.8:
r . 44 NN 44 o 44 44 NN 44 o
. . 4 4 4 4 14 -+ co; Ti T1- 4 4 4 f 4 #1 o94
.8N4.4 .8N4.4
_1
D
.94; 0 .434;
0
.3
S
a I:
84.4 W 544.4
33 -------
4 8844334 -- ------
¢4N.N .---.- émNN
m>¢0 448458.; 0:284:44 w>¢D 4:95:84 8:44:43
44 NN 44 o 44 44 NN 44 o
4 4 4 4 4 4 4 4 894 4 4 4 4 r 4 4 4 co;
.er.4 .éNm.4
1 -,
0
:0va 0 6'9.—
0
I3
8
(IV! \ 3
/ea - 544.4 808 84.4
1-1---- 3
- ..-.-- 4 8884334
N 4944.334 iémNN ..omw.~

 

38038 30 001

38038 30 001

 

72

Alternative There will be an increase in the linear trend
Hypothesis: for scores on Speed of Closure following the
point of treatment intervention during the
Treatment phase.
The daily plotted scores for the Repetitive Psychometric Measures
test for Speed of Closure for all subjects are presented in Figure 4.6.
The median scores obtained during Baseline and Treatment phases are
given in Table 4.4. There was an increase in number of items scored
correctly across phases for Subject 2 of 3.3 points, for Subject 3
of 5.4 points, and for Subject 4 of 2.8 points. The median score
decreased by .8 points for Subject 1. Binomial tests (Table 4.5) indi-
cated that none of the overall increases in scores were significant at
the a = .05 level. Therefore, the null hypothesis for no differences
in the linear trend was not rejected in favor of the alternative in
any of the four cases. There were no improvements in subjects' ability

to correctly identify four-letter words in rows of random letters fol-

lowing the treatment intervention.

VII. Null Hypothesis: There will be no difference in the linear trend
for scores on Visualization between Baseline
and Treatment phases.

Alternative There will be an increase in the linear trend
Hypothesis: for scores on Visualization following the
point of treatment intervention during the
Treatment phase.
Daily plotted scores on the Repetitive Psychometric Measures test

for Visualization are shown in Figure 4.7. Results from Table 4.4

indicate increases in the median scores for number of items scored

correctly from Baseline to Treatment for Subjects 1, 3, and 4 of

2.8, 12.2, and 2 points, respectively. There was a decrease of 1 point

for Subject 2. Probabilities from the binomial tests (Table 4.5) show

 

 

 

.843404u 4o nomam :0 484084 4448o No.3 mczm44

 

 

73

 

 

 

 

 

 

 

 

 

 

 

w>¢D 44404546.; 043.44.844.44 w>¢D 448.54.80.44 48:44:44.8:
44 NN 44 o 44 44 NN 44 o
4 4 t- 4 4 4 4 4 094 T. 4 4 4 4 f 4 4 394
..ONN.4 ..oNN.4
41 lb 7
3 m 3 3
.44; 4/k M¥11111 .44;
38 ,//3N ----
Q/ I: \x-\
8 \-
3 ---
-- 844.4 808 \- $8.4
xx. 3 4 4848.334 \\
\\ \\\
3 4884334 ..x- --.--
\.\ irwm.4 P500.—
W>¢D 4:953: 3:44.843 w>¢D 4:44:58: 94.47.4424
44 NN 44 o 44 44 NN 44 o
4 4 4 411 414 4 4 4 888.4 4 4 4 .4 4 4 4 4 no.4
JSNN.4 .8NN.4
3.
0
.844.4 n8 844.4
0
J.—
S
3
x 484.4 w 84.4
\\ 3
\\
N 488.4.an \ s. 4 888.4434 - x - -
\ L58.4 -.- 8844.4

 

38038 30 001

38038 30 001

 

711

wv

mv

<4

mm

q»

w>¢o 45458:
«N 44

b h
u 1

44

.1
#

m:4_mmea

P
1

.cowpmN444344>

0
cm.

 

db

4 4444434

mm

p

 

w>¢D 4553.;
um 44

4:44w4mm

.bwm.

lév~.~

:owv.~

 

kowh.4

o

 

 

N 4444434

b!
d

 

r
d

P
1

Dow.

Jév4.4

Léw¢.~

 

:owh.—

co 4m4044 >44mo 44.4 mgzm44

38038 30 007

38038 30 007

w>¢D 4:958:

mv mm NN 44 o
4 L3 4 4 cm.

w:__w4c:

-P
4
-r
4

 

6N0.

4Ov~.4

40mv.w

 

 

 

 

4 4444434 4 4ooh.4
\
\
x
W >¢D ”4:04.569; 0:? 7.44QO
wv mm NN 44 o
41 4 4 4 411 4 14 14 cm.
/,, léuo.
l/l
III
,4 :044.4
.6m¢.—
\
puuwmnzm Lamb.—

 

 

38039 JD 007

38039 30 001

75

that the overall increase in scores was statistically significant

for Subject l at the p_= .002 level. The increases for Subjects 3

and 4 were not significant. Therefore, the null hypothesis for no
difference in linear trend was rejected in favor of the alternative in
one out of four cases. One subject improved in the ability to cor-
rectly visually track the path of a line embedded in a set of tangled
lines following treatment intervention.

Hypotheses VIII, IX, and X: Differences in
Scores on Self-Rating Scales of Tension

 

VIII. Null Hypothesis: There will be no difference in the linear
trend for ratings of general level of tension
on Self—Report Rating Scale 1 between Baseline
and Treatment phases.

Alternative There will be a decrease in the linear trend

Hypothesis: for ratings of general level of tension on
Self-Report Rating Scale l following the point
of treatment intervention during the Treatment
phase.

The plotted daily scores and celeration lines for Self-Report
Rating Scale l for Baseline and Treatment phases are presented in
Figure 4.8. The results from the binomial tests, summarized in
Table 4.6, show that ratings on Scale l decreased significantly for
Subjects 2 and 3. These changes were significant at the p_= .002
level. Even though the linear trendS'ﬁn"Subjects l and 4 showed a
decelerating pattern, overall ratings during Treatment did not decrease
below the extended linear trend of Baseline. Therefore, the null
hypothesis for no differences in the linear trend for ratings of

tension was rejected in favor of the alternative in two of four cases.

The results indicate that two subjects rated themselves lower on

 

 

76

w>mo

.44 @4444 mc4pmm14444 co 444044 444mg

 

 

 

4:934:44 4:: 7443.
mm mm «4 D
4 4 4 J. 4 4 4 c
..GNN.
.va.
..cwm.
zoom.
4 4444434
w>¢D 4:44.43: .44: 4443.

D

  

.

 

N 4444434

    

-

 

50¢v.

.omm.

émw.

 

38038 30 001

38038 $0 001

”w.q 044m44

 

 

 

W>¢D 4:44.55: E: 744::
mV mm NN 44 o
..DNN
:qu
4
ﬂoww.
\\ Iéom.
4444334 \.\
W>ED 4:958: 44:34am
9. mm NN : o
. , . . 4 . . . L

 

4 4 4 . 4 .

/

//////
////me///UWA

//
I

44,, e

4 4444434

 

 

. C

38038 30 001

30 001

 

 

 

77

subjective levels of tension following the point of treatment

intervention.

Table 4.6: Binomial Tests of Phase Changes for Research Hypotheses
Involving Self-Report Rating Scales

 

 

Baseline- Baseline- Baseline-
Subject Treatment Treatment Treatment
(H8) (H9) (H10)
1 l.000 .019* 1.0
2 .002* .002* 1.0
3 .002* .002* l.0
4 1.000 .002* 1.0
Across all 2 of 4 4 of 4 0 of 4
subjects significant significant significant

 

 

*Significant at .05 level.

IX. Null Hypothesis: There will be no difference in the linear trend
for ratings of the ability to modify levels of
tension on Self-Report Rating Scale 2 between
Baseline and Treatment phases.

Alternative There will be an increase in the linear trend
Hypothesis: for ratings of the ability to modify levels of
tension on Self-Report Rating Scale 2 following
the point of treatment intervention during the
Treatment phase.
The plotted daily scores on Self-Report Rating Scale 2, indicat-
ing the ability to modify levels of tension, are shown in Figure 4.9.
Statistical tests shown in Table 4.6 indicate that ratings on Scale 2
significantly increased for all subjects from Baseline to Treatment

phases. Therefore, the null hypothesis for no differences was

 

 

 

.mx m4wom mcwpmmlw—wm co mmgocm >44mo no.4 mcsmwd

 

 

 

   

 

 

 

 

 

 

 

 

   

 

 
 

w>¢D 4:253: 9.53::
mm NN : . o w>¢D 4:958: 9.233.
.r 4 4 4 4 . . 0/ D J. c m. mm NN : o
. ,4
..ONN.
..ONN.
e > \I/ \II) J
4i
. 0 .I
_ . $3. 0 0
. . Pd
0 évv
I1 0
..J
S
_ 3 S
D 3
..wa.
\ U0 . 0
3 bomb 8
. 3
6
yum. m uumnosm
a ”43.33 .me.
w>¢D “5:53: 3:35 w>¢D 4:253: ocﬁmmmm
mm NN H— 0 mt mm NN 4.4. O
E 4 w
4. _. ,
4 4 . . /
4 ,_ _ éNN. / .aww.
. . . _ /
_ _ a _ /
44 n i .4 ,/
4| / 4|
4 4 4 o ,, o
_, $3. 0 ,,. :03. O
a O , 0
.4 4 I: .d
_ S
_, w e w 3
, . 0 D
_. \ $8 8 H
,. .4 . 3 3
t ,\,
ﬂ \
4
4 .Omm.
N 3933 _,
. _ uuwmngm

 

 

 

79

rejected in favor of the alternative in four out of four cases. All
subjects consistently rated themselves higher in their ability to
modify their tension levels following the point of treatment inter-

vention.

X. Null Hypothesis: There will be no difference in the linear trend

for_ratings of the ability to detect changes in
levels of tension on Self-Report Rating Scale 3
between Baseline and Treatment phases.

Alternative There will be an increase in the linear trend

Hypothesis: for ratings of the ability to detect changes
in levels of tension on Self-Report Rating
Scale 3 following the point of treatment inter-
vention during the Treatment phase.

Plotted scores on Self-Report Rating Scale 3 for Baseline and
Treatment phases are shown in Figure 4.l0. Even though celeration
lines during Treatment showed an increasing trend for Subjects 1 and 2,
results from the binomial tests (Table 4.6) indicated that changes in
ratings from Baseline to Treatment were not statistically significant
for any of the four subjects. Therefore, the null hypothesis for no
differences was not rejected in favor of the alternative. None of
the subjects increased their ratings, indicating that they did not
improve in their ability to detect changes in levels of tension follow-

ing treatment intervention.

Clinical Observations

 

End of Session Resting EMGs

 

During Baseline and Treatment phases, at the end of every session,
l0 minutes of resting EMG levels were recorded. It was expected that

EMGs would decrease from Baseline to Treatment phases as subjects

 

m»¢o
mm mm

p
n-

b
1

db

 

m:__mmca

 

 
 

 

80

4 somhnsm

w>¢o

mm NN

P
4 4

q»
#
b

 

1

 

m:__omcm

h
4

600mg.

éoowm.

éoooc.

 

accem.

 

N spawnsm

5‘ __

--—‘4‘__;_'____

 

 

boow~.

éoowm.

boomv.

éoovm.

 

38033 30 001

38038 30 001

.M4 m4mum 0:44am-44mm co mmcoom xpwmo ”o—.w mczmwl

m>¢o

 

 

 

 

m:_4mmam
me mm NN
llllgvmxw
m uuwnnam
w>¢D chwmmm
mv mm mu

[-7

o

éoom—.

boowm.

boocc.

 

boovw.

o

 

4 gummnzm

4P

-
d

 

d

 

léooou.

boowm.

éooov.

Soo4m.

 

38039 JD 001

38038 30 001

8l

learned to lower EMGs during biofeedback training. Figure 4.ll shows
the daily EMG levels and celeration lines for the l0-minute resting
EMGs across Baseline and Treatment phases. The split-middle method
of trend analysis with binomial tests of significance (a = .05)
indicated that there were statistically significant decreases in

EMG levels for Subject 3 with p_= .019 and Subject 4 with p_= .002.
The probabilities are shown in Table 4.7. The Treatment phase linear
trend shows slight increases in EMG levels for Subjects 1 and 2.
Therefore, in two out of four cases, subjects were able to maintain
significantly lower EMGs for l0 minutes at the end of the training

sessions which occurred during the Treatment intervention phase.

Table 4.7: Binomial Tests of Phase Changes for End of Session
lO-Minute Resting EMGs

 

 

Subject Baseline-Treatment
l .9l0
2 .998
3 .0l9*
4 .002*
Across all 2 of 4
subjects significant

 

*Significant at .05 level.

Final Four Biofeedback Training Sessions
A comparison was made between EMG levels in the first nine and

final four biofeedback training sessions. The median EMG value for

 

 

 

 

€32

mv

mm

D
d.

 

.cowmmmm 4o ccm pm .CwE o— 404 m4m>w4 u2m x44mo

Sam
3. .

“coagcocp

"a

ocwpwmmm

o

 

cu-

 

4

Cb?

 

h!
‘

h
d

6NN.

6".

6mm.

 

 

 

 

4 sawhosm ..t... econ.
w>¢D “c258: 9.232.

4 4 nm« ll» -4 l4. 444 w. o
1rCNN.
tOtv.
46mm.

N uuwhnzm L

 

 

dom.

l'lOA-OHOIH .40 001

iWOA-UBOIN 30 00']

mm

u._.4 aa=m_a

w>¢o
3

b!

ucwEumocp

4“

b

m:w_ommm

h

o

 

 

wv

T
L

m gooﬁngm

mm

4

m>¢o

NN

P
d

 

ucmsucmcp

b
J

n

4

p

d

d

oc4_omcm

P
J

Léww.

6".

boo.

 

4690.

o

 

'F'

«
4h
#

 

4 uumnnzm

P
d

 

ASNN.

{Ovv.

.éww.

 

g.ooo.

110A-0831N :10 001

l'lOA-OUJIN :10 001

 

83

each of these phases was used in order to determine the differences.
The maximum reduction from the first nine to the last four sessions
for any subject was 0.2 Mv, or two-tenths of a microvolt. Since this
represented such a small change, it was concluded that there were no
differences in EMG levels between these two phases.
EMG Changes During the Pre- and
Post-Treatment Interviews

EMGs were recorded during three events (rest, TMAS administra-
tion, rest) for the Pre-and Post-Treatment Interviews. In examining
the fluctuation in the median values across these events, it was
observed that these fluctuations were variable for subjects and
seemed to follow no predictable pattern. The averages are shown in
Table 4.8. For example, it was expected that during the Pre-Treatment
Interview, there would be an increase in EMG levels from the first
event (rest) to the administration of the TMAS. This only occurred
for Subject 2. During the Post-Treatment Interview which was given
at the end of Treatment, it was expected that EMGs would remain low
from the first event of rest to the test administration. Instead,
there were increases for Subjects l, 2, and 4. The change for
Subject 4 was small enough, however (from 4.2 Mv to 4.9 Mv), that it

was considered insignificant.

Taylor Manifest Anxiety Scale

 

Comparisons were made for scores on the Taylor Manifest Anxiety

Scale between the Pre-and Post-Treatment Interviews. The test score

indicated the number of items marked positive for anxiety out of a

 

84

 

.mp_o>o2ows n >2

 

 

>2 o>.v >2 om.¢ >2 om.2 2
. . . Acowmmmm 44
>2 ON m >2 om N >2 ON m m 2a>>2mp22
. . . “cospmmch
>2 ON 2 >2 oo 0 >2 on p N lumen
>2 mm.N >2 Fo.m >2 mN.N _
>2 o2.¢ >2 N¢.m >2 om.¢ 2
. . . . 4:o_mmmm 44
>2 CV m >2 ON m >2 On Q m 3ww>mecH
>2 mm.4 >2 om.m >2 m>.¢ N peaspmwu>
. 4
>2 Fo.m >2 mo.m >2 mo.m _
uzm m<2h 40 G24
m=>>ma2 .242 OF eo>>aapm>:_22< newpmam .:_2 o> poa42>m

 

m2m>>2mp24 pcprmwcpnpmoa 22m -mcm mcwcso mpumnnzm 204 mm24w> 92m cmwum2 ”w.¢ mPQNH

 

85

total of 50. These scores, as well as the percentage of change from
the Pne to Post sessions are shown in Table 4.9. Results show that
Subjects 1, 2, and 3 decreased the number of items scored positive

for anxiety from the Pre to Post sessions by 26%, 50%, and 38%,
respectively. There was an increase for Subject 4 of 6%. Thus, in
three out of four cases, scores for anxiety improved after biofeedback

training.

Table 4.9: Percentage of Change for Subjects' TMAS Test Scores

 

a

 

 

, Test Scores Percent
Subject of Change
Pre-Treatment Post-Treatment
1 3l l8 -26%
2 38 l0 -56%
3 4l 22 -38%
4 l3 16 + 6%

 

aTest score indicates the number of items scored positive for
anxiety out of a total of 50.

Progressive Relaxation

 

Subjects were asked to use cassette tapes of modified Progressive
Relaxation exercises in daily practice at home. They were asked about
their experiences with the exercises and the consistency of practice.
Their reports indicated that Subject 1 practiced approximately 80%
of the time; Subject 2, 78% of the time; Subject 3, 58% of the time;
and Subject 4, 75% of the time.

 

86
Subjects' Experiences During
Biofeedback Training

During the biofeedback training sessions, subjects reported on
techniques which they found helpful in terms of producing relaxing
feelings and decreasing EMGs. Redirecting and focusing of thoughts
to nonstressful ideation seemed to be critical for all subjects. For
example, one subject was able to relax when thinking of a very small
circle or a straight line, while another thought of playing tennis.

A third subject thought of being "dark, empty, and soft inside." No
matter what images helped subjects to relax, they all found it diffi-
cult to maintain low EMG levels when certain perseverative thoughts
came into focus. For one subject these were self-punishing state—
ments; for another they were thoughts of failure. Training sessions
were frequently interrupted when subjects felt they could no longer
maintain relaxation because of threatening ideation. By the end of
training, subjects were able to maintain relaxation without interrup-
tion for at least l5 consecutive minutes.

All subjects reported at various times throughout the training
sessions that they had experienced being able to alter feelings of
physical tension at different times in the day using techniques
learned in treatment. At these times, they were able to identify
situations outside of treatment which seemed to produce tense feelings.

Subject 2 reported being able to sleep without her usual seda-
tive after the eighth biofeedback training session. She continued to
sleep without medication periodically and stated that this was the

first time in six years that she was able to do so. This was verified

87

with her physician. Subjects l and 4 also had antipsychotic medica-
tions altered or discontinued during treatment after at least six

months of continuous usage.

Summary

Ten hypotheses were tested to examine changes in repeated meas-
ures of EMG values, changes in scores on cognitive tests, and changes
in self-ratings of various aspects of tension. The split-middle method
of trend analysis with one—tailed tests of significance at the a = .05
level was applied to an A-B, single-subject time series research design
to test for Baseline-Treatment phase differences. The treatment under
consideration was a combined program of electromyographic biofeedback
with home practice of modified Progressive Relaxation exercises.

The following is a summary of the results for each hypothesis
test:

l. Hypothesis I predicted a decrease in EMG levels following
the point of treatment intervention with comparisons made between
20 minutes of rest during Baseline and 30 minutes of biofeedback
training during Treatment. The null hypothesis was rejected in four
of the four cases.

2. Hypothesis II predicted a decrease in EMG levels during the
administration of cognitive tests following the point of treatment
intervention. Comparisons were made for subjects between the admin-
istration of tests during Baseline and test administration during
Treatment. The null hypothesis was rejected for one of the four

subjects.

 

 

 

88

3. Hypotheses III through VII predicted that scores on each
of five Repetitive Psychometric Measures tests would increase fol-
lowing the point of treatment intervention. Comparisons were made
for subjects between scores obtained during Baseline and Treatment
phases.

Hypothesis III: The null hypothesis was not rejected for any
of the four subjects for scores on Flexibility of Closure.

 

Hypothesis IV: The null hypothesis was rejected in favor of
the alternative hypothesis in two of four cases for scores on
Number Facility.

 

Hypothesis V: The null hypothesis was rejected in favor of
the alternative hypothesis in one of four cases for scores on
Perceptual Speed.

 

Hypothesis VI: The null hypothesis was not rejected in favor of
the alternative hypothesis in any of four cases for scores on
Speed of Closure.

 

Hypothesis VII: The null hypothesis was rejected in favor of
the alternative hypothesis for one of four subjects for scores
on Visualization.

 

4. Hypothesis VIII predicted that scores on Self-Report Rating
Scale 1, indicating general level of tension, would decrease following
the point of treatment intervention. Comparisons were made for sub-
jects between scores obtained during Baseline and Treatment phases.
The null hypothesis was rejected in four out of four cases.

5. Hypothesis IX predicted that scores on Self-Report Rating
Scale 2, indicating the ability to modify levels of tension, would
increase following treatment intervention. Comparisons were made for
subjects between scores obtained during Baseline and Treatment sessions.
The null hypothesis was rejected in favor of the alternative hypothesis

in four of four cases.

 

89

6. Hypothesis X predicted that scores on Self-Report Rating
Scale 3, indicating the ability to detect changes in levels of ten-
sion, would increase following the point of treatment intervention.
Comparisons were made for subjects between scores obtained during
Baseline and Treatment sessions. The null hypothesis was not rejected
in favor of the alternative hypothesis in any of the four cases.

In Chapter V a summary of the study is presented. The findings
of the study are discussed and conclusions presented. Limitations

of the study and implications for future research are also discussed.

 

 

 

CHAPTER V

SUMMARY AND CONCLUSIONS

The purpose of the study was to evaluate the effectiveness
of a combined treatment program of electromyographic biofeedback and
modified Progressive Relaxation exercises upon specific behaviors of
persons diagnosed as schizophrenic. The treatment intervention was
assessed using EMG measures, scores on a series of cognitive tests,
subject self-rated scales of tension, and scores on a test of anxiety.
In this chapter a summary of the study will be presented. Conclusions
based on the results of the analysis of the data will be included
along with a discussion of the results of the study. Limitations of

the present study and implications for future research will be given.

Summar

The study was based on several areas of theory and research.
The first of these was the study of abnormal skeletal—muscle response
tendencies in schizophrenics. Related to this were theoretical for-
mulations concerning the psychophysiology of schizophrenia, which
attempted to integrate abnormal physiological arousal with the pro-
duction and maintenance of symptoms of schizophrenia. A third area
involved the use of electromyographic biofeedback and Progressive

Relaxation as a clinical intervention technique.

90

 

 

 

 

 

 

 

91

A common theme which was stressed in all three areas of research
was that physiological responses represented in the skeletal-muscle
system were reflective of responses occurring in other systems involv-
ing cognitive and emotional behaviors.

A review of the literature was presented in which this theme
was evident. For example, in previous research it was revealed that
schizophrenics were abnormally reactive on skeletal-muscle measures
during the experimental events of rest, the presentation of stressors,
and during task performance. Authors concluded that a clear corres-
pondence existed between degree of physiologic disturbance and the
variables of anxiety and psychological maladjustment. In addition,
it was observed that skeletal-muscle tension increased for schizo-
phrenics prior to psychotic episodes and that when clinical condition
improved, arousal decreased but continued to be abnormally variable
from day to day. These findings were related to psychological vari-
ables involving cognition and emotions. The theories which were
reviewed summarized that chronic states of physiological disequilib-
rium, evident in the skeletal-muscle system, impacted on the schizo-
phrenic's ability to process and filter information. It was hypothe-
sized that the results were seen as the inability to focus thoughts,
increased attention to irrelevant stimuli, and cognitive disorgani-
zation.

Finally, literature reviewed in the areas of electromyographic
biofeedback and Progressive Relaxation revealed that skeletal-muscle
arousal was maintained, in part, through subjective experiences rep-

resented in cognition. These combined techniques were shown to

 

92

effect positive changes in emotional symptoms related to stress and
anxiety in highly anxious individuals by teaching them to modify the
stress-related symptoms of excessive muscle tension. One study indi-
cated that other combinations of relaxation therapy, specifically
Quieting Response Training, resulted in positive outcomes for psychi-
atric inpatients. The literature reviewed provided the rationale for
the present study, which proposed that relaxation therapy techniques
be explored with persons who were diagnosed as schizophrenic. It

was felt that this specific population could benefit from learning to
modify excessive skeletal-muscle tension and, in turn, might experi-
ence some of the same improvements in areas of emotional and cognitive
functioning noted for other patient populations.

One male and three females were accepted for the study based on
screening criteria which included clinical signs of schizophrenia
without paranoid ideation; functional stability; and the experience of
physical tension. The experimenter required that individuals have
the consent of their psychiatrists in order to participate in the
study. Each subject had a history of at least three psychiatric
hospitalizations and was being maintained on an antipsychotic medi-
cation. They varied in terms of educational and marital experiences
and in terms of their daily routines.

Prior to acceptance into the study, potential subjects were
interviewed. During the interview, the biofeedback equipment was
demonstrated, and the purpose and conditions of the study were intro-
duced. Four out of five individuals interviewed were selected to

participate.

93

The following session, the Pre-Treatment Interview, was the
first phase of the experiment. Subjects were asked to respond to a
Subject Information Form which elicited summary information concern-
ing social, educational, psychiatric, and medical history. Additional
information obtained from this session was repeated on the last ses-
sion of the study following biofeedback training and was referred to
as the Post-Treatment Interview. Data obtained in these two inter-
views included, in the order presented, scores on three Self-Report
Rating Scales of tension, l0 minutes of resting EMG, scores on the
Taylor Manifest Anxiety Scale, and a final lO-minute resting EMG.
Items from the TMAS were read to the subjects by a third party while
EMGs were recorded by the investigator. The rationale for using an
interview format for this test was that subjects might experience
some tension, reflected in EMG measures, in having to respond to ques-
tions about themselves to a stranger.

Findings from these two sessions were included in a discussion
of clinical observations and were not part of the hypotheses tests.
The results were not statistically analyzed because the TMAS was not
constructed for use as a continuous measure, which was one of the
requirements for the research design. Observations were made in the
fluctuations of EMG levels across the events in each session. Scores
on the Taylor Manifest Anxiety Scale were compared from the pre to
post interviews.

The next seven sessions following the Pre-Treatment Interview
were for purposes of establishing a baseline. After the administration

of the three tension rating scales, EMG levels were recorded for the

 

 

 

94

remainder of the sessions during (a) 20 minutes of rest, (b) the
administration of five cognitive tests known as the Repetitive Psycho-
metric Measures, and (c) l0 additional minutes of rest.

The treatment intervention began after baselining and continued
for l3 sessions. Only information from the first nine of these ses-
sions was included in statistical analysis and hypothesis testing.

The final four sessions were offered because it could not be deter-
mined at the outset of the study, the number of sessions which might
be needed for the subjects to feel that they had obtained maximum
benefits from relaxation training. Hhile nine sessions were adequate
for the actual data collection, there was a concern that maximum
gains be experienced by the subjects.

During the first nine Treatment sessions, the format was iden-
tical to Baseline, with the exception of the first 20 minutes of rest-
ing EMG levels. In place of this, 30 minutes of EMG biofeedback
training was given while EMG levels were recorded. Subjects were
also given cassette tapes of a shortened version of Progressive
Relaxation exercises for daily use at home. During these sessions,
EMGs were recorded during (a) 30 minutes of biofeedback training,

(b) the administration of the RPMs, and (c) To minutes of rest.

Research hypotheses were formulated based on information obtained
from the Baseline and first nine Treatment sessions. Comparisons
were made between EMG values during the first 20 minutes of rest
from Baseline and 30 minutes of biofeedback training during Treatment;

between EMG values during the administration of cognitive tests from

Baseline to Treatment; between scores on each of five cognitive tests

 

95

from Baseline to Treatment; and between scores on the three rating
scales of tension from Baseline to Treatment.

The tenth through thirteenth Treatment sessions involved admin-
istration of the Self-Report Rating Scales, 30 minutes of biofeed-
back training, and l0 minutes of rest. Differences in the median
EMGs obtained during the 30 minutes of biofeedback training from the
first nine and last four Treatment sessions were computed in order to
determine if EMG levels continued to decline. These results were
discussed in the Clinical Observations section.

The Post-Treatment Interview was given following the last Treat-
ment session. The format and information obtained were identical to
those of the Pre-Treatment Interview which was discussed.

Ten research hypotheses were formulated for the seven Baseline
and first nine Treatment sessions. The first two hypotheses were based
on previous findings of chronic overarousal expressed in the skeletal-
muscle system of schizophrenics. It was hypothesized that during the
Treatment phase, EMG levels would decrease. This would occur during
30 minutes of biofeedback training and during the administration of
cognitive tests. Positive results would indicate that schizophrenics
could learn to modify skeletal-muscle tension during biofeedback
training and during task performance.

Five hypotheses were formulated based on theory which predicted
that abnormal physiological arousal, observed as excessive skeletal-.
muscle tension, contributed to the schizophrenic's behavioral responses
of increased distraction, excessive stimulus generalization, and dis—

organization at the cortical level. It was hypothesized that during

96

the Treatment phase, subjects' scores on five cognitive measures would
increase. It was believed that improved performance on these measures
would be indicative of decreased distraction and increased ability to
focus at the cognitive level.

The final three hypotheses related to research in EMG biofeed-
back techniques which emphasized the importance of combining relaxa-
tion exercises with EMG biofeedback training in order to produce
generalization of results outside of the training sessions. It was
therefore hypothesized that during Treatment, subjective ratings of
general levels of tension would decrease and that ratings of the
ability to modify levels of tension and of the ability to detect
changes in levels of tension during the day would increase. Changes
in these scores would indicate that the treatment had an effect on
subjects' increased awareness and control over aspects of tension.

An A-B single-subject time series research design was employed
in the study. Recordings were charted daily during Baseline and
Treatment on the following measures: (a) EMG levels during rest and
biofeedback training; (b) EMG levels during the administration of the
five Repetitive Psychometric Measures of Flexibility of Closure,
Number Facility, Perceptual Speed, Speed of Closure, and Visualiza-
tion; and (c) scores on three Self-Report Rating scales of tension.
The treatment effects were contrasted against the l0 dependent vari-
ables for four individuals at the same relative points in the treatment
schedule. Analysis of the data from the continuous measures was per-
formed using the split-middle technique of trend analysis with binomial

tests to determine the significance of Baseline-Treatment phase

97

changes. One-tailed tests of significance were conducted at the
a = .05 level. Following is a summary of the results of the hypothe-
sis tests:

l. Significant differences were found between phases for §M§_
values during 30 minutes of biofeedback training for all subjects
(P. 2 .002).

2. There was a significant difference between phases for EME-
values during the administration of the Repetitive Psychometric
Measures for Subject 4. at the 9.2 .002 level. No differences were
found for Subjects l, 2, and 3.

3. There were no significant differences between phases for

scores on the RPM test for Flexibility of Closure for any of the four

 

subjects.
4. Significant differences were found between phases for scores

on the RPM test for Number Facility for Subjects l and 4 (p_g .0l9

 

and p_z .002). No differences were found for Subjects 2 and 3.
5. A significant difference was found between phases for scores

on the RPM test for Perceptual Speed for Subject 1 (p_z .002). No

 

differences were found for Subjects 2, 3, and 4.
6. There were no significant differences between phases for

scores on the RPM test for Speed of Closure for any of the subjects.

 

7. There was a significant difference between phases for scores

on the RPM test for Visualization for Subject l (p_g .002 level).

 

There were no differences for Subjects 2, 3, and 4.
8. There were significant differences between phases for

scores on Self-Rating Scale #1 for Subject 2 (p_z .002) and Subject 3

 

 

98

(p_z .002). No differences were found for Subjects l and 4. The
scale indicated overall general levels of tension.
9. Significant differences were found for all subjects between

phases for scores on Self-Rating Scale #2. Significance levels

 

ranged from p_z .002 to p_z .0l9. This scale indicated the ability
to modify levels (If tension during the day.
l0. There were no significant differences between phases for

scores on Self-Rating Scale #3, indicating the ability to detect

 

changes in levels of tension throughout the day.

The following clinical observations were reported:

l. The trend analysis was used to determine the significance
of differences of 10 minutes of resting EMGs recorded at the end of
Baseline and Treatment sessions. Although this was not part of the
hypotheses tests, it was found that Subjects 3 and 4 decreased EMGs
during Treatment (p_z .0l9, p_z .002) for the end-of-session EMGs.

2. Fluctuations in EMG levels during the Pre- and Post-Treatment
Interviews were observed. It was expected that across the pre-
session, the median EMG would increase from the first l0 minutes of
rest to the administration of the TMAS. This was true for Subject 2.
It was also expected that median EMGs during the post-session would
remain low from the first l0 minutes of test to the test administra-
tion. There were, however, increases observed for Subjects l, 2,
and 4.

3. A comparison of the first nine and final four treatment
sessions, using median EMG values, revealed that subjects did not con-

tinue to decrease EMG levels after the first nine sessions.

 

 

 

 

99

4. Scores from the Taylor Manifest Anxiety Scale indicated
that the percentage of items scored positive for anxiety decreased
for Subjects 1, 2, and 3 from the first to secOnd testings follow-
ing biofeedback training. There was a slight increase in items
scored positive for Subject 4.

5. Further clinical observations indicated that subjects found
it difficult during biofeedback sessions to focus thinking away from
stressful or threatening ideation. It was observed that they could
not maintain low EMG levels with the intrusion of such thoughts. As
a consequence, subjects learned to identify thoughts which seemed to
be compatible with relaxation. By the end of treatment, each subject
was able to maintain consistently low EMGs for at least l5 minutes.

6. Following treatment, Subject 2 could periodically fall
asleep at night without her usual sedative, Subject 1 was prescribed
a different antipsychotic medication, and Subject 4 discontinued
medication altogether.

7. Subjects' reports of the frequency of assigned daily home
practice of relaxation exercises during the Treatment phase indicated
that Subject l practiced approximately 80% of the time, Subject 2
practiced 78%, Subject 3 practiced 58%, and Subject 4 practiced 75%

of the time.

Conclusions

 

The outcome data, which included physiologic, cognitive, and
self-report measures as well as various clinical observations, indi-

cated in which ways the treatment intervention was clinically

 

100

beneficial. In the following discussion, conclusions regarding the
impact of treatment are drawn based on changes which occurred (or
did not occur) in each of the different types of variables.
Together, the results were supportive of the use of combined EMG
biofeedback and relaxation exercises for persons who are diagnosed

as schizophrenic and who are in stable clinical condition.

Electromyographic Levels

 

Findings which were related to the changes in EMG levels through-
out various phases of the study indicated in which ways subjects were
able to modify physiological tension, represented in the skeletal-
muscle system, as a result of the treatment intervention. Research
findings repeatedly emphasized that schizophrenics were chronically
over-aroused and that this was consistently observable in skeletal-
muscle measures. In addition, hypotheses put forth in the psycho-
physiological theories of schizophrenia suggested that physiologic
over-arousal was predispositional in nature. Implied in these state-
ments is the notion that muscular tension in schizophrenics is rela-
tively refractory to change. It was evident from the present findings
that subjects experienced considerable success during the feedback
sessions in modifying levels of skeletal-muscle arousal. From
20-minute resting baseline measures to 30-minute recordings during
biofeedback training, subjects decreased skeletal-muscle tension
across phases by not less than 40% and in one case of up to 73%.

In addition, they were able to achieve significant decreases within

nine treatment sessions, and by the end of treatment were able to

lOl

maintain consistently low EMGs for at least l5-minutes duration.
These data are noteworthy, given that the individuals being treated
had experienced long-term difficulty with tension levels. It was
demonstrated that some schizophrenics can begin to learn to alter
chronic arousal within a period of time which does not exceed practi-
cal limits. A description of EMG changes during other events in the
research schedule is important in that there were some unpredictable
trends which related to the clinical significance of the treatment
intervention and raised questions as to general statements which have
been made about schizophrenics by other researchers.

EMG levels recorded during Baseline and Treatment phases while
cognitive tests were administered showed two distinct patterns of
change. For Subjects 1 and 4, there was an expected increase in the
linear trend during Baseline and a decreasing linear trend during
Treatment. The change across phases was significant in the case of
Subject 4. The reverse pattern was observed, however, for Subjects 2
and 3. In these cases, there was a decreasing linear trend during
Baseline which increased at the beginning of Treatment and then began
to decline again. Clearly, the treatment did not have the desired
effect on skeletal-muscle measures during task performance in these
two cases. The fact that EMGs actually declined during baselining
does not follow from studies by Malmo and his co-workers and Goldstein,
who found that during the presentation of stressors and during task
performance, EMG levels in schizophrenics increased or remained ele-
vated from those recorded immediately prior to a task. Not only did

EMG levels decline during baseline sessions in two cases, they were

 

 

102

lower than the 20-minute resting measures taken prior to the task

in all four cases. This observation, then, contradicts those made by
other researchers. Subjects were not highly aroused during task
performance, and the conclusion reached given this information needs
to be qualified. It can be stated that even though EMGs were rela-
tively low during task performance, one of the subjects sustained
even lower EMGs following treatment intervention.

There are several explanations which might account for these
patterns. First, recall that in the clinical observations, subjects
were able to effect a decrease in muscular tension only when thoughts
were directed to nonstressful ideation. The Repetitive Psychometric
Measures are relatively concrete, simple tasks which require some
concentration but probably present little personal threat to the indi-
vidual. Therefore, during Baseline and Treatment sessions, with
attention focused to a nonthreatening task, a decrease rather than an
increase in muscle tension naturally occurred.

A second explanation, which is not necessarily inconsistent with
the first, is that predictions for the present study were based on
data from group studies of inpatient and outpatient samples. Indi-
viduals in the present study could have been quite different from
other randomly selected groups of schizophrenics both in their clini-
cal condition and in the criterion used for diagnosis. Most of the
studies which focus on the psychophysiologic responses of schizo-
phrenics and which are frequently quoted in the literature were per-
formed in the 19505 and 19605. Perhaps it is time to reexamine certain

data on schizophrenia within the context of current therapeutic

 

103

environments. With the introduction and refinement of the use of
antipsychotic medications in the late 19505, changing attitudes
about mental illness, and improved treatment milieus, the variables
which could profoundly affect the progression of any mental illness
have changed dramatically. Findings from the present study indicated
that the diagnosis of schizophrenia alone is not sufficient informa-
tion from which to make global statements about arousal response
tendencies. In this respect, physiologic arousal most likely follows
on a continuum which at times more resembles normal populations, and
it is possible that this might not have been the case 20 years ago.

There was a discrepancy between the EMG patterns observed dur-
ing rest in the present study and resting EMGs reported by Whatmore.
The schizophrenics in his study, who were reported to be in excellent
clinical condition, exhibited EMG patterns which varied widely from
day to day. While subjects in both studies were clinically similar,
those in the present study had EMG patterns during baselining which
were relatively stable, with daily fluctuations which did not exceed
l.8 Mv across days for any one subject. This phenomenon was shown
in the plotted data for 20-minute resting baseline measures. Clearly,
patterns of responding in the skeletal-muscle system are not the same
for all persons who have been diagnosed as schizophrenic, and caution
should be used when generalizing the results from one study to popu-
lations in general.

EMG measures taken during 10 minutes of rest at the end of each
session and during the administration of the TMAS at the end of

Treatment indicated that some of the subjects were able to maintain

 

104

lower EMGs during other events presented in the research schedule

as a result of relaxation training. Subject 3 had lower EMGs during
the 10 minutes of rest and during the test administration while this
was true for Subject 4 during the 10 minutes of rest only. The find-
ings indicated that subjects were beginning to generalize relaxation
from the 30 minutes of biofeedback training. It also could be sug-
gestive that generalization of relaxation was occurring at other times
during the day as well.

In summary, although EMG measures did not always follow predict-
able trends, the treatment intervention was clinically significant
with respect to decreasing skeletal-muscle tension in the following
ways: (a) All subjects were able to effect a decrease in tension
during biofeedback training, and they were able to sustain decreased
EMGs for at least l5 minutes duration within nine treatment sessions;
(b) One subject significantly decreased EMGs during task performance;
and (c) Generalization of relaxation to other events in the research
schedule occurred during the Treatment sessions for two subjects and

continued to occur for one subject at the end of the study.

Repetitive Psychometric Measures

 

The psychophysiological theories of schizophrenia hypothesized
that disturbances in attention and learning were related to physio-
logic over-arousal. Mednick proposed that as physiological arousal
increased, the schizophrenic experienced a diminished capacity at the
cognitive level to discriminate relevant from irrelevant stimuli.

Broen hypothesized that increasing arousal resulted in sensory

105

overload at the cognitive level with a diminished ability to utilize
relevant environmental cues. Tecce and Cole also theorized that as
arousal increased, a decrement in cognitive performance occurred.

It was expected that cognition, as measured by test performance,
would be one of the variables affected by changing EMG levels. There
was no apparent relationship, however, between EMG levels at the time
of testing and improved performance on tests for subjects in the
present study.

During treatment, Subject l did not decrease EMGs during testing,
but he did improve scores on three of the five tests presented, while
Subject 4 decreased EMGs but improved scores on only one of the tests.
Although it could not have been predicted, the finding that there was
no apparent relationship between cognitive performance and EMGs was
not surprising in view of the fact that EMGs were relatively low dur-
ing the test administration across Baseline and Treatment sessions.
Integrating these data with the theoretical formulations, it could be
that at lower levels of arousal, disturbances in cognition are not
evident in test performance. Although data provided in the present
study do not necessarily negate the theories, they point to a weakness
of the theories to take into account the fluctuating clinical condi-
tion of the schizophrenic, specifically, when symptoms improve.

While theorists emphasized the effects of increasing arousal on behav-
iors, they did not discuss the nature of the interaction of variables

at what appears to be the opposite end of a continuum of behaviors.

It was assumed in the theories that the schizophrenic was chronically

over-aroused, and therefore, it would be expected that even at lower

 

106

levels of arousal there would continue to be an interactive effect
with cognition which is measurable. A more accurate statement

might be that schizophrenics are prone to excessive physiological
arousal which can result in changes at the cognitive level, but in

the absence of symptoms, they resemble normals on formal cognitive
functioning. If this is the case, then relaxation training could be
effectively used as a preventative measure, teaching the schizophrenic
to become more aware of changing states of tension and to become cog—
nizant of the internal and external events which affect tension levels
so that when tension began to increase, individuals would be familiar
with techniques to minimize its effects.

Even though there was no discernible relationship between arousal
and test performance, there could be a relationship between elements
of the treatment intervention and improved scores having to do with
thought control. During biofeedback training, subjects became aware
that it was necessary to focus attention in order to decrease arousal.
Even though they were in excellent clinical condition, they still
experienced difficulty with perseverative, stressful ideation.
Improvements in the ability to exclude extraneous stimuli by focused
and narrowed attention resulted in an increased ability to focus dur-
ing tests. It is possible that techniques learned in the treatment
intervention were used by two of the four subjects, evidenced by the
improved ability to perform on cognitive tests as a result of learning
to focus attention on the task.

The test construction itself guards against improvements in

scores occurring because of repeated testings. The 20 forms of each

 

 

107

test were equated in terms of level of difficulty and had adequate
test/retest reliabilities, so changes could not be attributed to
factors of instability or nonequivalence of the tests themselves.

It was unclear as to why there were improvements on some tests and
not on others. The authors reported that each test represented a
different factor of cognitive abilities, but they did not elaborate
on this information in stating which factors might be more sensitive

to changes.

Self-Report Rating Scales

Because the biofeedback and relaxation exercises were combined
as a treatment intervention, it was not possible to determine the
separate effects each might have had on the dependent variables.
Research indicated, however, that individuals experienced more posi—
tive changes in subjective ratings of tension with relaxation exercises
than with biofeedback. It was therefore expected that daily home prac-
tice of relaxation exercises would enhance awareness of tension and
would contribute to increased control over tension outside of the bio-
feedback sessions.

Subjects l, 2, and 3 showed improvements on two out of three of
the rating scales while practicing 80%, 78%, and 58% of the time.
Subject 4 practiced 75% of the time and improved on one of the scales.
If subjects had practiced more consistently, they might have experi-
ﬁanced a greater degree of awareness and control over levels of tension.

Examination of the linear trends for daily ratings of tension

( Scale #1) indicated that for Subjects 2 and 3, whose ratings of

 

 

 

108

tension decreased significantly across phases, the linear trend
increased during Baseline and decreased during Treatment, as expected.
In the case of Subjects 1 and 4, the linear trend began to decrease
during Baseline sessions, which indicated that before Treatment they
began to experience less tension. It is possible that their Baseline
ratings were reflecting an increasing degree of comfort with the
experimental situation itself. These ratings increased at the begin-
ning of biofeedback training and then decreased as the sessions pro-
gressed. The fluctuating linear trend could again have been an indi-
cation that during Treatment, subjects felt some initial discomfort
when feedback about tension levels was introduced, and over time,
adjustments were made to the experimental conditions, resulting in
lower ratings. 0n the other hand, this might have indicated that
subjects rated themselves more realistically on tension as they began
to discriminate tension from relaxation during Treatment. Unfortu-
nately, it is not possible to determine which interpretation best
described these trends. In all cases, as treatment progressed, sub-
jects' ratings of tension decreased and followed a pattern which was
consistent with their decreasing EMG levels. In two of these cases,
the treatment intervention resulted in a significant positive change
in subjective feelings of tension.
Positive changes occurred for each subject in their perceived

aibility to decrease tension when they were aware of being tense.
This finding was consistent with the fact that with feedback during

tzraining, they were able to significantly decrease EMGs. Clearly,

 

 

109

the treatment intervention resulted in subjects feeling that they had
more control over tension.

There were no changes in subjects' ratings of their awareness
of changes in levels of tension during the day (Scale #3). Since
daily EMG levels were relatively stable, subjects probably did not
experience the dramatic changes in tension which were expected, and
therefore, they found it difficult to report on any minor fluctua-
tions which might have occurred. Scale #3 is probably the most diffi-
cult of the scales in that it requires thateaperson be sensitive to
fluctuating degrees of tension during various activities when atten-
tion is not necessarily focused on physical sensations. This degree
of sensitivity may require more extensive training in discriminating
degrees of tension and relaxation so that the process becomes inter-
nalized. Consistency and duration of the practice of relaxation
exercises might not have been sufficient in order for improvements to
occur on this scale.

It was stated that subjective feelings of tension might not be
consistent with skeletal-muscle arousal because of the effects of anti-
psychotic medications. Lader and Matthews pointed out that these
drugs decreased autonomic arousal and therefore, individuals may feel
that tension is lower, even though the drugs have no effect in reduc-
ing somatic or skeletal-muscle arousal. Results from the first two
r~ating scales indicated that as treatment progressed, subjects felt
<:onsistently less tense as EMG levels decreased and they were aware

(Jf’their increased ability to control tension. The combined treatments

 

110

were effective in that subjects seemed to be able to realistically
assess changes in muscular tension and were aware of their increased

ability to alter tension levels.

Clinical Observations

 

At the end of treatment, Subjects l, 2, and 3 reported fewer
symptoms of anxiety indicated by scores on the TMAS. Prior to treat-
ment, the number of items they scored positive for anxiety ranged
from 3l to 4l out of a total of 50. Their scores decreased from 26%
to 56% at the end of treatment. In contrast, Subject 4 scored l3
items positive for anxiety prior to testing and increased her score
by three items at the end of treatment. Subject 4, then, had a rela-
tively low level of anxiety at the beginning of the study and experi—
enced a relatively small change in symptoms compared to the others.
The slight increase is inconsistent in view of the progress she made
during treatment. She decreased EMGs by 73%, maintained consistently
lower EMGs during the end—of—session resting measures and during task
performance, improved scores on one test, and reported an increased
ability to modify tension levels. A critical factor during Treatment
which could have effected the increase in scores was that Subject 4
was the only individual in the study who discontinued antipsychotic
medications. She was also one of two subjects who did not report
feeling less tense by the end of treatment. The change in medication
could have been followed by increased levels of autonomic arousal,

thereby increasing her experiences of tension and symptoms of anxiety.

 

111

Because subjects continued to attend scheduled therapy sessions
during the study, it cannot be ruled out that changes in test scores
were a result of extraneous variables. However, since decreased
anxiety occurred in three out of four cases, and the slight increase
for Subject 4 was likely due to medication effects, it is more prob-
able that the treatment intervention contributed to the improvements
in anxiety.

The importance of cognitions, particularly excessive rumination
and preoccupation, in the maintenance of tension was stressed in the
biofeedback literature. It was stated that in the process of reduc—
ing the physiological symptoms of stress through biofeedback, the
individual's attention shifted to the subjective mechanisms which
maintained stress. In addition, it was pointed out that during the
biofeedback process, individuals become aware of the importance of
self-control, which gives them the capacity to alleviate their own
symptoms.

The process by which subjects learned to modify tension clearly
involved becoming aware of and learning to control throughts. As
subjects became aware of fluctuations in their EMG levels during
biofeedback training, they began to identify persistent thoughts
which prevented them from relaxing. They discovered that a conscious
effort was required in order to persist in controlling stressful
ideation, and consequently, in order to maintain lowered tension. It
was a major accomplishment that, by the end of treatment, they could
rnaintain a mind set free from these preoccupations for a minimum of

l 5 minutes. Certainly, these discoveries were a significant gain

 

112

for all subjects, and the treatment process allowed significant
insights regarding the impact of cognitions on functioning.

Since medications were not controlled for in the study, it
cannot be established that there was a relationship between the
intervention and the medical decision to discontinue Subject 4's
antipsychotic medication. Each subject, at various times in the
study, expressed a desire to stop taking medications. This "goal"
seemed to be present prior to the study, and therefore, it was note-
worthy that this subject had been maintained on the same type and
dosage of medication for one year prior to the study. Perhaps the
insights related to cognitions and stress and an increased sense of
self-control contributed to the decision to discontinue medications.

Subject 2, who stopped taking her nightly sedative, stated that
she could relax and focus thoughts to nonstressful ideation. Her
psychiatrist confirmed that for six years prior to the study she had
been unable to sleep without medication. The issue of whether or not
the treatment intervention contributed to the decreased need for
antipsychotic medications in clinically stable schizophrenics is one
which cannot be answered within the present study. It was estab—
lished, however, that elements of the intervention were directly
related to one subject's decreased reliance on sedatives.

Researchers provided evidence that Quieting Response Training
was successful for use with a mixed group of psychiatric inpatients
in alleviating somatic complaints. Most importantly, they found that
r~elaxation techniques did not increase emotional instability as others

fiad predicted. The results of the present study provided evidence

 

113

that another form of relaxation training can also be used without
producing negative emotional reactions in schizophrenic outpatients
who are in stable clinical condition. In fact, the treatment inter—
vention was clinically beneficial in that subjects were able to
modify skeletal-muscle tension which was believed to be chronic and
predispositional in nature. In addition, there was some evidence
that the ability to decrease skeletal-muscle tension was generalized
to other situations in the research schedule for some individuals.
Other improvements were noted for some in the ability to perform
concrete cognitive tasks, in symptoms of anxiety, and in subjective
feelings of tension. The issue of a decreased need for antipsychotic
medications and sedatives was discussed in conjunction with elements
of the treatment intervention. Finally, the treatment process allowed
individuals to realize the relationship between tension and ideation
and provided a setting within which they could practice redirection
and refocusing of cognitions. The initial premise that the treat-
ment be used as an adjunct to other accepted treatment modalities

seems well founded in view of the present findings.

Limitations
Some of the major limitations of this study concern the research
design itself. The most obvious limitation in the single-case study
design involves the generalization of results. It is not known if
the results from cases presented in the present study would be rele-
vant to other individuals. The subjects were not a representative,

r~andom sample, and it cannot be inferred that they were characteristic

 

114

of any given population of schizophrenics. Therefore, in the absence
of random sampling of subjects, the generality of findings to other
patient populations is restricted. Since the treatment procedures
were repeated with four individuals, it is possible to make infer-
ences to other patients with similar characteristics, the most impor-
tant being the diagnostic criteria and the clinical condition of the
patient.

Similarly, the lack of randomization in terms of therapists and
treatment settings also restricts the generality of findings. It
cannot be assumed that other clinicians in different treatment settings
would achieve the same results. The investigator was the same indi-
vidual who provided the clinical treatment to subjects, and there is
the possibility that changes in behaviors were a result of increasing
comfort subjects might have experienced with the investigator.

A more serious problem might exist if the outcome of measures was
influenced by experimenter bias. This seems unlikely because EMG
measures, cognitive tests, and true/false tests of anxiety only need
to be recorded, leaving the interpretation to the scores themselves.
Scores on the self-rating scales could have been more easily influ—
enced by the investigator in some way communicating to the subjects
that certain results were desired.

Because of the lack of post-treatment follow—up and patient
controls in this study, there were threats to internal validity which
necessitate that conclusions about the effects of treatment on behavior
be interpreted with caution. For example, there was no information

provided about the natural course of the behaviors studied. It is

   

 

 

115

possible that variables other than the treatment intervention, such

as history and maturation, could have contributed to the changes

which were observed. These threats were, in part, minimized by
repeating baseline measures over seven sessions in order to estab-
lish the course of changes in behaviors which might have naturally
occurred. But the baseline measures alone do not firmly establish
that changes in the Treatment phase occurred because of the treatment
intervention. Therefore, statements of causality cannot be made,

and inferences about the results should be made with some reservation.

Another limitation lies within the self-rating scales which were
developed for this study. Even though they appeared to have face
validity, no reliability coefficients were determined prior to the
study for these scales. Therefore, any significant results reported
from the scales should be interpreted with caution.

Finally, it is important to emphasize that awareness and modi—
fication of muscular tension requires extensive practice and training.
Although changes in behaviors were noted during the study, it cannot
be established that in the absence of additional training, individuals
would continue to experience improvements in symptoms of anxiety and

in their ability to control tension levels.

Implications for Future Research
In this study, data were presented which suggested that a com-
bined treatment program of EMG biofeedback with Progressive Relaxation
can be a viable therapeutic intervention for clinically stable schizo-

pfirenics. There are several ways in which future research could expand

 

 

 

116

on these preliminary findings. First, the study should be replicated
with adequate controls for factors which could affect outcomes. It
needs to be established with certainty to which degree physiological,
cognitive, and emotional behaviors are affected by the proposed
treatment. This could be accomplished by extending the study for
post-treatment follow-up and by altering the research design so that
other subjects serve as controls.

In addition, the clinical applicability of the proposed treatment
intervention could be tested by using the same techniques with a wider
variety of subjects, particularly those who are psychologically less
stable, and with other therapists and treatment settings.

Finally, the relationship between schizophrenics' ideation and
physical tension seems to be a viable topic for research. It was
suggested in this study that states of physical tension in schizo-
phrenics are reinforced and maintained by perseverative, stressful
ideation. Further studies could explore the benefits of using EMG
biofeedback training to enhance awareness of such ideation and to teach
schizophrenics a process for altering or refocusing cognitions. In
this respect, combinations of relaxation therapies such as Quieting
Response Training or autogenic training, such as Open Focus Exercises
developed by Fehmi, might be used to enhance the physical, mental,

and emotional integration of this specific population.

 

 

APPENDICES

117

   

 

APPENDIX A

SELF-REPORT RATING SCALES

118

 

 

APPENDIX A

Following are three questions which subjects were asked to rate

according to the way they presently felt.

1.

What is your present general level of tension? Rate yourself
on a scale from l-lO, with 1 being very low, 5 moderate, and

10 very high.

When you feel tense, how well do you feel you can lower your
level of tension? Rate yourself on a scale from l-lO, with
l being no ability to lower tension levels, 5 moderate
ability, and 10 being very good ability to lower levels of

tension.

Are you aware of the times when tension is increasing or
decreasing during your daily routine? Rate yourself on a
scale from l-5, with l being never aware, 2 being almost
never aware, 3 being sometimes aware, 4 being often aware,

and 5 being almost always aware.

119

 

 

APPENDIX B

REPETITIVE PSYCHOMETRIC MEASURES

120

 

APPENDIX B

INSTRUCTIONS
(V)

Your task is to follow a single line from beginning to end through a
whole group of lines. Look at the figure below. At the left -hand

side you will see a set of numbered boxes. A line begins at the side
of each one of these boxes and ends at an empty box on the right. You
are to write the correct number in the box on the right. Look at the
line beginning at the side of box 1 and follow it _w_i_th_ your eyes to the
point where it ends at a box on the right, Notice that the figure 1 has
been written in the proper box. For practice, fill in the other boxes.
Remember, the lines must be followed with your eyes; do not follow
with your finger or pencil.

 

Copyright by H.” ”Motion for Mental Health, 1959.

121

122

V—Form 2 Follow each line with your eyes to its proper cell on the right and
write the number of the line in the cell.

 

Subiect Date Examiner

 

 

123

O

INSTRUCTIONS
(SC)

Look at the two rows of letters below.

Amos--.TEYKZCIwQEOZNPE
@0T@VA¥FSMI--YUNTRI

Notice that a circle has been drawn around groups of four consecutive
letters that spell out a common English word. Only four-letter words
are marked. The same letters are used in only one word at a time.

Below are more rows of letters on which to practice. Draw a circle
around each set of four adjacent letters which spell out a common
English word. Go ahead. Do not wait for any signal. Read along the
rows from the left side of the page to the right side.

KIYOURGAC'SPIHANDVZQJUXDIEDOLROOM
VORSZREADUOFTEBEENAQPAUNTAQSERTL
XJAKCOWBNOSELIQXESTFASTWDDMTRIZE
E‘LBIEBESTBTREXDOWNGUTNELASTWITCI

When the signal is given (not yet). turn the page and draw a circle
around each set of four adjacent letters in a row that spells out a

common English word. Go from left to right in each row. Work

fast. Mark only foul—letter words.

 

 

 

 

124

>3NDH¥GWN>¢O¢KONSOC§O3NO>C§

“NKHZWQ43OXZOMQHDO>QCDG
WOHHHNUNxWNDmiii-tOMNIDZHE-t
mmemoc-zwuzaomnwquzdzaw
OZDﬁ>BNBOBhNDODQXO¥QNm

nmm>mmaquamZstcmo>Ho
<oacmscoom3><smozmmm<4<3=>m
HIQQSOASOZIM>ZOZ>UHH34CQ

30339

:3 <3 2: <3 *3 <3 0: :n r: <4 >< :3 2: ca 3: << 3: :3 a: <3 a: '3

<3 u: :3 #4 :n C) :3 << n: :3 a: *3 >1 c: .3 c: :3 a. :4 a. :4
<< :3 <3 *3 a. 64 Ec <3 r4 &< r: '3 a: ~< <3 3. <3 3: :3 2: :3 ::

mmmmmmzmmmmmmmmonxmam
c3 '3 <3 Eu 64 E: :> to << :3 <3 3» :3 2: <3 r4 2: <3 03 <3 :3 :>
<n E4 9! <3 3: «a >< :3 a: :3 «I <3 c3 '3 3: <3 on <3 C: <3 3. <3
<3 #4 :n 3: '1 a) 1< at :4 a. C? :< 3. <3 #1 '3 F4 .3 :3 <3 c: <3

0 m

@303 on» :3

.3 :> '3 :a :3 C? r: < a: a: :3 3' <3 < c: <3 < a: :3 >< <3
<3 3. In a: :3 2: <3 *3 2: Ed :3 c: :n >< :3 a. :n 2: <3 <3 r: :3
xzeoncxawmmmw>nn>m>m>
<3 c> <3 c: 54 :z *3 <3 :3 :3 E0 a: <n z: :3 a: :3 a: :3 2: :n
on ca >< <3 2: :4 w< <3 3: <3 :3 c> <3 <3 c3 <n c3 *3 <3 e< E4
:3 c: :3 at Id c3 2: v3 2: '1 r4 c: c: 3: :3 >~ >< *4 >3 E: .3
:n a. *3 <3 a) <3 3» #4 <3 <3 >< :3 <3 3. c3 :3 a: <2 < n1 :3
.3 «2 <3 :3 c3 :3 a: on t» 2: <: < c3 <3 2: .3 bl r3 < »1 <3 2.
< 2: <2 2: 64 <3 0: <3 r: <3 a: <3 >' <3 <3 2: <3 :3 «i <= '3
2: c: 2: <3 C? >. 54 '3 s: :z 3: <3 I: 3: <3 #4 :4 Ea 54 <3 0: .3

m H S < o m H

ago: nouuoa<psou some

ommmHmzmzxxmawxmc-JDDO
Abzoxxmcaonzmmmmmzqmn

>< E~ ch ‘4 a: :n c: .3 c: c: .3 60 E4 co 2: :3 < a: :z 3: <3
H
<3 no :3 a: <3 n. :z a. 6* <3 3: :n s: to &< co 64 :z a: :< a: co
m 2: a: :4 a. <: <3 e< r: <4 ht a: <3 2: <3 4 a: E4 <1 <3 3» cr
:3 :> c: e< <3 cr :3 z: >< a: :3 c3 2: .3 :3 <3 0: .3 >< <3 3: co
a: :3 c3 :2 a: :3 c: v: e: is *3 <3 r: <3 2: <3 c> <4 a. o: v: c:

m N x

canons 030330

3: :3 3: <3 => <3 it <3 bl <0 2: <3 >< <3 r3 <3 2: E< >< 2: <3 >¢
m<uaummqmzzaammmzomoos

{U

GHQROM<GNMZMHmhﬁ><QOQE
mxm<mmzmzz>mo==monHmnm

swan

mocmcmszmmmmmmspzmax>

mHmmmaomonmm03>3manqn
BWZOZNhHh>QBBQQKﬂ>ND<Qm

nmqmamnzwmmqwozmomazHa

zaxzmbm<ucoom>xozamazm

<m<3zne<n==3ma=e<om<3m3m><mnwe

WZZHNAMWZBMWZZBWMHQSOQW

wnwxmoonmmDonmn>zzneo

3 >< :2 a. :3 a. »< :3 <3 :3 <5 <3 2: <4 <3 >< 3: <3 <3 «a :n 3

 

Examiner

 

 

Subject

3
t
a
D
asom-om

 

125

INSTRUCTIONS
(NF)

This is a test to see how quickly and accurately you can add. It is not
expected that you will finish all the problems in the time allowed. Work
steadily and as quickly as you can without sacrificing accuracy.

You are to write your answers in the boxes below the problems. Examples
of the problems are given below with the first one correctly worked. Prac-
tice for speed on the others.

Practice Problems:

4 7 12 84 7 34 17 45 31 80
9 6 5 54 38 81 50 41 52 78
l 15 67 . 72 80 51 74 89 19 15

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘F
-—-1

 

 

Copyrtﬁt by No” Foundation for Mental Health, 7959.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

126

NF -Fo rm 1

35 24 10 16 20 33 32 51 26 38
38 23 16 86 38 42 38 97 1 50
31 96 25 91 47 96 44 33 49 13
66 67 4O 67 14 64 5 71 95 86
14 90 84 45 11 75 73 88 5 9O
68 5 51 18 20 33 96 2 74 19
20 46 78 73 9O 97 51 4O 14 2
64 19 58 97 79 15 6 15 93 20

5 26 93 7O 60 22 35 85 15 13
7 97 10 88 23 9 95 42 99 64
68 71 86 85 85 54 8? 66 47 54
14 65 52 68 74 87 37 78 22 41
17 53 77 58 71 71 59 36 50 72
9O 26 59 21 19 23 41 61 33 12
41 23 52 55 99 31 52 23 69 96
26 99 61 65 53 58 4 49 80 7O
46 98 63 71 62 33 26 16 7 45
42 53 32 37 32 27 7 36 51 80
32 9O 79 78 53 13 55 38 58 59

5 3 72 93 15 57 12 10 14 21
31 62 43 9O 90 6 18 44 32 53
17 37 93 23 78 87 35 20 96 43
77 4 74 47 67 21 76 33 50 25
98 10 50 71 74 12 86 73 58 7
52 42 7 44 38 15 51 29 13 42
49 17 46 9 62 90 52 84 77 27
79 83 86 19 62 6 76 50 3 10
Subject Examiner Date

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

127

INSTRUCTIONS

(PC)

In this test you are to copy each of the figures in the dotted space to the

right of it.
for every corner.

Figure A has been
copied as an example

The little circles show youyhere to begin.

Copy figure B in
the dotted space.

There is a dot

Copy ﬁgure Copy ﬁgure D
C in the dotted in the dotted

 

in the dotted space to Begin at the small space in the space.
the right of it. circle and use the same
heavy dots as manner.
guides. ‘
A I c o
- .. . ....o ....o .....
yy. ..... ..... 9....
‘ .... .....

 

 

 

 

 

 

Copy each of the following figures in the same way. Begin at the small

circles. Your lines do not have to be straight but they should begin and

end on dots. Make corrections it you wish, but do not waste time erasing.
. Go right ahead. Do not wait for any signal. -

.0. .---- ..

When the signal is given (not yet) turn the page and copy more ﬁgures of
the same kind.

 

 

 

 

 

 

 

Work rapidly because your mark will be the total number of figures cor-
rectly copied.

You may not be able to ﬁnish in the time allowed.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

129

INSTRUCTIONS
(PS)

Look at the rows of numbers below. A ring has been put around the first
number in each row. The numbers in the row that are like the circled number

are crossed out.
@721
63/28
@0863;
@69/{1

Cross out the numbers in the following rows that are like the circled numbers
at the beginning of each row.

@19523
.67158

@95482
@45352

Here are more practice rows. Cross out each number in a row that is like the
circled number at the beginning of each row.

O
O

0‘
U)

.7603521058479326585380118
@5910794200834208616935861
@5825973468532019674567243
@3580181074257587432790187
@5369122537580386053421975
@8347182027436801863647965
@4585745873268459110784276
@6454758018325457968017539
@0795334791704023 815379-0

When the signal is given (not yet), turn the page and cross out each number in
a row that is like the circled number. Work Fast. Find as many numbers as
you can in the time allowed.

 

PS-Form 1a

@6169

3

l

8

@®0®@®®@®®@G@®®®®®@®®®@®

6

Subject

Cross out each number in a row that is like the circled number.

130

772844423478

02895
019407535 0470
3094032212950
695017717 8140
985358523 3308
465087640 6253
169948958 8077
670076740 2752
897403497 3095
9955018017581
796094927 8981
766427742 3516
842234425 3425
3601147551161
788374427 0919
234099548 6341
237866676 0893
079410934 46.93
944232182 5007
524744673 7792
693429414 9714
6502130791358
646058090 6845
6981874218684
050209498 8007

Date

 

7

6

5

8

8

2

3

2

1

6451049

3

8

l

8

O

2

0508

 

8

8

9

6

8

8

3

9

Examiner;

0

2

1

6

4

3

4

 

   

PS-Form la
6 l 6 9

3019

@®0@©®®@@®@0@®®®®®@®®®®@®

Subiect

130

Cross out each number in a row that is like the circled number.

0

2895772

0

0

7

5

2

3

2

5

l

2

0

2

844423478

4

7

 

0

Date

7

5

8

8

2

2

l

5

 

64510494
38880147

80882635

3 0 7 5 1 6 1 4
1 2 9 3 7 2 O 5
0 8 6 9 3 1 4 6

Examiner:-

 

 

PS-Form lb

6

0

(990696")@@®®®®©®@@®@®@@®®O@©

Subiect

6

7

3

8

131

Cross out each number in a row that is like the circled number.

7

7

9

2

9

0

3

0

6

5

7

1180729

4

8

8

Date

1

9

2

8

1

8

4

3

5

6

1

20744156

1

4

l

Examiner

 

 

7

9

6

5

8

3

l

2

7

0

6

9

5

   

APPENDIX C

TAYLOR MANIFEST ANXIETY SCALE

132

 

 

 

APPENDIX C

Instructions

 

The examiner reads the items to the subject. The subject is
instructed to answer TRUE if the statement is true (M" mostly true
for him. He is to answer FALSE if the statement is false or mostly
false for him.

Questions

1. I do not tire quickly.

2. I am often sick to my stomach.

3. I am about as nervous as other people.
4. I have very few headaches.

5. I work under a great deal of strain.
6. I cannot keep my mind on one thing.

7. I worry over money and business.

8. I frequently notice my hand shakes when I try to do
something.

9. I blush as often as others.

10. I have diarrhea ("the runs") once a month or more.
ll. I worry quite a bit over possible troubles.

12. I practically never blush.

13. I am often afraid that I am going to blush.

14. I have nightmares every few nights.

15. My hands and feet are usually warm enough.

16. I sweat very easily even on cool days.

l7. When embarrassed I often break out in a sweat which is
very annoying.

133

 

18.

19.
20.
21.
22.
23.
24.

25.
26.
27.
28.
29.
30.
31.

32.
33.
34.

35.

36.

37.

38.
39.

l34
I do not often notice my heart pounding and I am seldom
short of breath.
I feel hungry almost all the time.
Often my bowels don't move for several days at a time.
I have a great deal of stomach trouble.
At times I lose sleep over worry.
My sleep is restless and disturbed.

I often dream about things I don't like to tell other
people.

I am easily embarrassed.

My feelings are hurt easier than most people.
I often find myself worrying about something.
I wish I could be as happy as others.

I am usually calm and not easily upset.

I cry easily.

I feel anxious about something or someone almost all of
the time.

I am happy most of the time.
It makes me nervous to have to wait.

At times I am so restless that I cannot sit in a chair for
very long.

Sometimes I become so excited that I find it hard to get
to sleep.

I have often felt that I faced so many difficulties I could
not overcome them.

At times I have been worried beyond reason about something
that really did not matter.

I do not have as many fears as my friends.

I have been afraid of things or people that I know could
not hurt me.

 

4o.
41.
42.
43.
44.
45.
46.
47.
48.
49.

50.

135

I certainly feel useless at times.

I find it hard to keep my mind on a task or job.
I am more self-conscious than most people.

I am the kind of person who takes things hard.

I am a very nervous person.

Life is often a strain for me.

At times I think I am no good at all.

I am not at all confident of myself.

At times I feel that I am going to crack up.

I don't like to face a difficulty or make an important
decision.

I am very confident of myself.

 

APPENDIX D

SUBJECT'S CONSENT FORM FOR ELECTROMYOGRAPHIC
BIOFEEDBACK STUDY

136

 

 

1.

APPENDIX D

I have freely consented to take part in a scientific study
being conducted by: Carol A. Keating
under the supervision of: William Hinds. Ph.D.

 

 

Academic Title: Professor of counseling. Personnel segyﬁoee,

and Educational Psychology

I understand that to participate in the study, I will be aksed to:

a. reply to questions about myself which will be presented
orally.

b. allow the examiner to place surface electrodes on the frontalis
muscle of my forehead in order to record the tension of that
muscle.

c. perform short paper-and-pencil tests.

practice relaxation exercises with the aid of a tape-recorded
program.

I understand that I will be participating in a treatment program
approximately four days per week, with each session lasting
approximately one hour.

I understand that this program will be conducted over a period
of time of not less than three weeks and not more than ten weeks.

I understand that I will not be required to discontinue any
ongoing therapy, Such as medication therapy or individual or
group therapy, during the study.

I understand that I am free to discontinue my participation in the
study at any time without penalty.

I understand that the results of the study will be treated in
strict confidence and that I will remain anonymous. Within these
restrictions, results of my own test data will be made available
to me at my request.

137

138

I understand that my participation does not guarantee any
beneficial results to me, but that there is a probability that

this may occur.

I understand that, at my request, I can receive additional
explanation of the study after my participation is completed.

 

 

Signed Date

 

 

Witnessed by Date

APPENDIX E

PERSONAL DATA INVENTORY

139

 

APPENDIX E

 

 

 

 

 

Name

Birth Date Sex

Marital Status (single, married, widowed, divorced)
Education

Occupation

 

Current Living Situation

 

Onset of Psychiatric Illness

 

Previous Hospitalizations

 

 

Medications Taken in Past 6 Months

 

 

Current Medications and Dates of Changes

 

 

 

Persons directly involved in study: Name and Function

 

 

 

 

140

 

APPENDIX F

PHYSICIAN'S CONSENT FORM FOR ELECTROMYOGRAPHIC
BIOFEEDBACK STUDY

141

 

-—J
0

APPENDIX F

I freely consent to allow my patient:

 

to take part in a scientific study being conducted
by: carol A. Keating

 

under the supervision of: William Hinds, Ph.D.

 

Academic Tit1e: Professor Qf'Cbunseling, Personnel services,

 

and Educational Psychology

 

I have been fully informed of the requirements of my patient for
participation and believe that he/she is medically and psycho-
logically able to participate in the proposed treatment program.

I agree to assess the psychological well-being of my patient
during this study and will withdraw my consent if, in my judgment,
the program is detrimental to his/her well-being.

I understand that my patient will continue with any current psy-
chopharmacological and psychotherapeutic treatment which I consider
beneficial, and that it is preferable that such treatments not be
altered during the course of the study.

If alterations in current treatments (outlined in statement 4)
are necessary, I will inform the examiner of such changes.

I understand that, at my request, I can obtain results of test out-
come data and will treat such materials with utmost confidentiality.

I understand that participation in this study does not guarantee
any beneficial results to my patient, but that there is a proba-
bility that this will occur.

 

 

Signed Date

 

 

Witnessed by Date

142

 

REFERENCES

143

 

REFERENCES

Alexander, A. B. An experimental test of assumptions relating to
the use of electromyographic biofeedback as a general relaxation
technique. Psychophysiology, 1975, 1216), 656-662.

 

Basmajian, J. (Ed.). Biofeedback. Principles and practice for
clinicians. Baltimore: Williams & Wilkins Co., 1979.

 

Benson, H. The relaxation response. New York: William Morrow & Co.,
Inc., 1975.

 

Broen, W. E., Jr. Schizophrenia: Research and theory. New York:
Academic Press, 1968.

 

Brown, B. Stress and the art of biofeedback. New York: Bantam
Books, Inc., 1978.

 

Canter, A., Kondo, C., & Knott, J. A comparison of EMG biofeedback
and progressive muscle relaxation training in anxiety neurosis.
British Journal of Psychiatry, 1975, 121, 470-477.

 

Coursey, R. Electromyograph feedback as a relaxation technique.
Journal of Consulting and Clinical Psychology, 1975, 43,
825-834.

 

Dahlstrom, W., Welsh, G., & Dahlstrom, L. An MMPI handbook: Volume II.

 

Minneapolis: University of Minnesota Press, 1975.

Feighner, J., Robins, E., Guze, S., Woodruff, R., Winokur, G., &
Munoz, R. Diagnostic criteria for use in psychiatric research.
Archives of General Psychiatry, 1972, 26, 57-63.

 

Fenz, W., & Velner, J. Physiological concomitants of behavioral
indexes in schizophrenia. Journal of Abnormal Psychology, 1970,
2611), 27-35.

 

Ford, M., Stroebel, C., Strong, P., & Szarek, B. Quieting response
training: Treatment of psychophysiological disorders in psy-
chiatric patients. Proceedings of the Biofeedback Society of
America. Twelfth Annual Meeting, March 1981, Louisville,
Kentucky.

Freedman, A., Kaplan, H., & Sadock, B. (Eds.). Comprehensive textbook
of psychiatry. Baltimore: Williams & Wilkins Co., 1976.

 

144

 

145

Fuller, G. Biofeedback: Methods and procedures in clinical praCtice.
San Francisco: Biofeedback Press, 1977.

 

Gaardner, K., & Montgomery, P. Clinical biofeedback: A procedural
manual. Baltimore: Williams & Wilkins Co., 1977.

 

Gellhorn, E. Motion and emotion: The role of proprioception in the
physiology and pathology of the emotions. Psychological Review,
1964, 2116), 457-472.

 

Goldstein, I. The relationship of muscle tension and autonomic
activity to psychiatric disorders. Psychosomatic Medicine, 1965,
27, 39-52.

 

Green, E., & Green, A. Beyond biofeedback. San Francisco: Delacorte
Press, 1977.

 

Hersen, M., & Barlow, D. Single case experimental designs: Strategies
for studying behavior chéhge. New York: Pergamon Press, 1976.

 

 

Jessup, B., & Neufeld. Effects of biofeedback and "autographic relaxa-
tion" techniques on physiological and subjective responses in
psychiatric patients: A preliminary analysis. In J. Stoyva,

J. Kamiya, T. X. Barber, N. Miller, & D. Shapiro (Eds.), Biofeede
back and self-control. 1977/78. New York: Aldine Publishing
Co., 1979.

 

Kazdin, A. Methodological and interpretive problems of single case
experimental designs. Journal of Consulting and Clinical'Psy:
chology, 1978, 6614), 629-642.

 

Kratochwill, T. (Ed.). Single subject research: Strategies for
evaluating change. New York: Academic Press, Inc., 1978.

 

 

Lader, M. H., & Mathews, A. M. Electromyographic studies of tension.
Journal of Psychosomatic Research, 1971, 163 479-486.

 

Lang, P., Rice, 0., & Sternbach, R. The psychophysiology of emotion.
In N. Greenfield & R. Sternbach (Eds.), Handbook of psycho-
physiology. New York: Holt, Rinehart & Winston, 1972.

 

Lehman, H. Psychopharmacological treatment of schizophrenia.
Schizophrenia Bulletin, 1975, 16, 27-45.

 

Lezak, M. Neuropsychologjcal assessment. New York: Oxford Univer-
sity Press, 1976.

 

Malmo, R., & Shagass, C. Physiologic studies of reaction to stress
in anxiety and early schizophrenia. Psychosomatic Medic1ne,
1949, 11, 9-24.

 

146

Malmo, R., Shagass, C., Davis, J., & Eng, M. Electromyographic studies
of muscular tension in psychiatric patients under stress.
Journal of Clinical and Experimental Psychopathology, 1951, 12(1),
45-66. __

Malmo, R., Shagass, C., & Smith, A. Responsiveness in chronic schizo-
phrenia. Journal of Personality, 1951, 16, 359-375.

 

Mednick, S. A learning theory approach to research in schizophrenia.
P§ychological Bulletin, 1958, 66(5), 316-327.

 

Moran, L., & Kimble, J., Jr. Repetitive psychometric measures:
Equating alternate forms. ngchological Reports, 1964, 16,
335-338.

 

Moran, L., & Mefferd, R., Jr. Repetitive psychometric measures.
Psychological Reports, 1959, 6, 269-275.

 

Olton, D., & Noonberg, A. Biofeedback: Clinical applications in
behavioral medicine. Englewood Cliffs, N.J.: Prentice—Hall,
1980.

 

Raskin, M., Johnson, G., & Rondestvedt, J. Chronic anxiety treated
by feedback—induced muscle relaxation. Archives of General
Psychiatry, 1973, 66, 263-267.

Reinking, R., & Kohl, M. Effects of various forms of relaxation
training on physiological and self-report measures of relaxation.
Journal of Consulting and Clinical Psychology, 1975, 66(5),
595-600.

 

Schwartz, G., & Beatty, J. Biofeedback: Theory and research. New
York: Academic Press, 1977.

 

Spohn, H., Lacoursiere, R., Thompson, K., & Coyne, L. Phenothiazine
effects on psychological and psychophysiological dysfunction in
chronic schizophrenics. Archives of General Psychiatry, 1977,
66, 633-644.

 

Taylor, J. A personality scale of manifest anxiety. The Journal of
Abnormal and Social Psychology, 1953, 66(2), 285-290.

 

Taylor, M., & Abrams, R. A critique of the St. Louis psychiatric
research criteria for schizophrenia. American Journal of
Psychiatry, 1975, 166(2), 1276—1280.

Tecce, J., & Cole, J. Psychophysiologic responses of schizophrenics
to drugs. Psychopharmacologia, 1972, 66, 159-200.

147

Tecce, J., & Cole, J. The distraction-arousal hypothesis, CNV, and
schizophrenia. In D. Mostofsky (Ed.), Behavior control and
modification of physiological actiVity. Englewood Cliffs, N.J.:
Prentice-Hall, 1976.

 

 

Thoresen, C., & Anton, J. Intensive experimental research in counsel-
ing. Journal of Counseling Psychology, 1974, 6116), 553-559.

 

Townsend, R., House, H., & Addario, D. A comparison of biofeedback-
mediated relaxation and group therapy in the treatment of chronic
anxiety. American Journal of Psychiatry, 1975, 166, 598-601.

Tuttle, W., & Schottelius, B. Textbook of physiology. St. Louis:
C. V. Mosby Co., 1969.

Whatmore, G. Some neurophysiologic differences between schizophrenia
and depression. American Journal of Psychiatry, 1966, 166,
712-716.

Whatmore, G., & Ellis, R. Some motor aspects of schizophrenia: An
EMG study. American Journal of Psychiatry, 1958, 116, 882-889.

Whatmore, G., & Kohli, D. The phy§igpathology and treatment of
functional disorders. New York: Grune & Stratton, Inc., 1974.

 

 

White, 0. R. A manual for the calculation and use of the median
slope--A technique 0? progress estimation and predictiOn in the
single case. Working Paper #16. Eugene, Oregon: Resource
Center for Handicapped Children, 1972.

 

White, 0. R. The "split middle"--A "quickie" method of trend esti-
mation. Experimental Education Unit, Child Development and
Mental Retardation Center, University of Washington, 1974.

 

 

 

 

"I1114411144