.r n . .
A15 Edng/W(...¥ I.

.c S u4 35:5]:
1.21:3
.6?

{1135);} I:.{.....el.‘ti..a. K'-
33. 1511.§‘2{r!| II refill: 5})-
(fi?! 97!.) t. (:5!- :3..‘.....vul.
uv . a :95). at}: f

 

. . S 1.5.!
r :1 :1 5:3
(1).

‘15) :51
.N! .7! 1):: $1.1. (u... 1| (I). . .rtlix
333(v........ 1:; 9
. 3.1;}... it. )1;

. 5...: (L...

at. J 33.1%; 3:).

.1 SlSVa‘Sis :

. 5:: il<lttiizc
s. )1.

2:}.(1. 4.).»

1r. .3.
$1.... 41:).

1.72:1!5 .1311)... .
1| 6.2;». A 8!!

(r5: {Suit-ye
.9 . a?) 3523
D;
.74vt: ,
A1531) 1.

4((‘npl3l’fnl1l
.11.); .3)...

1.

2.: Ir.A...r 5:025. )
.. .115} . 1 ,1. 1
rt (1.x . r{.f;)‘c..:\.

 

. ‘
a (5:3; {531251)
ritixgyzc 1.}: .(. :54) ;
33x. («.5

m“ a

i .. x, ‘5‘ {.7 x
{a}; (I<r.<(/.<r ‘
:Pia‘iré: . 12.12.)...4
I S»: yaimtci... 121
.J <; 5). )1:

I!“

i

s ..\..fi(7\4lt
(<5). 5);)!

an: r... I 1. 4 . 1.!
u...” n... , V .2... .. . .3. 5):". I...
‘ .1); 13.3.7125 . iii),

 

MICHIGAN STATEU

I!!!“ ”ll ”MIMI!NIH!!!”11111111!!!

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

81 3971
LIBRARY
Michigan State
University
This is to certify that the
thesis entitled "1

The Effects of Preferred Music,
Nonpreferred Music, and Silence on
Anxiety, Relaxation, and Muscle Tension

presented'by

Bonnie Faye Chan

has been accepted towards fulfillment
of the requirements for

Master's Music Therapy

degreein

 

 

MXW

Major professor

,1),th 26,, Wa

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

( ‘1‘ \ 934%4
99

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Affirmative Action/Equal Opportunity Institution

THE EFFECTS OF PREFERRED MUSIC, NONPREFERRED MUSIC,
AND SILENCE ON ANXIETY, RELAXATION,
AND MUSCLE TENSION

BY

Bonnie Faye Chan

A THESIS

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

MASTER OF MUSIC, MUSIC THERAPY

School of Music

1990

+-+
(T)
RD
m

K.

x
r‘.

ABSTRACT

THE EFFECTS OF PREFERRED MUSIC, NONPREFERRED MUSIC,
AND SILENCE ON ANXIETY, RELAXATION,
AND MUSCLE TENSION

BY

Bonnie Faye Chan

The purpose of this study was to determine the effects of
preferred music, nonpreferred music, and silence on measures
of state anxiety, relaxation, and muscle tension. Nine
subjects were selected from a pool of 90 college students
after passing criteria for trait anxiety and musical
experience. Each of the nine subjects was tested
individually for a total of three testing sessions, using one
condition (preferred music, nonpreferred music, or silence)
per session. Pretests-posttests of state anxiety and
relaxation were administered during each condition, and
muscle tension was measured using an electromyogram. Results
were calculated with a repeated measures analysis of variance
and Pearson product-moment correlations. A main effect was
discovered between subjects in the measure of state anxiety,
as well as a difference between conditions on the measure of
relaxation. It was found also that state anxiety and
relaxation were correlated significantly under silence.

State anxiety and BMG muscle tension were related inversely
under preferred music, while other EMG correlations were
asociated with low correlation coefficients. Lastly, a trend
appeared for preferred music to induce lesser anxiety and

greater relaxation than nonpreferred music.

 

ACKNOWLEDGEMENTS

I would like to express my sincere appreciation to my
thesis advisor, Dr. Dale Bartlett, for his patient guidance
throughout this study. Appreciation is given also to
Professor Emeritus Robert Unkefer and Professor Roger
Smeltekop for their support during my years at Michigan State
University.

Special thanks is offered to Professor David Novak of
Lansing Community College who loaned the use of the
electromyogram biofeedback machine, and who gave generously
of his time and advice.

For their assistance in judging categories of the Music
Preference Test, the author gratefully acknowledges Royce
Tevis, John Ware, and Ronnie Wooten.

I would like to thank Jill Hempfling, Sue Johnson, and
Kathryn Saretsky for volunteering their time during the pilot
study. I am grateful also to Kathryn for her support and
assistance in locating subjects.

My special thanks goes to my Michigan family —— Rob, Mary,
Kris, and Erika Cunningham -- for their support and unceasing
faith in me.

This study could not have been accomplished without the
valued assistance and unending support I have received from
my friend, Kris Cunningham. Her patient ear and constant
encouragement gave me the needed strength and motivation to

complete this thesis.

LIST OF
LIST OF

Chapter
I.

II.

TABLE OF CONTENTS

TABLES
FIGURES

THE PROBLEM

Introduction

Need for the Study
Purpose of the Study
Research Questions
Assumptions
Limitations

Overview
REVIEW OF RELATED LITERATURE

Anxiety

Electromyography

M ic and Anxiety

Music and Musculature

L/Music and Psychophysiology

Variables Affecting Psychophysiological
Responses to Music

Measurement of Music Preference

Music Studies with Preferred/
Nonpreferred Music

Summary
L//

Page

vi

(DQmONLDNt—J

.15
19
23

.28

3O
34

37
41

III. METHODOLOGY

Design

Sample

Setting
Instrumentation
Procedure

Pilot Testing
IV. RESULTS

V. CONCLUSION, DISCUSSION,
AND RECOMMENDATIONS

L/Summary
A/anclusions

ZrDiscussion

J/Becommendations
APPENDICES

REFERENCES

43

.43
.43

43
44
46
50

.51

76

76
77

.79

82

.85

.98

LIST OF TABLES

Table Page

Subjects' Preferred and Nonpreferred Music . . . . .47
State Anxiety Difference Scores Under Conditions of

Preferred Music, Nonpreferred Music, and Silence . .51
3. Relaxation Difference Scores Under Conditions of

Preferred Music, Nonpreferred Music, and Silence . .53
4. EMG Difference Scores Under Conditions of

Preferred Music, Nonpreferred Music, and Silence . .54

Analysis of Variance Summary Table . . . . . . . . 55

Analysis of Variance for EMG:
First, Middle, and Last Five-Minute Segments . . . 56

7. Pearson Product-Moment Correlations . . . . . . . . 57

vi

Figure

10.

LIST OF FIGURES

Page

Difference Scores Across Measures Under
Conditions of Preferred Music, Nonpreferred Music,
and Silence: Subject 1

Average EMG Microvolts Under Experimental
Conditions: Subject 1

Difference Scores Across Measures Under
Conditions of Preferred Music, Nonpreferred Music,
and Silence: Subject 2

Average EMG Microvolts Under Experimental
Conditions: Subject 2

Difference Scores Across Measures Under
Conditions of Preferred Music, Nonpreferred Music,
and Silence: Subject 3

Average EMG Microvolts Under Experimental
Conditions: Subject 3

Difference Scores Across Measures Under
Conditions of Preferred Music, Nonpreferred Music,
and Silence: Subject 4

Average EMG Mocrovolts Under Experimental
Conditions: Subject 4

Differences Scores Across Measures Under
Conditions of Preferred Music, Nonpreferred Music,
and Silence: Subject 5

Average EMG Microvolts Under Experimental

Conditions: Subject 5

6O

6O

61

.62

63

.64

66

.66

67

.67

Figure Page

11. Difference Scores Across Measures Under

Conditions of Preferred Music, Nonpreferred Music,

and Silence: Subject 6 . . . . . . . . . . . . . . 69
12. Average EMG Microvolts Under Experimental

Conditions: Subject 6 . . . . . . . . . . . . . . 7O
13. Difference Scores Across Measures Under

Conditions of Preferred Music, Nonpreferred Music,

and Silence: Subject 7 . . . . . . . . . . . . . . 71
14. Average EMG Microvolts Under Experimental

Conditions: Subject 7 . . . . . . . . . . . . . . .71
15. Difference Scores Across Measures Under

Conditions of Preferred Music, Nonpreferred Music,

and Silence: Subject 8 . . . . . . . . . . . . . . 72
16. Average EMG Microvolts Under Experimental

Conditions: Subject 8 . . . . . . . . . . . . . . .73
17. Difference Scores Across Measures Under

Conditions of Preferred Music, Nonpreferred Music,

and Silence: Subject 9 . . . . . . . . . . . . . . 75
18. Average EMG Microvolts Under Experimental
Conditions: Subject 9 . . . . . . . . . . . . . . .75

viii

 

CHAPTER 1
THE PROBLEM

mm.

In contemporary psychology, "anxiety" is used customarily
to denote "a transitory emotional state or condition
characterized by feelings of tension and apprehension and
heightened autonomic nervous system activity" (Spielberger,
1971, p. 24). Stress, as defined by flebster;§_flew_ggllegiate
Dictionary (1977), is "a physical, chemical, or emotional
factor that causes bodily or mental tension and may be a
factor in disease causation." Anxiety and stress have been
known to be major contributors to a variety of medical
disorders. Hanser (1985) identified some of these disorders
that are, in part, due to stress and anxiety: hypertension,
gastrointestinal problems, skin disorders, headaches,
coronary artery disease, female reproductive dysfunction,
respiratory ailments, and other life-threatening diseases.
In the field of psychopathology, high rates of depression and
suicide have been linked to anxiety and stressful life
events. In short, anxiety and stress are important
etiological components of many physical and psychological
problems.

Varied studies seem to indicate a positive correlation
between increases in anxiety and muscle tension elevation
(Goldstein, 1972). Similarly, decreased muscle tension, as
well as decreases in other physiological measures such as
heart rate, blood pressure, and respiration rate, generally

indicates reduced anxiety and stress levels (Jacobson, 1938).

Diversified coping interventions to assist in the reduction
of anxiety and stress levels, as well as in the decrease of
muscle tension, incorporate such measures as diet, drug
therapy, psychotherapy, biofeedback, massage therapy,
relaxation techniques, and activity therapies such as art,
dance, recreational, and occupational therapies. Music
therapy, as one of the activity therapies, demonstrates the
capability to facilitate the expression of, and changes in,
emotion, thereby alleviating stress. Furthermore, music
therapy relaxation techniques can aid in the reduction of

muscle tension.

N§§Q_i91_ih§_fiindx

There have been a number of studies involving research on
a variety of physiological responses to music, such as
studies on heart rate, respiration, blood pressure, galvanic
skin response, and muscle tension. However, Dainow (1977)
maintained that these studies were incomplete, possibly due
to the many methodological issues involved. Important issues
which he discussed included instructions to subjects, volume
of the musical stimuli, subject attention, suppression of
movement due to fear of disturbing electrodes, and
differences between musicians and nonmusicians.
Additionally, Dainow observed that the diversity of
physiological variables and variety of musical stimuli create
a multitude of measurement and interpretation problems.
Increased SOphistication of instrumentation has made
measurement more reliable; however, interpretation problems
remain due to inconsistencies in research methodology.

As examples of research with inconsistent results,

Hodges (1980) summarized the music research related to heart

rate: seven studies found that heart rate decreased in

response to sedative music and increased to stimulative
music; five indicated that heart rate increased in response
to any music (stimulative or sedative); two studies stated
that heart rate changed in response to stimulative or
sedative music, but that the changes were not predictable;
and seven found that music had no effect on heart rate.

Hodges concurred with Dainow that inaccurate
measurements and confounding variables were factors which
contributed to the inconsistent data. Another reason was
that "the definitions of stimulative and sedative music may
be too general and not allow for a clear enough distinction
between two kinds of music" (Hodges, 1980, p. 396).

Gaston (1951) defined stimulative music asjmusig_which

 

 

 

 

_is primarily rhythmical, percussive,_andpcgntgigs detached

m1 uwtq-v—rv u: u- _ .
a». "" ’- . wararpnn.\aun

notes. Sedative music, conversely, lacks rhythmic and

_ “EIWJ‘PL‘M.Wfl‘fi/Iv.c‘i"-Vi v

0". . _
5-. “us“.w- .,

percussive elements and is, instead, music which contains

 

sustained melodic passages. As noted previously, many
different studies have utilized music according to these two
classifications. However, Hanser (1985) advised caution in
selecting musical pieces to represent the entire class of
either stimulative or sedative music. She stated that
untested music may contain elements that might evoke
differential responses in listeners. Taylor (1973) also
suggested that sedative and stimulative music should not be
classified solely on the basis of the characteristics of the
music alone; the response of the listener should be taken
into consideration as well.

The use of precategorized music, such as stimulative and
sedative, lends to the "belief that intersubject responses to
musical and other sensory stimuli are similar, predictable,
and generalizable" (Davis & Thaut, 1989, p. 169). However,
Lacey (1950) found that individuals exhibit highly
idiosyncratic psychophysiological response patterns to

stimuli, an idea supported by Lacey and Lacey (1970) as
expressed in their concept of "response stereotype" or
"response specificity." The idea that each individual
responds uniquely to a given stimulus is also supported by
Harrer and Harrer (1977) who identified three variables which
may determine intrasubject response patterns to music:
a). . .the lability or stability of the autonomic
regulatory processes. This, in turn, is influenced by
constitution (predisposition), age, sex, mode of life,
physical fitness, general state of health, or such
temporary factors as fatigue, drinking alcohol or coffee,
and so on; b) emotional reactivity; and c) attitudes

toward music, the importance of music in the subject's
life. . . (p.202).

Studies on attitudes and preferences toward music appear
to have been reviewed extensively by Kuhn (1981) and Wapnick
(1976). Although numerous studies have researched music
attitudes and preferences, few seem to have measured the
effects of preferred music on psychological experiences
and/or physiological responses. Hodges summarized the
results of music research related to physiological responses,
such as research on heart rate (as mentioned previously),
blood pressure, respiration, galvanic skin responses,
muscular and motor responses, and brain waves. These
studies, however, utilized precategorized music and did not
appear to take into consideration individual preferences.
Similarly, some research studies (Logan & Roberts, 1984;
Miller & Bornstein, 1977; Peretti & Swenson, 1974; Smith &
Morris, 1977; Stoudenmire, 1975) measured the effects of
music on anxiety or relaxation, but also employed
precategorized music.

It appears there is a need to add to the limited amount of
literature on the effects of preferred/nonpreferred music on
psychological experiences, such as anxiety, and physiological

responses, such as muscle tension. The results may shed

further light on the significance of individual responses to
musical stimuli, thereby increasing ways to address the

problems of physical and psychological tension.

W

The purpose of this study was to examine the effects of
preferred music, nonpreferred music, and silence on measures
of state anxiety, relaxation, and muscle tension. Music has
been shown to have an effect on anxiety, relaxation, and
muscle tension; however, review of the literature has not yet
determined what kinds of music will produce what specific
effects.

Nonpreferred music was chosen as an experimental condition
to contrast preferred music. Dainow (1977) recommended to
experimenters that "to promote even greater response,
subjects should select as part of the stimulus material music
that they are familiar with and either like or dislike"

(p. 218). Davis and Thaut (1989) demonstrated that preferred
music significantly decreased state anxiety in a group of
college students, although the physiological data indicated
that "the music aroused and excited rather than soothed
autonomic and muscular activity" (p. 168). It is the attempt
of this study to explore the relationship between preferred
and nonpreferred music, to determine the significance of
correlations between these conditions and silence, and to
examine the relationships between psychological (subjective)
and physiological (objective) measures of anxiety and

relaxation.

W
ousstion 1: Will there be significant differences between
preferred music, nonpreferred music, and
silence on the measure of state anxiety?
Qusstion 2: Will there be significant differences between
preferred music, nonpreferred music, and
silence on the measure of relaxation?
Qusstion 3: Will there be significant differences between
preferred music, nonpreferred music, and
silence on the measure of muscle tension?
Question 4: Will there be significant relationships
between measures of state anxiety, relaxation,
and muscle tension under conditions of
preferred music, nonpreferred music, and

silence?

Assumptions

For the purpose of this study, the following assumptions
were made:

1. The Spielberger State-Trait Anxiety Inventory is a
valid and reliable measure of state and trait anxiety.

2. Subjects responded to the State—Trait Anxiety Inventory
and the Subjective Relaxation Test honestly, in a manner most
consistent with their affective state.

3. The Electromyogram Myosone 409 measures the sum of
numerous action potentials, and accurately displays digital
microvolt readings. It is as technically accurate as surface
electromyogram recordings can be, in comparison to the more
accurate intramuscular recordings (Buchtal, 1957).

4. Subjects responded accurately to the Musical Experience
Questionnaire, honestly reporting the length of musical
training they had received.

5. Subjects responded honestly to the Music Preference
Test, indicating their preference, on a scale of one to
seven, for each of the brief musical excerpts.

6. Of the subjects who were asked to bring their music to
a testing session, those subjects truthfully chose, within
limits of their resources, music which represented their
preferred relaxing music.

7. Subjects reported truthfully that they were not under
the influence of alcohol and/or drugs during the experimental

sessions.

Limitstions

1. Of the 90 subjects who were administered the music
preference test, only nine subjects qualified for the
remainder of the experiment after being screened for musical
training and anxiety. This small number necessitated the
need to label this thesis a "pilot study."

2. Motor responses to music stimuli were not examined,
although they are related closely to muscle tension and
electromyogram research studies.

3. Much literature abounds in music preference research
which examines the influence of various factors on music
preference. Although these variables were briefly commented
upon, it was not the intent of this research paper to explore
the following types of studies in depth: (a) the influence of
musical characteristics on music preference; (b) the
influence of personality types, sex, age, race, aptitude, and
socioeconomic background on music preference; and (c) the
influence of other factors such as familiarity and repetition

on music preference.

v v' w

Chapter II presents a review of related literature. The
first two sections deal separately with anxiety and
electromyography. The latter half reviews the music
research: music and anxiety, music and musculature, music
and psychophysiology, variables affecting psychophysiological
responses to music, the measurement of music preference, and
finally, music studies using preferred/nonpreferred music.

Chapter III discusses the methodology used to study the
effects of music and silence on anxiety, relaxation, and
muscle tension. The design, sample, instrumentation, and
experimental procedures are presented, along with a
description of the Music Preference Test.

Chapter IV presents the data in tables and graphs, as well
as an interpretation of individual results. Chapter V
includes a summary and conclusion of the study, an
interpretive discussion, and recommendations for further
research. The appendix contains the various forms utilized
in this experiment, a list of selections used in the Music

Preference Test, and tables of raw data.

CHAPTER II
REVIEW OF RELATED LITERATURE

Anxiatx

According to Freud (1936, p. 85), anxiety was the
"fundamental phenomenon and the central problem of neurosis."
It was Freud who first recognized the condition of anxiety
and the importance of its role in the etiology of
psychoneurotic and psychosomatic disorders. Research on
anxiety before 1950 was limited due to the "complexity of
anxiety phenomena, the ambiguity and vagueness in theoretical
conceptions of anxiety, the lack of appropriate measuring
'instruments, and ethical problems associated with inducing
anxiety in laboratory settings" (Spielberger, Gorsuch,
Lushene, Vagg, and Jacobs, 1983, p. 1). However, since 1950,
advances have been made in two of these areas: anxiety has
been clarified more clearly as a theoretical construct, and
various scales have been devised to measure anxiety.

The concepts of state and trait anxiety, which were first
introduced by Cattell and Scheier (1961), have helped to

clarify anxiety as a theoretical construct. State anxiety

may be defined as

a transitory emotional state or condition of the human
organism that varies in intensity and fluctuates over
time. This condition is characterized by subjective,
consciously perceived feelings of tension and
apprehension, and activation of the autonomic nervous
system . . . . Trait anxiety refers to relatively stable
individual differences in anxiety proneness, that is, to
differences in the disposition to perceive a wide range of
stimulus situations as dangerous or threatening, and in
the tendency to respond to such threats with A-State
[state anxiety] reactions (Spielberger, 1972, p. 39).

Spielberger et al. (1983) state that persons who are high
in trait anxiety interpret a broader range of situations as
dangerous orthreatening more often than persons with low
trait anxiety. Although physically dangerous situations do

not seem to be threatening differentially to high or low

1O

trait anxiety subjects, a fear of failure seems to be

characteristic of people with high trait anxiety. Sarason

reports:
the bulk of the available findings suggest that high
anxious 35 are affected more detrimentally by motivating
conditions or failure reports than are 88 lower in the
anxiety score distribution. . . . It is interesting to
note that high anxious 83 have been found to be more self-
deprecatory, more self-preoccupied, and generally less

content with themselves than 83 lower in the distribution
of anxiety scores (1960, pp. 404-405).

A valid and reliable scale which measures state and trait
anxiety is Spielberger's State-Trait Anxiety Inventory
(STAI), first published in 1970 and revised in 1983. The
inventory consists of two scales: Trait Anxiety and State
Anxiety. The Trait Anxiety Scale contains 20 questions which
asks subjects to indicate how they feel in general (i.e.,
tendencies toward anxiety). The State Anxiety Scale also
contains 20 questions, but it asks subjects to describe how
they feel at a particular moment in time. Spielberger
correlated the two scales by administering them to college
students during a testing session when they were exposed to
varying amounts of stress. Results showed that the Trait
Anxiety scores remained constant, while the State Anxiety
scores increased during times of greater stress and decreased
under more relaxed conditions, thus corresponding to the
definitions of "relatively stable" trait and "transitory"
state anxiety.

Cognitive and somatic anxiety are two other distinct
concepts of anxiety. Davidson and Schwartz (1976) offer this
example of cognitive anxiety: "a person who is physically
tired and somatically relaxed to lie down, [but] unable to
fall asleep because his 'mind is racing'" (p. 402).
Conversely, somatic anxiety is characterized by bodily
tension and autonomic stress without concomitant cognitive
symptoms. For example, a neophyte meditator who may be able
to induce a calm state of mind might still report muscular

tension and somatic aches and pains throughout his/her body.

11

Davidson and Schwartz conjectured that anxiety may be mode
specific, that is, experienced cognitively or somatically
(physiologically), or a combination of both.

In measuring cognitive anxiety, the scale, questionnaire,
and inventory, such as the State—Trait Anxiety Inventory,
remain the most popular forms of measurement. Somatic
anxiety may be measured in various ways using physiological
indicators to measure such factors as heart rate, blood
pressure, respiration rate, and muscle tension.

Through the 19303 and 19403 the search to associate
specific physiological responses with psychological variables
continued. It was not until the 19503, however, that
emotions could be diffentiated on the basis of physiological
responses. Ax (1953), in his classic study, obtained several
physiological measures from normal subjects under conditions
designed to elicit fear and anger. He found significant
differences in physiological reactions between fear and
anger, differences which suggested that the response patterns
of anger were similar to those produced by injections of
norepinephrine and epinephrine combined and the response
patterns of fear as being similar to injections of
epinephrine. Martin (1961) summarized the results of four
studies and concluded also that, despite some inconsistencies
among the studies, anger was associated with a mixture of
norepinephrine and epinephrine responses, and anxiety was
associated with an epinephrine reaction.

The fundamental question of Ax, which emphasizes the

classical use of physiological reactions by psychologists,

asks: "Can . . . physiological reactions serve as an
emotional . . . indicator during psychological observation?"
(1953, p. 433). Do significant correlations exist between

physiological and psychological measures of anxiety?

Martin (1961) did not appear to believe 30: "Research
gives little ground for optimism that these variables will
correlate very highly, if at all" (p. 243). Schachter seemed

to agree:

12

. precisely the same physiological state --- an
epinephrine-induced state of sympathetic arousal --- can
be manifested as anger, euphoria, amusement, fear, or

no mood or emotion at all. Such results are virtually
incomprehensible if we persist in the assumption of
identity between physiological stress, but they fall
neatly into place if we specify the fashion in which
cognitive and physiological factors interact (1967,

p. 124).

Schacter suggested that, given a state of physiological
arousal, a person will label feelings associated with that
arousal by way of the social interpretations one gives to the
situations in which these feelings are experienced.

Crabbs and Hopper (1980) did not find significant
relationships between physiological (electromyogram and
finger temperature) and self—report (STAI) measures of
anxiety. However, there was a significant positive
relationship of each measure with itself during the three
testing phases of their study. Crabbs and Hopper
hypothesized that their findings supported the research of
Davidson and Schwartz, which suggested that anxiety may be
experienced cognitively and/or somatically. Subjects seemed
to be aware of one type of measurement exclusive of the
other, although other subjects appeared to be aware of both
dimensions.

In the study by Reinking and Kohl (1975), subjects seemed
to be aware of both cognitive and somatic dimensions, at
least during the last part of the experiment. Four groups
were compared with each other to observe the effects of
different kinds of relaxation training: (a) classic
Jacobson-Wolpe instructions, (b) EMG feedback, (c) EMG
feedback plus Jacobson-Wolpe instructions, and (d) EMG
feedback plus a monetary reward. Each group, in addition to
a control group, had their EMG levels recorded and were asked
to report their levels of subjective tension on a scale of l
to 10. Results showed that all groups, except for the
control, indicated increased physiological relaxation over a
total of 15 sessions. The correlations between EMG and

13

self—report measures improved frcmi L = .38 in the first three

experimental sessions to .57 during the last 3 sessions.

Shipman et al. (1964) tested for anxiety and muscle
tension with psychiatric patients under three varying
conditions: (a) "Affect Arousal. The purpose . . . was to
evoke as much intense affect as possible through a stress
interview" (p. 333); (b) Self Control. The patient was
asked to describe how he controlled his feelings and was
encouraged to maintain an attitude of strong self control;
and (c) Neutral Interview. In an attempt to decrease
apprehension, the patient was given advance knowledge of what
was to occur on that day and was asked to avoid any
personally disturbing topics. One difference this study made
was that the patients did not rate their own emotional state,
but were, instead, rated by two independent observers.
Results indicated that muscle tension was found to be similar
in all three conditions, and that the link between state
anxiety and muscle tension elevation did not correlate. On
the contrary, the most anxious patient was shown to have the
least rise above EMG baseline. The authors noted that,
although group averages showed no significant differences,
individuals behaved in varying ways (e.g., some tensed with
anxiety; others went "limp"), emphasizing the importance of
personality differences. .

Leboeuf (1977) incorporated personality differences into
his study by comparing the effects of EMG feedback training
on anxiety in introverts and extroverts; introversion—
extroversion was defined by the Eysenck Personality
Inventory. The introverted/extroverted subjects were chosen
for anxiety only if their anxiety scores fell at least one
half of one standard deviation above the trait anxiety mean
of Spielberger's State-Trait Anxiety Inventory, as published
in the test manual. Additionally, only subjects who
demonstrated some degree of muscle tension, as defined by the
author, were accepted. Although both introverts and

extroverts showed significant decreases in muscle tension,

14

only introverts displayed significant decreases in state
anxiety. Interestingly, 6 out of 16 extroverts reported an
increase in anxiety. For future research, Leboeuf suggested
identifying other techniques for the anxious extrovert that
would be capable of inhibiting anxiety.

Stoudenmire (1972) also compared introverts and extroverts
by testing for the effects of muscle relaxation training on
anxiety. All subjects were screened for introversion/
extroversion and trait anxiety using the same scales and
standards as Leboeuf. Results indicated that no significant
decrease was found in trait anxiety; however, introverts
demonstrated a significant decrease in state anxiety
measures, a finding similar to Leboeuf's study.

Lastly, Goldstein (1964) compared normal women and
chronically anxious women during periods of rest and a white—
noise stimulus. Significant differences were found in muscle
tension between the two groups, with the anxious group
responding with greater muscle activity.

Varying results have been obtained in studies of anxiety.
One explanation for this is based on Lacey's (1950) concept
that individuals evince highly idiosyncratic response
patterns, and that "for a given individual some physiological
measures may be much more sensitive indicators of change in
anxiety level than others" (Martin, 1961, p. 244). Another
reason may be due to the difficulties involved in defining
anxiety itself. Goldstein (1972) noted that researchers have
usually depended upon the diagnoses of experts or upon some
psychological test of anxiety. Goldstein cited Malmstrom's
(1968) work in which two different adjective check lists --
the Zuckerman Affect/Adjective Check List (AACL) and the
Nowlis Mood Adjective Check List (MCL) ——— were used
together. It was found that frontalis muscle tension
correlated positively with the AACL but negatively with the
MCL. Goldstein surmised that, similar to other tests of
anxiety, the AACL and MCL have broached only limited aspects
of the psychophysiological state of anxiety.

 

15

BMW

The term "muscle tension" has been used quite freely,
especially in the older psychophysiological literature
(Hassett, 1978). It has ranged from Jacobson's (1938)
measurements of kneejerks to the recording of muscles used in
different tasks, such as in the task of striking typewriter-
like keys under changing conditions (Morgan, 1916). Other
measures of muscle tension are more direct, including the
assessment of posture of a limb, and the rigidity of a
certain muscle (Goldstein, 1972). Care, therefore, must be
taken when comparing studies.

The electromyogram (EMG) provides a direct measure of
electrical activity associated with muscular contraction.

The EMG actually measures the sum of numerous action
potentials, but does not directly measure muscular
contraction, which is a mechanical event that occurs
immediately after the electrical firing of a group of muscle
fibers (Goldstein, 1972).

What, then, is the physiological basis of an action
potential? Goldstein (1972) and Hassett (1978) provide
concise explanations of muscles and muscle action potentials.
A muscle is a mass of tissue which consists of millions of
individual muscle fibers bound by a sheet of connective
tissue. Muscle fibers function as a group known as a motor
unit. The number of fibers in each motor unit depends upon
the type of muscle. For example, muscles of the eye may have
only 10 fibers in a motor unit, while muscles concerned with
larger motor movements, such as the postural muscles, may
have 3000 muscle fibers in each motor unit. Each motor unit
is innervated by a single nerve fiber (neuron), the cell body
of which is located in the spinal cord for the muscles of the
body and in the brain stem for the muscles of the head.

During the resting state each muscle fiber possesses a
negative intracellular potential which forms the basis of the

electrical occurrence of the action potential. By way of the

16

potassium-sodium pump theory, a cell's selectively permeable
membrane allows potassium and chloride ions to diffuse more
freely than sodium ions through the membrane. Therefore,
there exists a greater concentration of sodium ions outside
the cell, and more potassium ions inside the cell. When
potassium ions diffuse out of the cell, a positive charge is
created outside the cell and an accompanying negative
intracellular charge, thus causing the cell to polarize.

A depolarization, or reversal, of the resting membrane
potential occurs when the cell becomes active and increases
its permeability to sodium, resulting in a temporarily
greater concentration of sodium inside the cell and a
correspondingly lesser amount of potassium. This
depolarization is known as the muscle action potential. The
resting level returns in a millisecond when the membrane
potential becomes internally positive, allowing increased
permeability to potassium and decreased permeability to
sodium.

Excitation of the cell not only results in a
depolarization of the membrane potential but also causes a
wave of contraction to spread over the fiber with continual
velocity. Muscular contraction is a function of the number
of motor units activated and their frequency of firing. In
summary, because muscular contraction is proportional to the
electrical activity of the motor units firing, the EMG
provides an index of muscular contraction or tension.
Quantification of tension is expressed in microvolts; for
example, the lower the microvolt level, the more relaxed the
muscle.

EMG activity is usually detected by electrodes attached to
the skin. Intramuscular electrodes are used also, but
primarily when there is concern with the activity of
individual muscle units. Surface electrodes are utilized
commonly when interest centers on groups of muscles
(Goldstein, 1972). Hassett (1978) notes that a particular

problem with the use of electrodes is electrode polarization,

17

when "electrodes begin to act as miniature batteries
generating a potential of their own" (p. 27). However,
commonly used silver-silver chloride electrodes (electrodes
made of silver and coated with silver chloride) resist
polarization. ‘

Selection of muscle sites is an important variable in the
consideration of recording EMG activity. The frontalis and
forearm muscles have been used commonly in many research
studies as indicators of general muscle tension. However,
are they predictors of overall body tension, or do they
specifically measure tension in their respective muscles?
Goldstein, in reviewing the literature, found muscle tension
to be centered about the limb musculature. This tension,
though, changed under varying conditions. For example, Voas
(1952) found that the trapezius and masseter muscles were
most prominent during stress and frustration, and that the
best indicators of tension during mental work were the arm
muscles. Voas reported that only the frontalis muscle,
however, maintained a high level of reliability throughout a
test period of approximately nine days. Fair (1983)
discovered that the forehead and neck regions are good
indicators of generalized muscle tension, and are "associated
with anxiety, back pain, and tension headaches" (p. 173).
Fair suggested that a common sense approach in the selection
of muscle sites was to ask the subject to report which area
of the body most frequently became tense. However, this
approach would make it difficult to standardize treatment
which is necessary in experimental research. Finally,
Alexander and Smith (1979) seemed skeptical that EMG
measurement from a single site could indicate overall
tension:

the relationship between skeletal muscle tension in
general, let alone at a single site, and relaxation in the
osyohophysiological [sit] sense has yet to be explicated.

It will not do to simply make the assumption that muscle

biofeedback experiments, in which all that is recorded is
EMG from a single site, are studying relaxation in any

18

form other than in the limited sense of the_ohysiolooioal
relaxaLuzLruLa_circumscrxmmiJmuuusegronp,(p. 115).

There are other variables that need to be controlled in
EMG experiments. Age has been the variable most fully
investigated, although Goldstein reported inconsistencies in
the literature in correlating age with muscle tension.
Education, IQ, and sex appear to be unrelated to changes in
muscle tension. Goldstein stated that a few studies which
reported variations between the sexes found differences in
strength to be the causative factor.

Temperature has been found to affect muscle tension.
Goldstein reported that muscular activity was effective in
raising intramuscular temperature, and that "atmospheric
temperature was found to be positively correlated with
general muscle tension factors obtained during rest"

(p. 337). By maintaining a constant room temperature level
and restricting the activity of subjects, she suggested that
it is possible to control for some of the effects of
temperature.

The above variables, i.e., age, sex, education,
temperature, and IQ can be classified as "artifacts."
According to Kallman and Feuerstein (1986),

"artifacts" are defined here as any change in a psycho-

physiological response system that is not attributable to

a psychobiologically relevant stimulus. This includes not

only the traditionally defined sources of artifacts such

as movement or electrical interference but the more subtle
environmental variables that can produce a false-positive

interpretation of the psychophysiological record . .

These artifacts can be divided into two major categories:

artifacts of instrumentation and recording, and artifacts
of interpretation (p. 340).

The subject and the measurement environment are two
sources of instrumentation artifacts. The primary subject
artifact is movement, such as between the subject and
electrode/electrode paste. Even eye blinks may cause changes
in the recording of frontalis muscle tension.

Transient electrical fields in the test room are the main

environmental sources of instrumentation artifact. Sources

19

such as wall outlets, fluorescent light fixtures, and
electromechanical equipment can cause interference, but these
may be eliminated by grounding the subject or by electrically
shielding the test room, as even small electrical signals can
interfere with the recording.

Artifacts of interpretation include variables already
mentioned (sex, age, education, IQ, and temperature), as well
as other variables such as noise, light, odors, tactile
stimulation, and medical conditions, all of which could lead
to false-positive readings. Finally, habituation to a novel
stimulus is an additional factor to consider. Kallman and
Feuerstein define habituation as "a decrease in a
physiological response with repeated stimulus presentations"
(p. 340). Assessment over several sessions was recommended
as a way to ensure that a response was due not only to the

novelty of the stimulus.

MusiansLAnxieix

There have been numerous research studies focusing on the
effects of music on anxiety. Although studies indicate that
music does affect anxiety, the data are inconsistent as to
what kinds of music will produce what specific effects. It
is believed generally that sedative music decreases anxiety,
while stimulative music increases anxiety. Questions still
remain unanswered, however, regarding the effects on anxiety
of: (a) precategorized music such as stimulative—sedative,
happy-sad, calm—exciting; and (b) specific characteristics of
music such as tempo, intensity, pitch, and rhythm. Biller,
Olson, and Breen (1974) utilized happy and sad music to
examine the effects of both listening and participation in
music on state and trait anxiety. The two songs used were
"Sunshine Superman" and "Catch the Wind" by Donavan, rated
independently to be classified as happy and sad,
respectively. College students listened either passively or

actively (played tambourine) or heard no music.

 

20

No differences were reported in trait anxiety, which
supported the definition of trait as measuring relatively
stable emotional differences (Spielberger et al., 1983);
however, subjects who were exposed to the sad song showed
significantly lower state anxiety. On the other hand,
participation in playing the tambourine did not seem to have
a noticeable effect on state anxiety.

State and trait anxiety were measured also in a study by
Stoudenmire (1975) who compared sedative music and muscle
relaxation training. One hundred-eight female college
students were defined as "anxious" as their anxiety scores
fell at least one-half of one standard deviation above the
STAI trait anxiety mean. Pretreatment and posttreatment
scores of the STAI were compared, indicating that both
conditions were effective in lowering state anxiety levels,
but not trait anxiety levels.

In a somewhat different vein, music classified as
stimulative and soothing was used to investigate the effects
of background music upon client-counselor interaction.
Though Prueter and Mezzano (1973) found no significant
differences, there seemed to be a trend for soothing music to
promote more client-counselor and affective interaction than
did stimulating music or a lack of music. The authors
cautioned that their research should be viewed as a pilot
study, however, due to the small number (6 counselors and 18
clients).

The study by Rohner and Miller (1980) also demonstrated a
trend for sedative music to decrease state anxiety more
effectively than stimulating music. One hundred fifty-six
subjects were selected, and defined as "high-anxiety" with
their scores falling above the mean of the normative group on
the administration of the Eight State Questionnaire. The
subjects were assigned randomly to one of four musical
conditions: (a) familiar-stimulating, (b) familiar-sedative,

(c) unfamiliar-stimulating, or (d) unfamiliar-sedative.

21

A no-music condition made up the control group. No
significant differences were found, and the musical
conditions did not reduce anxiety any more than the condition
of silence, although sedative music tended to lower anxiety
more effectively than stimulating music, as stated above.
Familiarity with the music had little effect upon the
subjects.

Logan and Roberts (1984) utilized two different kinds of
music which have been claimed to induce a state of
relaxation: (a) new age music by Steven Halpern entitled
"Comfort Zone," and (b) "superlearning" music, a learning
capacity enhancing music developed by Superlearning
Corporation. The authors compared the effects of these
musical types and silence on subjective tension level. On
eight separate occasions throughout a relaxation training
session, twenty-five college students were required to record
their present tension level on a predetermined scale
developed by the authors. Logan and Roberts claimed that the
procedure of testing on eight separate occasions would help
to record more accurately a subject's momentary tension level
and would minimize problems related to tension level changes
over long periods. No statistically significant differences
were found in tension levels produced by the two kinds of
music, but the no music group had the lowest overall tension
level.

Smith and Morris (1976) tested 66 college students under
conditions of stimulative music, sedative music, or no music
during a psychology examination. The examination consisted
of five sections, and one of the five following types of
music was played during each section of the test: classical,
jazz and blues, country/bluegrass, easy listening, and rock.
Subjects rated their worry, ability to concentrate,
emotionality (physiological-affective arousal), expectancy of
performance, and like-dislike of the music. It was found

that stimulative music significantly increased worry and

22

emotionality as compared with sedative music. Preferred
music did not seem to have an effect on a subject's emotional
state.

In a later study, Smith and Morris (1977) differentiated
between music majors and psychology majors when testing for
the effects of stimulative and sedative music on anxiety,
concentration, and performance. Subjects were assigned
randomly to five music groups (similar to their 1976 study)
while performing a test. Concentration, worry, and
emotionality were again rated by the subjects, in addition to
their like or dislike of the music. Results were varied.
Worry increased more under stimulative music conditions than
sedative music, but neither condition differed significantly
from no music. As expected, concentration was higher with no
music, and higher for sedative than for stimulative music.
Differences, which reached borderline significance, were
found between psychology majors and music majors. Music
majors found stimulative music more distracting, while
psychology majors found both stimulative and sedative music
distracting. Only easy listening music produced better
concentration than no music for psychology majors. Though
all subjects liked sedative music better than stimulative,
music majors liked all types of music better than psychology
majors. Smith and Morris concluded that

the effects of music are to be understood in terms of

cognitive processes such as worry, expectancy, and

concentration, rather than primarily on the basis of the
arousal or reduction of physiological-affective responses
to musical stimuli. Any person variable which affects

these cognitive processes. . . should affect one's anxiety
and performance under music conditions (p. 1052).

Smith and Morris recommended the use of specific kinds of
music in connection with musical preferences, as their
correlations of the like/dislike variable indicated.
Additionally, differences between psychology majors and music
majors were recommended as deserving further research
attention.

23

Peretti and Swenson (1974) also compared music majors and
nonmusic majors, and males and females to determine the
effects of preselected music on anxiety, as measured by the
galvanic skin response (GSR). Fifty male music majors and 50
female music majors, and an equal number of nonmusic majors
of each sex were administered a pencil—maze task while GSR
was being recorded. Music was later introduced into the
experimental situation under similar conditions. Results
indicated that music majors (both sexes) demonstrated a
greater decrease in anxiety levels than nonmusic majors.
Furthermore, female music and nonmusic majors showed greater
changes as compared with males of both groups, and female
music majors had a greater decrease in anxiety when compared
with female nonmusic majors. The authors hypothesized that
the greater decrease in anxiety of female music majors may
have resulted from tendencies to be more influenced by the
music, or from more unstable or flexible anxiety states.

Fisher and Greenberg (1972) also explored the effects of
gender by employing college females to listen to either
exciting, calm, or no music, and then to indicate their mood
and rate the music. Several personality scales, including a
measure of femininity (as measured by the 1964 Gough
California Inventory), were taken during the music listening.
The exciting music produced significantly higher anxiety and
aggression scores, and calm music produced the least anxiety.
Interestingly, it was found that femininity was related
inversely to the amount of change observed, but no
relationship was found between other personality measures and

the effects of music on mood change.

WW
Sears (1952, 1958, 1960) was instrumental in stimulating
interest in the relationship between music and postural
change/muscle tension. His 1952 study measured the effects

of stimulative music and sedative music on posture. While

24

the music played, 12 subjects were photographed through a
one-way mirror, providing evidence of the postural angle, or
erectness, of a subject, and also indicating muscular
tension. Sear's findings reported that musical stimuli can
influence posture significantly.

In another study, Sears (1958, 1960) utilized the
electromyograph and measured the effects of stimulative and
sedative music on degree of muscle tension, comparing
musicians and nonmusicians, males and females. Subjects'
tension levels were recorded from their forearm flexor and
extensor muscles. Stimulative music was played if a
subject's degree of tension was low, and sedative music was
played when a subject's tension was high. Results indicated
that sedative music was effective in reducing tension nearly
100% of the time, although stimulative music was less
effective in increasing tension. Sears' findings that
changes were produced more effectively in females correlate
with similar findings of Peretti and Swenson (1974).

Preferred music was used as feedback which was contingent
on maintaining low frontalis muscle tension levels in a 1974
study by Epstein, Hersen, and Hemphill. This contingent
preferred music feedback was the treatment for a 39-year-old
male who was plagued by chronic tension headaches. Frequency
and intensity of the headaches, as well as medication needs,
were evaluated by this male on an inpatient and, later,
outpatient basis. Decreased headaches and lowered medication
levels were observed during the feedback phase, but the
headaches and medication levels increased after feedback was
terminated. The authors surmised that "feedback was not a
sufficient condition to produce long-lasting clinical
effects" (p. 63) and that more controlled studies were needed
to determine the necessary elements that would lead to
longer—lasting effects.

Harrer and Harrer (1977) have observed that music
listening may lead to changes in autonomic responses, even

when sounds are not consciously perceived, as with incidental

 

25

music for films, and functional music in places like shopping
malls, elevators, and factories. Harrer and Harrer
graphically demonstrated an increase in the number and
amplitude of muscle action potentials as a subject listened
to music. They noted also that qualitative differences exist
between various muscles. For example, greater frontalis
muscle activity was found during an arithmetical task, while
muscle action potentials of the legs increased sharply during
dance music. Another interesting finding reported
fluctuations of muscle activity as a subject listened to
Bach's WW4. It is intriguing that the
changes occurred at the same passages and were reproducible
by repeat performances. Lastly, Harrer and Harrer showed
that lullabies decreased and marches increased muscular hand
strength as measured by an ergometer.

Prager-Decker (1978) compared the influence of four
relaxation techniques (music listening skills, progressive
muscle relaxation -- PMR, EMG biofeedback, and EMG
facilitated PMR) on EMG muscle tension levels. Eighty-one
college-aged males were categorized separately as either
internally-controlled or externally-controlled individuals.
Seven relaxation training sessions were administered,
followed by exposure to six repetitions of an industrial
accident film. Biofeedback was shown to reduce EMG levels
significantly more than music or PMR training. An
interesting finding was that externally controlled subjects
reduced their EMG levels more and faster than did the
internally controlled subjects. In Leboeuf's (1977) study,
both introverts and extroverts significantly decreased their
muscle tension with the use of EMG biofeedback training.

Only introverts, however, displayed a significant reduction
in state anxiety. These examples seem to typify the
disparity between physiological and psychological measures of
anxiety.

—/ Scartelli (1982, 1984) has pursued research actively on

the effects of EMG feedback and sedative music on relaxation.

26

In his 1982 study, he compared EMG biofeedback training with
biofeedback training plus a sedative instrumental music
background in six spastic cerebral palsied adults (three in
each group). The dependent variable was the amount of
tension in the finger extensor muscles. Results strongly
favored the music condition which showed a 65% mean decrease
of muscle tension, in contrast to a 32.5% mean decrease in
the other group. Scartelli cautioned, however, that definite
conclusions cannot be drawn because of the small sample size.

Scartelli added a third condition to his 1984 experiment:
a sedative music only condition. Thirty music majors were
assigned randomly to one of the three conditions, with the
degree of frontalis muscle tension acting as the dependent
variable. Precautions were taken prior to the experiment as
each subject underwent a brief interview to determine whether
any physical discomfort (e.g., headache) was being
experienced that might contribute to an unusually high
tension reading. In addition, an EMG measurement was taken
prior to the experiment to, again, safeguard against
abnormally low or high muscle tension levels. Results
indicated that the EMG.biofeedback group did not obtain a
significant mean decrease in EMG levels, which the other
groups obtained. Although no significant differences were
demonstrated among the three conditions, the biofeedback/
sedative music condition seemed to be the most effective in
reducing frontalis muscle tension.

In a third study, Scartelli and Borling (1986) approached
the use of EMG biofeedback training with sedative music in a
different vein. The biofeedback training and music were
presented sequentially instead of simultaneously, as
Scartelli's previous two studies had done. Three conditions
were examined for their effects on frontalis muscle tension:
EMG biofeedback with simultaneous presentation of sedative
music; sedative music followed by EMG biofeedback training;
and EMG biofeedback followed by the sedative music. Although

significant differences were not obtained, the sequenced

27

application of biofeedback training and sedative music was
found to have a more dramatic reduction in microvolt levels.
Also, sequenced presentations appeared relatively more
productive than simultaneous applications of the music and
biofeedback training. During the simultaneous condition,
confusion was reported by subjects (in the present study and
in Scartelli's 1984 study) of having to decide to which
stimulus to attend primarily —— the music or the auditory
biofeedback.

Pollack (1988) explored the effects of stimulative and
sedative classical music on the muscle tension, or tone, of
spastic cerebral palsied children. Eight children were
treated individually to eight 5-minute music listening
periods, including one no-music condition. Muscle tone was
measured not with an electromyogram, but by viewing
videotapes of pretests-posttests of resting posture and
performance of an arm motion task in each session. An
overall decrease in muscle tone was shown, although
individual differences revealed varying effects. Stimulative
and sedative music seemed equally effective in decreasing
muscle tone. Pollack concluded that music preference was
influential in inducing a variety of responses to the music,
as observed through affect, verbal expressions, and behavior
during the listening periods.

" A final study which utilized music in determining
relaxation/tension was done by Kibler and Rider (1983).
Seventy-six subjects were divided into three-groups which
received one of these conditions: (a) sedative music (M);
(b) progressive muscle relaxation (PMR); or (c) music plus
progressive muscle relaxation (M + PMR). Sedative music
consisted of two works by Debussy -- _£toluoo_to_ths
Aftotnoon_of_a_fiouu and Clouds. Finger temperature, or
vasoconstriction, was used as a stress indicator. All
conditions were found to decrease finger temperature
significantly; however, no condition was more significant
than the other, although the M + PMR group's mean increase

28

was greater than the mean increases of the other two groups.
This finding supports the related research literature which
demonstrates that the use of music in conjunction with other
relaxation techniques enhances relaxation more than either

music or other technique used alone.

W

In Chapter I, it was noted by Hodges (1980) and Dainow
(1977) that the incredible diversity of physiological
variables, inaccurate measurements, and lack of
methodological control contributed to inconsistent data
regarding the effects of music on psychological and
physiological measures of anxiety and relaxation. Standley
(1986) appeared to review the literature extensively and
found 81 studies exploring the effects of music or sound in
medical and dental treatment. Though the trend was usually
that sedative music decreased physiological responses and
stimulative music heightened responses, Standley emphasized
that statistically significant changes were not always
demonstrated and that varying individual responses occurred.
Dainow contended that since a greater number of research
studies opposed the relationship between subjective and
objective data than supported it, "it must be concluded that
little if any correlation exists between the subjective and
objective responses to music" (p. 217).

Barger (1979) found that sedative music (Bach's Ait_on_o
fi_§ttino) produced significant differences in a self-report
measure of arousal between conditions of music and silence,
and between conditions of music and music plus verbal
suggestions of relaxation. None of the conditions, however,
were more effective than silence in reducing heart rate.

Discrepancies were found also between subjective and
objective measures of tension in Jellison's (1975) study.
"Calming" music (Bach's Ait_on_o_§_Stting) and "exciting"
music (Dvorak's Symphony No. 2,4th movement) were compared

29

with each other and with conditions of white noise and
silence. No significant differences were found in
physiological measures, but both types of music produced
significant decreases in state anxiety, while white noise
showed significant increases in state anxiety.

The contention that subjective and objective responses are
in agreement was supported by Miller and Bornstein (1977) who
found that subjective and objective measures of anxiety
correlated positively with each other. Eighty subjects were
assigned randomly to one of these 30-minute relaxation
procedures: (a) progressive relaxation, (b) muscle and mental
relaxation, (c) mental relaxation, (d) self-relaxation (no
instructions on how to relax), plus classical music paired
with each of the foregoing conditions. Relaxation was
assessed with self-report anxiety scales and EMG recordings.
Significant changes under all conditions were found in the
direction of relaxation on both the EMG measurement and the
anxiety scales. Music was not shown to add to the overall
effectivness of treatment, but the authors suggested that the
inclusion of more relaxation sessions may have made a
difference.

Rider, Floyd, and Kirkpatrick (1985) seemed to take their
study a step further by measuring not just the "typical"
physiological parameters such as heart rate, blood pressure,
or muscle tension, but the adrenal corticosteroids, or
"stress hormones." The authors stated that several of these
steroids are released from the adrenal cortex during stress.
The effects of music (music that was used in Kibler and
Rider's 1983 study), progressive muscle relaxation, and
guided imagery in a taped induction were tested on 12 female
nurses. During an initial briefing period, nurses were
administered a series of psychological tests. A tendency was
demonstrated for high trait anxiety nurses to be affected
more positively by the tape. Other results included a
significant decline in circadian amplitude and a significant

increase in degree of corticosteroid rhythm entrainment. The

30

mean corticosteroid level also decreased during the listening
periods, although nonsignificantly. The authors implied
that a technique exists to change body chemistry positively
by utilizing music, and recommended further exploration of

the role of music in facilitating relaxation.

H . 1] EEE I' E 1 1 . J . 1
WW

When a particular piece of music causes the heart rates of

subjects to increase, what does this mean? It means only

that the heart rates increased. It is not known whether

the heart rates increased because the subjects enjoyed the

music, were excited by it, were made anxious by it. . .,

or because of any other unknown reason (Hodges, 1980,
p. 396).

Hodges' keen observation illustrates the difficulties
inherent in research studies, studies which sometimes appear
overwhelmed by the immensity of variables and individual
differences. Lacey's (1950) concept of "individual
stereotypy" exemplifies the need to consider individual
differences and preferences, especially in design and
interpretation phases of an experiment. Harrer and Harrer
(1977) stressed that

the system of maximal response. . . depends mainly on

(a) the character of the subject's individual autonomic

response. In some persons psychological stimuli such as

stress give rise to respiratory changes predominantly,
whereas in others marked circulatory. . . alterations are

elicited by the same type of stimuli; (b) the type of
music which is being played (p. 204).

The "type of music" could include music categorized as
stimulative-sedative, happy-sad, exciting-calm, or other,
more typical classifications such as jazz, rock, or
classical. Characteristics of the music, such as volume,
tempo, rhythm, and pitch also deserve attention.

Hassett (1978) concisely summarized the need for control

of variables in psychophysiological experiments: "The total

31

pattern of physiological response to a given situation is a
function both of the situation and of the individual"

(p. 154). Careful consideration shown for individual
characteristics and other variables would ensure more
controlled research studies.

DeJong, Van Mourik, and Schellekens (1973) researched the
effects of tempo on various physiological parameters (heart
rate, respiration rate, skin conductance, and galvanic skin
response). Forty-five subjects were divided into three
groups: (A1) amateur musicians; (A2) nonmusicians; (A3)
students from the Amsterdam Conservatory of Music. The
musical stimuli consisted of three pieces of different tempi
(fast, medium, slow) within each style period (Baroque,
Romantic, Modern). Subjects were asked to rate the music
from "very beautiful" to "very ugly" on a scale of one to
seven. Varied results indicated that: (a) heart rate was
faster for group A3 than for A1; (b) heart and respiration
rates (HR & RR) were faster for fast music, but RR was faster
for slow than for medium music; (c) HR was faster during
beautiful music for groups A1 and A3; (d) RR was faster
during beautiful music, and skin conductance (SC) and
galvanic skin response (GSR) were higher during beautiful
music; (e) group A2 showed significant HR decelerations,
accompanied by increases in RR and SC; (f) group A3 had a
significant HR acceleration during the beautiful music; and
(9) group A1 showed a HR deceleration during the "ugly"
music. In summary, different tempi were found to produce
highly significant differences for HR, RR, SC, and GSR.

Dainow (1977) reviewed the literature on the effects of
volume and discovered that the intensity and sometimes
duration of reactions corresponded strongly with the loudness
of the music. Wilson and Aiken (1977) tested 52 nonmusic
majors for the effects of intensity level of rock music upon
physiological (GSR, heart rate, & respiration rate) and
psychological (an adjective check list) responses.

Decreased GSR and heart rate, and increased respiration rate

32

indicated a response associated with arousal and attention to
the music. However, intensity level only slightly affected
physiological responses, possibly because the two decibel
levels (79 & 95 decibels) were not sufficiently
differentiated. Conclusions were limited to those whose
preference was hard rock music.

Abeles (1980) cited the work of a few authors who reported
that repetitions of musical selections increased preference.
Bartlett (1973) found that greater awareness of musical
structure induced positive affective response from repeated
listenings of Schubert's Eifth_§ymnhony in subjects whose
most preferred music was not classical. Upon comparing the
first and later sessions these subjects demonstrated a
decline in preference for the initially most-liked music.

Familiarity of music, a subject closely akin to music
preference, is another variable which has been explored in
research studies. In the experiment by Fontaine and Schwalm
(1979), 35 subjects were instructed to perform a vigilance
task under conditions of familiar rock, familiar easy-
listening, unfamiliar rock, unfamiliar easy-listening, and no
music. Though type of music had no effect, familiar music
produced significantly higher levels of arousal (as measured
by heart rate) than unfamiliar music. Concurrently, subjects
who listened to the unfamiliar music showed significantly
higher arousal levels than subjects who listened to no music.
In contrast, Jacobson (1956) and Rohner and Miller (1980)
found that the familiarity or unfamiliarity of music had
negligible effects upon anxiety.

Keston and Pinto (1955) and Abeles (1980) concluded that
gender is not likely to be a factor in the influence of
music preference. However, Sears (1960) and Peretti and
Swenson (1974) found that females demonstrated greater
changes than males in muscle tension and anxiety,
respectively.

Abeles also reviewed the literature on the effects of

musical training and found that training influences music

33

preference, specifically by the length of time spent in
musical participation. Scartelli and Borling's (1986) study
seemed to be one of the few which specified training
restrictions for subjects: "2 years or less [of] formal
music study and no formal music activity participation

in the previous 3 years" (p. 159). The limitations were
applied in an attempt to utilize subjects who would process
the music in a nonanalytical manner. Bradley (1971) measured
the effects of analytical training versus repetition on the
music preferences of seventh graders, and also showed that
training significantly influenced music preference. Landreth
(1974) concluded in her study that training, or "increased
knowledge and understanding of a musical score" (p. 12),
affected heart rate response to music. Nonmusicians were
divided into groups for the experiment, and physiological
recordings were taken while subjects listened to Beethoven's
Symphony No. 5. Significant differences were found between
the group who listened each week to an audiotutorial tape of
Beethoven's Symphony No. S and the control group who
participated in a general music class, demonstrating that
training may be an important variable in the influence of
physiological responses to music.

As an adjunct to research on the effects of musical
training, differences between musicians and nonmusicians have
been the focus of a number of studies. Dainow (1977) defined
the term "musicians" as "those with greater knowledge of
experience, whether playing music or listening to music"

(p. 217). Lack of a standard definition has probably
contributed to the discrepant findings which Hodges (1980)
noted: 14 studies indicated significant differences between
musicians and nonmusicians, while 8 studies showed no
differences. In studies already reviewed, DeJong et al.
(1973), Peretti and Swenson (1974), Sears (1960), and Smith
and Morris (1977) found differences between musicians and

nonmusicians.

34

Abeles (1980) concluded that there seems to be
"insufficient evidence to clearly identify interactions"
(p.133) between all of the above variables and psychological/
physiological responses to music. Though the effects of these
factors appear unknown and warrants further research,
necessary precautions should still be exercised in order to

control for possible confounding variables.

W
The measurement of attitudes in music research has been
concerned with constructs such as preference, attitude,
opinion, and taste. In attempting to systematize the
terminology, Kuhn (1981) defined each precisely:

Attitudes are positive or negative feelings associated
with psychological objects; attitudes are expressed
verbally as opinions, and behaviorally as preferences.
Because attitudes. . . are predispositions. . . , they are
not directly observable or measurable.

Taste has been shown to be a product of prior training,
and it represents a learned disposition to approach or
avoid certain generic musical styles.

An opinion is a verbal expression of attitude made in
the absence of a stimulus object and without contextual or
situational referents being given.

Preference is the act of choosing, esteeming, or giving
advantage to one thing over another. . . (pp. 4-6).

Although Kuhn states that an opinion "is a verbal expression
of attitude," "preference" will be defined here as the
verbal, written, and behavioral expression of attitude, for
the purposes of this research paper.

How are music attitudes measured? Kuhn provides a
comprehensive review of various kinds of measurement
techniques used in music, and cites studies which
incorporate each technique. The most common type of
measurement used in music preference research is the self-
report mode (LeBlanc, 1984). In this mode, verbal or written
responses are required of subjects to honestly report their

feelings toward a stimulus. Self-report measures include

35

open-ended questions, paired comparisons, multiple choice
scales, ranked preferences, rating scales, pictographic
scales, summated ratings, and semantic differentials (Kuhn,
1981). Construction of these scales as valid forms of
measurement appears to be a formidable task:
It is difficult to design and interpret measures of
attitude and interest . . . . Reliability and validity
are exceedingly difficult to determine, and equivalent
forms are virtually impossible to construct. . . . There
are no right or wrong answers. The subject can often
sense what the test writer is seeking and can modify his
responses accordingly if he wishes. The examiner must

depend upon the examinee to be truthful. Curiously,
however, the item, "I have never been in trouble with the

law " on the Cal1forn1a_Rsxchelcgical_1nyentorx is marked

"true" by 20 per cent of the prison inmates tested
(Lehman, 1968, pp. 78-79).

Regardless of the difficulties inherent in self-report modes,
these measures have been used extensively in a number of
music preference studies. Kuhn commented that "if a
researcher is using a dependent measure of opinion or
preference for groups of subjects, then a self-report rating
measure is the most efficient and, in all likelihood, is
quite adequate" (p. 14).

The measurement of observed behavior on music preference
includes observational measures, single stimulus listening
time, reward value, multiple stimulus listening time, and
manipulative response measures, as categorized by Kuhn
(1981). According to LeBlanc (1984), two popular recording
instruments in the measure of behavioral responses include
the Operant Music Listening Recorder (OMLR) used by
R. Douglas Greer and associates and the Music Selection
Recorder (MSR) used by Terry L. Kuhn and associates. With
these instruments, simultaneously available musical stimuli
are offered on different channels, and music preference is
considered analagous to the time spent listening to the
stimuli.

The Music Selection Recorder was utilized in a study by
Kuhn, Sims, and Shehan (1982). Fifty-five college students

36

rated three music selections on a like-dislike scale; these
three selections were then made available on different
channels of the MSR. Results indicated a low correlation
between degree of liking and listening time. Although some
subjects reported a high degree of preference for a
particular piece, the reasons they gave for not listening to
that selection included: "boredom, intrigue with new style,
imagery generated by music, liking for all three selections
with no preference, dislike for all the selections with no
preference, [and] curiousity about the nonpreferred
selections. . ." (p. 187).

This study points to the discrepancy between attitude and
behavior. Attitudes are of value to the extent in which they
predict observable behavior. However, Flowers (1981) noted
that the majority of studies have not found a direct nor
obvious connection between the two indices. Berlyne (1974)
emphasized the need for further research in this area:

It can hardly be reiterated too often that we need

extensive study of ways in which nonverbal behavior can be

predicted from verbal judgements of the same or other
subjects. Until such studies have been completed, totally
unwarranted assumptions on this point are bound to be
rife. Specialists in other areas besides psychological
aesthetics, notably the social psychology of attitudes and

market research, have assumed far too freely that what
people do is in line with what they say (p. 328).

Kuhn remarked that more highly controlled experiments,
advanced measuring techniques, and increased complexity of
statistical analyses "should make it possible to demonstrate
a cause—and-effect explanation of music attitude acquisition
in the future" (p. 15). In the meantime, LeBlanc (1984)
recommended the use of more than one kind of data, e.g.,
written and behavioral, so that each may serve as cross-

validation checks for the other.

37

It appears that the majority of studies which researched
the effects of music on psychological experiences and/or
physiological responses have utilized some form of
precategorized music, while relatively few have used the
subject's preferred music or incorporated a like-dislike
variable into their research design. Some of these studies
which used preferred music or a like-dislike variable have
already been reviewed (Bartlett, 1973; Epstein, Hersen,

& Hemphill, 1974; Smith & Morris, 1976; Smith & Morris, 1977;
Wilson & Aiken, 1977). Additional studies will be reviewed
here in the remaining section of this chapter.

In detailing the use of music therapy with burn patients,
Christenberry (1979) suggested that familiar or preferred
music may be useful in reorienting a burn patient after the
initial trauma. A familiar song may help to decrease
confusion and disorientation and increase order in the
patient's life through familiarity and repetition. The
nonthreatening quality of the music could also aid in
restoring a sense of security, thus furthering reorientation.
Secondly, a burn patient may have difficulty accepting
disfigurement and/or loss of function of a body part, and may
manifest symptoms of low self-esteem. The use of music which
a patient prefers and enjoys could be structured into a
session to ensure successful, pleasurable experiences as a
means of increasing self-esteem.

Predetermined musical selections were used in a study by
Ries (1969) in which subjects (extroverts and introverts)
were asked to rate the music on a five-point like-dislike
scale and on a five-point scale ranging from Very Much Effect
to No Effect. Galvanic skin response and respiratory
reactions were recorded as physiological data for each
subject. Although no correlation was found between GSR and
the like-dislike scale, a correlation was discovered between
GSR and the effect the music had on subjects, specifically

extroverts. Also, the more a subject liked the music, the

38

deeper her/his breathing became, indicating a significant
correlation between music preference and breathing amplitude.

Rider (1985) compared five taped music conditions plus two
control conditions to determine their effects on the EMG and
pain levels of 23 spinal cord injury patients. Five of the
music conditions were preceded by nine minutes of muscle
relaxation instructions followed by a one-minute instruction
to use the music as a guide towards pain relief. Two of the
five music types were Steve Reich's "minimalistic" music; two
pieces were "impressionistic": Debussy's Eroluoo_to_tho
amiotnoon_ofi_o_fisun and Pat Metheny's "If I Could." The
fifth condition, "entrainment" music, consisted of
synthesized and acoustic guitar music "which exhibited a
definite mood shift from unpleasant to pleasant. . ."

(p. 186). Each of the musical selections were scaled on a
self-report music preference questionnaire. A no music
control contained 20 minutes of muscle relaxation/pain relief
imagery, and a no relaxation/no imagery control consisted of
20 minutes of a subject's most preferred music.

Results showed that all conditions, except the music of
Reich and Debussy, were found to have significantly lower EMG
levels, but the entrainment condition was the most effective
in lowering both EMG and pain levels. All conditions were
more effective than preferred music in lowering pain.
However, the comparison between preferred music and the other
taped conditions might be more valid if preferred music were
paired with the relaxation/imagery induction. Also of
interest was the finding that the two loost preferred music
types, the minimalistic music and the entrainment music, were
the most effective in lowering pain levels.

Curtis (1986) also examined the influence of music on pain
relief in terminally ill cancer patients during the last six
months of their lives. Three conditions were compared:

(a) no intervention, (b) background hospital sounds, and
(c) a lS-minute tape of calm, preferred, instrumental music.

Statistically significant differences were not found,

39

although an analysis of individual responses suggested that
the music may have been more effective than the nonmusic
conditions.

The study by Hanser, Larson, and O'Connell (1983)
demonstrated that preferred music reduced pain during labor
in all seven expectant mothers who participated in the
experiment. An individualized music program was developed by
the music therapist:

Selections were based on the mother's preferences and on

observations of the tempo or pace of her breathing.

Musical excerpts of gradually faster tempi were recorded

to correspond with the mother's breathing rates. .

During the course of labor, the music therapist observed

the mother's breathing tempo and played the corresponding
recorded music on a portable cassette player (pp. 52—53).

Despite the study's impressive results in reducing pain in
100% of the mothers, the authors cautioned that the
experiment represented only an initial attempt in examining
the role of music in the birth process with its small sample
and lack of objective subject samplings.

Stratton and Zalanowski (1984) compared the effects of
five types of music (soothing classical, stimulating
classical, romantic, atonal, and Muzak arrangements),
silence, and the degree of liking upon self-reported
relaxation. Thirty—six college students were asked to rate
their liking for the music on a scale of 1-10. It was found
that the degree of liking was the factor most closely related
to relaxation. Subjects with the greatest degree of liking
were found significantly to be more relaxed than subjects who
liked the music the least, and subjects who disliked the
music were significantly less relaxed than the subjects who
received no music. However, subjects who liked the music the
most did not report greater relaxation than the subjects with
no music. The authors stressed that the subjects were not
given an opportunity to choose the music they heard, and

recommended further research in this area.

4O

In_measuring the influence of preferred music on anxiety,
relaxation, and physiological responses, Davis and Thaut
(1989) allowed the subjects to provide the music to which
they most preferred to relax. Eighteen college-aged
nonmusicians were individually tested three times as they
listened to their most preferred relaxing music. The State
Anxiety Inventory (Spielberger et al., 1983) and a
seven-point Likert relaxation scale were administered before
and after each testing session, while physiological
indicators measured heart rate, vascular constriction, muscle
tension, and finger skin temperature. Results indicated that
relaxation increased while anxiety, as measured subjectively,
decreased. Physiological data did not indicate decreased
anxiety, however. It was shown that autonomic and muscular
activity were aroused and excited rather than soothed by
preferred music. This seeming inconsistency between
psychological and physiological measures may be explained by
Berlyne's (1971) theory of hedonic value and arousal:

"First, . . . when arousal approaches the upper extreme, a
decrease to a lower arousal level is pleasurable and
rewarding. Second, a limited rise in arousal, which is not
enough to drive arousal up into the unpleasant range, can
apparently be pleasurable" (p. 82). Davis and Thaut
suggested that the increased relaxation and decreased anxiety
were facilitated by increases in physiological arousal which

accompanied the enjoyment of the music.

 

41

Summary

Anxiety has been measured in both cognitive
(psychological) and somatic (physiological) terms. The most
popular forms of measuring cognitive anxiety are scales,
questionnaires, and inventories, such as the State-Trait
Anxiety Inventory. The concepts of state and trait anxiety,
defined as "transitory" and "relatively stable" anxiety,
respectively, have been incorporated into many research
studies which measure cognitive anxiety. Somatic anxiety can
be measured using various physiological indicators to measure
such factors as heart rate, galvanic skin response,
respiration rate, and muscle tension. A widely used
instrument which measures the degree of muscle tension, or
more precisely, the sum of numerous action potentials, is the
electromyograph (EMG). An issue of debate in the use of the
EMG is whether it measures overall body tension or just the
muscle site to which the electrode is attached. Also, a
number of variables, such as age, gender, temperature, and
other instrumentation and interpretation artifacts, need to
be controlled in EMG experiments.

There appears to exist a lack of correlation between
psychological and physiological measures of anxiety. One
reason may be due to the difficulties in defining anxiety
itself. Another explanation may be based on the concept that
individuals respond in highly idiosyncratic patterns, and
that changes in anxiety levels could be indicated more
sensitively with differing physiological measures, depending
on the individual.

Research on the effects of music on psychophysiological
measures of anxiety or tension have usually employed
precategorized music, such as stimulative-sedative, happy-
sad, or calm—exciting. On the whole, sedative music seemed
to decrease state anxiety and muscle tension while
stimulative music appeared to heighten such responses. The
results of such studies, however, were not always

significantly different and varying individual responses

42

occurred. Many variables, which need to be considered and
controlled, could affect the outcome of these studies; these
variables include, but are not limited to, tempo, volume,
repetition, familiarity, gender, and training.

Music preference is measured commonly using the verbal or
written self—report mode such as multiple choice scales,
ranked preferences, or rating scales. Behavioral measures
frequently utilize the Operant Music Listening Recorder or
the Music Selection Recorder, both of which consider music
preference as analagous to the time spent listening to
musical stimuli. The discrepancy between the self-report
mode and behavioral measures is often seen, however, pointing
to the need to further examine the relationship between
attitude and actual behavior.

Studies which explore the relationship between preferred
music and anxiety or tension have shown, overall, that
preferred music may decrease anxiety. However, the decrease
in anxiety is usually demonstrated through psychological
measures. The lack of correlation between subjective
(psychological) and objective (physiological) measures is
again apparent in the music research literature, emphasizing
the value of accurate measurements and the need for

methodological control.

CHAPTER III
METHODOLOGY

Design
This study adopted a repeated measures design using
pretests-posttests of the dependent variables, state anxiety
and relaxation. The independent variables consisted of
preferred music, nonpreferred music, and silence, each of
which was used in one of three 15-minute listening sessions.
A third dependent variable, degree of muscle tension, was

recorded during each of the 15-minute conditions.

Sample

The sample population for the initial portion of this
study consisted of 90 undergraduates, ages 18-22, selected
from introductory psychology classes from a large, four-year
midwestern university through informed consent procedures.
Nine of the 90 subjects (five females, four males)
participated in the latter portion of the experiment after
passing these selection criteria: (a) must have scored -
approximately one-half of the standard deviation above the
trait anxiety mean of the State-Trait Anxiety Inventory
(STAI) as published in the test manual: 42.89 for males,
45.47 for females (Spielberger et al., 1983); (b) must have
had two years or less of formal musical training or music

activity participation, e.g., band or choir.

Setting
The group setting for the administration of the Trait
Anxiety Scale of the STAI, the Musical Experience
Questionnaire, and the Music Preference Test was held in a

large classroom at the University's School of Music, where

43

44

subjects were seated at individual desks to record their
answers on paper. The setting for the individual testing
periods was held in a single room of the Music Therapy
Clinic/Psychology of Music laboratory, where subjects were
seated in a recliner chair facing a curtain with their backs
to the experimeter. Care was taken to ensure that
fluorescent lights were not utilized during the operation of

the EMG.

Instrumentation

The music preference test was taped and played on a Sony
Stereo Cassette Deck TC—FX6. Also utilized for the test were
a Hafler Pre-Amplifier, an Adcom Power Amplifier Model GFA—
555, and Altec Speakers. Taped music during the individual
sessions was played on a Kenwood Stereo Cassette Deck KX-620,
amplified through a Sony STR-AV330, and heard through
Celestion Speakers. A General Radio Company Sound-Level
Meter, Type 155l-C was used to control the decibel level of
the music. Muscle tension data were collected with an
Electromyogram Myosone 409 by Bio-Logic Devices, Inc., using
Signa electrode gel on silver-silver chloride surface
electrodes. State and trait anxiety were assessed with the
Spielberger State-Trait Anxiety Inventory, while relaxation
was measured with the Subjective Relaxation Te3t (Appendix A)
using a 7-point Likert scale (1 = extremely relaxed;

7 = extremely stressed).

Music preference was assessed by a music preference test
that was developed by Bartlett (1973) and revised for this
current research (see Appendix B). The test consisted of 56
taped musical excerpts, each of which was approximately 30
seconds long; about eight seconds of silence separated each
excerpt. Subjects were asked to rate each excerpt on a scale
of 1 to 7 (1 = strongly like; 7 = strongly dislike).

There were eight musical categories; each category contained

45

seven examples of each which were presented in random order
with the exception that no single type of music was offered
twice in a row. The eight categories were: country western,
jazz, folk, classical, atonal, hard rock, light rock, and new
age. Atonal music was chosen as a category separate from
classical music as a result of the 1984 study by Stratton and
Zalanowski, which was reviewed in chapter II. Comparing the
effects of five types of music, the authors found that atonal
music lead to significantly less relaxation. Inclusion of
atonal music into the Music Preference Test was based on the
expectation that it might be chosen as a least preferred type
of music.

Taped instructions for the music preference test were
quoted verbatim from Bartlett's test:

The purpose of this test is to determine what kinds or

types of music you like and the degree to which you like

them. You will hear a series of short musical excerpts.
After the completion of each excerpt, you are asked to

rate how much you like that particular excerpt to the best

of your ability by circling the appropriate number on the
rating scale provided. Each scale consists of a number
which corresponds with a degree of dislike or like. For
example, if, after listening to a musical excerpt you
decide you like "mildly" that selection, then you should
mark the appropriate rating scale in the following manner.
There will be a brief pause after each excerpt to mark

your rating. Do not attempt to change a previously marked

rating. First, one musical example will be given to help
your understanding of the test. Are there any questions?

At the completion of the test, further instructions were
given:

That completes our test. Now, you will find on the next
page a list of musical types: classical, atonal, country
western, light rock, hard rock, jazz, folk, and new age.
Please place the number one next to the musical type that

you like best, the number two next to the musical type you

like next best, the number three beside the type you like
third best, and so forth for all eight musical types.
Thank you.

Prior to the experiment, the music preference selections

were rated by three judges (three doctoral music degree

46

candidates) to determine the test's face validity in terms of
the categorizations of music. A majority decision by the
judges' panel confirmed the choice of selections as
categories.

For the individual sessions, preferred music was provided
by the subjects, and specifically, was that music which was
determined to be their most preferred as shown by the music
preference test (see Table 1). The types of preferred music
and the number of subjects who chose each type as their most
preferred include: (a) hard rock--4; (b) light rock--3;

(c) new age--1; and (d) jazz-—1. Nonpreferred music was
provided by the experimenter. The categories of music which
were specified as nonpreferred and the number of subjects who
selected each category as their least preferred music

include: country western-~3, atonal--5, and hard rock-—l.

Brmsdure

Subjects met together in a large classroom where forms
were handed out initially: the consent form, the Musical
Experience Questionnaire, the Trait Anxiety Scale of the
STAI, and the answer sheet to the Music Preference Test
(Appendix A). After ensuring that the consent form was read
and understood, the subjects were asked to complete the
Musical Experience Questionnaire and the Trait Anxiety Scale.
After completion of the scale and the questionnaire, the
Music Preference Test was administered, with the volume
adjusted to a comfortable level. Anonymity of each response
was accomplished by a predetermined number which wasassigned

to each subject.

47

Table 1
MW

 

Preferresilmsic

A-Ha: WW Journey: Wits
Berlin: "No More Words" Pat Metheny: Still_Lifio
Duran Duran: "The Chauffeur" Orchestral Manoeuvres In The
Frankie Goes To Hollywood: Dark: Ihe_£ooifio_Aoo

"Relax" Pink Floyd: Iho_Woll
Amy Grant: Iho_§ollootion Shadowfax
Janet Jackson: "Control," Styx: "Mr. Roboto"

"Nasty," "What Have You Done
For Me Lately?"

n M 1
QQEDLI¥_H§§LQLE
Willie Nelson: "I'd Trade All Of My Tomorrows,"
"Fraulein"
Emmy Lou Harris: "Heartbreak Hill"
Lester Flatt and Earl Scruggs: "Folsom Prison Blues,"

"Long Road To Houston"
Patsy Cline: "Your Cheatin' Heart"
Atonal
Bernd Alois Zimmerman: Potspootiyos
Arnold Schoenberg: Itio_1Sttinglt_qus_15 (1946)
Ha2d_EQQK
Metallica: "Hit The Lights," "The Four Horsemen,"

"Phantom Lord"

 

48

Each of the nine subjects who participated in the latter
portion of the experiment was tested individually for a total
of three testing periods, at a minimum of three days apart.
The presentation order of preferred music, nonpreferred
music, and silence were counterbalanced for all the subjects.

At the beginning of each individual session, standard

questions were asked of each participant:

These questions are purely for research purposes and your
answers will remain strictly confidential. Are you
presently experiencing any pain in your body, e.g.,
headache, stomachache? Are you taking any muscle-relaxing
drugs or other kinds of medication? Are you under the
influence of any narcotic substance? Have you consumed
alcohol within the past 24 hours? If so, how much?

The Subjective Relaxation Test and the State Anxiety Scale
were then administered. After completion of these scales, an
explanation of the EMG was given during a subject's first

individual session:

This machine is run on batteries. There is no chance of
getting shocked. These are the electrodes which will be
attached to your shoulders and the back of your neck. You
will feel no pain from them and you will soon forget they
are on. All you may feel is the sticky electrode gel
which is used to provide a better recording for the
electrodes. The gel is water-soluble, and there should be
no trace of it afterward, on you or your clothing. Do you
have any questions?

Two surface electrodes were attached onto the subject's upper
left and right trapezius muscles, with a ground electrode
attached at the base of the neck. Further instructions were
then given:
Make yourself as comfortable as possible. The chair
reclines into two more positions; use whichever one feels
most comfortable. Do not go to sleep. I will record
readings for the next five minutes. Please try not to
move. If you have to move, e.g., cough or sneeze, do not

be too concerned about it; I will just begin recording
that segment over again.

The EMG was set to display digital readings at 30-second

analysis time intervals, using a wide band pass filter and a

49

slow meter averaging time constant. Baseline readings were
taken for the next five minutes, with the lights dimmed
slightly.

At the completion of the five-minute period, the subject
was allowed to move. If preferred music was used, the volume
of the music was adjusted to a level comfortable to the
subject during this time. If nonpreferred music was the
experimental condition, the decibel levels of both the hard
rock and country western music were preset at approximately
60-70 decibels (dB). Because the atonal music varied more in
volume, its decibel levels ranged from 50 dB to 80 dB.
Whether music or silence was the condition presented, the
subject was told to make herself/himself as comfortable as
possible. Instructions to remain awake and stationary for
the next 15 minutes were repeated. EMG readings continued to
be monitored.

At the end of the lS-minute period, the music was turned
off and the subject was again allowed to move if he/she
chose. The State Anxiety Scale and the Subjective Relaxation
Test were readministered, and the electrodes were removed
after completion of the tests. The subject was instructed to
refrain from discussing the experiment with classmates or
friends until after the end of the University's term, in
order to control for possible biasing of other subjects. All
of the above individual testing procedures, with the
exception of the explanation of the EMG, were repeated two

more times for each subject.

50

Pilot Testing

The individual experimental procedures were
practiced with five normal subjects of varied ages to
provide an opportunity for the researcher to use the EMG
and to gain consistency and continuity in the procedures.
Dimmer lighting resulted from suggestions made by two
subjects; the lights were dimmed to possibly enhance
relaxation.

Chapter IV
RESULTS

Difference scores for the dependent variables, state
anxiety and relaxation, were obtained by subtracting the
posttest scores from the pretest scores for each condition.
EMG difference scores represent the differences between the
average number of microvolts for each condition and their
respective average-microvolt baselines. These results are
presented in Tables 2-4, while the raw data scores for each
subject are reported in Appendix C. Positive scores indicate
an increase in relaxation levels and a decrease in anxiety
and muscle tension levels. Conversely, less‘positive or more
negative scores indicate decreased relaxation and increased

anxiety and muscle tension.

Table 2

. - f 1-‘ D - ‘1 - err ne‘ one o! of

 

 

 

Preferred Nonpreferred
Subject Music Music Silence
1 8 10 11
2 -1 1 5
3 5 -1
4 6 6 13
5 1 —2
6 11
7 4
8 8 —4
9 6 -3 10
M 5 3 2.4 5 8

51

52

As seen from the mean values for each condition, preferred
music and silence seemed to induce less anxiety and greater
relaxation than nonpreferred music, but nonpreferred music
indicated less muscle tension than the other conditions.
Also, silence appeared to produce the greatest degree of
muscle tension. However, the EMG results may be skewed due
to the scores of Subject 3 under nonpreferred music, and
Subject 2 under silence.

An analysis of the effects of each measure upon conditions
and subjects was performed using a repeated measures analysis
of variance (Table 5). None of the F values were significant
at the .05 level, although two E values were significant at
the .10 level. Table 5 shows that there is a main effect
between state anxiety and subjects at the 10% level, with a
table value of 2.09 at 8 and 16 degrees of freedom,
indicating a statistical difference between subjects on the
measure of anxiety. The other significant effect at the .10
level is between relaxation and music preference, with a
table value of 2.67 at 2 and 16 degrees of freedom. This
seemed to indicate that the conditions of preferred music,
nonpreferred music, and silence influenced relaxation, but
only at the 10% level.

 

53

Table 3
E E l M . H E 1 M . 3 5.]

 

 

 

Preferred Nonpreferred

Subject Music Music Silence
1 1 2 4

2 1 0 1

3 2 1 0

4 2 O 3

5 0.5 O 0.5
6 2 2 2

7 2 O 2

8 2.5 -O.5 O

9 1 1 1
(M 1.5 O 5 1

The two table values for state anxiety-conditions and EMG-
conditions were associated with a p,of only .25. In other
words, the conditions of music and silence appeared to have
little effect on state anxiety and muscle tension. The table
value for the effect between EMG and subjects was associated
with a p,of .25, while the probability for the effect between
relaxation and subjects was only .50, indicating few
differences between subjects on measures of muscle tension

and relaxation.

54

 

 

 

Table 4
EMS Qiffsrsnos Sootss Qnosr Conditions of

' N ' n ' n

Preferred Nonpreferred

Subject Music Music Silence
1 0.06 0.91 0.05
2 -0.27 -0.37 -3.80
3 0.96 10.43 -0.19
4 -0.33 -0.29 0.36
5 0.96 0.58 -0.10
6 -2.26 -0.49 0.20
7 0.29 0.02 0.19
8 0.58 3.45 0.25
9 -0.18 0.22 0.45
M -0.02 1.61 -0.29

Another analysis of variance was computed to determine if
division of the EMG scores into five—minute segments would
indicate a difference that was not shown when scores were
averaged for the full 15 minutes. This analysis was
performed to observe if time had an effect on the degree of
muscle tension. The data, as reported in Table 6, were not
found to be significant at the .05 level. Values for the
first five-minute EMG period were associated with a p of .25.
Values for the middle five-minute period were associated with
a p of .50 for the EMG-conditions effect, and a p of .25 for
the EMG-subjects effect. Finally, p values of .25 and .50
were associated with the EMG-conditions effect and the EMG-
subjects effect, respectively, for the last five-minute
period. Therefore, time, as measured in three five-minute
segments, does not appear to have a significant effect on

muscle tension.

Table 5

55

E J . E V . S I 1]

 

 

 

Source dF 38 MS p
I. State Anxiety
Conditions: C 61.55 30.78 2.26 .05
Subjects: 8 251.33 31.42 2.31 .05*
Interaction: CS 16 217.79 13.61
II. Relaxation
C 5.05 2.53 2.70 .05*
S 8 10.67 1.33 1.42 .05
CS 16 14.95 0.93
III .EMG
C 18.93 9.46 2.23 .05
S 52.82 6.60 1.55 .05
CS 16 68.11 4.25
* significant at the .10 level

56

 

 

Table 6
Fir i n
W
Source dF SS MS F p
I. First 5 Mins.
Conditions: (2 2 37.25 18.62 2.06 >.05
Subjects: S 8 120.06 15.00 1.66 >.05
Interaction: CS 16 144.50 9.03
II. Middle 5 Mins.
C 2 11.07 5.53 1.49 >.05
S 8 43.73 5.46 1.47 >.05
CS 16 59.43 3.71
III. Last 5 Mins.
C 15 81 7 90 1 98 > 05
S 38 80 85 1 21 > 05
CS 16 63.76 3.98

 

Pearson product-moment correlations were calculated
between conditions and measures (see Table 7). Only one
correlation was found to be significant at the .01 level:
state anxiety under silence with relaxation under silence.
The probability for all other correlations was greater than
.05. Interestingly, the correlations between state anxiety
and relaxation under the remaining conditions of preferred and
nonpreferred music also appear to be related substantially
(t = .540 and .580, respectively). This would seem to
indicate that as relaxation increased, anxiety decreased, and
vice versa, strengthening the relationship between the self-

report modes.

 

 

57
Table 7
r n Pr —M m n rr i
Egsfisrtso Musio Nonprsfstrso Musio Siisnos
SA BEL EMG A RE E E EM
matured
11151: .
SA --- .540 -.510 .333 -—- -—— .118 --- ——-
REL --- --- -.159 --- -.159 --- —-- -.100 ---
EMG —-- -—- —-- --- --- .378 --- --- .024
N n f r M 1
SA --- .580 —.115 .440 —-— ———
REL --— —-- .035 --- .467 --—
EMG -—— -—— --- -—— —-—- 027
Silonos
SA ——— .811*.164
REL -—- ——- .174
EMG --—- --- ---
* p < .01

SA = state anxiety
REL = relaxation

EMG = electromyogram

Moderate correlations were found between relaxation/
nonpreferred music and relaxation/silence (t = .467), and
between state anxiety/nonpreferred music and state anxiety/
silence (t = .440). These correlations may suggest that the
measure of relaxation was similar under conditions of both
nonpreferred music and silence, and that the same could be
implied of state anxiety. However, the significance of these
relationships may appear misleading. A correlation

coefficient implies a relationship between two variables of

58

commonality. The amount of relationship is expressed in
percentage as computed by squaring the coefficient. For
example, a coefficient of .440, as seen above, indicates that
a relationship exists between two variables equal to 19%.
Therefore, if 19% represents the amount of relationship that
can be accounted for, 81% remains unaccounted for.
Consequently, a .440 relationship, although seemingly
moderate, is far from direct.

One last correlation of interest exists between state
anxiety and EMG under the condition of preferred music.

A negative correlation was found (t,= -.510), indicating an
inverse relationship. In other words, muscle tension levels
did not decrease as state anxiety decreased, as expected, but
instead, moved in the reverse direction. However, this
occurrence happened only under the condition of preferred
music. This implies that, though preferred music may lower
state anxiety, preferred music could also, conversely,
increase muscle tension.

The next sets of graphs present the data for each subject.
The odd-numbered figures (e.g., 1, 3, 5, 7) compare
difference scores for each measure under conditions of
preferred music, nonpreferred music, and silence. Constants
were added onto the actual difference scores (+4 for EMG;
+5 for state anxiety and relaxation), first, for purposes of
analysis, and second, to make the graphs more easily
readable. Although the graphs vary in scale, they are
presented to indicate directional differences in tension/
relaxation. The even-numbered figures (e.g., 2, 4, 6, 8)
display the average number of microvolts for each 30-second
interval during baseline and condition. Discussion of
quantitative results accompany each set of graphs.

Figure 1 indicates that silence induced the greatest
degree of relaxation and the least amount of anxiety in
Subject 1, and that preferred music had the opposite effect.
The EMG, however, showed that preferred music and silence

were nearly equal in difference scores, while nonpreferred

59

music demonstrated the greatest difference in muscle tension.
A look at Figure 2 reveals that nonpreferred music contains
the highest baseline, in terms of number of microvolts. The
subject mentioned that a slight cramp was felt in his right
side due to running prior to the experiment. This may
account for the greater degree of muscle tension during the
nonpreferred music baseline, and subsequent large

difference score when the EMG levels returned to levels
similar to that of the other conditions.

There seems to be a discrepancy in Subject 2's anxiety and
muscle tension scores, as seen in Figure 3. Silence seemed
to induce the least anxiety, but also the greatest muscle
tension; conversely, the condition of preferred music
contained the greatest anxiety and the least muscle tension.
Figure 4 may aid, partly, in explaining this disparity.
Immediately noticeable in this figure are the extremely wide
leaps in tension during silence. Subject 2 had difficulty
remaining awake during the silent condition; a few reminders
to keep her head upright were required, beginning at the
sixth 30-second interval of the 15-minute period. Although
the EMG machine was reset whenever the experimenter noticed
gross head movement or sleep, it was hard to control for
slighter movement (i.e., nod of the head). During the
preferred music condition, a few reminders to remain awake
were again needed, e.g., the leap in tension at interval six
of the 15-minute period. Ideally, these data should have
been disregarded due to movement artifacts.

Subject 3's anxiety and relaxation scores seem to be in
agreement, with preferred music showing greater
relaxation/lesser anxiety and silence indicating the opposite
(see Figure 5). The EMG scores, though, graphically show
that nonpreferred music brought about the greatest degree of
muscle tension reduction. A closer look at Figure 6
discloses one reason for the large difference in EMG scores:
a high degree of tension in the baseline of the nonpreferred

60

 

 

 

 

 

 

 

 

 

Anxiety Relaxation EMG
20 T 10 .. 5
18 -4 9 -1- l A.
16 < 8 1 4 -
14 - / 7 ..
12 1 6 j 3 1
10 - 5 .
8 .. 4 . 2 .
6 — 3 . _
4 ~ 2 . 1 —
2 _ l -. .
PM NPM SIL PM NPM SIL . PM NPM SIL

Figure 1. Difference Scores Across Measures Under Conditions of

Preferred Music, Nonpreferred Music, and Silence: Subject 1.

° Preferred Music

D Nonpreferred Music

4 A Silence

Microvolts
N

H

 

A A 1 l I n ‘ r 1 L L L J
1 v r - v

Baseline 30-Second Intervals
Figure 2. Average EMG Microvolts Under Experimental Conditions:

Subject 1.

 

61

 

 

 

 

 

 

 

 

 

 

Anxiety Relaxation EMG
20 . 10 5
18 9«
14: 7. 1
12 : . 6< 3‘
8: 4.. 2.
6i 3- .
4 : 2.. 1-
2i 1 .
PM NPM SIL PM NPM SIL T3; NPM SIL

Figure 3. Difference Scores Across Measures Under Conditions of

Preferred Music, Nonpreferred Music, and Silence: Subject 2.

Microvolts

i-‘H
OHHNNwwkhmmmmflmkDOI—J
I I

62

' Preferred Music

40.0 - D Nonpreferred Music

30.0
20.04 ‘ A Silence

19.01
18.0-
17.0»
16.0-
15.0»
14.0»
13.0.

12.0-

0 O D I O O I
O O
: e - :

‘
UtO U10U'IOU10U'IOU'IOUIOOO

 

i

 

 

 

I l . L

 

 

Baseline 30-Second Intervals
Figure 4. _Ayerage EMG Microvolts Under Experimental Conditions:

Subject 2.

Anxiety
zol
18‘
16-
14—
12-

10.

 

L I

 

 

PM NPM SIL

10.

 

63

Relaxation

 

 

1 n
l | r

PM NPM SIL

15.0,

 

EMG

A; l

 

PM NPM SIL

Figure 5. Difference Scores Across Measures Under Conditions of

Preferred_Music, Nonpreferred Music, and Silence:

Subject 3.

 

 

Microvolts

or—It—‘NNWWJ-‘J-‘U‘MO‘O‘NCDQ

64

- Preferred Music

D g Nonpreferred Music

A. Silence

)— i—
t—‘ N
. .
O O
. .
in“

10.0.

U‘IOUIOUIOUIOUIOUIOUIOOO
I IA 1 I :I

o

 

 

. . 11.. ‘IAAAALAJ
7", v.yvrvvrrrfirvr V ' V "’

 

 

Baseline 30-Second Intervals
Figure 6. Average EMG Microvolts Under Experimental Conditions:

Subject 3.

65

music condition. During the actual listening period,
however, the tension decreased to levels similar to the
condition of silence. In general, Figure 6 shows that
preferred music induced the lowest tension levels, similar to
the anxiety and relaxation findings shown in Figure 5.

Psychological and physiological data appear to agree with
each other in the condition of silence for Subject 4 (Figure
7). During this condition, greater relaxation and lesser
anxiety and muscle tension was observed. The conditions of
preferred and nonpreferred music also seem to be correlated
under anxiety and EMG measures, both indicating similar
tension levels. Figure 8, however, shows that preferred
music has a slightly higher tension level than nonpreferred
music.

Figure 9 indicates that the condition of silence brought
about the greatest change in anxiety for Subject 5, but
showed little change in EMG tension. Also, anxiety levels
differed between preferred and nonpreferred music: preferred
music seemed to induce less anxiety. Preferred music also
brought about a greater change in muscle tension, although
nonpreferred music seemed to have a similar EMG difference
score. Relaxation scores did not vary greatly between
conditions.

Figure 10 demonstrates a partial reason for the small
difference in tension during silence: the baseline was
already low. A decrease in muscle tension, therefore, would
not have shown much difference. Also, preferred and
nonpreferred music both show decreases in tension. The
higher baseline of preferred music, though, allows for a
slightly greater difference between baseline and condition.

Figure 11 demonstrates quite opposite changes for Subject
6 in measures of anxiety and EMG tension. Preferred music
induced the least anxiety but the greatest amount of muscle
tension; silence, on the contrary, showed the greatest
anxiety and the least muscle tension. The differences in

nonpreferred music scores were more similar to the silence

 

Microvolts
N U.) J>

H

66

Anxiety Relaxation EMG

20 1 101 5
18 l' 91

16 4 81 4i /
14 . 7 .

12 . 6-. \ 3.
10 o 5.

 

 

 

 

 

 

 

 

 

8 4 2-
6 . 3 _
4 2 1‘
2 - 1
PM NPM SIL PM NPM SIL I PM NPM SIL

 

Figure 7. Difference Scores Across Measures Under Conditions of

Preferred Music, Nonpreferred Music, and Silence: Subject 4.

. Preferred Music
D Nonpreferred Music

A Silence

 

Baseline 3OJSecond Intervals
Figure 8. Average EMG Microvolts Under Experimental Conditions:

Subject 4.

Microvolts
N La)

H

67

 

 

 

 

 

 

 

 

 

Anxiety Relaxation EMG

20* 10 - 5

18‘ 9 ' \\
16-- 8 .- 4

14» 7 . i

12* 6 3

10« 5. .T‘ro”" .

81 4. 2-

6. 31

40 2.. 1.

2. 1 .

PM NPM SIL PM NPM st — PM NPM SIL

Figure 9. Difference Scores Across Measures Under Conditions of

Preferred Music, Nonpreferred Music, and Silence: Subject 5.

. Preferred Music

D Nonpreferred Music

A Silence

 

1 .

Baseline 30—Second Intervals

Figure 10. Average EMG Microvolts Under Experimental Conditions:

Subject 5.

 

68

scores than to the scores for preferred music. There were no
differences in relaxation.

Figure 12 shows a dramatic increase in tension level at
the onset of the preferred music listening period, which
tapered off to lower levels during the latter half.
Nonpreferred music also demonstrates an increase in tension
from baseline to the music listening period. These scores,
however, remain high and do not taper off as do the preferred
music scores, except at the last six 30-second intervals.
Subject 6 reported difficulty in remaining awake during the
last 3—4 minutes of the nonpreferred music listening period,
which seems to account for the lower microvolt levels. The
condition of silence indicates lower tension levels than the
other conditions, both during baseline and during the 15-
minute period. The subject stated that Motrin was taken for
a headache approximately five hours prior to the experiment
in which silence was the experimental condition. Perhaps the
medication was a factor in reduced muscle tension levels.

Subject 7's EMG levels did not seem to vary greatly,
although changes in anxiety and relaxation can be observed in
Figure 13. Silence appeared to effect the greatest change in
anxiety reduction, and preferred music the least change.

Both silence and preferred music seemed to induce equal
relaxation changes, while nonpreferred music brought about
the least relaxation. A look at Figure 14 reveals relatively
similar EMG levels, with nonpreferred music indicating
slightly lower tension than preferred music or silence. The
subject reported a need to cough during the last half of the
nonpreferred music baseline; suppression of the cough may
have resulted in greater muscular tension.

Figure 15 indicates that preferred music induced greater
relaxation and lesser anxiety for Subject 8, and nonpreferred
music induced the opposite: greater anxiety and lesser
relaxation. Nonpreferred music, however, showed the greatest
positive change in muscle tension. Figure 16 demonstrates a

high baseline for nonpreferred music, permitting a greater

 

69

 

 

 

 

 

 

 

 

 

Anxiety Relaxation EMG

20 . IOT 51
18 . 9- “
16: 3.. z, .
14 T 7- o————o—-—c
12 : 6-- 3
10:. 5. -
8: 4. 2"
6 3‘ ’
4 :' 2 - 1
2 1 a ..

PM NPM SIL PM NPM SIL PM NPM SIL

Figure 11. Difference Scores Across Measures Under Conditions of

Preferred Music, Nonpreferred Music, and Silence: Subject 6.

 

Microvolts

 

70

‘ Preferred Music

D Nonpreferred Music

[5 Silence

 

 

 

Baseline 30-Second Intervals

Figure 12.

Subject 6.

Average EMG Microvolts Under Experimental Conditions:

 

Microvolts

Anxiety
20 -
18«
16 n
14

12 0
10 a .’,,,0///‘

NbO‘G)

 

 

1

PM NPM SIL

 

71

 

 

 

 

 

 

Relaxation EMG
1o» 5 ‘
9- -
30 4 n
7. .
61- V 3 "
5. n
4? 2 n "-o———-0
3., ..
2,. 1 n
1.. .
PM NPM SIL ’ PM NPM SIL

Figure 13. Difference Scores Across Measures Under Conditions of

Preferred Music, Nonpreferred Music, and Silence: Subject 7.

. Preferred Music

D Nonpreferred Music
A Silence

 

 

Baseline

Figure 14. Average EMG

Subject 7.

30—Second Intervals

Microvolts Under Experimental Conditions:

 

72

 

 

 

 

 

 

Anxiety Relaxation
20 n 10- 10
18 : 90 9
16 : 81 8
' O
14 < 7 7
12 : 6» 6
10 : 50 5
8 : 4. 4
6~ 3» 3
4 - 2” 2
2 - 1__ 1
PM NPM SlL PM NPM SIL

 

EMG

 

v

PM NPM SIL

Figure 15. Difference Scores Across Measures Under Conditions of

Preferred Music, Nonpreferred Music, and Silence:

Subject 8.

 

 

 

Microvolts

Ov-‘P-‘NNwWJ-‘J-‘UMO‘O‘NmkD

10.0“

o o o
l l
I

l
v

5/

UIOU'IOUIOLDOU‘OUIOUIOOO
AA A I ll

 

Figure 16.

Subject 8.

Baseline

Average EMG Microvolts Under Experimental Conditions:

 

73

0 Preferred Music
D Nonpreferred Music

ZS Silence

lAAlllllll
IIYTT‘YI’Y‘VI

30—Second Intervals

 

 

74

decrease in tension and a larger increase in the difference
score. The condition of silence did not seem to have much
effect on muscle tension, anxiety, or relaxation. One last
note for Subject 8: a rise in muscle tension during the last
four intervals of preferred music is indicated in Figure 16.
The subject reported throat irritation which occurred during
this time period; the irritation may have been responsible
for the slight increase in tension.

Silence induced the least anxiety and nonpreferred music
demonstrated the greatest amount of anxiety for Subject 9, as
shown in Figure 17. Silence also showed the greatest
decrease in muscle tension, while preferred music
demonstrated the greatest increase in tension. Silence, in
this case, seems to be correlated positively with both
anxiety and EMG data. There were no differences in
relaxation. Figure 18 lends support to the EMG data in
Figure 17. Silence shows the greatest reduction in tension,
and the overall rise from baseline during the preferred music

condition indicates an increase in tension.

 

Microvolts
N b.)

p—a

 

 

75

 

 

 

 

 

Anxiety Relaxation EMG

20, 10 5
181- 9 -
16" 8 4 /
14.. 7 ~
12__ 6 -. o——o-—o 3
104_ 5 .
8-- 4 — 2
6-. 3 -
4- 2 . 1
2. 1 —

PM NPM SIL PM NPM SIL PM NPM SIL
Figure 17. Difference Scores Across Measures Under Conditions of

Preferred Music, Nonpreferred Music, and Silence: Subject 9.

9M

. Preferred Music

Nonpreferred Music

Silence

 

 

 

fifrfi?§{IIIIvriillllvlilliilfiillfi
Baseline 30—Second Intervals
Figure 18. Average EMG Microvolts Under Experimental Conditions:

Subject 9.

 

 

CHAPTER V
CONCLUSIONS, DISCUSSION, AND
RECOMMENDATIONS

Summer!

The purpose of this study was to determine the effects of
preferred music, nonpreferred music, and silence on measures
of state anxiety, relaxation, and muscle tension. Nine
subjects were selected from a pool of 90 college students
after passing criteria for trait anxiety and musical
experience. Each of the nine subjects was tested
individually for a total of three testing periods. One
condition (preferred music, nonpreferred music, or silence)
per session was used, with the presentation order
counterbalanced for all the subjects. Pretests-posttests of
state anxiety and relaxation were administered during each
condition, and muscle tension was measured using an
electromyogram. Results were obtained using a repeated
measures analysis of variance to analyze the effects of each
measure upon conditions and subjects. Pearson product-moment
correlations were calculated to determine relationships
between conditions and measures. Quantitative results were
reported in tables and graphs and explained in the
accompanying text. The tables presented results from the
analysis of variance and Pearson correlations, and also
reported difference scores for each measure under conditions
of preferred music, nonpreferred music, and silence. The
graphs presented difference scores for each subject, as well

as individual EMG data under all conditions.

76

 

77

Malena

The following conclusions are paired with the research
questions asked in chapter I:

Question ;: Will there be significant differences between
preferred music, nonpreferred music, and silence on
the measure of state anxiety?

The analysis of variance showed that the main effect
between conditions and state anxiety was not significant at
the .05 level.

Question 2: Will there be significant differences between
preferred music, nonpreferred music, and silence on
the measure of relaxation?

The analysis of variance revealed that the main effect
between conditions and relaxation was not significant at the
.05 level. However, a significant effect between conditions
and relaxation was found at the .10 level.

Questig 3: Will there be significant differences between
preferred music, nonpreferred music, and silence on
the measure of muscle tension?

The analysis of variance showed that the main effect
between conditions and muscle tension was not significant at
the .05 level.

Question 4: Will there be significant relationships
between measures of state anxiety, relaxation, and
muscle tension under conditions of preferred music,
nonpreferred music, and silence?

One significant correlation was found at the .01 level:
state anxiety/silence and relaxation/silence. The
probability for all other correlations was greater than .05.
The relationship between state anxiety and relaxation under
conditions of preferred and nonpreferred music also appeared
to be correlated substantially (; = .540 and .580,
respectively), indicating that relaxation increased as
anxiety decreased, and vice versa.

A negative correlation (r = -.510) was established between

state anxiety and muscle tension under the condition of

 

78

preferred music, pointing to an inverse relationship in which
state anxiety decreased as muscle tension increased. This
relationship was not found under conditions of nonpreferred
music or silence.

There seemed to be little relationship between measures of
relaxation and muscle tension. Correlations were low under
all conditions: preferred music (1 = -.159); nonpreferred
music (1 = .035); and silence (r = .174). This appears to
imply that as psychological relaxation decreased, a
corresponding decrease in physiological relaxation as
measured by the EMG was not found.

On the measure of state anxiety, Pearson correlations
revealed that there was little correlation between preferred
music and silence (r = .118) and between preferred and
nonpreferred music (r = .333). A moderate correlation
appeared to exist between nonpreferred music and silence
(r = .440). On the measure of relaxation, only a small

inverse correlation appeared between preferred and

nonpreferred music (; = —.159) and between preferred music
and silence (; = —.lOO). A moderate positive correlation was
calculated between nonpreferred music and silence (r = .467)

on the measure of relaxation. Finally, low correlations were
discovered between preferred and nonpreferred music

(1 = .378), preferred music and silence (; = .024), and
nonpreferred music and silence (I = .027) on the measure of
muscle tension.

In summary, no significant E values were found at the .05
level. Two main effects, however, were found to be
significant at the .10 level: state anxiety across subjects
and relaxation across conditions. Pearson correlation
computations found that state anxiety and relaxation were
correlated significantly under the condition of silence.
State anxiety and relaxation were correlated also, although
nonsignificantly, under preferred and nonpreferred music. In
addition, a moderate relationship was established in state

anxiety under conditions of nonpreferred music and silence,

 

79

and in relaxation under the same conditions. State anxiety
and EMG muscle tension were related inversely under preferred
music, while other correlations of EMG were associated with
low correlation coefficients. Lastly, a trend appeared for
preferred music to induce lesser anxiety and greater

relaxation than nonpreferred music.

Win

The initially tested 90 subjects produced 16 subjects who
passed both of these criteria: a) must have scored
approximately one-half of the standard deviation above the
trait anxiety mean of the STAI, and b) must have had two ;
years or less of formal musical training or music activity fi
participation. Of these 16 subjects, only nine participated i
in the experiment. The small number of subjects necessitates
the need to limit the results of the study to this specific
sample only. The high standards set for selecting subjects
may appear too stringent, and might have prevented greater
subject participation. On the other hand, a second
possibility for the small subject number could be due to the
population from which the subjects were chosen (introductory
college psychology classes). These college students might
have had, for example, more musical training than individuals
without college education.

A variety of music, including rock, new age, jazz, and
Christian, were selected by the subjects as being their most
preferred relaxing music (Table 1). Although the majority of
related research studies seem to have utilized precategorized
music such as stimulative or sedative, attempts at placing
these subjects' preferred music into similar categories may
be difficult due to a variety of stylistic features such as
tempo, volume, and rhythm. Approximately 77% of the chosen
music contained lyrics. It was not the purpose of this
study, however, to explore the specific effects of vocal

music versus the effects of instrumental music.

80

The decision of choosing categories for inclusion in the
Music Preference Test was difficult. A number of categories
which now exist in the musical repertoire have developed
since the time of Bartlett's (1973) study. For instance, in
the general category of "rock," further divisions can be
found: pop, progressive rock, funk, heavy metal, disco, and
new wave. New age music was mistaken occasionally for new
wave; this is an example of confusion among categories, and a
need for clearer categorical definitions. One type of music
which consistently required definition (at the end of each
Music Preference Test) was atonal music. To briefly
demonstrate characteristics of atonal music, e.g., tone
clusters, wide leaps, and lack of a tonal center, a piano was
utilized, in addition to verbal explanations. Other
categories of music which were not included in the Music
Preference Test, besides the ones mentioned above, are
reggae, fusion, bluegrass, polkas, waltzes, marches, movie
music, Broadway tunes, and rhythm and blues. Clearly, a
music preference test that was designed to utilize as many of
these categories as possible would lengthen the time required
for completion of the test. Thus, the categories selected
for this study were chosen to represent a variety of musical
styles. For example, "light rock" could mean pop, ballads,
and easy listening rock; "hard rock" would include heavy
metal, progressive rock, new wave, and funk.

A common problem in a study which uses a repeated measures
design to measure physiological responses is that of
establishing a subjects's baseline. Would an increase or
decrease from that baseline necessarily indicate a response
due to the experimental stimulus, e.g., a decrease in muscle
tension due to preferred music? Or could a decrease in EMG
tension from an unusually high baseline be a natural result
of progression over time? For example, Subject 3's baseline
for nonpreferred music was quite high, beginning at 32
microvolts and staying in a range of 10-16 microvolts for

about 3 minutes (see Figure 6). During the listening

 

81

portion, the levels remained approximately in the 2.5-3.5
microvolt range, similar to levels under other conditions.
Was the large difference score (14.43) due to the
nonpreferred music, or to time progression? In such a case,
should the nonpreferred music session have been repeated
until the subject's baseline was within the "normal" range
(similar to baselines from other conditions)? These
questions point to a few of the difficulties inherent in
physiological research.

The low correlations associated between muscle tension and
state anxiety or relaxation concur with Dainow's (1977)
observation that little correlation exists between the
psychological and physiological responses to music. One
reason that may account for this disparity is attention to
the music. Hanser (1985) noted that a reaction comparable to
stress may be produced by attending to the music. Davis and
Thaut (1989) remarked also that physiological responses could
be indicative of either pleasant or unpleasant experiences.
Additionally, Fontaine and Schwalm (1979) demonstrated that
familiar music produced higher levels of arousal than
unfamiliar music. These factors may offer an explanation for
the inverse relationship found between state anxiety and
muscle tension under the condition of preferred music, which
is demonstrated in Figures 11 and 12. The trend for muscle
tension to increase while state anxiety decreases is similar
to the findings of Davis and Thaut. The explanation for this
inconsistency between psychological and physiological data
may be explained by Berlyne's (1971) theory of hedonic value
and arousal which states that decreases from high arousal
levels as well as increases from low levels of arousal can be
perceived as pleasant. It appears that a reduction in
anxiety and an increase in physiological arousal accompanied
a pleasurable listening experience which was induced by
preferred music.

Preferred music appeared to induce lesser anxiety and

greater relaxation than nonpreferred music. The implications

 

82

of this for music therapy suggest that a client's musical
preferences are important considerations in the determination
of appropriate music therapy activities designed to alleviate
stress or facilitate changes in emotion. In the process of
developing individualized strategies, a music therapist
attempts to establish a directional stimulus-response
relationship (Thaut, 1990). In other words, music is
utilized to facilitate desired responses. Selection of
appropriate music, then, must be determined not only by
musical characteristics, but by an individual's unique
background, which includes factors such as preference,
familiarity, cultural context, and past experiences (Davis &
Thaut, 1989). Significant psychophysiological changes may

then occur if attention is paid to these important variables.

Becommendatig s

A number of improvements can be made in this study.

1. Criteria for selection of subjects could be less
restrictive to allow for greater participation. For example,
the requirements for musical experience would be expanded to
that of Scartelli and Borling's (1986) study: "2 years or
less [of] formal music study and no formal music activity
participation. . . in the previous 3 years" (p. 159). This
would allow for individuals who played in junior high school
band, for example, to participate as long as the requirements
for formal music study were met. In a similar manner,
selecting subjects from a different population (e.g.,
subjects with only a high school education) might permit

increased subject participation.

 

83

2. Anxiety was measured in an atmosphere that was more
conducive to relaxation than to anxiety. Since Kallman and
Feuerstein (1986) noted that physiological responses seem to
be related to specific situations, it is recommended that
future research designs incorporate some type of anxiety-
invoking situation to make the environment and measurements
more "psychobiologically relevant."

3. Intensity has been shown to affect responses to music
(Dainow, 1977). Although the subjects were allowed to
control the volume of their preferred music, perhaps a future
study could increase the volume of nonpreferred music past
the decibel levels set for this study (70-80 dB). This
suggestion is based on a subject's comment that distraction
from the nonpreferred music would have been more difficult
had the music been any louder.

4. An analysis of variance was performed on five—minute
increments of EMG levels to determine if time had an effect
on muscle tension. A different approach, similar to one used
by Landreth (1974), would be to divide the musical stimulus
not into time periods, but into musical segments which are
characterized by tempo and volume, for example. Landreth
discovered that varying musical settings induced significant
heart rate changes, but that these changes would not have
been noticeable had the summation of heart rates been taken
during the entire musical selection. This procedure may be
useful in identifying specific musical elements which may
influence psychophysiological responses.

5. Specific elements of nonpreferred music may also need
to be identified. A few subjects who disliked country
western reported, at the end of the nonpreferred music
listening session, that their tolerance for Willie Nelson's
music was higher than for the music of someone like Patsy
Cline or Emmy Lou Harris. The subject whose nonpreferred
music was hard rock stated that she especially disliked the

sound of loud electric guitars. Therefore, this study could

 

84

be improved by eliciting responses from individuals, prior to
the music listening sessions, regarding specific qualities of
the music they dislike.

6. Reinking and Kohl (1975) found that the low correlation
between psychological and physiological measures of anxiety
may increase with practice of self-measured relaxation, using
EMG biofeedback as the tool. The authors hypothesized that
"the increasing correlations may show subjects' improving
ability to perceive and discriminate the internal cues
reflecting autonomic activity" (p. 599). Hanser (1985)
recommended that music therapy researchers further
investigate the area of biofeedback, and suggested that
"future investigations . . . explore the potential to monitor
continuously the listener's physiological reactions to
different musical selections" (p. 200). These musical
selections might include subjects' preferred music and/or
nonpreferred music.

7. A variety of responses occurred between subjects, as
demonstrated graphically in Figures 1-18, while only one
correlation was found to be significant at the .05 level.
These differential responses and lack of significant
relationships seem to indicate a need to pursue
psychophysiological research using a within subject approach,
i.e., examining individual response patterns to a variety of
musical stimuli using a repeated measures design (Davis &
Thaut, 1989). This procedure would deviate from the more
traditional research approach which categorizes music (e.g.,
stimulative or sedative) according to supposedly consistent
response patterns, and instead, focus on idiosyncrasies of
individuals under specific conditions (e.g., preferred

music).

 

 

S
E
C
I
D
N
E
P
m

APPENDIX A

FORMS

 

CONSENT FORM

You indicate your voluntary agreement to participate by completing
and returning this questionnaire.

The purpose of this study is to examine the effects of various kinds
of music on relaxation and muscle tension. You will be asked to:

1) complete a short music questionnaire, 2) complete a self-evaluation
questionnaire, and 3) listen to various musical excerpts as you record

your degree of liking for them on a piece of paper. For the second half

of the experiment, if you willingly choose to participate, you will be

seen individually for three sessions which will be approximately a half
hour each. For each session, you will be asked to complete a self—
evaluation questionnaire and also to indicate your current level of
relaxation. You will be seated in a recliner chair with surface electrodes
attached to your upper shoulder areas as you either listen to music or sit
in silence. The electrodes will be attached to an instrument that measures
muscle tension. There should be no discomfort involved.

All results will be treated with strict confidence and you will remain
anonymous in any report of research findings. Results may be made available
to you on request and within these restriction results.

Your participation is voluntary. You may choose not to participate
at all, may refuse to participate in certain procedures or answer certain
questions, or may discontinue the experiment at any time.

This is a statement to the effect that the experiment has been explained
to you, and that you understand it, including any inherent risks and/or
discomforts.

You may contact me (Bonnie Chan) at 393-8984 if you have any questions
or concerns that may be raised by participating in the study.

85

 

86

MUSICAL EXPERIENCE QUESTIONNAIRE

 

NAME: NO.
(last) (first)
Instruments you play Private
or have played: Yrs. Study? yes no Yrs.
Yrs. yes no Yrs.
Yrs. yes no Yrs.
Yrs. yes no Yrs.
Yrs. yes no Yrs.

 

Musical organizations you have performed in at junior and senior high schools:

Yrs.

 

Band_____yrs.______ Glee Club____ yrs._____
0rch.____yrs._____ Other
Choir____yrs._____

Record/tape/CD collection: small ____ moderate ____ extensive

Types of records/tapes/CD's collected:

 

(Folk, New Age, Jazz, etc.)

 

 

 

 

Other music courses taken in high school or college:

Describe:

 

Describe:

 

Describe:

 

 

87

Trait Anxiety Scale

Sample Questions

(taken from the State-Trait Anxiety Inventory by Spielberger,
Gorsuch, Lushene, Vagg, and Jacobs, 1983)

Almost Almost

Never Sometimes Often Always
l. I feel nervous and l 2 3 4

restless.
2. I feel like a failure. 1 2 3 4
3. I am happy. 1 2 3 4
4. I lack self—confidence. l 2 3 4
5. I am content. 1 2 3 4
State Anxiety Scale
Sample Questions
Not At Moderately Very
All Somewhat So Much So

1. I feel calm. 1 2 3 4
2. I feel upset. I 2 3 4
3. I feel indecisive. l 2 3 4
4. I feel content. 1 2 3 4
5. I feel confused. I 2 3 4

 

88

 

 

_J_IA_A_J¥
r—.-—.-—1v—1v—u
co
N

HHH
q
N

 

 

 

 

 

 

.9144.
.uum.1.—.r1.y.fir—1
H

1

: CH

 

 

HHH
\DNCDG‘

 

 

.a...
r-qxvsrxgu)

 

 

 

 

(\NNNNf\NNNBNNBNNNNNNHNNNNNNNNNN
\D\D\O\O\O\O@0000‘0000000000000000000
mmmmmmmmmmmmmmmmmmmmmmmmmInmmnnm
mmmmmmmmmmmmmmmmmmmmmmmmmmmmmm

.g_194
fluyrwr—lrx

 

.O

QQQQQQQGQQQQQ'Q'QQ'QQQQQ‘QQQQ’\fxTxTfiq
NNNNNNNNNNNNNNNNNNNNNNNNNNNNNN

He oaasmx

r\
\O
Ln
\‘f
m
N
H

 

m o m e m N H
mxfiamfio wxfiamwn oxaamfla mxfiamwo uoz mxflq exam mafia
hawconum macaw: axed umcufimz macaw: . mamcouum

umbszz ummgm umamc< ume mucouwmmum owmaz

89

 

 

.
\l

v—1

\0
a“)

 

..
\I

v—I

Ln
U‘I

 

‘
\I

A. -. --

fl

 

\‘f

m
If)!“

 

\7

‘~‘ \V ‘w hv \‘

IleI

 

\‘J‘

 

..
V

 

wwwwwwvw

\7

I:|_1‘J_I_J_.l

 

O
\Ttfiuvul

 

‘5

 

\T

 

1

 

‘ \‘nv ~w ~v¢

I A: An RI

 

U)uv£nuvu1u1u1uyu»l.nl.nl.n

mmmmmmmmmmmm

 

wfifirwu—vv—vv.

 

 

 

 

 

 

 

 

QQVQQGQQQQQ

g

 

\T

\1\7\lk1\l£1£1\l\1

I...“

J

 

\‘T

 

u‘u u) u) u) u» ul U) Luv) u) u)

 

\T

.

 

 

\‘r‘

 

 

1;. AI

 

 

l\l\[\NNNNNNNNNNNNNNNNNNNNNNN

 

wwwvw‘vwwvwwvw‘aww

VII-nth

\TQ

 

£1K)(v§1\1

1

 

g
V'IIrIIVrIrI'I'fi

 

n
oxwamwa
hawcouum

a
”shaman

m
oxaamao
sHeHHz

s
mxaamua “oz
mxau pueuamz

m
mead
saeafiz

waA

use;
hawaouum

.0

MUSICAL

Classical
Atonal

Country Western
Light Rock

Hard Rock

New Age

Jazz

Folk

90

CATEGORIES

91

 

m e n c m N H
nommouum vommmuum pommmuum ucouwum«ucH poxmfiom pmmemm momeom
zawEouuxm >uw> mama“: SHvHHz >uo> Samsouuxm

A.umbesc m oHouHo ommmamv

wusoEoE mfibu um Hmow 30> op 30m

uwoa coaumxmawm 0>Huomhbsm

92

EMG LAB REPORT

Number

EMG Session #

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Date
Condition
Meter Readings
(Microvolts)

l. 17.

2. 18.

3. l9.

4, 20.

5. 21.

6. 22.

7. 23.

8. 24.

9. 25.
10. 26.
ll. 27.
12. 28.
13. 29.
14. 30.
15. 31.
16. 32.

 

Comments:

 

 

 

APPENDIX B

 

MUSIC PREFERENCE TEST ITEMS

MUSIC
KEY:
CW = Country Western A =
= Jazz HR =
F = Folk LR =
C = Classical NA =

H
O

11.
12.
l3.
14.
15.

16.
17.
18.
19.
20.
21.
22.
23.
24.
25.

\OGJQO‘UIDLAJNH

CATEGORY ABILSILQQMPQSEB

CW Tanya Tucker
David Sanborn

F Joan Baez

C Franz Schubert

Karlheinz Stockhausen

HR ZZ Top

LR Wham

CW Oak Ridge Boys

NA Andreas Vollenweider

F Judy Collins

C Antonio Vivaldi

LR Billy Ocean

A Arnold Schoenberg

J McCoy Tyner

LR Glenn Medeiros

NA George Winston

CW Randy Travis

C Nicolai Rimsky-Korsakov
Louis Armstrong

F John Denver

A Iannis Xenakis

LR Miami Sound Machine

NA Vangelis

HR Van Halen

CW Willie Nelson

93

PREFERENCE TEST ITEMS

Atonal
Hard Rock
Light Rock
New Age

ELE
"Rainy Girl"
"Darn That Dream"
"There But For Fortune"

. WA

"Heartbeat"
"01' Kentucky Song"
"So Early, Early
in the Spring"
EIEELfififlEQfli
"Without You"

Eive Eieeee Fer Qreheetre
"Lazy Bird"
"Nothing's Gonna Change
My Love For You"
"Thanksgiving"
"Forever Amen"

: . . E 1
"Garden Song"
Metastasis
"Give It Up"

"Why Can't This Be Love?"

"Good Hearted Woman"

26.
27.

28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.

39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.

55.

56.

CW

“'1

NA
LR

HR
CW

q

HR

NA
LR

LR
LR
CW
NA

HR

94

Gordon Lightfoot

Dolly Parton, Linda

Ronstadt,
Amadeus Mozart

Duke Ellington

Emmy Lou Harris

Kitaro
Akiyoshi/Lew Tabakin
Peter Tchaikowsky
David Bowie
Keith Jarrett
Pete Seeger
Pierre Boulez
Suzanne Cianni

Barbra Streisand

Ludwig von Beethoven
U2
Charlie Daniels Band
Lukas Foss
Pat Metheny
Howard Hanson
Def Leppard
Peter, Paul, & Mary
David Lanz & Paul Speer
Billy Joel
Anton von Webern
Journey
Whitney Houston
Emmy Lou Harris
Ray Lynch

Kingston Trio

Elliott Carter

Cheap Trick

"Ordinary Man"

"Pain of Loving You"

(Piano Quintet)

"Silk Road"

m n li

"Too Young To Go Steady"

"Freight Train"

"Velocity of Love"
In And Out or is“:
Your Life"

Sympheny Ne. 5

"Comin'

Sympheny Ne. 2
"Rock! Rock!"
"Early Mornin' Rain"
"Rainforest"
"Uptown Girl"
gentete Ne. 1

"All At Once"

"Deep Breakfast"
"Where Have All The
Flowers Gone?"
Deeble geneerte f9;
H . 1 1 E'
w' w r

"The House is Rockin'"

APPENDIX C

RAW DATA

 

Table C.1

 

 

 

Anx' Pr -P R w r

Sables; Preferred_Mu§ic Neanreferred_Mu§ic Silence
1 29—21 38-28 40-29
2 41-42 48-47 39-34
3 42-37 42-39 43-44
4 28-22 49-43 60-47
5 22-21 23-25 27—24
6 35—24 42-36 40—36
7 24—20 34—29 28—21
8 35—27 28-32 21—20
9 28—22 24—27 , 32—22

Table C.2

 

Sundae: Preferred_Muaic Nonpreferred_Muaic Silence
1 2-1 5—3 6—2
2 4—3 5-5 3-2
3 4—2 4-3 5—5
4 4—2 4—4 6-3
5 2 5-2 3-3 2 5-2
6 3—1 5-3 4—2
7 3-1 3—3 3-1
8 5-2 5 3—3.5 2—2
9 2—1 4—3 3—2

 

95

96

 

 

 

 

Table C.3
v EM Mi r v 1 r f r B 'n n n i i n
Preferred Music Nenpreferred Meeie Silenee
Subject Base-Condition Base—Condition Base-Condition
1 1.90-1.84 2.90-1.99 1.87—1.82
2 2.66—2.93 2.67-3.04 3.52-7.32
3 3.16—2.20 13.43—3.00 3.03-3.22
4 2.19—2.52 1.90—2.19 2.42—2.06
5 3.07-2.11 2.10-1.52 1.55-1.65
6 2.81-5.07 3.54—4.03 2.55-2.35
7 2.42—2.71 2.06—2.04 2.49—2.30
8 3.57—2.99 6.37—2.92 1.91-1.67
9 2.42-2.60 2.71-2.49 2.72—2.27 - E
Table C.4
v E v r
v - n
51.11233: W W52 Silence
1
First 1.60 1.95 1.64
Middle 1.74 2.17 1.66
Last 2.12 1.84 2.16
2
First 3.52 3.03 14.86
Middle 2.47 3.02 4.35
Last 2.83 3.07 3.27
3.
First 2.42 3.13 3.67
Middle 2.15 2.99 3.02
Last 2.03 2.89 3.03

 

A

First 2.69 1.92 1.98
Middle 2.45 2.29 2.07
Last 2.45 2.31 2.15
5

First 2.74 1.79 1.60
Middle 2.03 1.44 1.51
Last 1.63 1.36 1.83
6

First 10.06 4.91 2.88
Middle 3.29 4.71 2.34
Last 2.50 2.47 1.53
1

First 2.41 1.98 2.46
Middle 2.81 2.04 2.42
Last 2.88 2.09 2.05
§

First 3.24 3.30 1.62
Middle 2.79 2.89 1.63
Last 2.98 2.61 1.77
2

First 2.47 2.43 2.24
Middle 2.57 2.69 2.44
Last 2.73 2.33 2.13

REFERENCES

 

Refegeneee

Abeles, H. F. (1980). Responses to music. In D. A. Hodges
(Ed.), Handhnnk_nf_Music_Psxchlegx (pp. 108—140).
Lawrence, Kansas: National Association for Music Therapy,
Inc.

Alexander, A. B. & Smith, D. D. (1979). Clinical
applications of EMG biofeedback. In R. J. Gatchel &
K. P. Price (Eds. ), ' '
(pp. 112—133). New York:

Pergamon Press.

Autogenic Systems, Inc. (1975). Instrugtign_manual_fru11he
Autggen___QQ

11 . Berkeley, CA: Author.
Autogenic Systems, Inc. (1976). n k f h ' l i
Feedbeek (Vol. 1). Berkeley, CA: Author.

Ax, A. F. (1953). The physiological differentiation between
fear and anger in humans. Psychnanmatic1Medicine, 15,
433- 442.

Barger, D. A. (1979). The effects of music and verbal
suggestion on heart rate and self-reports.

Mus1c_lherapx 16, 158— 171.

Bartlett, D. L. (1973). Effect of repeated listenings on
structural discrimination and affective response. Jeernel

Qf_Besearnh_in_Music_Educatinn.21, 302-317.

Berlyne, D. E. (1971).
New York: Appleton-Century-Crofts.

Berlyne, D. E. (1974). Studiea_1n_the_neu_eznerimental
h ' ' w ' 'V h
h i r i i n. Washington, D. C.: Hemisphere
Publishing Company.
Biller, J. D., Olson, P. J. , & Breen T. (1974). The effect
of "happy" versus "sad" music and participation on
anxiety- J9urnal.rfiLBm§1c_Theran¥I ll, 68- 73

Bio-Medical Instruments, Inc. BMI Instructign Manual;
Binlngic_Devises_492_EMG Warren, MI: Author.

Bradley, I. L. (1971). Repetition as a factor in the

development of musical preferences. r 1 R r h
inLBM§1c_EducatiQn 12, 295- 298.

98

 

99

Buchtal, F. (1957). An_1ntraduct1an_t9_electramxasranh¥.

Copenhagen: Nordisk Forlag.

Cattell, R. B., & Scheier, I. H. (1961).

The_mean1ng_and
maaauramant_af1nauraticism_and_anx1etx. New York: Ronald

Press.

Christenberry, E. (1979). The use of music therapy with burn
patients. Jaurna1_af_Music_Iheranx, 16, 138- 148

Crabbs, M. A., & Hopper, G. (1980). The relationship between
cognitive and somatic measures in the assessment of

anxiety- Bullet1n_af.tha_Es¥channm1c_Sacietx,15.218-220.

Curtis, S. L. (1986). The effect of music on pain relief and
relaxation of the terminally ill.

Therapy, 23, 10— 24.

Dainow, E. (1977). Physical effects and motor responses to
music. Jeernel ef Reeeereh in Meeie Edeeatien, 25,
211-221.

Davidson, R. J., & Schwartz, G. E. (1976). The psychobiology
of relaxation and related states: A multi-process theory.
In D. I. Mostofsky (Ed.), Behevier eegtrel ene
meeifieetien ef phyeielegieel eetivitx (pp. 399-442).

New Jersey: Prentice—Hall, Inc.

Davis, W. B., & Thaut, M. H. (1989). The influence of
preferred relaxing music on measures of state anxiety,
relaxation, and physiological responses.

may, 2_6_[ 168—187-

DeJong, M. A. , Van Mourik, K. R. , & Schellekens, H. M.
(1973). A physiological approach to aesthetic preference

—— music. PsychQtherap¥_and_Bsychngmatics, 22, 46- 51
Epstein, L. H., Hersen, M., & Hemphill, D. P. (1974). Music

feedback in the treatment of tension headache:
experimental case study. Jeernel ef Behavierel Therapy
and_Exper1mental_Psxch1atry 5, 59-63.

Fair, P. L. (1983). Biofeedback-assisted relaxation
strategies in psychotherapy. In J. V. Basmajian (Ed.),
Biefeedbeek; Prineiplee end Ereetiee fer Clinieians,
(2nd ed., pp.170-191). Baltimore, MD: Williams and
Wilkins.

 

100

Fisher, 8., & Greenberg, R. P. (1972). Selective effects
upon women of exciting and calm music.

MQLQL_Shill§, 34.987‘990-

Flowers, P. J. (1981). Relationship between two measures of
music preference. Contr1butions_to_Mus1c_Education, a,
47-54.

Fontaine, C. W., & Schwalm, N. D. (1979). Effects of

familiarity of music on vigilant performance. Eereeptgel
and_Mator_Skills.19,71-74-

Freud, s. (1936). The_problem_af_anx1etx. New York:
w. W. Norton.

Gaston, E. T. (1951). Dynamic music factors in mood change.
Music_Educatara_Journal 31.42
Goldstein, I. B. (1964). Physiological reSponses in anxious

women patients. Arch1yes_of_fienara1_£s¥£h1atr¥ 10,
382—388.

Goldstein, I. B. (1972). Electromyography: A measure of
skeletal muscle response. In N. S. Greenfield & R. A.

Sternbach (Eds.), Handhaok_afJauummmm¥siolag¥ (pp. 329-

365). New York: Holt, Rinehart & Winston, Inc.

Hanser, S. B., Larson, S. C., & O'Connell, A. S. (1983). The
effect of music on relaxation of expectant mothers during

labor. Jeernel ef Meeie Therapy,ZQJ 50—58.

Hanser, S. B. (1985). Music therapy and stress reduction

research. Journa1_of_Muaia_Tharapy, 22,193— —206.

Harrer, G., & Harrer, H. (1977). Music, emotion, and
autonomic function. In M. Critchley & R. A. Henson
(Eds. ), Mueie_end_the_brein. London: William Heinemann
Medical Books (pp. 202—216).

Hassett, J. (1978). A primer ef eeyeheehysielegy.
San Francisco: W. H. Freeman and Co.

Hodges, D. A. (Ed.). (1980). Handbook_nf_Musio_Paxcholaa¥
Lawrence, Kansas: National Association for Music Therapy,
Inc.

Jacobson, E. (1938). Progress11e_relaxatien. Chicago, 111.:

The University of Chicago Press.

 

101

Jacobson, H. L. (1956). The effects of sedative music on the
tension, anxiety, and pain experienced by mental patients
during dental procedures.

Asaoc1at1on_for_Music_Therap¥ 5(3), 9.

Jellison, J. A. (1975). The effect of music on autonomic
stress responses and verbal reports. In C. K. Madsen,
R. D. Greer, & C. H. Madsen, Jr., Beeea;eh_in_mueie
behevier in the eleeereem (pp. 206-219). New York:

Teachers College Press.

Kallman, W. M., & Feuerstein, M. J. (1986).
Psychophysiological procedures. In A. R. Ciminero, K. S.

Calhoun, & H. E. Adams (Eds.), H n k f B h vi r 1
Assessment (2nd ed., pp. 325-350). New York: John Wiley
& Sons.

Keston, M. J., & Pinto, I. M. (1955). Possible factors
influencing musical preference. Jeurnal_ef_fienetie

Psychelagy, 85,101—113.
Kibler, V. E., & Rider, M. S. (1983). Effects of progressive
muscle relaxation and music on stress as measured by

finger temperature response. Qeernel ef glinieel
Psychology, 39,213-215.

Kuhn, T. L. (1981). Instrumentation for the measurement of
music attitudes. gentribetiene te Meeie Edeeetien,§,
2-38.

Kuhn, T. L. , Sims, W. L. & Shehan, P. K. (1982).
Relationship between listening time and like- dislike

ratings on three music selections. Qeernel ef Meeie
mm, 1.8., 181-192.

Lacey, J. I. (1950). Individual differences in somatic
response patterns. n m 'v n

Phasialogica1_2sycholos¥ 43.338- -350.

Lacey, J. I., & Lacey, B. C. (1970). Some autonomic CNS
inter-relationships. In P. Black (Ed. ), Bhyeielegieel
(pp. 205— —227). New York: Academic
Press.

Landreth, J. E. (1974). Effects of music on physiological
reSponse. Jeurnal_of_Raaaarch_1n_Muaic_Education 22,4- 12

LeBlanc, A. (1984). Selecting a response mode in music
preference research. Cantr1buLions_to_Music_Educatinn 11,
1— 14.

 

Leboeuf, A. (1977). The effects of EMG feedback training on
state anxiety in introverts and extraverts. Jenrnal_ef

Clinical_P§¥;hologx 33, 251- 253

Lehman, P. (1968). n m ' m
Englewood Cliffs, N. J.: Prentice-Hall.
Logan, T. G. , & Roberts, A. R. (1984). The effects of

different types of relaxation music on tension level.

Journal_af_Mus1c_Iheran¥ 21, 177— 183.

Malmstrom, E. J. (1968). The effect of prestimulus
variability upon physiological reactivity scores.

Psychophysiologx 5.149-165

Martin, B. (1961). The assessment of anxiety by
physiological behavioral measures. Eeyehelegieel_finlletin,
58,234- -255.

Miller, R. K., & Bornstein, P. H. (1977). Thirty-minute
relaxation: A comparison of some methods. rn f

Behay1or_Therapx_and_Exper1mental_2sxch1atr¥, 8, 291— 294.

Morgan, J. J. B. (1916). The overcoming of distraction and

other resistances. Arehivee ef Peyehelegy,3§.

Peretti, P. O. , & Swenson, K. (1974). Effects of music on
anxiety as determined by physiological skin responses. .

Jburnal_of_Basaarch11n1Music_Educatian 22.278- -283
Pollack, N. J. (1988). Tha_affact_of_music_listen1ns_on

v m
ehileren. Unpublished master's thesis, Michigan State
University, East Lansing, MI.

 

Prager-Decker, I. J. (1978). The relative efficacy of
progressive muscle relaxation, EMG biofeedback and music
for reducing stress arousal of internally vs. externally
controlled individuals.

Internationalhia.3l778.

Prueter, B. A. , & Mezzano, J. (1973). Effects of background
music upon initial counseling interaction. Jeernel ef

Music_Therap¥, 10, 205— 212.

Reinking, R. H. & Kohl, M. L. (1975). Effects on various
forms of relaxation training on physiological and self-
report measures of relaxation.

Clinical__sxohoiosx.ii.595- 600.

103

Rider, M. S. (1985). Entrainment mechanisms are involved in
pain reduction, muscle relaxation, and music- mediated

imagery. Journal_afdmnu111meran¥ 22, 183- 192

Rider, M.S., Floyd, J. W., & Kirkpatrick, J. (1985). The
effect of music imagery and relaxation on adrenal
corticosteroids and the re-entrainment of circadian

rhythms. Journal_of_Music_Iherapx.22.46—58.

Ries, H. A. (1969). GSR and breathing amplitude related to

emotional reactions to music. Beyehenemie_591enee,1$,
62— 64.

Rohner, S. J., & Miller, R. (1980). Degrees of familiar and
affective music and their effects on state anxiety.

Jeernel ef Meeie Ihereey,21,2~15.
Sarason, I. G. (1960). Empirical findings and theoretical

problems in the use of anxiety scales.
Bulletin,5_,403- -415.

Scartelli, J. P. (1982). The effect of sedative music on
electromyographic biofeedback assisted relaxation training E
of spastic cerebral palsied adults. geernel ef Meeie

Iherapx.12,210-218.

Scartelli, J. P. (1984). The effect of EMG biofeedback and
sedative music, EMG biofeedback only, and sedative music
only on frontalis muscle relaxation ability.

Mus1c_l_erapx 21. 67- 78.

Scartelli, J. P., & Borling, J. E., (1986). The effects of
sequenced versus simultaneous EMG biofeedback and
sedative music on frontalis relaxation training.

Janrnal_of_Music_Iherapx.23, 157-165.

Schachter, S. (1967). Cognitive effects on bodily
functioning: Studies of obesity and eating. In D. C.
Glass (Ed ), Naurophxsiolos¥_and_emotian(pp 117-144).
New York: Rockefeller University Press.

 

Sears, W. W. (1952). Postural responses to recorded music.
In E. G. Gilliland (Ed.), Music_Iherap¥L1951(pp- 197-
198). Chicago: Allen Press.

Sears, W. W. (1958). The effect of music on muscle tonus.

In E. T Gaston (Ed ), Music_Iherap¥1_1911 (pp 199- -205).
Lawrence, Kansas: Allen Press.

104

Sears, W. W. (1960). A study of some effects of music upon
muscle tension as evidenced by electromyographic

recordings. D1ssartation_Abstracts, 21,208- -209

Shipman, Wm. G., Oken, D., Goldstein, I. B. , Grinker, Sr. ,
R. R., & Heath, H. A. (1964). Study in psychophysiology
of muscle tension: II. Personality factors. Arehiv vee ef

Genera1_Ps¥ch1atry,11,330- -345.

Smith, C. A. & Morris, L. W. (1976). Effects of stimulative
and sedative music on cognitive and emotional components i
of anxiety Psychological_3aports, 38, 1187- 1193

Smith, C. A. , & Morris, L. W. (1977). Differential effects
of stimulative and sedative music on anxiety,
concentration, and performance. Eeyehelegiee1__eperte, _1,
1047— 1053.

Spielberger, C D (Ed. ). (1972). Anxiat11__§urrant_trands
in th eer y ener reeeer eh (Vol. 1). New York: Academic

Press .

 

Spielberger, C. D., Gorsuch, R. L., Lushene, R., Vagg, P. R.,
& Jacobs, G. (1983). Manual_for_the_State:trait_Anx1etx :
lnyentery. Palo Alto, CA: Consulting Psychologists
Press.

Standley, J. M. (1986). Music research in medical/dental
treatment: Meta—analysis and clinical applications.

Journal_of_Music_Iharapx.23.56-122.
Stoudenmire, J. (1972). Effects of muscle relaxation

training on state and trait anxiety in introverts and

extraverts. dourna1_of_Personal1ty_and_Soc1al_Esychology,
25,273-275.

Stoudenmire, J. (1975). A comparison of muscle relaxation
training and music in the reduction of state and trait

anxiety. Jaurnal_of_Cl1n19al_RsychologL 31. 490- 492

Stratton, V. N., & Zalanowski, A. H. (1984). The
relationship between music, degree of liking, and self-

reported relaxation. Jeernal ef MeeieT herepy, 21,
184—192.

Taylor, D. B. (1973). Subject responses to precategorized

stimulative and sedative music. Jen;ne1_ef_Musie_Iheraey,
1Q,86—94.

105

Thaut, M. H. (1990). Physiological and motor responses to
music stimuli. In R. Unkefer (Ed. ),
New York:
Schirmer.

Voas, R. B. (1952).
' v . Unpublished doctoral dissertation,

University of California, Los Angeles.

Wapnick, J. (1976). A review of research on attitude and
preference. ' ' ' M

Education.1&.l-20-

' ' ' ' . (1977). Springfield,
MA: Merriam-Webster.

Wilson, C. V. , & Aiken, L. S. (1977). The effect of
intensity levels upon physiological and subjective
affective response to rock music. rn i

Iherapx,lA,60— 76.

 

 

‘illll fl‘l'I‘ll I

 

.J

1|1111111111211111111111110111111111111114111111311111111111111111