1/7

tJ/de"

ABSTRACT

ELECTRICAL SKIN CONDUCTANCE AND CARDIAC RESPONSES
OF REPRESSERS, MIDLINES, AND SENSITIZERS TO
AFFECTIVE STIMULUS FILMS

BY

Yonkel Goldstein

Personality theorists have paid much attention to
psychological defensiveness. One conceptualization of
defensiveness which was reviewed here is Byrne's notion
that defensiveness may be viewed as a behavioral continuum
ranging from repressing and ignoring behaviors to sensi-
tizing and distorting behaviors. PsychOphysiologists have
devoted much attention to the correlates of an organism's
accepting and/or rejecting stimulus input. This study was
an attempt to investigate the relationship between Byrne's
conceptualization of defensiveness and hypothesized psycho-
physiological correlates of shutting out or accepting
environmental stimulation.

Two physioloqical measures--heart rate and electrical
skin conductance--were analyzed. Literature was reviewed
which suggested that heart rate increases when an organism
shuts out stimuli; conversely, when an organism attends to

stimuli heart rate decreases. Thereforerone hypothesis of

Yonkel Goldstein

the study was that when a subject was involved with a filmed
vignette, heart rate should decrease. When a subject was
not involved with a vignette heart rate should increase.

Evidence from the literature was also reviewed suggest-
ing that when subjects are defensive electrical skin conduc-
tance (ESC) activity maintains its rate or increases. When
a subject accepts environmental input ESC activity decreases.
Therefore, a second hypothesis of the study was that when a
subject was involved with a filmed vignette, ESC should de-
crease; when a subject was uninvolved, ESC activity should
remain the same or increase.

The literature also suggested that repressers show more
physiolOgical arousal than sensitizers, both before and dur-
ing stimulation. This, then, became a third hypothesis.

A fourth hypothesis tested in this study was that re-
pressers' physiological patterns when they are involved with
vignettes differs from sensitizers' patterns when they are
involved and that repressers' physiological patterns when
they are uninvolved differs from sensitizers' patterns when
they are uninvolved.

A fifth hypothesis was that subjects' involvement with
the vignettes could be predicted from their personality type
as measured by the Byrne Repression-Sensitization Scale,
cardiac change scores, and ESC change scores.

Forty-five female subjects from Education 484 classes
were picked, on the basis of their scores on the Byrne Re-

pression-Sensitization Scale, such that fifteen could be

Yonkel Goldstein

considered repressers, fifteen midlines, and fifteen sen—
sitizers. Each subject was shown twelve stimulus vignettes
while her heart rate and ESC were being recorded; simultan-
eously, she was videotaped. After watching the vignettes
and taking a memory test about them, she watched the video-
tape of herself viewing the film, stopping the tape to
describe her experiences while watching the film. A tape
recording of this "recall" was then rated by two judges.
From their ratings two vignettes per subject were designa-
ted as "involved"--meaning that the subject verbalized an
emotional reaction to them--and two are designated as "un-
involved"--meaning that the subject failed to verbalize

any reaction to them.

A multivariate analysis of variance was carried out on
these data. The results indicate that repressers are more
labile than other groups--ie. they had larger electrical
skin conductance standardized standard deviations than sen-
sitizers or midlines--during the base rate and while watch-
ing the films. The results also indicated that subjects have
a lower mean skin conductance level when uninvolved with the
vignettes than when they are involved with them. No statis-
tically significant interaction was found in this analysis.

A multiple regression analysis based on the same six
vignettes for all subjects was performed, attempting to
predict overall involvement ratings from Repression-Sensiti-
zation scores, electrical skin conductance change scores, and

cardiac change scores. For all subjects approximately

Yonkel Goldstein

one-third of the involvement variance was accounted for by
Repression-Sensitization and ESC change scores. However,
when separate analyses were run on the different personal-
ity groups, ESC change scores accounted for approximately
76% of the involvement variance for repressers, 6% of the
involvement variance of midlines, and 1% of the involvement
variance for sensitizers.

The implications of these findings were discussed.
These results gave no support to the interpretation of car-
diac activity on which these hypotheses were based; however,
there were sufficient differences between this study and
others so that neither do they damage the basis of those
hypotheses. These results confirm the findings reported in
the literature that repressers are more physiologically re—
active than other personality groups. In spite of the lack
of a significant interaction between involvement and person-
ality group in the MANOVA, results are interpreted, based
primarily on the multiple regression, suggesting that per-
sonality groups do differ in their physiological correlates
of involvementé-specifically as evidenced by ESC change
scores. A clear pattern of a decreasing number of ESC
responses were found in vignettes with which repressers
became involved; no such clear pattern was found for the

other personality groups.

ELECTRICAL SKIN CONDUCTANCE AND CARDIAC RESPONSES
OF REPRESSERS, MIDLINES, AND SENSITIZERS TO
AFFECTIVE STIMULUS FILMS

BY

c 0“
Yonkel Goldstein

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Counseling, Personnel Services
and Educational Psychology

1976

To Mom and Dad

To paraphrase Sholem Aleichem -
if I reach higher than my parents
it's only because I'm standing on
their shoulders.

ii

ACKNOWLEDGMENTS

Were it not for these people, this dissertation
would never have been written.

Norm Kagan - who in no small way has made my stay
here such a productive one. To him I owe thanks for a
million different things: for encouraging me with my
first clients, fdr trusting me with his IPR, for provid-
ing financial support, for giving me free reign in his
lab, and most of all for treating me with such dignity
and respect.

Maryellen McSweeney - who has such a rare combina-
tion of technical competence and dedication to teaching
that she has become, for me, a model to be emulated.

Tom Adams - who always gave unselfishly of his time
and knowledge about personality types and assessment.

Bob Wilson - who always allowed himself to be cornered
somewhere between his office and mine and whose clear think-
ing significantly helped this sometimes-befuddled experimen-
ter.

Joe Kertesz - for his technical assistance, for giving
me access to his students and mostly for his warm support.

Peggy Stansberry - who warmed up the machines, sub-
stituted for me when I was called to the phone, and most

iii

of all who kept me company during some very long days.
All the Winter Term IPR students - who showed up
on time (more or less), let me stick electrodes on them,

took my tests, and showed a real interest and excitement

for what I was doing.

iv

TABLE OF CONTENTS

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

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

CHAPTER

I

II

III

IV

THE PROBLEM . . . . . . . .

Introduction
Need

Purpose
Hypothesis
Theory
Overview

REVIEW OF THE LITERATURE . . . .

Heart Rate
Electrodermal Activity (EDA)
Summary and Conclusions

WTHODOLOGY O O O O O O O 0

Physiological Measures
Psychological Measures
Stimulus Films

Sample

Procedure

Design

Experimental Hypotheses
Analysis

Summary

MALYSIS O O O O O O O O 0

Analysis of the Results
Summary

SUMMARY, CONCLUSIONS, DISCUSSION AND
'IMPLICATIONS . . . . . . . .

Summary

Conclusions

Discussion

Subjective Observations
Implications for Future Research

Page
vii

viii

. 49

. 90

APPENDIX . O C O O O O O O O O 0

Instructions for Rating Scale

B Memory Test

C Background Sheet

D Vignette Typescript

E Procedures for Operating the Polygraph
BIBLIOGRAPHY . . . . . . . . . .

vi

Page
105

113

LIST OF TABLES

Table Page

3.1 Mean Involvement Ratings and Analysis
of Variannce . . . . . . . . . . . 54

4.1 Descriptive Statistics for Physiological
Variables . . . . . . . . . . . . 79

4.2 Individual Cell Values . . . . . I . . 81

4.3 Multivariate and Univariate F tests Ho:l . . 82
4.4 Multivariate and Univariate F tests H0:2 . . 83
4.5 Multivariate and Univariate F tests HO:3 . . 84
4.6 Multivariate and Univariate F tests H0:4 . . 85

4.7 Cell Values Used for Testing H :5 . . . . 86

0
4.8 Multiple Regression Analysis . . . . . . 86
4.9 95% Confidence Intervals for Beta Weights . . 86

4.10 Simple Regression of Involvement on ESCC
by Group . . . . . . . . . . . . 87

vii

LIST OF FIGURES

Figure Page
3.1 Design Diagram for MANOVA . . . . . . . . 72

5.1 Records of an Involved and Uninvolved Vignette
for a Represser and a Sensitizer . . . . . 101

viii

CHAPTER I

THE PROBLEM

Introduction

 

Human behavior, in one form or another, makes up a
part of the subject matter for a myriad of academic dis-
ciplines ranging from biology to aesthetics. However,
traditionally in academia the various sub-specialties have
had little to do with each other, as if those who were
concerned with the appreciation of beauty had no object of
study in common with those who sought to unravel the mys-
tery of the optic nerve. Increasingly, as the arbitrary
boundaries have become obscured, the need has been felt
for inter-disciplinary study, and hybrid fields have devel-
oped; psychoPhysiology is an example of such a case. Of
course, in the case of psychophysiology, both "parental"
fields are themselves quite diversified and, predictably,
this "inter-marriage" occurred between the most compatible
parts--experimental psychology and that part of physiology

which deals with intact homo sapiens. However, once the

 

precedent had been set, it was only a matter of time until
the "second cousins," such as personality, clinical and

counseling psychology, began to view their particular fields

from a physiological perspective. (On the other side of

the fence, of course, neurological and endocrine researchers
began to look past their dissected Specimens, speculating
about complex, living humans.) The advantages of this
academic intercourse seem obvious--the broadened perspec-
tives can lead to new conceptualizations and fresh strategies
in attacking old, persistent problems. The pitfalls, unfor-
tunately, are staggering: there is twice as much basic
information to be assimilated, there are "translation"
problems from one discipline to the other, and special measure-
ment problems which come from adding apples and oranges.

This study is one that belongs to that second set of
"inter-marriages"--specifically between personality psychology,
based on clinical assessment,and psychOphysiology; it is a
piece of research concerned with translation. Physiologists
have for years dealt with "orienting reflexes" and "defensive"
reactions; psychophysiologists have conceived of various
physiological patterns as being geared towards receiving
information from the environment or shutting it out. Person-
ality psychologists have classified pe0ple as being introverted,
extroverted, defensive or open. Psychotherapists spend a
great deal of energy in trying to assess when an individual
is being defensive. On an intuitive level it would seem that
there should be a fair amount of overlap between many of these
concepts. If some equivalence could be established and
specific patterns identified, this could both help to clarify

a part of the relationship between mind and body and be a

helpful indicator to the practising psychotherapist. This
study is in the area of interdisciplinary translations. In
its broadest sense this study is an examination of the rela-
tionship between the psychOphysiologist's environmentally-
oriented organism and Specific defense patterns known to

personality theorists.

Need

 

The general need which this study serves is that of
obtaining more data about the relationship of the "mind" to
the body. Socrates (Plato, c 390 B.C.) spoke of a body and
soul, emphasizing the latter because it is that which gives
humans the ability to discern good from evil. Plato was a
dualist seeing the soul as an independent entity attached
to the body for the extent of the human life. Aristotle
(c 340 B.C.), on the other hand, believed that body and
psyche were indivisible. Hundreds of years later Descartes
(1649) amplified on the concepts of Aristotelian psyche and
body, but changed the emphasis so as to separate them. He
considered mind and body to be totally separate, picturing
mind and body to be entities in two distinct worlds,with
the body being subject to natural laws, while the mind was
free of such constraints and was alone capable of exercising
free will. The immediate problem posed by this dualistic
conception of reality was how--if at all--did the two enti-
ties interact? Some, such as de Malebranche and Guilincx,

denied that there could be any such interaction (Watson, 1968).

Hobbes (1651) took an extreme materialist position, ex-
plaining everything in terms of molecular interaction.
Spinoza (1677), in contrast, took the view that "the order
and connection of ideas is the same as the order and con-
nection of things." He admitted that mind and body
correlated with each other, but denied that events in

one sphere could cause things to happen in the other. This
view was later expounded upon by Fechner (1966) who viewed
mind and body as differing aspects of one unity. He sought
to quantify the relation of the two in order to show that
increases in mental sensation correspond to proportional
changes in physical stimuli. Nietzsche took an even more
extreme view saying, "Body am I entirely, and nothing more;
and soul is only the name of something in the body"
(Edwards and Pap, 1965).

The above is not intended to reduce to one paragraph
the work of some of the greatest minds in the Western world.
It is, however, an attempt to establish the fact that the
relation of mind to body has been a matter of major intel-
lectual concern for hundreds of years. This study will
proceed from the philosophical position that there is a
relationship between mind and bOdy; it is an attempt to
specify one part of that relationship.

There is a second, more practical need which this study
bears upon. Psychotherapists spend a great deal of energy
trying to be aware of, and making their clients aware of,

defenses and resistances. Freud (1924) wrote:

A vehement effort must have been exercised to prevent
the mental process in question from penetrating into
consciousness and as a result it has remained uncon-
scious; being unconscious it had the power to construct

a symptom. The same vehement effort is again at work
during analytic treatment, opposing the attempt to

bring the unconscious into consciousness. This we per—
ceive in the form of resistances. The pathogenic process
which is demonstrated by the resistance we call REPRES-
SIGN.

Having thus said that resistance to therapy is a mani-
festation of a person's defenses (ie. repression), he under-
scores the importance of resistance.

Every step of the treatment is accompanied by resistance;
every single thought, every mental act of the patient's,

must pay toll to the resistance and represents a compro-

mise between the forces moving towards the cure and those
gathered to Oppose it.

Neo-Freudians such as Reich (1933) have devoted great impor-
tance in their therapeutic technique to dealing with these
resistances and defenses. If physiological patterns indica-
tive of defensive orientations could be found, it is possible
that this pattern could be useful in makingboth clients and
therapists aware of the defensive posture, thereby improving
or accelerating therapy.
Archer 3E El: (1971) wrote:

If one of the basic goals of counseling and psychotherapy
is to help clients to become aware of the incongruities
between what they perceive and feel, on the one hand, and
what they are willing or able to actually acknowledge and
admit to themselves on the other, then it seemed logical
and desirable to provide a client not only an opportunity
to study his physical and verbal reactions to a real or
simulated interpresonal engagement but to literally per-
mit him to see what his internal responses were as well...
it seemed reasonable that provision to the client of
measures of his physiological responses during the coun-
eling session would aid him even more in the process of
self-discovery and change.

If one accepts this reasoning it seems logical that
it could be potentially of even more benefit to the client
to have some interpretive help in understanding his physio-

logical "read out."

Purpose

There are two major purposes of this study. The first
of these is to test and extend the original psychophysiolog-
ical work differentiating heart rate and skin conductance
patterns in organisms attending to and shutting out the
environment. The original work, done primarily by Lacey and
his associates, in the case of heart rate, and Edelberg,
in the case of electrodermal activity, primarily used
psychophysical stimuli such as photic flashes and cold pres-
sor tests; it is unclear to what extent the results can be
generalized to vastly more complex interpresonal stimulation;
this study will attempt such an extension.

The second reason for doing this study is that it will
clarify whether, in this area, the personality psychologist,
who talks about "styles of interpersonal coping," and the
physiological psychologist, who talks about "externally" and
"internally-oriented organisms" are simply speaking two diff-

erent languages or are really referring to unrelated phenomena.

Hypothesis

 

It is expected that there will be a moderate positive

correlation between physiological patterns indicative of

openness toward the environment and the level of involve-
ment in brief scenes in which actors give a variety of
messages to the subject. It is hypothesized that there
will be a low positive correlation between scores on the
Byrne Repression-Sensitization Scale and physiological
response patterns. It is hypothesized that there will be
a significant interaction effect in immediate involvement
between physiological response patterns and sensitizer-
represser classification on the Byrne scale, ie. a sensi-
tizer's physiological responses when involved will differ

from the involved physiological responses of a represser.

Theory

This study bears directly on two separate theories--
a psychological one concerning defense patterns and a
psychophysiological one concerning cardiac and electrical
skin conductance response patterns. The theoretical im-
plications from personality theory will be considered first.

According to psychoanalytic theory a neurotic conflict
occurs when there is "a tendency striving for discharge and
another tendency that tries to prevent this discharge"
(Fenichel, 1945). It is to defend the ego from a realiza-
tion of the abhorrent instincts of the id that defense
mechanisms come into play. In that sense all defense mech-
anisms are restrictive in goal--they seek to keep something

from the ego. Fenichel (1945) writes:

In fainting, the organism, which has been overwhelmed

with too intense stimuli, shuts out the influx of

further stimuli. The complicated defense mechanisms

of the psychoneuroses may be looked upon as partial

faintings. In fainting, all perception is blocked;

in repression, some selected perceptions are blocked.
He goes on to say:

Neurotic characters, instead of reacting to their exper-

ience adequately, respond more or less rigidly with the

same reaction patterns. They are not only fixated to

certain levels of instinctual demands but also to cer-

tain mechanisms of defense.

Various typolOgies have been employed to characterize
typical response patterns. Horney (1937), for example,
speaks of four basic means of protection--securing affection,
being submissive, obtaining power, and withdrawing.

Byrne (1964) sees a continuum of defensive behaviors
which at one extreme seek to avoid anxiety-producing stimuli
(eg. repression, denial) and at the other seek to control and
manipulate the stimulus, thereby avoiding the anxiety (eg.
intellectualization, obsessive behaviors, and rumination).
Behaviorally, the former involve an avoidance of the stimulus,
while the later are characterized by an approach which dis-
torts; they are all defense mechanisms in that they seek to
reduce the ego's experience of the anxiety.

Other theorists, examining different concepts have
described a dichotomy similar to the one postulated by Byrne.
Hudson (1966), discussing differences between "convergers"
and "divergers" says that the former's “defenses lie in the

direction of inhibition," while the latter's "defenses lie

in a subtle distortion of true openness."

 

Byrne (1964) reviews a host of studies which find differ-
ential recognition thresholds or differential accuracy
scores for a variety of neutral and conflictual stimuli
including words, drawings, Blacky pictures, and tape-
recorded sentences. In these studies ease of recognition
was varied by using different exposure speeds, illumination
intensities, loudness of recordings, or by using successive
carbons to decrease legibility or by increasing background
White noise. A variety of defense indicators were used
in&21.uding readiness to verbalize conflictual material on
tilei 'TAT and sentence completion tests, ability to recall
failures vs. successes in an experiment, recall of Blacky
pictures, California F scale scores, ratings of Rorschach
iarldi Machover Figure-Drawings, and psychiatric evaluations
beifiied on interviews and case histories. The general find-
:irlsys of all these studies were that subjects who perceive
t1‘lreatening stimuli as accurately or more accurately than
11$utral stimuli respond in other situations with approach
behavior, intellectualization, and sensitization. Those
‘Vrlo perceive conflictual stimuli less well than neutral
£3timuli tend to repress and avoid in other situations.

It is important to note that according to Byrne's
t11'1eory both the repressing and sensitizing ends of the con-
‘tsznuum are considered to be pathological. Byrne (1964)
whites:

If each end of the repression-sensitization continuum

represents an extreme of the respective defense modes,
scores on the R-S scale would be expected to have a

n

r"? r)

r1

 

AA

in»-
"fight
.

.

IIA’.

“div.
.

'0‘
a

Q)
(I)

n u
as...“
'1.-
l

1“ v
flu“ i

n,

x T.
h...°

:‘n
51‘

)5
l’\

".

'V‘J

!

h' “I
O
h.‘

“T"a
6.5:

 

10

curvilinear relationship with various indices of psy-
chological adjustment. Neither over-intellectualiza-
tion of conflicts nor denial of them should result in

optimal adjustment .
In summary, then, the relevant points from person-

ality theory discussed here are: 1) there exists a class

<3f Iaehaviors--defense mechanisms--which block the ego's
Perception of anxiety-arousing stimuli, 2) peOple often
'adC>E>t typical response postures with rigid reliance on a
JPaJthicular set of behaviors being indicative of neurotic

adj ustment, and 3) one particular way of conceptualizing
these behaviors is along a continuum from always controlling,
approaching, and distorting the stimulus to another extreme
of always avoiding the stimulus.

The second theoretical set of underpinnings this
est11:1dy bears upon is the relationship betweeen emotional
Eirldi physical events in the human body. The first point in
this theory is that there is a relationship. There is ample
E3\ridence for such a connection ranging from the subjective
feelings such as "butterflies in the stomach," headaches,

61:15 physical tenseness reported by individuals in response
i=<> stressful situations, to more formal academic findings
EiEMmciating physiological changes with a variety of emotional
sgituations. In his review of the field Oken (1967) cites
‘SBIJbstantial evidence showing such diverse biological changes
Eiss the secretion of l7-hydroxycorticosteroids by the adrenal
'c=<3rtex, aldosterone secretion, catecholamine (ephinephrine
'E‘IId norepinephrine) secretion, and a variety of autonomic

‘Tfiasponses such as differences in muscle tension, heart rate,

1
Mr-
UVH‘I

I..'

in)
O

6

 

K I)
i it.

A»
I

‘-A
'v

”A

WI

'r—

(I)

“‘5

13‘

 

ll

systolic and diastolic blood pressure, respiration, skin
conductance, and finger temperature in response to changes
in the psychological environment.

Unfortunately, the next step in building this theory--
determining the precise relationship between the physiolog-
ical and psychological variables--is not nearly as clear-cut.
Indeed, Dykman gt 31. (1968) write: "Psychophysiological
research by competent investigators is remarkable for re-
Su-‘|-‘t:.s which are inconclusive, contradictory, or paradoxical--
thus attesting to the complexity of the data and the diffi-
cality of posing soluble problems." Although philosophers
have dealt with this problem at least from the time of
Plato, modern psychology only began addressing this problem
with the work of William James (1884) who posited that
prllzsiological states were evoked directly by affective stim-
uli and that what the individual experienced as emotion was
due to his own processing of bodily states. Cannon (1927)
cllallenged this notion with his demonstration that partially
8evering the central nervous innervation of the viscera
did not eliminate emotional behavior. Cannon hypothesized
3- chain of communication going from the viscera to the thal-
a~Il‘tus, to the hypothalamus and, from there, to the cerebral
<=<Drtex and the viscera. In this system he posited that the
QQrtex would play an inhibitory role over the lower centers.
AS Lang (1971) traces in his sununary of psychophysiological
1lilstory, the basic "chain of command" hypothesized by Cannon

was later expanded by Papez, who showed that the cingulate

 

12

gyrus served as the cortical reception center from lower
brain impulses, and by MacLean, who suggested that the
amygdala is of major importance in the parasympathetic
regulatory system and that the hippocampus is of primary
importance in sympathetic, emotion activity. Lindsley
(1951) suggested that the degree of cortical arousal was
responsible for the emotion which was felt. The problem
this notion faced was that there was only one dimension--
intensity of arousal--along which emotion could vary, and
that: made it very difficult to explain qualitative differ-
enCzes in emotion. It was as if the theory could account
for feeling slightly peeved or very furious, but could not
deal with the difference between anger and sadness. Hebb
(1955) remedied this theoretical problem by proposing a
tVITO-factor model. The first factor is one of arousal--
identical to Lindsley's theory; the second factor was what
he called a "cue" function which gave the aroused feeling
a behavioral direction. For example, an emotion such as
fear was thought to be made up of a high arousal component
aJud a cue component picking avoidance as the direction. It
is significant that prominence is given in this theory to
l'Ieurological as Opposed to visceral mechanisms. Schacter
( 1964) modified the two-factor theory somewhat, suggesting
that emotions are composed of an undifferentiated autonomic
pattern which is characterized by a cognitive state which
jLuterprets the emotion according to the context. In one

e><periment subjects were given epinephrine injections

13

which resulted in intensifying whatever emotion was appro-
priate in the social context (ie. those watching comic
movies found them funnier, those with a hostile experi-
mental stooge became more hostile, and those with a playful 7
accomplice were more playful) (Schacter -& Wheeler, 1962) .
However, this data can be seen to reflect only part of the
picture, for if one examines the work of Ax (1953) one finds
a clear differentiation between visceral reactions to exper-
imentally-induced fear and anger. Specifically Ax reported
that diastolic blood pressure increases, heart rate decreases
and the frequency of skin conductance responses were greater
in the case of anger than fear, while the magnitude of skin
conductance increases and respiration increases were greater
in the case of fear than in the case of anger. In light of
this: one must question Schacter's "undifferentiated auto-
nomic pattern."

It is this general arousal theory of emotion with its
1atter variants which form the springboard from which various
Other theorists take off, either seeking to confirm or dis-
credit the arousal theory. There are two general strategies
employed: one--devising two or more different emotional
sitnations and with sophisticated instrumentation recording
a Wide range of physiological variables seeking to show
differing reactions--is best illustrated by Ax's work (1953)
referred to above, while the second is illustrated by the
w°rk of Lacey (1959) and involves finding some general char-

acteristic of emotion which accounts for some percentage of

Ava:
Vein

Afl’n
EV“.

.
"l'n

ND.

‘95
o...
I

II A
bah
A»...

Q
a:-
0".

0y
9’ I A
9.5“

'1‘
w

I:
(I.
(n

.
’I
up

D» A- _
(I,

F:
"-

 

14

the physiological variance not accounted for by arousal
theory. The particular characteristic Lacey has focused
on is the external versus internal orientation of the
organism. He reports monitoring heart rate and palmar skin
conductance while subjects were engaged in four tasks:
visual attention, empathic listening, thinking, and endur-
ing physical pain (Lacey, 1959) . In the first two it is
necessary to orient one's attention outwardly; in the
latter two it is helpful to shut out outside input. His
findings were that skin conductance increased across all
four tasks; however, heart rate accelerated while the person
was thinking of experiencing the physical pain and deceler-
ated while he was attending visually or listening empathic-
ally, Support for this idea that the heart rate either
increases or decreases depending upon whether the organism
is Seeking to attend to or reject the external stimulation
is most clearly seen in Davis' (1957) work presenting
Pietures to male subjects. He identified a "P" pattern
WhiCh consisted of considerable sweat gland activity, per-
ipheral vascoconstriction, and a decreasing heart rate;
this pattern was most reliably obtained from male subjects
vieWing female nudes. However, when the visual stimulus
was a starving man the cardiac response was reversed.

Lacey (1959) considers these findings to be devasta-
ting to any concept of emotion based solely on arousal
beClause of the difficulty in simultaneously explaining .in-

creasing arousal, as evidenced by the increasing skin

 

15

conductance (indicative of increased sympathetic activity)
and decelerating heart rate (indicative of decreased sympa-
thetic activity, increased parasympathetic activity or both).
He goes on to hypothesize that these cardiovascular changes
have a direct effect on central nervous system activity.
An increase in pulse rate is picked up by pressure-sensitive
receptors in the carotid sinus, the aortid arch, and possibly
in the heart itself which, by way of Hering's nerve, inhibits
cortical electrical activity and spinal motor-neuronal acti-
Vity (Callaway & Dembo, 1958) . Lacey and Lacey (1970) cite
evidence that direct stimulation of Hering's nerve or an
increase in pressure in the carotid sinus decreases muscle
tone in an anesthetized animal. The distension of the caro-
tid sinus causes cortical inhibition as evidenced by the
Shift from low voltage fast activity to high voltage slow
wave activity. As pressure on the carotid sinus increases,
response thresholds also increase. (The contention that
incBreased cardiac activity has this inhibitory effect on
att-‘-ention has been questioned by Obrist e_t_:_ _a_1_. (1965) who
haVe shown that pharmacologically blocking this pathway does
not; change performance as expected.)

From a somewhat different perspective Edelberg (1972)

joins the attack on the general arousal theory of emotion.
He

he

points out that electrodermal activity is considered to
an indication of sympathetic activity which is interpreted
to be an equivalent of arousal level. This interpretation

13 based on what Edelberg terms an "unwarranted" assumption

A.“

I.

U.

(I!

.i

 

16

of a direct relationship between sympathetic activity and
behavioral correlates. (Furthermore, as Lang (1971) points
out, measures of electrical skin activity provide an atypical
picture of sympathetic activity in that the sympathetic
synaptic transmitter is typically norepinephrine, while in
the case of sweat gland innervation the synaptic transmitter
is acetylcholine.) To further complicate the already be-
clouded physiological picture Davis (1957) writes, "the
parasympathetic nervous system would now have to be recog-
nized as capable of producing excitatory states..." Support
for this statement can be found in Gellhorn's work (1968,
1969 , 1970) which classifies individuals as having a domin-
ant sympathetic or parasympathetic system. The propensity
for increased sympathetic activity--ca11ed ergotropic tuning,
is associated with cortical desynchronization, high muscle
tone, and behavioral arousal, while the tendency for increased
Parasympathetic activity--known as trophotrOpic tuning--is
a380ciated with cortical synchronization, decreased muscle
tone and sleep. The important point here, though, is that
at; extreme levels of either ergotrOpic or trophotropic
tunings one finds reversal phenomena. For example, in an
e"'Qotropic system, a novel stimulus evokes an increase in
alpha (a trophotropic response) instead of the expected
cortical desynchronization.

Other_evidence which can be considered damaging to the
idea of equating sympathetic activity with psychological

a‘Z’Ousal is Martin & Edelberg's (1963) demonstration that

 

17

electrodermal activity may be diminished when subjects show
considerable irritation. Along the same lines, Johnson's
and Lublin's (1966) finding that electrodermal activity
can be greater during stage four sleep than when the same
subject is performing a task in the waking state, points to
a more complex phenomena than the simple equation of elec-
trodermal activity and general arousal. Edelberg (1972)
concludes that it is inapprOpriate to speak of increased
nervous activity or heightened arousal; he prefers to try
to be more specific, speaking of cortical, amygdaloid,
hippocampal, or pontine activity and, behaviorally, of
defensive arousal, cognitive-activation, and alertness for
information. He postulates that electrodermal activity is
an index of reticular activation.

Whatever the precise relationship of electrodermal
activity to neurological and psychological phenomena, it
does have some very clear physical effects. Edelberg (1972)
gives the following list of possibilities:

(a) thermoregulation to compensate for cutaneous vas-
coconstriction, (b) thermal preparation for a heat
load of muscular origin, (c) the adjustment of physical
properties to favor manipulative contact, (d) the ad-
justment of physical properties as a defense against
abrasion, (e) the enhancement of tactual acuity either
mechanically or neurochemically, or (f) the secretion
of a characteristic odoriferous substance that facili-
tates tracking by other members of the species.
Reading over the list one cannot help but to be impressed
by the possibility that electrodermal activity is well-

Suited to affect the interaction between the organism and

the environment. Martin and Edelberg (1963) have explored

18

this possibility in a limited way by having subjects whose
galvanic skin response was being monitored indicate by press-
ing a key their positive and negative feelings as they were
listening to tape recorded readings. They found that when
the subjects expressed negative feelings ("irritation")

via pressing the key, the galvanic skin response was dimin-
ished. They expected, but failed to find, an increase in
Galvanic skin response activity associated with positive
feelings. Thus, this study only partially supports the
hYpothesis that galvanic skin response is part of a screen-
ing mechanism employed by the organism. The authors
suggest that this failure to confirm fully the hypothesis
might be due to the low valence attached to what they con-
Sidered positive readings. The implication here about the
relationship of galvanic skin response activity to psycho-
logieal involvement is similar to Lacey's (1959) report of
work by Furer and Hardy where after repeated presentation

of a painful stimulus, the galvanic skin response habituated,
mm was made to reappear when the situation varied slightly.
When the subject is attending galvanic skin response activity
appEErs to be present.

This whole problem has been approached from a slightly
different perspective, primarily by Russian investigators.
BerJ-Yne (1966) discusses Pavlov's early work with the
OrielIting reflex (OR). Pavlov's dogs would cease any on-
going activity, turn their heads, and look toward the source

of any unusual event. Pavlov classified this as an

19

unconditioned reflex, which was subject to extinction and
disinhibition much like any conditioned reflex. Sokolov
(1963) has extended the notion of this response to novel
stimuli by positing a generalized response system, known
as the orienting response, which through its autonomic
feedback 100ps facilitates peripheral receptor sensitivity.
He postulates that a neuronal model of a stimulus is built
up in the central nervous system after repeated presentation.
The smaller the discrepancy between the neuronal model and
the real-life stimulus, the smaller the OR. Opposed to this
OR.he posits a defensive reflex (DR) which is also a gener-
alized response system with feedback mechanisms, but is
distinguishable from the OR in that with repeated stimula-
tion OR's habituate, whereas DR's do not habituate and may
even become stronger. Furthermore, OR's are elicited by
low intensity stimulation, whereas DR's are elicited by
high intensity and/or noxious stimulation. On the basis then
of this Pavlovian-Sokolovian theory, it can be predicted
that if a person were "Open" to external (non-defensive)
stimulation they would orient towards a stimulus and elec-
trodermal activity would diminish; conversely, if the subject
were closed to such stimulation, electrodermal activity should
maintain steadily or increase.

Up until this point research has been cited which
suggests that both electrodermal and cardiac activity are
in some way related to the organism's attending to the ex-

ternal environment. Darrow (1929) presents evidence which

20

has to do with the inter-relationship between these two
variables, suggesting that a high pulse rate's inhibitory
effect on cortical activity can be overcome by the excita-
tory effect of skin conductance activity. They showed
that palmar electrical skin conductance (ESC) increases
were associated with alpha blocking, while blood pressure
increases were negatively correlated with alpha blocking.
They had subjects perform a task which involved attending
to lights and found that ESC and heart rate increases were
associated with false responses. They concluded that skin
conductance activity, which is excitatory, can overcome the
inhibitory effects of a high heart rate. However, when heart
rate is low, the organism is already in an excitatory state,
and skin conductance activity has no additional effect.
Dykman EE,§£- (1968), on the basis of work with psychiatric
patients and normals describe four basic patterns. When
examined for cardiac and electrodermal activity components,
the patterns are: l) alerting - moderate to large skin
resistance decreases, moderate heart rate increases, 2)
closed - negligible or no skin resistance decreases, moderate
heart rate increases, 3) cpen - moderate galvanic skin
response and no difference or decreasing heart rate, and
4) non-responding - no change in either direction.

Thus, whether or not the organism is opened or closed
to its environment should account for some percentage of
the variance of these measures above and beyond that accounted

for by simple arousal. One should keep in mind, however,

“qt

Had»

n A 9

$39-
C

I V
In.

id.

‘ I
”.5”
h...

‘
Ion...
abdi

’5:"
may

'r!

21

that a major function of these biological systems concerns
vegetative bodily processes; therefore, one expects small,
but significant amounts of the variance to be associated
with these variables.

In summary, then the physiological theory which this
study is based on comes from two separate sources: 1) a
reaction against the classical arousal theory of emotion and
2) an examination of Pavlovian orienting responses. Logical
arguments from both theories predict that cardiac and elec-
trodermal responses should differentiate internally vs.
externally oriented postures. The question addressed in
this study is whether there is a relationship between the
psychophysiological theories summarized here and the person-

ality theory summarized earlier.
Overview

In Chapter II, the literature most releveant to
relating psychological defense to physiological responses
is reviewed. Methodology, instrumentation, procedures,
analysis techniques, and research hypotheses are described
in Chapter III. The results of the study, with the appro-
priate statistical analyses applied are given in Chapter
IV. In Chapter V the conclusions and implications of this

study are discussed.

CHAPTER II

REVIEW OF THE LITERATURE

The purpose of this chapter is to review the pertin-
ent findings regarding the two physiological variables
most relevant to this study--cardiac and electrodermal
activity (EDA)--paying particular attention to how they
might relate to psychological defensiveness. These two
variables were picked because the literature seems to im-
plicate them, more than any other, vis-a-vis subjects'
openness and closedness to environmental stumuli. The
best way to proceed is to divide the chapter into two parts--
the first reviewing the literature pertaining to the inter-
pretation of heart rate and the second reviewing the liter-

ature concerned with electrodermal activity.
Heart Rate

Lacey's (1959) contention that heart rate decelerates
when the organism is oriented externally and that it accel-
erates when the organism is oriented internally has already
been presented in Chapter I. The literature is replete with
examples which confirm his essential findings, although at

times, other conclusions are drawn from the data. In their

22

23

review of the literature concerning heart rate, Graham and
Clifton (1966) conclude that pulse rate deceleration is a
part of the orienting response and that pulse rate accel-
eration is part of a defense reflex. Porges and Raskin
(1969) found that pulse rate decelerates when subjects
attend to a light, while when they try to estimate their

own heart rates, the pulse accelerates. Hodges and Speil-
berger (1966) measured heart rate of high and low anxious
subjects under two conditions--threat of shock and no threat.
They found that heart rate acceleration was greater in the
threat condition, supporting the contention that heart rate
acceleration is part of a defensive reaction; furthermore,
those subjects reporting moderate to high fear of shock two
months before the experiment exhibited greater cardiac
acceleration than those reporting low fear of shock. In

a similar study Hare and Blevings (1975) divided female sub-
jects into two groups, those who were afraid of spiders and
those who were not. They found that in the fear-of-spiders
group heart rate accelerated when presented with slides of
spiders, while in the not-afraid-of-spiders group there was
no significant acceleration. Israel (1969), studying the
differential cardiac reaponse of "levelers" and "sharperers"
(among other characteristics "sharepeners" tend to be very
attentive to external detail and focus on differences between
object while "levelers" tend to make global judgments, are
less attentive to external detail, and tend to focus on sim-

ilarities between objects) found marked cardiac deceleration

24

in "Sharpeners" when attending to stimuli as compared with
"levelers." Elliott gt 31. (1970) found that heart rate
decelerates as the difficulty of an externally oriented
task increases, while heart rate does not decelerate in
re8ponse to shock. Obrist (1963) found a deceleratory heart
rate in response to continuous environmental input even
though sympathetic tonus (as evidenced by decreasing skin
resistance) was increasing. Malmo and Davis (1956) found
that as heart rate increased, performance time in a mirror-
tracing task decreased. Bull and Lang (1972) found that
cardiac response was a function of both the subjects' abil-
ity to discriminate stimuli and the physical intensity of
the auditory stimuli used.

Lang and Buss (1965), reviewing literature concern-
ing deficits in schizophrenics, cite evidence that schizo-
phrenics, as compared with normals, tend to exhibit an
increased level of cardiac activity. If one assumes that
schizophrenics tend to shut out a greater proportion of the
external environment than normals, this finding is in accord-
ance with Lacey's theory. (One important qualification to
remember in modern studies comparing schizophrenics with
other groups is that the effect of phenothiazines, the class
of medications most commonly prescribed for schizophrenics,
is to increase heart rate and decrease heart rate variability
and electrical skin conductance (Gruzelier and venables,
1975). Spohn 33 El. (1971) report that when schizophrenics

are taken off medication, heart rate remains elevated.)

25

Venables (1966) writes, "the general consensus of opinion
is of a diminished responsiveness of the autonomic nervous
system in the chronic schiZOphrenic." The finding that
schizophrenics have elevated heart rate levels is consis-
tent with this because of Wilder's Law of Initial Values
which states that the response of an autonomic variable
decreases as the prestimulus level increases (Wilder, 1967).
However, Gray (1975) found that schizOphrenics decreased
their inhibition of anticipatory responding as signal in-
tensity increased, while controls did the opposite. From
this and reports he cites in the literature, he questions
the position that schizophrenics exhibit a diminished
autonomic re3ponsiveness due to an elevated physiological
basal level. Thetford gt gt. (1974) failed to find heart
rate difference in groups of paranoid vs. nonparanoid
schizophrenics, acute vs. chronic schizophrenics, or groups
of those with good vs. poor pre-morbid adjustments.

Investigating this same matter of the relation be-
tween psychological coping ability and physiological re-
ponse, but from a different perspective, Roessler gt gt.
(1963) report a trend that as ego strength (defined as the
"capacity to integrate environmental adaptive demands with
psychological economy") increases, cardiac acceleration to
auditory stimuli increases.

Tan's data (1964), comparing anxious subjects with
normals supports this finding in that those with high anxiety

had higher resting pulse rates and mean pulse rates in

26

response to stimulation. Deane (1966) found that the heart
rate accelerated during the time the subject anticipated a
shock; the acceleration decreased during and after the time
the subject acutally received the shock. Schnell and
Catania (1975) required subjects to detect the threshold
level of a visual stimulus and found that heart rate decel-
erations during the "warning period" were greater on "hit"
trials than on "miss" trials.

Lacey and Lacey (1974) report attempts to study heart
rate when tasks involve both internal and external orienta-
tion. In one task which involved mental arithmetic (inter-
nal orientation) and signal detection (external orientation),
heart rates simply increased as would have been expected
had there been no externally oriented part to the task.
However, in another task, called "Rules of the Game" which
also involved both an internal and external orientation
(ie.-attending to outside stimuli and internally memorized
rules), they found that the heart rates could be described
as a "vectorial resultant" (or arithmetic average) of the
increase one would expect from an internally-oriented task
and the decrease one would expect from an externally-oriented
task.

In Lacey's (1959) early studies he found that heart
rate accelerated when a subject immersed a limb in very cold
water (known as the cold pressor test). This appeared to
be consistent with the idea that cardiac acceleration was

associated with shutting out input from the environment.

27

However, Obrist gt gt. (1965) have shown that in a classical
conditioning paradigm, heart rate will exhibit a biphasic
response with an initial acceleration and subsequent decel-
eration when a noxious UCS is presented. Furthermore, they
found that the initial heart rate increase is a respiratory
artifact. However, they conclude that this finding is not
necessarily inconsistent with Lacey's hypothesis in that
"...deceleration of heart rate is thought to facilitate
sensory intake. Therefore, the deceleration observed during
conditioning could involve similar processes and have a like
function." Thus it would seem that the critical difference
'between the direction of heart rate change in the cold
pressor test and the conditioning experiment is that the
latter was a learning situation where it was advantageous
for the organism to attend to the environment, while in the
former nothing was to be gained from any such attention.
Lacey and Lacey (1974) agree with Obrist's position that
"attention to unpleasant, aversive, and potentially threat-
ening stimuli may be biologically advantageous." "Unpleas-
antness," they conclude, "does not preclude the intention

to note and detect." In situations where attending to the
noxious stimulation accomplishes nothing, heart rate accel-
eration is found. For example, Gang and Teft (1975), using
dental drilling sounds as a stimulus, found greater cardiac
acceleration in subjects who rated the stimulus as being
more unpleasant.

A further criticism of Lacey's hypothesis is that heart

28

rate acceleration is an artifact of respiratory activity
(Westcott and Huttelocher, 1961). Wood and Obrist (1964)
report that with respiration controlled, the acceleratory
cardiac component was not found, while the deceleratory
activity was unchanged. Germana and Klein (1968) found only
a cardiac deceleration to an auditory CS when respiration
was controlled. Deane (1964) has presented contradictory
evidence to suggest that breathing is not totally respon-
sible for cardiac acceleration. However, even if it is
true it does little damage to that portion of Lacey's
hypothesis most relveant here; it says nothing about the
idea that deceleration is associated with increased exter-
nal attention. It seems that it is of crucial importance
to the physiological question of what system(s).mediate
cardiac changes, though it says nothing about the reliabil-
ity of those differences.

Campos and Johnson (1966) failed to find that an orien-
tation varible was significant in determining heart rate.
Instead, they found that the requirement to verbalize pro-
duced heart rate increases. This finding is suggestive of
the idea that it is the demand characteristics of a poten—
tial response and not differential orientation which is
responsible for the pattern that Lacey reports. Support
for this notion may be found in a study by Obrist gt gt.
(1970) in which subjects are required to perform a reaction-
time task while heart rate, electromyelograph bursts recorded

from the skin, and eye movements are recorded. Their findings

29

are that cardiac deceleration is associated with a gener-
alized inhibition of task irrelevant activities; in other
words, the deceleratory response is interpreted as being
due to the organism's inhibiting functions which are ex-
traneous to the given task rather than because the organism
has assumed some orientation posture. Elliott (1972),in
reviewing a number of studies designed to test Lacey's
hypothesis,finds Obrist's reliance on metabolic require-
ments to be the superior explanation of the actual phenom-
enon. Epstein EE.El° (1975), comparing heart rate and
electrical skin conductance, conclude that heart rate is
more reliably related to motor output than is electrical
skin conductance, whereas electriCal skin conductance is
more reactive to stimulus input than motor output. Lacey
and Lacey (1974) have responded to Obrist's and Elliott's
position In; agreeing that task-irrelevant motor activity
is probably inhibited in preparation for reaction-time trials.
However, they do not believe that this is sufficient to
account for the cardiac deceleration. They criticize
Obrist's sampling of other motor activity (chin muscle and
eye movements) as being too small to generalize to overall
activity level and suggest that muscles not sampled in the
research might have been increasing in tension. Furthermore,
they point to one of Christ's own findings (Obrist gt gt.
1969) that reports that when subjects were required to con-
tinuously tap their fingers while being conditioned to a

negative sitmulus, electromyelogram activity increased while

30

heart rate decreased. In addition, they refer to a study
by Chase gt gt. (1963) similar to the Obrist study except
that the reaction-time task was a leg lift rather than a
finger tap. This presumably increased the metabolic demand
placed on the heart. The finding was that because of the
increased metabolic demand there was a cardiac acceleration.
However, "superimposed on this response was a statistically
significant cardiac deceleration." Lastly, in support of
their position, the Lacey's point out that the metabolically
relevant variable is cardiac output--not pulse rate. They
quote the physiologist, Guyton:
...the decreased cardiac output is not proportional to
the decrease in heart rate because the degree of filling
of the ventricles becomes enhanced during the prolonged
diastolic filling periods, which increase the stroke
volume, and in this way offsets to a great extent the
decrease in output that one might expect.
Lacey gt gt. (1963) suggest another factor, involve-
ment, which is important in predicting cardiac response.
They found that the greater the childhood involvement with
scholastic tasks, the greater the cardiac response, as adults,
to problem-solving tasks. They concluded that as the motiva-
tional relevance of a task increases, so will an individual's
cardiac response. The importance of an involvement factor
is supported by Lazarus gt gt. (1963) who found that when a
task included a sense of personal involvement heart rate
increased even when the subject's attention was externally

focused, while when attention was directed externally without

a sense of personal involvement, heart rate decreased.

31

This review of the cardiac literature would give an
inaccurate picture if it gave the impression that all the
experimental data correSponded with some particular theory
of cardiac change. Klorman gt gt. (1975) divided female
subjects into groups of high and low fear with respect to
mutilation. Subjects were then presented with slides from
three categories: neutral (ie. someone in typical photo—
graphic pose), incongruous (eg. bald man with lemon attached
to him) and mutilation (ie. burns and accidents). As pre-
dicted by Lacey's theory the fearful subjects had cardiac
acceleration, while the non-fearful subjects had cardiac
deceleration, in response to the mutilation pictures. Both
groups exhibited cardiac deceleration to the incongruous
stimuli. Surprisingly, they also both showed heart rate
increases in response to the neutral pictures. This result
is not explicable in terms of any theory reviewed by the
present author.

A similar example of seemingly inexplicable experimen-
tal findings are those of Uno and Grings (1964) who found
cardiac deceleration associated with auditory tones of 60
and 100 decibels, but cardiac acceleration associated with
tones of 70, 80, and 90 decibles. Maximal acceleration
occurred 4-5 beats after the onset of stimulation, while
maximal deceleration occurred 6-7 beats after stimulus on-
set. Average acceleration was 2.5 beats; average decelera-

tion was 2 beats. No similar findings have been reported.

32
Electrodermal Activity (EDA)

Before launching into a review of the pertinent EDA
literature, it will be useful to clarify some of the ter-
minology used in reporting experiments. In the old liter—
ature the terms psychogalvanic reflex '(PGR) and galvanic
skin response (GSR) are used almost exclusively. Vigoroux
(1879) was the first to report electrical variations of the
skin. Féré (1888) discovered that if a small current were
run between two points on the skin a resistance to that
current appeared and that that resistance varied in re—
sponse to emotional stimuli. This is the classic galvanic
skin response; it is usually measured by placing a bridge
circuit between the subject and a recording amplifier,
thereby making the subject an arm of the bridge while a
variable resistor in the other arm is adjusted so as to
balance the voltage at some null point, thus allowing resis-
tance fluctuations in the subject to be recorded (Lang, 1971).
One should note from this description that the GSR differs
from cardiac reSponses, discussed earlier, in that the
cardiac response is generated entirely by the subject,'
whereas in the case of GSR there are resistance changes
generated by the subject in response to an electric current
being applied by the experimenter. There does exist a sub-
class of EDA responses--skin potential responses--which are
entirely generated by the subject (largely by sweat gland

activity) (Lang, 1971). Because the term GSR came to be

33

used (incorrectly) to denote almost any sort of EDA, modern
writers prefer to use terms which Specify exactly what is
being measured--electrical Skin resistance (ESR) and its
reciprocal, electrical Skin conductance (ESC). Both of
these, in turn, are divided into measures of level--refer-
ing to average value of baseline between responses--and
measures of response--referring to change of amplitude

from base level to peak (Edelberg, 1973). In this review,
where possible, the specific terminology will be used.

Where the term GSR is used, it was not possible to be more
specific about the measure, but in the context of the liter-
ature it seemed reasonable to assume that the term was being
employed in its original sense, as described by Féré. Often
measures of conductance are preferred over those of resis-
tance because they tend to be more normally distributed,

and thus more suitable for many parametric tests (Lacey,
1947; Lacey & Siegel, 1949).

The biOphySical significance of EDA is yet to be com-
pletely explained. Lang (1971) summarizes the following
facts regarding possible underlying events. The distribution
of sweat glands on the human body and where resistance
changes take place overlap to a great degree. He reports
correlations ranging from .4 to .9 between the number of
active sweat glands and resistance levels. This seems to
implicate the sweat glands as the major factor responsible
for ESR activity. However, the correlation between sweat

and skin resistance is low. Furthermore, skin resistance

34

changes appear just before sweat can be detected, suggesting
that it is the sweat gland activity, rather than the pre—
sence of the actual sweat which is important. The fact that
atropine, a drug which blocks cholinergic transmission (the
mechanism of sweat gland innervation) increases skin resis-
tance supports the importance of sweat gland activity.
However, another factor--the permeability of an epidermal
layer--has also been shown to be important in affecting ESR.

McCurdy (1950) writes:

From the earliest period of PGR experimentation it has
been asserted that magnitude of galvanometric deflection
is intimately associated with the subject's estimate of
the intensity of his experience--whether emotionally,
affective, conationaly, or otherwise. This clear state-
ment has been beclouded by Sundry differences of opinion
as to the exact quality of experience usually involved;
but so far as the actual evidence goes there has appar-
ently never been an instance in the literature when the
central issue has been in doubt.

Although EDA'S association with the "intensity" of experience

may not have been questioned, the precise nature of the

association has been hotly debated.

Raskin (1973) reviewed the general psychological Sig-
nificance ascribed to EDA. The most common position taken
is that of Duffy who emphasized the overall level of energy
mobilization as a major aspect of emotion and who saw skin
conductance levels as indicative of that energy level.
Flanagan (1967) cites studies which Show a strong relation
of GSR to startle, a study showing greater emotional response

to be associated with smaller GSR's, and a study showing

that adrenalin--secreted in emotional situations--diminishes

35

GSR,in an attempt to marshal evidence that indicates that
GSR is associated with attention and not emotion. He pre-
sents evidence from his own work, having subjects rate.
photographs for their emotional involvement and their
attention-getting value and finds support for his conten-
tion that attention is the best intervening variable to
explain GSR activity. He writes:
Experienced GSR researchers have repeatedly indicated
this conclusion. [that GSR is associated with attention
and not emotionl] However, those interested in personal-
ity have continued to interpet GSR as an index of emotion
of anxiety. One thing GSR cannot indicate is continued
strong emotion or anxiety because if adrenalin is involved,
the GSR is suppressed!
Johnson and Lubin (1966) found that there was more spontan—
eous EDA during sleep than when awake. Although caution is
urged about generalizing from sleeping to waking states, they
conclude that EDA is not an indicator of arousal.

Lazarus and Alfert (1964) found that Subjects who were
pre-disposed to deny affective disturbance, as judged by the
K scale on the MMPI, exhibited high ESC levels of reactivity,
in response to films depicting upsetting scenes. Conversely,
people who were classified as low deniers, exhibited low ESC
levels. Gordon and Glass (1970) classified female subjects
as repressers and sensitizers and had them make choices about
the color of a pen in two situations--one where their decision
was judged as being indicative of their femininity and one
where it was not. Their pertinent findings were that when

GSR was measured repressers were more active than sensitizers

physioloqically. Learmonth gt gt. (1959), using a variety

36

of Rorschach measures and MMPI scales reported that the
increase of palmar potential fluctuations of student nurses
during stress is negatively correlated with expressivity.
Conversely, the increase of palmar potential fluctuations is
positively correlated with restraint and curtailment of un-
pleasant or prohibited feelings and actions. Raskin gt gt.
(1969) presented varying intensities of white noise to male
students and concluded that the magnitude of the positive
wave of the skin potential response and its failure to
habituate is related to the occurrence of a defensive reflex.
Koepke and Pribram (1966) found that the median number of
trials for GSR habituation was 5.5 for subjects classified as
physiological "stabiles" and 19.0 for those classified as
physiological "labiles." If failure to habituate is taken

as evidence of a defensive reflex (Sokolov, l963),then this
can be taken as evidence that people who have labile GSR
activity are more prone to defensive, as opposed to orienting,
reactions.

Carrier and Orton (1964) monitored ESC during complex
learning by groups of bright, normal, and retarded children.
They found a trend for increasing Skin conductance across
the sessions for the normal and combined groups and interpret
this to mean that ESC will increase as interest and task in-
volvement increases. Kintsch (1965) showed that in a paired-
associate learning task GSR'S began to habituate after
learning was completed. The GSR'S were strongest in the

intermediate stage of learning. Berlyne (1966) showed that

37

GSR magnitude with repeated exposure to one or more visual
pattern decreases.

In one study which seems to be at variance with those
reported until now, a comparison was made of subjects who
had the most reactive and least reactive GSR recordings;
the results showed that the reactive subjects were dependent,
dreamy, idealistic, and suggestible while the non-reactive
individuals were cool, evasive, opportunistic, and independent
(Block, 1957). Block goes on to discuss GSR lability as an
index of anxiety or affect in a given situation. He sees no
relationship between directionality of response and GSR.
Jones (1950) also classified subjects as having high or low
GSR reactivity and then looked for psychological traits they
had in common. He described his highly reactive adolescent
subjects as less assertive, less animated, less talkative,
less attention-seeking, calmer, more deliberate, more good-
natured, more c00perative, and more responsible than the low
reactors who were more irritable, excitable, and impulsive.

Lader (1967) measured GSR'S in five patient groups
(anxiety with depression, anxiety, agoraphobia, social phobias,
and specific monosymptomatic phobias) and found that those
patients with Specific phobias had fewer spontaneous fluctua-
tions and habituated more rapidly to auditory stimuli.
Assuming that patients with specific phobias have less free-
floating anxiety than the others in this study (because it
is stimulus bound) this argues for high spontaneous activity

and resistance to habituation being considered indicative of

38

anxiety. Syz and Kinder (1928) found that depressed and
catatonic patients had high initial resistance and few GSR
responses to a variety of stimuli including words, a car
horn, and a pin prick, while paranoid schiZOphrenics had
frequent GSR'S, a low initial resistance, a large number of
Spontaneous fluctuations, and a clear decrease in resistance
during the testing session. The pattern obtained for normals
was one of low initial resistance, like with paranoid schizo-
phrenics, but with far less change during the experiement.
These results suggest that psychopathology is associated
with both ends of the GSR reactivity continuum, with adjust-
ment being located in the middle.

Another study suggestive of the notion that adjustment
is located between two pathological extremes on a GSR reac-
tivity continuum, compares the conditioning of GSR to
electric Shock in anxious, chronic schizophrenic, and normal
subjects (Howe, 1958). Anxious subjects had a greater response
magnitude than the other two groups (no significant difference
between the schizophrenic and normal groups was found in this
dimension, possibly because of the large error variance in
the schizophrenic group), while the chronic schizophrenics
were less reactive than the other two groups. It should be
noted that if this interpretation of Howe's data is correct,
it suggests that different components of EDA (in this case
magnitude vs. frequency of response) are correlated with
different psychological phenomena. The work of Edelberg and

Wright (1964), finding a sweat gland component and an

39

epidermal component of GSR activity, and the work of Darrow
(1964) finding different processes occuring at different
levels of arousal, suggest that attaching different Signi-
ficance to the various EDA measures is a legitimate reading
of the data. Thayer and Silber (1971) compared schizophren-
ics who had been taken off phenothiazines for two weeks
(not long enough for their systems to be drug free) and
normals. If both groups were equated for tonic (ie. long-
term) arousal level as measured by skin resistance, no
physiological differences between the two emerged. However,
they did find that those subjects, regardless of diagnostic
category, with high tonic arousal levels evidenced more GSR
activity than those with low arousal levels.

In their review of schizophrenic deficits, Lang and
Buss (1965) stress that of all the autonomic measures
taken, only skin resiStance suggests that schizophrenics are
underaroused. However, Zahn gt gt. (1968) found that compared
to normals, chronic schizophrenics had higher ESC levels,
greater frequency of non-specific GSR'S and a slower EDA
habituation rate. Bernstein (1970) Showed that "confused"
psychiatric patients exhibited diminished reactivity to audi-
tory stimulation of low to moderate intensity. Using anxious
patients and normal controls, Tan (1964) found that there was
a greater decrease in ESR level in the anxious patients than
in the controls. However, Katkin and McCubbin (1969) found
no relationship between the classification of subjects into

high and low groups on the basis of the Taylor Manifest

40

Anxiety Scale and high and low GSR lability groups. (Fenz
and Epstein (1965) have suggested that the Taylor Anxiety
Scale confounds two separate anxiety variables--an autonomic
and a cognitive one--and suggest that this is a reason why.
many studies using it fail to find predicted results.)
However, they did find that autonomically stabile subjects
habituated to low and moderate intensity tones, while labile
subjects habituated only to low intensity tones. They inter-
pret this in the light of Sokolov's theory (1963) suggesting
that stabile subjects exhibit OR's at both stimulus intensities,
while labile subjects exhibit DR's at the higher levels of
stimulation.

Van der Valk and Green (1950) reported increased elec-
trical skin resistance in patients with essential hypertension,
peripheral vascular disease, ulcers, colitis, and bronchial
asthma--all physical diseasesvfiifllsignificant psychosomatic
components.

Attempting to investigate this problem from a positive
rather than a negative perspective, Roessler gt gt. (1963)
investigated the relationship between ego strength, as
measured by the Barron ego-strength scale, and skin resistance
responses to auditory stimuli. They found "...the high Es
[ego strength] group will Show the greatest response, the
low Es the least response, and the middle Es group an inter-
mediate response. The greatest differences among Es groups
will be evident at threshold intensity of stimulation and

the least difference at maximum intensity."

41

Stennett (1957) found that an inverted U relationship
existed between palmar conductance and performance level
on an auditory tracking task, suggesting that the organism
is most open to the environment at moderate ESC levels.

Edelberg (1970) measured EDA recovery rate and found
that subjects with a fast recovery rate during a task exhi-
bited Slower EDA habituation; he interprets this to be
adaptive for goal-directed behavior. Subsequently, he inter-
preted similar findings to suggest that "enhanced electro-
dermal activity with retarded recovery may signal a defen-
sive reaction" (Edelberg, 1972). The same study suggested
a relationship between recovery time and skin conductance
level. However, Crider and Augenbraun (1975) showed that
those subjects who exhibited slow habituation of electrodermal
responses perceived signals more accurately than those with
fast habituation. On the basis of an analysis of the errors,
they suggest that this was due to a difference in the response
criterion employed by the two groups (with slow habituators
being less cautious about reporting an event as a signal)
rather than group differences in attention.

Kaplan (1963) measured GSR activity during interaction
in a three person female group and found that reactivity
increased when what was being said had a negative tone and
that, for the two patients in the group, GSR increased as did
their participation. He concluded that GSR activity is a
measure of defensiveness. However, this result has not been

found uniformly. Martin and Edelberg (1963) failed to find

42

an increase in interest associated with an increase in GSR;
they did, however, report that GSR activity decreased as
subject's reported irritation increased. Furthermore, they
found a trend relating the change in the slope of EDA to
interest.

Edelberg (1972) examined EDA in four conditions-—
resting with the eyes open, cold pressor test, mirror-tracing,
and threat of shock. He found that the recovery limb time
constant decreased as performance and task complexity in-
creased. EDA recovery time increased with activation except
in the case of the cold pressor test. A decrease in the
recovery rate followed the threat of Shock. He concluded
that goal-orientation is the primary determinant of the
acceleration of recovery. He hypothesizes that when activa-
tion is for defense, an inhibition of the re-absorption
process causes a protective flooding of the skin. However,
when activation is for manipulative purposes, sweat secretion
and re-absorption are balanced, which then manifests itself
as an accelerated recovery rate. Orne-Johnson (1973) found
that the GSR habituation rate to auditory tones was faster
for meditators than for controls. Furthermore, meditators
had fewer spontaneous GSR'S than did controls. If one accepts
the claim that meditatorsare better able to respond to the
present situation than are non-meditators,this supports
Edelberg's contention that goal-orientation is of primary
importance in determining EDA recovery rate.

Johnson (1963), using helicopter pilots as subjects,

43

Showed that those subjects who had a large amount of Spon-
taneous GSR activity, had increased reactivity to an initial
stimulus and adapted less to repetitive stimuli.

Gordon gt gt. (1959) hypnotized female subjects and
instructed them to recall a conflict with their parents.
They used three different conditions-~a repression Situation
in which the subjects were told not to think about the con-
flcit, a suppression condition in which they were told to
think about, but not to discuss, the conflict, and a verbal-
ization condition in which they were told to discuss the
conflict. Under all conditions skin conductance increased
throughout the session; however, the Slope of the increase
for suppressers was Significantly steeper than for those who
were allowed to verbalize.

Panek and Martin (1959) found that skin resistance dips
were associated with "speech disturbances" (ie. pauses and
repetitions) during therapy.

Sternbach and Tursky (1965) compared Skin potential
reSponses of housewives from different ethnic groups to
electric shock. They relied on Zboroski's characterization
of "old Americans" as having a "phlegmatic, matter-of-fact,
doctor-helping orientation," Jews as being concerned about
what the cause of the pain is and as distrusting palliatives,
and Italians as simply desiring relief from the pain. After
twenty-nine shocks, they found that Italians had the highest
palmar skin potentials, Yankees had the lowest, and Jews and

Irish were in the middle. They explain the Yankee's skin

44

potential adaptation in terms of their matter-of-fact
attitude and the Italians' large reaction in view of their
significant concern about immediate pain.

In View of the foregoing, there seems to be a solid
case for concluding that both heart rate and EDA are inti-
mately involved in the intake or rejection of external
stimuli. The next question is how do they interact. Lacey
gt gt. (1963) state that when attending to visual and audi-
tory stimuli, heart rate decreases and ESC increases. How-
ever, they point out that Skin conductance changes tend to
be smaller than differences in heart rate. Darrow's gt gt.'s
(1942) research Showed that palmar conductance increases
were excitatory in that they were positively associated
with alpha blocking and that increases in blood pressure
were "homeostatic" (ie. had an inhibitory effect on cortical
excitation) since they were negatively correlated with alpha
blocking. Lacey and his associates hypothesized that a high
heart rate would be inhibitory for sensorimotor impulsivity
and that a low heart rate would be facilitory for sensor-
imotor impulsivity. Using a key-tapping task they found
this to be true with the qualification that an ESC increase--
which is excitatory-~could overcome the inhibitory effect of
a high heart rate. However, the reverse is not true; when
heart rate is low--which is already an excitatory state--there
is no detectable effect with an increase in skin conductance.
Both the above hypotheses and results--coming from Lacey's

laboratory--are strange at first glance. What he is saying

45

is that subjects were more likely to reSpond to irrelevant
stimuli, inspite of instructions to the contrary, when
their heart rates were low, rather than when they were
high. If one reasons that a decelerating pulse is indica-
tive of increased attending, and assumes that increased
attention should be associated with a decrease in false
responses, one would predict the wrong result. A careful
analysis of the data can resolve this apparent dilemma;
there are two sorts of errors that can be made--a false
response and a failure to respond. Associated with cardiac
deceleration is an increase in responding-~including in-
correct responding. Thus one may qualify Lacey's contention
that attending is associated with cardiac deceleration.

It is attending in the sense of a generally heightened
responsivity to the external environment and not attending
in any sort of discriminative sense.

Lacey gt gt. (1963) present what is some of the most
confusing findings in the published literature. They played
tape recordings dealing with three different subject areas--
aggression to peers, aggression to mothers, and sexual con-
flict, recording heart rate and ESC. They also had subjects
identify line drawings<offigures, engaged in some activity
relevant to the conflict areas. They only found significant
results in the aggression tapes; no satisfactory explanation
is offered about why sexual conflict did not yield the pre-
dicted results. Furthermore, during the warning period

before the aggression to peer tape, the correlation between

46

heart rate and ESC was .66. Both measures were positively
related to tachistoscopic recognition times, contrary to

the prediction. However, with the aggression-to-mother tape,
the two physiological measures were uncorrelated with each
other. ESC had a negative correlation with perceptual
threshold, whereas heart rate had a positive, but statistic-
cally insignificant correlation. These results are in the
predicted direction with decreasing ESC or increasing heart
rate indicative of defensiveness. One can speculate about
the different stimulus values of the two tapes (eg. perhaps
aggression toward mother was much more threatening for these
young men than aggression towards peers), but no hard evi-
dence has yet been presented to explain these results.

Lacey (1959) presents a great deal of evidence to make
the point that some people respond only in one particular
physiological modality. In two studies he reports that
with the cold-pressor test 10.6% of the subjects and 11.2%
evidenced a cardiac increase with no skin conductance response.
At the same time another 14.9% and 10.1% of the subjects
had a conductance response with cardiac response. In other
words, between 21% and 25% of the subjects only reacted in
one sphere. These findings underline the importance of not
relying on just one physiological variable in evaluating

openness and closedness to the envirnoment.

47

Summary and Conclusions

 

The literature reviewed concerning cardiac activity
gives substantial support to the idea that cardiac decel-
eration is associated with increased responsivity to the
external environment. This phenomenoni£;most reliably
found in response to psychophysical stimuli such as photic
and auditory stimuli. Other factors, most notably vegeta-
tive requirements of the organism, definitely determine
pulse rate changes, but a strong case can be made--based
on physiological differences between amount of cardiac out-
put and pulse rate and mathematical comparison of tasks
with differing metabolic requirements--for the contention
that an internal vs. external attentiveness factor accounts
for an otherwise unexplained part of the variance.

The research reviewed here concerning EDA suggests
that an organism which is externally oriented may have a
large initial conductance response followed by a reduction
in conductance activity as manifested by fewer responses
and a decreasing EDA level. Continued EDA activity or
failure to habituate is indicative of a defensive reaction.

Evidence was also presented suggesting that because
some people respond only in one physiological domain or the
other that evidence of these response patterns would be
sufficient evidence to suggest environmental orientation.
Some studies were cited suggesting that interest level is

related to both cardiac and EDA responses.

48

Evidence was reviewed which suggests that people who
tend to deny and/or repress also tend to be more reactive
. physiologically than those who tend to approach and mani-
pulate stress-producing stimuli. In addition, some examples
of empirical findings which appear to be inexplicable as far
as any theory reviewed here were given.

No consistent evidence however, was found Specifically
linking the particular patterns of increasing vs. decreasing
activity of heart and Skin conductance, which is predicted
by the physiological research, and the degree of involvement
in interpersonal situations. Some ambiguous findings of
Lacey's gt gt. (1963) are cited dealing with this. Other
studies which touched on the problem dealt only with the
quantity rather than the directionality of response. In
addition, often there was a reliance on just one physiological
measure. It is therefore concluded that a study relating
directionality of both heart rate and ESC to involvement

in interpersonal situations is needed.

CHAPTER III

METHODOLOGY

Physiological Measures

 

Three measures were used to describe heart rate
activity. The first cardiac measure was the mean heart
rate during each vignette and during a five minute rest
period prior to stimulation. This measure was chosen,
while others such as heart rate variability measures were
excluded, on the basis of the cluster analysis done by
Speisman gt gt. (1961) which showed that the other variables
included in their analysis were too highly correlated
with mean heart rate for them to add sufficient information
to warrant their use. One measure not included in their
analysis, however, is the autonomic lability score (Lacey,
1956). This measure attempts to take into account the Law
of Intitial Values (Wilder, 1967) by looking at the level
which the individual is at before a reSponse is made.
Lacey explains his measure in terms of a regression analysis,
but Benjamin (1963) has shown that the procedure is similar
to an analysis of covariance with the initial level being
the covariate. According to this then, a heart beating 80

beats per minute which increases to 87 beats per minute has,

49

50

in view of its base level, made a larger response than a
heart beating at 65 beats per minute which increases to
72 beats per minute. The formula Lacey gives to compute

a lability score is:

 

y - x r
Autonomic lability score == 50 + 10( z : Xfi)
(l - rxy)
where xz'= individual's pre-stimulus level, expressed in

standard score units

y2 = individual's stimulus level, expressed in standard
score units

rxy = correlation for sample between initial level and
stress level.

The correlation for the sample was computed indepen-
dently for each cell defined by personality group and level
of involvement (eg. correlations for involved sensitizers,
uninvolved sensitizers, involved repressers). The correla-
tions ranged from .75 to .80; there were no Significant
differences between the correlations at the .1 alpha level.
Therefore, it seemed reasonable to pool the rxy's in computing
the lability scores.

The final cardiac score which was used was the heart
rate change score. This was computed by subtracting the
mean heart rate level during the last eight seconds of a
vignette from the mean cardiac level during the first eight
seconds of a vignette.

Four measures were used to evaluate ESC. ESC was re-

corded in micromhos with the sensitivity set so that one

51

micromho equals one centimeter. Again, relying on the
cluster analysis done by Speisman gt gt. (1961) two measures
were selected so as to maximize the amount of information
being taken from this variable. They were the mean Skin
conductance level for vignettes and for the five minute

rest period before stimulation and the standardized standard
deviation of the ESC over the same time periods.

There is a debate in the literature about the appli-
cability of Lacey's autonomic lability scores to ESC, since
ESC has no counterbalancing parasympathetic component as
does heart rate. Hord gt gt. (1964) have concluded that for
the skin conductance response the opposite of the Law of
Initial Values seems to hold true. Edelberg (1967) has
shown that with a small skin conductance reSponse the change
in conductance equals the change in the resistance divided
by the resistance level squared. In other words, conductance
units automatically adjust for resistance levels. In accor-
dance with the work of Wenger gt gt. (1961) and Haggard
(1949), it was decided to compute the change in log con—
ductance for each vignette. The reason for using the log
transformation here was to increase the approximation to
the normal distribution. However, as will be shown in the
following chapter, the log transformation decreased the
approximation to the normal distribution; therefore, simple
mean conducatance values, in micromhos, were used.

The final ESC measure, known as the ESC change score,

which was used was the difference in number of responses,.

52

or bumps (defined as any deflection of at least one quarter
micromho) between the first eight seconds of the stimulus

vignette and the last eight seconds of the vignette.

Psychological Measures

 

Another measure which was taken on a vignette by
vignette basis for all subjects was a rating of the subjects'
involvement with each stimulus vignette. This involvement
rating is based on the work done by Kagan and Millhouse
(1975) in which judges rated a subject's involvement with
each vignette on a scale of 1 (high involvement) to 5 (low
involvement) after listening to a tape recording of an IPR
recall procedure. Kagan and Millhouse (1975) state that
their scale is based solely on "face validity." In their
study they failed to find statistically significant differ-
ences on the measures of involvement with the vignettes
between pre and post differences for a group which had under-
gone IPR training and a control group. They had two judges
rate each of the vignettes and inter-rater reliabilities
ranged from .09 to .79. Thinking of the pre and post
series of vignettes as separate scales, they found a Kuder-
Richardson reliability coefficient of .75 for the pre-test
and .64 for the post test. If one examines the scale, one
finds that the criteria for judgment vary in that some re-
quire considerable inference on the part of the judge (eg.
the category of "unconscious involvement" which is to be

inferred by "slips, defense mechanisms, distortions etc.)

53

while other categories rely more heavily on the subject's
own verbalizations (eg. the category of uninhibited involve-
ment.where the subject says something such as "I was really
scared...angry...frightened...etc.") Using this scale as

a basis, those categories of involvement that relied most
heavily on inference were excluded. This leaves the scale
with four clear categories--uninhibited involvement, in—
hibited involvement, uninvolvement (called "detached,
distancing" on the original scale) and a "cannot be scored"
category. The specifiic criteria for each category are
given in Appendix A.

The judges used in this study were two female graduate
students--one completing her doctorate in counseling at
Michigan State University and the other completing her
Masters in recreational therapy who had been thoroughly
trained in Interpersonal Process Recall. Before rating the
experimental data, the judges were trained using tape record-
ings of recall sessions. An inter-judge reliability coeffi-
cient of .87 was obtained on a practice tape immediately
after training. The inter-judge reliability coefficient
for the tapes acutally used in this study was .80.

Decreasing the amount of subjective judgment required
of the raters was intended to increase the reliability of the
instrument, but this did not help the problem of having to
rely solely on face validity. To deal with this, two validity
measures were taken. The logic behind the first validity

measure was that sensitizers ought to be involved more

54

frequently than repressers with the vignettes. Therefore,
if the measure of involvement here was a valid one it ought
to reflect this. This was tested by using a repeated mea-
sures design by groups with involvement as the dependent
variable. The results of this test and the analysis of

varance table are given in Table 3.1.

Table 3.1 Mean Involvement Ratings and Analysis of Variance

 

 

 

Group Mean
Repressers 1.62
Midlines 1.83
Sensitizers 2.28 Sum of Mean F
Source DF Squares Square Ratio
Groups 2 40.84 20.42 30.42*
Subjects x Groups 42 28.20 .67
Vignettes 11 1.99 .18 .25
Vignettes x Groups 22 2.77 .13 .17
Vignettes x Subjects 462 335.07 .73

x Groups

 

 

*
sig. at less than .01.

The difference reported for the groups is in the expec-
ted direction; this is taken to support the validity of the
involvement scale.

(One should also note from this ANOVA table that even
using the liberal degrees of freedom one fails to find a
vignette effect or a vignette by groups interaction. This

has no bearing on the validity of the involvement scale,

55

but is an important justification for treating the various
vignettes as equivalent; the Significance of this is dis-
cussed below.)

The second validity measure was a correlation between
the number of vignettes a subject was rated as being in-
volved with and the score She received on a simple memory
test (reproduced hnAppendix B) in which she was asked to
recall visual details about a vignette (eg. what color
blouse was the woman wearing?) after being reminded of the
verbal content of that vignette.

The memory test was constructed by administering four
items for each of the twelve vignettes to a pilot group of
ten people. Two items were chosen per vignette for the final
test such that one item had been missed by 10-20% of the
pilot group and the other had been missed by 60-70% of the
same group.

Contrary to expectations a correlation of -.58 was
found between recall of cognitive details and involvement
scores. The original thinking here was that peOple should
be able to recall vignettes with which they are involved
better than vignettes with which they are not involved.
However, a search of the literature on this point not only
failed to turn up any evidence to substantiate this line of
thinking, but actually suggests the exact Opposite. Klein-
smith and Kaplan (1963) report that if GSR is used as the
measure of arousal the low arousal group performs better

than the high arousal group if recall is tested immediately

56

after a paired-associate learning task. Walker (1958)

found that his low arousal group recalled things better
immediately after presentation than did his high arousal
group. Easterbrook (1959) concludes, "...emotional arousal
acts consistently to reduce the range of cues that an
organism uses..." Callaway and Dembo (1958) report that

high arousal states such as anxiety, pain, and sexual exci-
,tation and chemical agents such as amphetamines and carbon
dioxide serve to reduce people's attention. Bahrick (1954)
found that when he instructed high and low incentive groups
to learn geometric shapes and then tested them on color
recognition the high incentive group performed worse than the
low incentive group.

On the basis of the above-cited studies and because

of the dearth of published material which would support the
original hypothesis it seems reasonable to consider the
correlation of -.58 between involvement ratings and scores

on the memory test as support for the validity of the involve-
ment ratings. It further suggests that involvement, as tapped
by the involvement scale, is correlated with emotional

arousal.

The final psychological measure which was used in this

study was Byrne's Revised Repression-Sensitization Scale

(R-S Scale) (Byrne, gt_gt., 1963). As previously mentioned
Byrnezconceptualized defensive behaviors along a continuum
Whicfli at one extreme avoided anxiety-producing stimuli and

at tile other extreme approached the stimuli in an attempt to

57

control and manipulate them. From this theoretical vantage
point he proceeded to develop a questionnaire to differen-
tiate people who characteristically defend against anxiety
in these two ways (Byrne, 1963). He based his scale on
previous work done by Altrocchi gt gt. (1960) where a sen-
sitization score based on the depression, psychasthenia, and
Welsh anxiety MMPI sub-scales is computed, a repression
,score based on the L and K validity scales and the Hysteria
denial sub-scale of the MMPI is computed and then computed
a final index by subtracting the total score based on the
last group of the scales from the total score based on the
first group of scales. Because of sub-scale overlap, in-
dividual items sometimes received double or conflicting
weights so Byrne (1961) revised the scale so as to elimin-
ate these inconsistencies and disproportionate weightings.
He further refined his Repression-Sensitization scale by
eliminating all individual items which failed to yield
correlations with the total R-S scale significant at at
least the .001 level in two independent samples (Byrne,
Barry & Nelson, 1963).

Byrne (1964) reviews a variety of different types of
evidence in an attempt to establish construct validity for
his instrument. The first set of evidence he marshals in
support of his instrument is the high correlation between
clinician's judgments and the R-S scale. When clinicians
were asked to fill out a questionnaire as they believed a

represser would, there was a 90% overlap between their

58

judgments and the R-S scale. In a second area--selective
forgetting--he cites the work of Gossett to show that re-
pressers fail to remember threatening material as well as
sensitizers. A third area he finds support for the construct
validity of the R-S Scale is the area of perceptual defense.
An example of such a study is the work of Tempone who found
repressers had a higher recognition threshold for threat
words originally encountered in a failure experience (the
failure being the inability to solve an anagram) than did
sensitizers. Cohen (1967) replicated this finding. A

fourth supporting area is the individual's subjective aware-
ness Of anxiety. Here the work of Pomeranz is cited show-
ing that in response to movies sensitizers indicated more
emotional arousal than repressers. Merbaum and Kazaoka
(1967) found, as expected, that sensitizers were quicker

to admit negative affect than were repressers. Byrne and
Scheffield (1965) found that sensitizers were more anxious
than repressers when they were being read sexual

material; they found no difference in anxiety levels when
subjects were exposed to neutral stimuli. ‘Feder (1968)

found that repressers report a smaller discrepancy between
their ideal selves and real selves than do sensitizers. A
fifth area mentioned by Byrne as corroborating the validity
of the R-S scale concerns the appreciation of humor. Accord-
ing to Freudian theory, normals (ie. those with intermediate
scores on the R—S scale) should hthethe highest appreciation

of humor, sensitizers should be next, and repressers should

59

have the least appreciation. He reports that when subjects
were divided into groups according to degree of apprecia-
tion of humor, the low appreciation group was the most re-
pressing. A sixth category mentioned by Byrne--and the one
that is most releveant to this thesis--is the area of phy-
siological response to threat. Davison (1963) showed a
subincision film to male subjects while their heart rates
and ESC were being recorded. He found that sensitizers were
the least responsive to the film, the group that was in the
intermediate range on the R-S scale was most reSponsive
physiologically, and the repressers exhibited an intermediate
range of physiological responsivity. Simal and Herr (1970)
reviewing the previous physiological literature in relation
to the Byrne scale writes:
Physiological studies seem to agree that repressers Show
more physiological arousal to Stress or anxiety-producing
situations than do sensitizers. This phenomenon was ini-
tially observed by Davison (1963) and later confirmed by
Lazarus and Alfert (1964). These studies suggest that a
repression-sensitization dimension such as is defined by
Byrne's scale does in fact exist.
However, it should be noted that using sexually arousing
pictures, Simal and Herr failed to find GSR differences be-
tween repressers and sensitizers. (This is reminiscent of
Lacey's previously mentioned failure to find expected
differences with sexually arousing stimuli, although Lacey
did find differences with aggressive stimuli; this suggests
that there may be something about sexually arousing stimuli

which uniquely alters EDA and cardiac activity.)

In spite of the substantial support much of the

60

literature seems to give to the construct validity of the
Res scale, it would be inaccurate to give the impression
that this support is universal. Wiener gt gt. (1956) using
judges' ratings, failed to find any support for the notion
that there is some consistencytx>individual's defensive be-
havior in terms of an approach/avoidance continuum. More
damaging to the R-S scale is Hoffman's (1970) study, fail-
ing to find support for the hypothesis that as stress in-
creases repressers will increase avoidance and senstizers
will increase vigilance. Lefcourt (1966) made the observa-
tion that the strongest evidence supporting the validity
of the R-S scale comes from self-report measures lacking in
behavioral correlates. In an attempt specifically designed
to find a behavioral correlate Thelen (1969) compared the
R-S scores of students seeking psychotherapy with those who
were not. The findings indicate that "therapy-seekers"
were more likely to be sensitizers than repressers. Further-
more, males who remained in therapy more than four sessions
were more likely to be sensitizers. Thelen concludes, "As
evidenced by the behavior of seeking therapy, this study
suggests that the R-S scale does have the behavioral re-
ferrants, such as avoidance and sensitization, which have
been questioned by Lefcourt."

Another attack on the validity of the R-S scale comes
from studies claiming that the R-S scale measures anxiety
because of the high inter-correlation between scores on the

Taylor Manifest Anxiety Scale and the R—S scale. Golin gt gt.

61

(1967) concluded the results of the two tests were prac-
tically interchangeable, with scores being determined by
two factors--emotionality and defensiveness. There is,
however, evidence to suggest that this high correlation is
not due to a mis-labelling of the R-S scale. Kaplan (1966),
writing about a study which used the Taylor Manifest Anxiety
scale in order to classify subjects, says:
Despite their low scores on the anxiety scales, the low
anxiety subjects in this Study did not appear to be .
individuals who are free from the threat of experiencing
significant amounts of anxiety (also Truax, 1957). Rather,
on the whole, they appear to be individuals who attempt
to ward off the threat of anxiety and other distressing
affects by heavy reliance on the defenses of repression
and/or denial. The high anxiety subjects, on the other
hand, appeared to be individuals who deal with anxiety
by thinking and talking about it. That is, their style
of defense permits conscious awareness of anxiety-arousing
ideas (also Spielberger, DeNike, and Stein, 1965).
Dublin (1968) has suggested that sensitization may be
a "measure of defensive failure resulting in a response
characterized by lack of ego adaptiveness which takes the
form of anxiety...and which has concomitant observable be-
haviors indicating ineptness, hesitancy and indecisiveness"
because when put in an ambiguous situation sensitizers, as
compared with repressors, were "taken aback." However, this
interpretation is clearly not consistent with the data pro-
vided by Petzel and Gynther (1968) who found that repressers
solved fewer anagrams when they received "ego-oriented" in-
structions than when they received task-oriented instructions,

while sensitizers exhibited the Opposite pattern. They con-

cluded, "Sensitizers...characteristically 'face up to' even

62

small stresses of everyday life and thus may be more likely
than repressers to deal effectively with the more severely
stressful situation."

It will be remembered from the earlier theoretical
discussion that Byrne considers there to be a curvilinear
relationship between his scale and psychological adjustment.
Some of the above validity studies are indicative of that
in that depending upon the group or behavior being studied,
varying negative characteristics are found at either end
of the continuum.

Evidence against the curvilinear hypothesis, and in
favor of a linear relationship between psychological adjust-
ment and repression-sensitization (indicating that repressers
are the healthier people) is found in the work of Byrne 2E.El°
(1968) (surprisingly!)vnu>found that students who were sen-
sitizers had a greater frequency and severity of physical
illness as measured by a health survey. In addition, male
sensitizers made more frequent visits to the health center
than male repressers. Along the same lines, Feder (1967)
compared psychiatric patients with medical-surgical patients
and found a higher frequency of sensitizers in the maladjusted
population and a higher frequency of repressers in the normal
population. Tempone and Lambi (1967) confirm this observa-
tion in a separate study finding more repressers in their
"adjusted" (non-patient) group and more sensitizers in the
"maladjuSted" (psychiatric patient) group.

In support of the curvilinear relationship Byrne cites

63

the work done by Ullmann, using his Facilitation-Inhibition
Scale. (Byrne [1964] considers the Facilitation-Inhibition
scale to be virtually "interchangeable" with his R-S scale,'
citing correlations of .76 and .88 between the two scales in
student populations and a correlation of .94 in a neuro-
psychiatric population. Furthermore, he points out that
approximately half the items on the F-I scale are included
in the revised R-S scale.) Ullmann (1962) compared male
neuropsychiatric VA patients' F-I scores with those of male
students. Presumably there should be a higher incidence of
pathologic defense in the patient population than in the
student pOpulation. As expected the patient population had
a significantly higher standard deviation than the Students.
(Student standard deviation = 6.50 compared with patient
standard deviations of 11.22 and 11.44.)

Other evidence mustered to support the curvilinear
relationship between adjustment and R-S scores can be found
in comparing a number of studies. Byrne (1964) discusses
Megargee and Mendelsohn's cross-validating an MMPI scale
designed to discriminate "extremely assaultive offenders
from other criminals and from normals" with the R-S scale.
They found that "extreme repressers are seen to fit into
the pattern of chronic overcontrol which includes rigidity,
conventionality, moralistic views, and (at least in the
criminal group) a greater probability of psychosis." On
the other side of the continuum Byrne (1964) cites the work

of Ullmann and McReynolds, reporting a Significant correlation

64

of -.50 between F-I scores and anxiety ratings in psychi-
atric patients, Joy's and Endler's work correlating MMPI
scales and R-S scale scores suggesting that in comparison
with repressers, sensitizers tend to be depressed, anxious,
introverted, discouraged, and tend to respond to the test
as would members of the opposite sex, and Byrne, Golightly
and Sheffield's correlation between the California Psycho-
logical Inventory and the R—S scale which indicated that
sensitizers were less well adjusted than repressers. Look-
ing over this sample of studies one can see that severe
maladjustment is associated with both extremes, with re-
pressers having a higher likelihood of acting out aggress-
ively and sensitizers being more prone to anxiety, depression,
and neuroticism. This argues for psychological adjustment
to be found in the middle of the R—S continuum.

Merbaum and Badia (1967) found partial support for
the interpretation that maladjustment is equally associated
with the two extremes of the R-S scale. They exposed male
and female subjects to varying intensities of shock and
measured their approach or avoidance of the stimulus. They
found a curvilinear trend for their female subjects in which
those who scored in the middle two quartiles on the R-S
distribution withstood higher shock intensities than those
at the two extreme quartiles. They concluded, "If, however,
one interprets low-shock tolerance within an adjustment frame
of reference, Byrne's (1964) prOposition...could be held

tenable at least for female subjects."

65

Due to the far greater number of studies which find
greater amounts of psychopathology at the sensitizing end
of the continuum than at the repressing end, Byrne (1964)
reluctantly concludes that greater psychopathology is found
at the sensitizing end of the scale. However, the present
author does not believe that this necessarily reflects a
serious flaw in the theoretical base for the R-S scale.
First, there is the possibility that the pathological con-
sequences Of repression are most clearly seen in extreme
groups such as the aggressive convicts in the Megargee and
Mendelsohn study (most other studies used less extreme
populations such as groups of students). More important,
however, is the fact that the plethora of studies cited
finding repressers to be healthier than sensitizers relies
on self-report measures to determine psychological adjust-
ment. By the very "nature of the beast" one would expect
sensitizers to admit with much greater frequency than re-
pressers, items which would detract from one's adjustment
rating. Therefore, thoughtfluaquantity of research clearly
suggests a linear relationship between adjustment and R-S

scores, the evidence is not compelling.

Stimulus Films

 

The procedure of using films to stimulate physiological
reactions is widely used in the past research (eg. Lazarus
& Alfert, 1964 and Davison, 1963). Typically these films

are pictures taken Of some particular event such as

66

sub-incision rites or motor vehicle accidents which the
experimenter believes will be anxiety-producing for the
subjects. Kagan and Schauble (1969), however, describe
the production of a somewhat different sort of filmed
stimulus:
In order to create the simulated emotional Situations
with which to confront the clients, professional actors
were trained to portray the four basic types of affect
with varying degrees of intensity, but to avoid use of
words which indicate a specific situation or 'story.’
Each actor was instructed to direct the emotion (re-
jection, affection, hostility, eth at an imaginary
individual directly behind the camera lens SO that the
viewer would see the resultant image as if the actor
were talking directly to him.
Thus, because of their deliberate vagueness and because they
treat the viewer as a participant, they require the subject
to participate actively in imagining him/herself in the
situation.
These films have been used as stimuli in one study done

by Kagan and Millhouse (1975), a study done by Smith (1975),

and in a series of pilot studies done by the present author.

Sample

Subjects for this study were drawn from the female
students taking the Interpersonal Process Recall (IPR)
classes (Ed. 484) during the Winter Term, 1976 at Michigan
State University. Members of the class ranged in age from
18 to 35 with the modal ages being 20 and 21. Because of
the various degree requirements, the class was largely

made up of pre-teaching and pre-nursing students and others

67

preparing to enter a variety of "helping professions" such
as social work, counseling, and music and recreational
therapy.

All members of the classes were required to take
Byrne's Repression-Sensitization Scale. The distribution
for the female class members on the Byrne R-S scale con-
formed very closely to the one reported by Byrne gt gt.
(1963), having a mean of 46.32 and a standard deviation of
21.47. Keeping within the limits used by Byrne all indivi-
duals scoring 69 or above were designated as sensitizers;
those scoring between 42 and 52 were considered midlines,
and those scoring 22 or below were designated as repressers.
Only those persons whose scores fell into one of these three
grouping were retained as subjects; all the data from the
other students were discarded.

One hundred three students from five classes went
through the entire experimental procedure. Five were elim-
inated as possible experimental subjects because they were
taking medication known possibly to affect heart rate. One
was eliminated because of poor mechanical recording and one
was eliminated because she was pregnant. One was eliminated
because she became so emotionally upset watching the first
two vignettes that the experimenter decided it was not in
her best interests to continue. This left ninety-five
potential subjects. Forty-five (fifteen per personality
sub-group) were selected because they met the criteria for

inclusion in one of Byrne's sub-groups and allowed the groups

68

to be of equal sizes.

Procedure

 

During their regularly scheduled IPR classes the pro-
cedure of the experiment was explained. It was pointed out
that the personality test and physiological feedback might
be useful to individuals seeking to know themselves better.
All questions concerning the purpose of the experiment were
deferred until after all the data had been gathered; a group
discussion about the experiment was scheduled for the end
of the term. It was also pointed out that though going
through the procedure was mandatory to get credit for the
class, inclusion of any data in the experimental analysis
was Optional. If anyone had any objection to being in-
cluded, they were instructed to inform the experimenter
after their experimental session and their data would not
be included; no one exercised this Option.

During the preparatory session a schedule was passed
around for class members to sign-up for one two-hour block
during the next six weeks. Because progesterone changes
occuring during the luteal phase of the menstrual cycle are
known to decrease electrical skin activity (Mackinnon and
Harrison, 1961) the women were asked to sign-up between five
and ten days after the end of their menstrual period.

For convenience in attaching the electrodes the students
were asked to wear a short-sleeve or sleeveless blouse and

not to wear panty hose, tights, or stockings.

69

At the end Of the preparatory session all the students
were asked to fill out a Byrne R-S questionnaire. The pur-
pose of the questionnaire was left ambiguous, but a complete
interpretation was promised later in the term.

When the subject arrived she was asked to wash her
hands thoroughly with Ivory soap and water. While she was
washing her hands the experimenter made sure that only two
fluorescent lights were on in the experimental room and
that the temperature was between 72° and 74° F. When the
subject returned to the room she was asked to fill out a
background information sheet (Appendix C). Next, two dis-
posable silver/silver chloride electrodes (manufactured by
American HOSpital Supply, No. 65375-030) were attached to
the outer surface of either arm, approximately four inches
below the shoulder, after the area was rubbed by a swab of
dry cotton. The electrodes were connected to an EKG tach-
ograph, Grass Model 7P4D. The electrical signal was ampli-
fied by a Grass Model 7DAE DE amplifier and recorded by one
pen on a four-channel Grass, Model 7WC8PA strip chart re-
corder at a speed of 100 mm per second.

After the cardiac electrodes were attached two more
disposable silver/silver chloride electrodes were attached,
one to the hypothenar eminence of the left hand after it was
rubbed lightly with a dry cotton swab, and the other to the
volar surface of the left forearm, after it was rubbed
vigorously with a dry cotton swab. The electrodes were then

connected to a Hagfors bridge (a constant voltage bridge)

70

which in turn was connected to a Grass, Model 7Plb, low level
DC pre-amplifier. Finally, the signal was amplified by a
Grass DC amplifier, Model 7DAE, and recorded on one channel
of a four channel Grass strip chart recorder, Model 7WC8PA,
at a paper speed of 100 mm.

In addition to these two sets of electrodes one 3/8
inch, silver/silver chloride electrode was attached, using
Unibase paste (produced by Parke-Davis) mixed with 50% by
volume with Normal saline, to the left leg of the subject
after the leg was rubbed with alcohol and connected to a
ground.

As each electrode was attached to the subject, its
purpose was explained. It was explained to the subject
that after the recorder was calibrated, a record of her
base resting level would be made. Therefore, she was in-
structed to relax with her eyes Open for the next several
minutes and to acclimate herself to the surroundings. Be-
fore the start of the rest period the subject was told over
an office intercom system: "You will now have five minutes
to relax. After that the films will appear on the screen
in front of you. Make an attempt to imagine yourself alone
with the person on the screen. If it helps to fantasize
a specific situation in which to place that person go ahead
and do so. But above all, do your best to react to that per-
son as if he or she were Speaking to you personally." The
vignettes were then played with variable lengths of blank

leader between the filmed segments. (The typescript is included

71

in Appendix D). The start and finish of each vignette and
of the rest period was marked on the physiological record
by the experimenter. The projector which was used was a
16mm Bell and Howell Autoload. It was located in the in-
strumentation room and projected on a translucent screen
approximately four feet in front of the subject. The
vignettes depicted hostility, manipulation, sexual advance,
rejection, and acceptance. As the films were being shown
the subject was video-recorded on one-half inch Ampex
equipment, so that the recording was a record of the film
sound track, any verbalizations the subject made, and the
face of the subject. (Exact procedures for running the
polygraph may be found in Appendix E.)

After the subject viewed all the vignettes the elec-
trodes were removed and she was given a chance to clean up.
She was then read a questionnaire asking Specific questions
about cognitive details of the films. After this the
videotape of the subject watching the vignettes was played
for her. She was encouraged to stOp the tape and verbalize
what she recalled experiencing, accOrding to the recall
process as described by Kagan gt gt. (1969). In the final
stage of the session, she was shown her printout of her
physiological responses and they were discussed with her

in view of her reported experience.

72

Design

This study is a predictive piece of research in that
the essential question being asked is does knowledge of a
subject's physiological pattern and the subject's class-
ification on the Byrne R-S Scale, enable one to predict
involvement or non-involvement with the vignettes? The
design, diagrammed in Figure 3.1, is a group by vignette

by subject by measure design.

Baseline V V

l 2’ ’ 12
M1 M2 M3 M1 M2 M3 M4 M5 M6 M7
Group 1 S1
(Repressers)
8
Group 2 15
(Midlines) 516
S
Group 3 30

(Sensitizers) S31

Where: Subject
Vignette

Mean heart rate

Mean Electrical skin conductance .
Standardized electrical skin conductance
standard deviation

autonomic lability score

cardiac change score

electrical skin conductance change score
Judges' rating as to involvement

s::s:z 3:3::<:m
~Jouncs (ABDP'
II II II

Figure 3.1 Design Diagram for MANOVA

73

It Should be noted that the judges' involvement
rating (M7) is dependent on the subject's response and
as such cannot be controlled by the experimenter. There-
fore, in order to make an analysis possible with this
measure two vignettes were chosen from each category--
involved and uninvolved. The number two was chosen so
as to allow the analysis to retain as much data as possible
per subject while maximizing the number of subjects re-
tained in the analysis. The statistical justification for
treating the vignettes as equivalent is that no difference
in involvement for any of the vignettes were found; neither
was there a vignette by personality group interaction (see

Table 3. 1).

 

Experimental Hypotheses

I. The groups of sensitizers (G1), Repressers (G3) and mid-
lines (G2) will differ in cardiac and ESC activity during
the base period.
A. Mean heart rates for the groups will not be the
same.
B. Mean ESC levels for the groups will not be the same.
C. Mean standardized ESC standard deviations for the
groups will not be the same.
II. The groups of sensitizers (G1), repressers (G3) and mid—
lines (G2) will differ in cardiac and ESC activity while

viewing the film.

A. Mean heart rates for the three groups will not be

74

the same.

B. Mean autonomic lability scores for the three
groups will not be the same.

C. Mean cardiac change scores for the three groups
will not be the same.

D. Mean ESC levels for the three groups will not be
the same.

E. ESC standardized standard deviations for the three
groups will not be the same.

F. Mean ESC change scores for the three groups will
not be the same.
III. The physiological variables will differ on those
vignettes with which the subject is rated as being involved
from those vignettes on which the subject is rated as being
uninvolved.

A. Mean heart rates for involved and uninvolved
vignettes will not be the same.

B. Mean autonomic lability scores for involved and
uninvolved vignettes will not be the same.

C. Mean cardiac change scores for involved and un-
involved vignettes will not be the same.

D. Mean ESC levels for involved and uninvolved
vignettes will not be the same.

E. ESC standardized standard deviations for involved
and uninvolved vignettes will not be the same.

F. Mean ESC change scores for the involved and unin-

volved vignettes will not be the same.

75

IV. There will be an interaction between personality types
and involvement ratings.

A. Repressers will have their highest scores on all
physiological variables when they are uninvolved with
vignettes. Their lowest physiological scores will be when
they are involved with vignettes. Sensitizers will have
their highest scores on all physiological variables when
they are involved with the vignettes; their lowest physio-
logical Scores will be when they are uninvolved with the
vignettes.

V. Physiological pattern, as defined by cardiac and ESC
change scores, and the Byrne Scale will together predict
a subject's involvement with the vignettes.

A. Subjects with high R-S scores and high cardiac
and ESC change scores Should receive higher involvement
ratings. Subjects with low R-S scores and low cardiac and

ESC change scores will receive lower involvement ratings.

Analysis

The four moments about the mean were initially com-
puted in order to determine the applicability of parametric
test statistics. I

Hypotheses I, II, III, IV were tested using a multi-
variate F test. When significant differences were found
univariate and step down F tests were also performed. In
addition, when significant differences were found in the

groups effect Tukey post-hoc comparisons were carried out.

76

The repeated measure design employed allows for testing a
group's effect, an involvement-non-involvement effect, and
a groups by involvement interaction.

Hypothesis V was tested with a step—wise multiple re-
gression analysis using the model:

y = B0 + Ble + BZX2 + B3X3

where y - level of involvement

X1 = R-S score
X2 = ESC change score
X3 = cardiac change score

It Should be noted that the data on which this last
analysis is based are different from the data used in testing
the first four hypothesis. Here, instead of using two in-
volved and two non-involved vignettes for each subject
(which would result in an equal involvement level for every-
one) six vignettes for which the physiological records were
clear for all subjects were chosen and the involvement and
physiological variables were averaged across the six vignettes.
Thus, this analysis relies on scores for each subject, as

opposed to scores for each vignette.
Summar

Forty-five female students from Interpersonal Pro-
cess Recall classes were chosen as subjects on the basis
of their scores on the Byrne Repression-Sensitization (R—S)
Scale. They were videotaped and had their heart rates
and electrical skin conductance levels recorded as they
watched twelve vignettes geared to provoke some emotion-

al reaction. The videotape of them watching the

77

films was then played back for them and they were encouraged
to describe their recollection of what they were experienc-
ing as they watched the tape. This was in turn tape-recorded
and the tape was rated by two judges. A multivariate re-
peated measures design was used to test for a group effect,
during the baseline, a group effect while watching the films,
an involvement effect while watching the films and a group by
involvement interaction. In addition, a multiple regression
was used to predict level of involvement on the basis of

the R-S Scale score, change in heart rate, and change in

skin conductance. The rationales for the validities of the

Byrne Scale and the involvement scale were also discussed.

CHAPTER IV

ANALYSIS

In this chapter the results of the study with the
appropriate Statistical analyses are presented. The gen-
eral outline is to present first the descriptive statistics
which were used to summarize the data and to determine
whether parametric or non-parametric tests would be most
appropriate. Next the results of the repeated measures
multivariate analysis of variance are presented, followed
where appropriate, by the results of post-hoc Tukey compar-
isons. Lastly, a regression analysis is presented in which
involvement levels are predicted from personality test
scores and electrical skin conductance change scores.

Table 4.1 lists the means, the variances, the skewnesses,
and the kurtoses of all the physiological variables. Means
and variances were calculated according to standard formulas.
Skewnesses and kurtoses were calculated such that a normal
distribution would have a score of zero.

It will be noted that the only variables which seriously
depart from a normal distribution are the log transformations
of the electrical skin conductance levels. Since the only

purpose of carrying out the log transformations was to

78

Table 4.1 Descriptive Statistics for Physiological Variables

79

 

 

Variable Mean Variance Skewness Kurtosis
Base Mean Heart Rate 86.556 72.071 .420 -.036
Base Mean Skin Con-

ductance 20.418 67.651 .027 -.217
Base Log Mean Skin

Conductance 1.262 .056 -l.682 3.661
Base ESC Standard

Deviation 12.531 60.024 .621 -.262
Involved Mean Heart

Rate 88.356 92.109 -.270 .544
Involved Mean Skin

Conductance 21.568 54.533 -.294 -.180
Involved Log Mean Skin ‘
Conductance 1.293 .043 -2.086 6.108
Involved ESC Standard

Deviation 11.531 57.993 .534 -.566
Involved Heart Rate

Lability 51.167 43.943 .116 .002
Involved Heart Rate

Change 2.289 15.221 -.736 .313
Involved ESC

Change .244 1.393 .051 -.440
Uninvolved Mean Heart

Rate 90.822 137.240 1.018 .342
Uninvolved Mean Skin

Conductance 19.046 38.432 -.l35 .872
Uninvolved Log Mean Skin

Conductance 1.238 .035 -l.764 4.114
Uninvolved ESC Standard

Deviation 11.162 41.423 .610 .004
Uninvolved Heart Rate

Lability 49.032 65.223 .074 .127
Uninvolved Heart

Change 3.333 19.614 -.510 .017
Uninvolved ESC

Change -.111 1.385 .344 .016

 

80

increase the approximation to the normal distribution a
decision was made to replace the log transformations with
the original skin conductance levels. Making this substi-
tution, it is then clear that nowhere is the assumption of
normality too badly violated and therefore a parametric
statistical analysis is appropriate.

A multivariate analysis of variance is carried out
next. Four questions are asked: 1) is there a difference
between groups on base levels of the physiological variables,
2) is there a groups effect during stimulation on the physi-
ological variables, 3) is there an involvement effect during
stimulation on the physiological variables, and 4) is there
a group by involvement interaction? The individual cell
values are presented in Table 4.2. The MANOVA tables are
presented in Tables 4.3, 4.4, 4.5 and 4.6. When apprOpriate
univariate add Step down F ratios are included, along with
Tukey post-hoc comparisons.

Because the four hypotheses mentioned abOve are being
tested, in order to keep the overall probability of making
a Type I error at or below .05 it is necessary to set indi-
vidual critical values at .0125. Furthermore, the overall
alpha level for a given hypothesis (.0125) must be divided
by the number of variables that contribute to the mean vector
(three in the first case--mean heart rate, mean skin conduc-
tance and standardized standard deviation-- and six in the
others--mean heart rate, mean skin conductance, ESC stand-

ardized standard deviation, heart rate lability, heart rate

81

Table 4.2 Individual Cell Values

 

 

 

Variable Total Repressers Midlines Sensitizers
BHR 86.5556 88.333 86.0667 85.2667
BMSC 20.4178 25.0067 17.4133 18.8333
STBES .0004 .8700 -.2513 -.6173
IMHR 88.3556 89.3333 86.7333 89.0000
UMHR 90.8222 92.8333 89.3667 90.2667
IMSC 21.5678 26.1800 20.3600 18.1633
UMSC 19.0456 21.9667 18.4800 16.6900
ISTSC -.0000 .8467 -.3360 -.5107
USTSC .0004 .8313 -.5693 -.2607
IHRL 49.999 56.4826 48.3429 45.1736
UHRL 50.0644 52.879 48.2309 49.0832
IHRC 2.2889 1.2333 2.6000 3.0333
UHRC 3.3333 3.7333 2.4333 3.8333
IESCC .2444 .8333 -.1000 0

UESCC -.1111 -.5667 .1000 .1333
Key: BHR = Base Heart Rate, BMSC Base Mean Skin Conductance,

STBES

standardized base electrical Skin conductance
ard deviation, IMHR = involved mean heart rate, UMHR = unin-
volved mean heart rate, IMSC = involved mean skin conductance,

UMHR = uninvolved mean skin conductance, ISTSC = involved
standardized Skin conductance standard deviation, USTSC =

uninvolved standardized skin conductance standard deviation,
IHRL = involved heart rate lability, UHRL = uninvolved heart

rate lability, IHRC = involved heart rate change, UMHRC =

uninvolved heart rate change, IESCC

involved electrical skin

conductance change, UESCC = uninvolved electrical Skin con-
ductance change

Stand-

82

lability, heart rate change and ESC change scores) leaving
a critical value. of .0041 or .0021. Finally, the reader
is cautioned in interpreting step down F ratios, that these
ratios provide residual tests of Significance so that when
statistical significance is reached in the case of one var-
iable one may not validly continue to test successive var—
iables. The reader is also cautioned that Tukey post-hoc
comparisons were performed at the .05 level of significance,
considerably above the overall alpha level described above.
They are included here as a data-snOOping technique to sug-
gest hypotheses for future research and are not intended as
formal tests of these hypotheses.

Hypothesis 1: Repressers, midlines, and sensitizers
differ on physiological measures mean heart rate, mean skin
conductance level, and standardized electrical skin conduc-

tance standard deviations during the base period.

Table 4.3 Multivariate and Univariate F Tests H0:1
F Ratio for Multivariate Test = 5.3537 D.F. = 6, 80
P less than .0002

 

 

 

. Hypothesis Univariate P less Step P less
Variable Mean Square F than Down F than
Bas HR 37.9556 .5150 .6013 .5150 .6013
Base ESC 244.4629 4.1272 .0231 3.6380 .0352
Standardized
Standard Dev. 9.0088 14.5475 .0001 12.6551 .0001
Tukey (.05 level) Sensitizers Midlines
Standardized
Standard Deviations -.6173 -.2513

Repressers

.8700

83

The results in Table 4.3 indicate that personality
groups do differ in their physiological base rates. Both
the univariate and step down F ratios indicate that the
standardized base ESC is a Specific physiological variable
which differentiates between groups. The Tukey comparison
suggests that repressers have a higher ESC standardized
standard deviation than sensitizers or midlines.

Hypothesis 2: Repressers, midlines, and sensitizers
differ on physiological measures mean heart rate, mean skin
conductance, standardized electrical skin conductance
standard deviation, heart rate lability, heart rate change,

and electrical skin conductance change during stimulation.

Table 4.4 Multivariate and Univariate F Tests H0:2

F Ratio for Multivariate Test = 3.8515 D.F. = 12, 74
P less than .0002

 

 

variable giggtgziiie univifiate Ptfiiis Difiipr Ptfiiis
Mean HR 138.1056 .3395 .7141 .3395 .7141
Mean ESC 698.0507 4.9063 ' .0122 4.6270 .0155
ESC Stand. 31.7270 12.9838 .0001 11.6389 .0002
Standard

Deviation

HR Lability 270.7167 2.0047 .1346 1.3181 .2791
HR Change 17.4389 .6509 .5268 4.1761 .0230
ESC Change .2667 .1080 .8979 .5277 .5944
Tukey .05 level Midlines Sensitizers

ESC Standardized -.452 -.385

 

Standard Deviation
Repressers
1.6780

5'

84

The multivariate F ratio given here indicates that
personality groups did differ while watching the films.

Both the univatiate and step down F's indicate that stand-
ardized electrical skin conductance is a particular variable
which registered this difference. The Tukey intervals
suggest that repressers have a higher standardized standard
deviation than the other two groups.

Hypothesis 3: There are differences on the physiolog-
ical variables mean heart rate, mean Skin conductance, stand-
ardized electrical skin conductance standard devitaion, heart
rate lability, heart rate change, and electrical skin con-
ductance change between those vignettes with which the sub-
jects were rated as being involved and those vignettes with

which the subject was rated as being uninvolved.

Table 4.5 Multivariate and Univariate F Tests H0:3

F ratio for Multivariate Test = 3.3683 D.F. = 6, 37
P less than .0095

 

 

. Hypothesis Univariate P less Step P less
Variable Mean Square 5‘ than Down F than
Mean HR 273.8000 4.1333 .0484 4.1333 .0484
Mean ESC 286.2722 16.1707 .0003 11.1645 .0018
Stand. ESC .0000 .0000 .9947 .2607 .6125
Standard
Deviation
HR Lability 14.4500 .1502 .7003 2.8771 .0979
HR Change 49.0889 1.1140 .2973 .1361 .7143

ESC Change 5.6889 2.0645 .1582 .7421 .3946

 

85

The F ratio in Table 4.5 indicates that there is a
significant difference between involved and uninvolved
vignettes. The step down and univariate F's indicate that
mean skin conductance is particularly sensitive to this
difference. It is unnecessary to perform Tukey post-hoc
comparisions because there are only two groups--involved
and uninvolved-—being compared here. The actual cell values
for the one significant variable--mean Skin conductance, are

21.5678 micromhos and un-

involved mean skin conductance

involved mean skin conductance 19.0456 micromhos.
Hypothesis 4: There is no significant interaction

between involvement ratings and personality types.

Table 4.6 Multivariate Test H0:4

F Ratio for Multivariate Test = 1.9631 D.F. = 12.74
P less than .0401

(No further statisitcs are given for this hypothesis because

the null hypothesis is not rejected.)

Judging by the F ratio in Table 4.6 the null hypothesis
cannot be rejected; therefore, one cannot conclude that there
is a significant interaction between personality groups and
involvement .

Next, the question "does a knowledge of a subject's R-S
score, cardiac change score, and ESC change score allow a
predicition to be made regarding the subject's involvement
with the vignettes" is asked. Table 4.7 contains the cell
values computed from the six vignettes used for the regression

analysis.

86

Table 4.7 Cell Values Used for Testing H0:5

 

Repressers Midlines Sensitizers

 

Involvement Rating .3467 .5780 .6593
Heart Rate Change 2.1000 1.9733 2.0867
ESC Change 1.6400 1.7800 1.6400

 

Hypothesis 5: The cardiac and ESC change scores and

the R-S scores will predict a subject's level of involvement.

Table 4.8 Multiple Regression Analysis

 

 

l . . . - 2 .

. F Signifi- Multiple 2 R Simple
Variable Ratio cance R R Change R
RS 12.630 .001 .476 .227 .227 .476
ESCC 8.185 .007 .594 .353 .126 .380
HRC .865 .358 .605 .366 .013 .173

 

From this analysis one may set up the following regression
equation, using RS scores and ESC change scores to predict
involvement level:

y = .2408 + .0045 (RS) + .0293 (ESCC)

where y = involvement 1evel,RS = Byrne Scale score,
ESCC = electrical Skin conductance change score

Table 4.9 95% Confidence Intervals for Beta Weights

Constant .0964 < 8 < .3852
RS .0020 < 8 < .0070
ESCC .0072 < B < .0513

87

If one treats R2 as an indicator of the percentage of
the variance accounted for by a variable, one will note that
by itself R-S scores account for over 22% of the involve-
ment variance; taken together with ESCC the two variables
account for over 35% of the involvement variance. HRC is
dropped because it does not appreciably add to the predice
tion of involvement.

If one examines the individual ESCC values for the
different personality groups (Table 4.2) one notes that
there is a much bigger difference between the involved and
uninvolved scores fOr repressers than for the other two
groups. Because of this observation, it was decided to
compute the regression for the various sub-groups, although
this was not originally planned. Because HRC did not contri-
bute significantly to the predicition of involvement and
the values in Table 4.2 gave no reason to think that HRC

varied across R-S groups, it was dropped from the computation.

Table 4.10 Simple Regression of Involvement on ESCC by Group

 

 

Group F Ratio Significance R R2 Beta
Repressers 41.206 .000 .872 .760 .0849
Midlines .927 .353 .258 .066 .0191
Sensitizers .180 .679 .117 .014 .0064

 

The szalues indicate that for repressers ESCC accOunts

for 76% of the involvement variance; for midlines ESCC accounts

88

for over 6% of the variance; for sensitizers ESCC accounts

for over 1% of the variance.

Summary

The results of the tests of five research hypotheses
were presented in this chapter. The first hypothesis was
that personality groups would differ in physiological base
rates. The multivariate test at the .0002 level indicated
that this was so. Step down F ratios indicated that the
standardized electrical skin conductance standard deviation
contributed most of any variable tested to this effect.

The second hypothesis tested was that personality groups
would differ in phySiological response to the vignettes.

A multivariate test at the .0002 level indicated that this
was SO. Step down F ratios indicated that standardized
electrical skin conductance standard deviations contributed
most heavily to this effect. Tukey post-hoc comparisons at
the .05 level suggested that repressers have a greater
standard deviation than the other two groups. The third
hypothesis tested was that there would be a difference in
physiological response between vignettes with which the sub—
ject was rated as being involved and those with which She
was rated as being uninvolved. The multivariate test at

the .0095 level indicated that this was so. Step down F
ratios showed that mean skin conductance in the involved
vignettes was significantly higher than mean skin conductance

in the uninvolved vignettes. The fourth hypothesis tested

89

was that there was an interaction between personality groups
and involvement with the vignettes. This was not confirmed
by the data. The fifth hypothesis tested was that a linear
multiple regression equation could be written which would
predict involvement from Repression-Sensitization scores,
electrical skin conductance (ESC) change scores, and heart
rate change scores. An equation was written accounting for
approximately one third of the involvement variance using

RS scores and ESC change scores.

CHAPTER V

SUMMARY, CONCLUSION, DISCUSSION AND IMPLICATIONS

Summary

Personality theorists have paid much attention to
psychological defensiveness. One conceptualization Of de-
fensiveness which was reviewed in detail here is Byrne's
(1961) notion that defensiveness may be viewed as a be-
havioral continuum ranging from repressing and ignoring
behaviors on the one hand, to sensitizing and distorting
behaviors on the other. PsychOphysiologists have devoted
a great deal of attention to the correlates of an organism's
accepting and/or rejecting stimulus input. This study was
an attempt to investigate the relationship between Byrne's
conceptualization of defensiveness and hypothesized psycho-
physiological correlates of defensiveness. Two specific
physiological measures--heart rate and electrical skin con-
ductance (ESC)--were analyzed. According to the work of
Lacey (1956), heart rate tends to increase when an organism
shuts out stimuli; conversely, when the organism attends to
stimuli heart rate tends to decrease. Therefore, it was
hypothesized that when a subject becomes involved with a

filmed vignette She is attending to the vignette, and therefore

90

91

her heart rate should decrease. When she is uninvolved
with a vignette she is likely not to be attending to it,
and her heart rate Should increase.

The work of Edelberg (1970, 1972) provided the back-
ground for making predictions about ESC activity. According
to this work when subjects are defensive and shut out in-
coming stimulation ESC activity maintains or increases its
rate. When a subject accepts environmental input ESC acti-
vity decreases. Therefore, it was hypothesized that when a
subject is involved with stimulus vignettes her ESC activity
should decrease; when she is uninvolved, ESC activity should
remain the same or increase.

The literature (Davison, 1963; Simal & Herr, 1970)
suggested that repressers should show more physiological
arousal than sensitizers. In an attempt to replicate this
finding, differences between personality groups as defined
by the Repression-Sensitization (R-S) Scale were hypothesized.

Forty-five female subjects from Education 484 classes
were picked, on the basis of their scores on the R-S Scale,
such that fifteen could be considered repressers, fifteen
midlines, and fifteen sensitizers. Each subject was shown
twelve stimulus vignettes while her heart rate and ESC were
being recorded; simultaneously, she was video recorded. After
watching the vignettes and taking a memory test about them,
she watched the videotape of herself viewing the film. She
was asked to stop the tape and describe her experiences

while the tape was being made, according to the recall

92

procedures described by Kagan gt gt. (1969). A tape re-
cording of this recall was then rated by two judges. From
their ratings, two vignettes per subject were designated
as "involved"--meaning that the sdbject had verbalized an
emotional reaction to them--and two were designated as
”uninvolved"--meaning that the subject had failed to ver-
balize any reaction to the vignettes. A multivariate
analysis of variance was carried out on this data. The
results indicated that repressers were more variable than
other groups--ie. they had larger standardized standard
deviations than sensitizers or midlines-~during the base
rate and while watching the films. The results also indica-
ted that subjects had a lower mean skin conductance level
when uninvolved than they did when they were involved. No
statistically significant interaction was detected in this
analysis.

A multiple regression analysis based on the same six
vignettes for all subjects was performed attempting to pre-
dict overall involvement ratings from R-S scores, ESC change
scores and heart rate change scores. For all subjects
approximately one-third of the involvement variance could
be accounted for by R-S scores and ESC change scores. How-
ever, when separate analyses were run on the different per-
sonality groups ESC change scores could account for
approximately 76% of the involvement variance for repressers,
6% of the involvement variance for midlines, and 1% of the

invlovement variance for sensitizers.

93

Conclusions

 

l. A significant difference exists between R-S groups on
the physiological measures tested during their base rates.
Variability of ESC, as measured by the standardized stand-
ard deviations, reflects this difference best. Repressers
seem to be more variable than the other groups.

2. A significant difference exists between R-S groups on
the physiological measures tested while they were watching
the films. Variability of ESC, as measured by the stand-
ardized standard deviation, reflects this difference best.
Repressers seem to be more variable than other groups.

3. A significant difference exists in physiological response
between those vignettes with which the subject was rated as
being involved and those vignettes with which the subject
was rated as being uninvolved. In particular, mean skin
conductance was Significantly lower when the subjects were
uninvolved than when they were involved with the vignettes.
4. No significant interaction between involvement and R-S
groups was found in the data based on the two involved and
the two uninvolved vignettes per subject.

5. R-S scores and ESC change scores account for 35% of the
involvement variance. Heart rate change scores do not account
for an appreciable percentage of the variance.

6. The data collected from the same six vignettes per sub-
ject suggests that ESC change scores are an extremely potent

predictor of involvement for repressers--accounting for 76%

94

of the involvement variance--while it is a poor predictor
of involvement for midlines and sensitizers--accounting for

approxiamtely 6% and 1% respectively.

Discussion

 

It seems clear on the basis of the findings reported
above that none of the hypotheses concerning heart rate
were supported. There can be two explanations for this.
The first possibility is that increasing or decreasing heart
rate is not a correlate of psychological involvement in
this case because of some idiosynchracy of this study.
Perhaps this is true because of the relatively low level
of stimulation (compared with real life situations or movies
of gruesome accidents) used in this research. Perhaps it
is true because the subjects were females or because of
some other demographic characteristic of the sample (one is
reminded that Lacey's most impressive findings were based
on male subjects [Lacey gt gt., 1963]). Perhaps this is
true because of the varied subject matter of the vignettes.
(The reader will recall that in one study Lacey obtained
different physiological reSponses depending on the nature
of the topic--sexual or aggressive--being dealt with
[Lacey gt gt., 1963]). Perhaps this is true because the
measurement procedures used here were not sensitive enough
to pick up the small .cardiaC‘ changes Lacey does. The
second possibility is that the hypothesis that increasing

heart rate is associated with rejection of the environment

95

is simply wrong. Perhaps Obrist gt gt.'s (1970) explana-
tion of cardiac rate in terms of metabolic requirements or
Westcott and Huttelocher's (1961) explanation in terms of
respiratory artifact or some as yet unformulated.theory
has greater validity. This research however, Sheds no
light on the probabilities of these possibilities. All
that can be said is that the results cannot be taken to
support Lacey's hypothesis; however, there are sufficient
differences in the characteristics of the experiments so
as not to be damaging to his hypothesis either.

Results of this study are considerably more positive
with regard to ESC. ESC variability does appear to be
greater for repressers than for other members of the pop-
ulation. This confirms the results of Davison (1963), and
Lazarus and Alfert (1964). This great variability in con-
ductance response on the part of repressers in at least
suggestive of the Freudian notion that it takes a great deal
of psychic energy to defend against something coming into
consciousness. Clearly, however, this hypothesis requires
further study.

Surprisingly, mean ESC level was greater during vignettes
when subjects were involved with the material than when they
were uninvolved. The original thinking here suspected the
Opposite--that when subjects were involved there would be a
decreasing ESC level, manifesting itself as a lower ESC mean.
In fact, looking at the data shows that that did happen. How-

ever, involved vignette scores also reached a higher maximal

96

level at the start of a vignette than did uninvolved vig-
nette scores, thus yielding the reported results. This
suggests that perhaps ESC slope might have been a superior,
more direct measure for the present purposes. This also
suggests a possible explanation for some apparent discrep-
ancies in the literature. For some groups ESC activity as
measured by mean level may be decreasing while ESC activity
as measured by responses (bumps) may be increasing.

When ESC change scores are examined in the multiple
regression analysis Edelberg's (1972) hypothesis receives.
its strongest support. Taken alone, ESC change scores
account for over 14% of the variance for all subjects.

Taken together with RS scores they account for over 35% of
the variance. However, if one just looks at repressers,
ESC change accounts for a whopping 76% of the variance.

This poses some interesting questions. The first of
these is why is there such a clear interaction between RS
and ESC change scores in the regression analysis, but not
in the MANOVA? The reader is reminded that rejection of
the null hypothesis was much more difficult in the MANOVA
because more hypotheses and more variables were being tested,
resulting in lower critical values. Secondly, the reader is
reminded that the two analyses were performed on different
sets of data. In the case of the MANOVA two involved and
'two uninvolved vignettes were arbitrarily selected for
subject. In the case of the regression the same six vignettes

were selected for each subject. This means that some

97

vignettes, such as ones with which the subject became only
reluctantly involved, were included in the regression anal-
ysis although they were excluded from the previous analysis.
This difference concerning the interaction suggests some
considerations which will be discussed under "Implications
for Future Research." However, because the regression anal-
ysis here provides the more powerful test and because the
direction of differences between the groups on ESC change
scores is the same in both sets of data, it is assumed

that an interaction does in fact exist.

The second question posed by the large difference in
percentage of involvement variance accounted for by ESC
change scores within the separate groups is why this is so.
One hypothesis is that the answer lies in the differential
meaning of involvement for repressers and sensitizers. A
represser who indicated she was involved was not likely to
have been defensive while watching the vignette. (Repressive
defenses would tend towards denying involvement.) A sensi-
tizer who indicated she was involved may or may not have
been defensive. (Sensitizing defenses might manifest them-
selves as over-involvement--perhaps not truly felt by the
subject, but nonetheless, rated as true involvement on the
rating scale employed.) The involvement scale does not
allow one to differentiate between true involvement and de-
fensive involvement. Therefore, when a represser is rated
as being involved, one can equate the work "involved" with

”non-defensive"; when a sensitizer is rated as being involved,

98

one cannot make such an inference.

If the data is seen in this light, it strongly sup-
ports Edelberg's (1972) contention that "enhanced electro-
dermal activity with retarded recovery may signal a
defensive reaction." When there is a clear indication of
psychological defense, ESC change scores behave as Edelberg
would predict--1arger ESC change scores for involved than
uninvolved vignettes. To test fully this hypothesis one
would want to Show that when sensitizers and midlines are
defensive their ESC change scores behave similarly. The
problem in this study is that for those two groups there is
no clear measure Of when they are being defensive. Again,
the reader is reminded that this interpretation is not
supported in that there was no significant interaction in
the MANOVA. Therefore even the extremely low critical alpha
levels, the relatively small number of subjects, and the
fact that the results are in the same direction as in the
regression analysis all suggest that this is a valid result,
caution is advised.

If this finding is correct, it also suggests a possible
explanation for discrepant results in the literature. Most
studies do not report the characteristics of their samples in
regard to Openness to stimulation. It is possible that sub-
jects in studies coming to differing conclusions differ in

psychological defensiveness.

99

Subjective Observations

 

The subjective observations of the experimenter fall
into three broad categories: remarks about the subjects,
comments about the procedure, and impressions of the data.

Sensitizers tended to be almost hysterical, somewhat
flighty, and highly enthusiastic. More often than not they
seemed to be nervous about the procedure and Often asked
questions to confirm that they really were not going to be
given "Shots" or to ascertain whether or not the movies were
"gross." Repressers, on the other hand, rarely verbalized
any of these concerns. Interestingly, however,of the five
people who missed their original appointments and needed to
be re-scheduled, four were repressers. The fifth, although
not a represser, as defined by the strict criteria in this
study, tended toward the repressing end of the R-S score
class distribution.

There are some important costs and benefits associated
with using the procedure employed in this study. Its biggest
drawback is its length and cumbersomeness. It requires
approximately two hours of time per subject and a substantial
initial financial outlay for the equipment. Furthermore, it
lacks highly discrete stimuli and relies on the Skill of a
clinical "inquirer" (the person asking questions of the
subject) which is difficult to keep constant across subjects.
However, it is precisely some of these "drawbacks" which are

simultaneously the greatest assets of the method. For example,

100

the lack of highly discrete stimuli is a function of its
relatively high approximation to the real world (as compared
with a photic flash). Its reliance on a Skilled inquirer
allows the subject to deal with what was on her mind. The
importance of this can be illustrated by the following
anecdote: one vignette was of a black male threatening,
with obscene language, the viewer. Most viewers saw this
as an aggressive theme and, if they reacted, reacted with
fear, anger, or some defensive posture. Two viewers focused
on the fact--unti1 then completely unnoticed by the exper-
imenter--that the man in the film had a gap between his two
front teeth. As both these subjects had similar character-
istics their reactions were ones of sympathy for the "poor,
ostracized" individual. In all likelihood more clean-cut,
precise experimental procedures would have missed these
sorts of nuances.

Finally, the records themselves deserve some comment.
As suggested by the findings already presented, clear visual
differences in ESC are found between the involved and unin-
volved represser vignettes. Visual inspection also differ-
entiates repressers from sensitizers. Examples, taken from
actual records are given in Figure 5.1. These examples
suggest that useful data may be obtained by using two addi-
tional measures of ESC activity: slope and a measure which
would differentiate between large and small responses (the
reader is reminded that in this study all "bumps" of k

micromho or greater were lumped together.

101

350m 35pm

 

Uninvolved Represser Involved-Represser

20pm .20um

Uninvolved Sensitizer Involved Sensitizer

Records of An Involved and Uninvolved Vignette for a
Represser and a Sensitizer

Figure 5.1

102

Visual inspection of the data also suggests ESC mean
levels differentiate between personality groups; they also
suggest that ESC mean level is sensitive to an interaction
between involvement and personality groups. The figures in
Table 4.2 tend to support this, although this is not sup-
ported by the statistical tests reported in the previous
chapter. One reason for this may be a large error variance
with respect to ESC mean level. This big error variance
suggests that perhaps the personality categories used in
this study can be further refined so as to reduce the amount

of individual variation within the groups.

Implications for Future Research

 

One need for further research is the need to confirm or
reject the specualtion that the large interaction suggested
in the multiple regression between ESC change scores and R-S
Scale scores is valid. This study should be replicated
using fewer physiological variables—-perhaps only ESC change
scores and ESC standardized standard deviations.

A theoretical question is raised about whether ESC
change scores correlate with psychological defense generally
or only with repression. Specifically, some measure of
sensitizing defense--which differentiates between true in-
volvement and defensive involvement--needs to be taken and
then correlated with ESC change scores. It would be ideal if
the rating of "non-involvement" could differentiate between

a non-defensive lack of involvement, due perhaps to disinterest

103

and a defensive non-involvement.

A second theoretical question about whether the greater
variability of represser's ESC levels correlates with Freud's
notion of the expenditure of psychological energy to main-
tain psychological defense deserves consideration. Measur-
ing ESC levels in a variety of situations in which one would
expect to find differential psychological energy expenditures
would be helpful.

Another question which is raised by this study is
whether further refinement of the personality categories will
yield stronger correlations between physiological measures
and psychological constructs. In addition, just as psycho-
logical categories can be further refined, so can the
physiological measures. Discriminating amount of ESC re-
sponse and looking at slope of ESC appear to be two such
possible refinements.

A variety of questions concerning heart rate and en-
vironmental screening remain. Would positive results have
been obtained with male subjects? Would dividing vignettes
according to theme yield different results? IS some other
translation of Lacey's "acceptance and rejection of envir-
onmental stimuli" such as "approach and "avoidance" of
stimuli, rather than involvement or non-involvement with
the stimuli, more accurate?

Finally, it would be interesting to learn if the ESC
change score patterns associated with represser's Openess

and defensiveness have any practical use. Using a represser

104

as a client, a future study might provide on-going feedback
about ESC activity to the therapist, the client, and/or both,
seeking to see if such feedback affects the outcome or rate

of therapy.

APPENDICES

APPENDIX A

Instructions for Rating Scale

 

Rate as involved if the person says he felt a par-
ticular feeling or wanted to do a particular behavior.

The only exception to this is if the only behavior the
person can identify is laughter or if the only feeling is
that it (the vignette) was humorous.

Rate as "hesitantly involved" any vignette where the
person labels a reaction cautiously--ie. I would have felt
scared--I kind of wanted to help him etc. Rate as unin-
volved any vignette where the person says "I was bored,"

I didn't get into it, etc. or where the only reaction is
laughter or humor. Also rate as uninvolved any vignette
the person can't recall or labels no reaction. Any vignette
not rateable in one of the above three categories rate as

"can't be rated."

105

APPENDIX B

Memory_Test

 

RECALL THE VIGNETTE WHERE A WOMAN ASKS WHETHER ANYONE HAS
EVER SAID ANYTHING ABOUT YOUR SHAPE.

What is the length of her hair?
What is the color of her blouse?

RECALL THE VIGNETTE WHERE A WOMAN CRIES, SAYING SHE THOUGHT
YOU LIKED HER.

What is her approximate age?
Describe her hair style.

RECALL THE VIGNETTE WHERE A MAN APOLOGIZES AND SAYS HE FEELS
BAD AND CRIESr

Describe any facial hair that he had.
What was he wearing?

RECALL THE VIGNETTE WHERE A MAN SAYS THAT HE'S NOT SCARED OF
YOU.

What was the color of his top (word that person
used in question before).

 

Was he wearing any jewelry?

RECALL THE VIGNETTE WHERE A COUPLE WAS LAUGHING.
Who is sitting on your right, the man or the woman?
Which one, if any, are wearing glasses?

RECALL THE VIGNETTE WHERE A WOMAN SAYS YOU DON'T WANT ME
AROUND HERE ANY MORE.

What was the color of her blouse?

106

107

Describe any jewelry she may have been wearing.

RECALL THE VIGNETTE WHERE A WOMAN SAYS YOU HAVE A GREAT
DEAL OF POTENTIAL.

How did her blouse close? eg. zippered?
How much of her could you see?

RECALL THE VIGNETTE WHERE A MAN SAYS HE REALLY LIKES BEING
WITH YOU.

Describe any facial hair he had.

What color was his shirt?

RECALL THE VIGNETTE WHERE A MAN SAYS HE WANTS TO FEEL YOU.
How long is his hair?

How did his shirt close?

RECALL THE VIGNETTE WHERE A MAN SAYS HE THOUGHT YOU REALLY
CARED.

Is he wearing a ring?

During part of the vignette he looks at you and during part
of the vignette he looks down. Which comes first?

RECALL THE VIGNETTE WHERE A WOMAN SAYS THAT NO ONE HAS
EVER BEEN SO NICE TO HER.

How long were her sleeves?-
What was the color of her dress?

RECALL THE VIGNETTE WHERE A MAN SAYS HE'S GOING TO KICK
YOUR.ASS.

Describe any facial hair he may have had.

Describe the pattern on his shirt.

APPENDIX C

Background Sheet

 

Name

 

Age

 

Marital Status

 

Are you taking any medication

 

If so, what?

 

When was the last time you ate

 

When was the last time you had coffee or tea _

When was the last time you smoked a cigarette

108

 

 

APPENDIX D

Vignette Typescript

 

Female (middle-aged): Anyone ever said anything to you
about your shape? Well, its different. Not that that's
not fine, but it is different.

Female (young): I thought you liked me. I really thought
you liked me. I guess I know what you're thinking. I ah
I thought that you really liked me and I uh oh wow...

Male (young, red sweater): I feel pretty bad. I can't.

I mean uh. I tried. I tried and I just uh can't do it.

I mean I'm sorry. I mean I'm sorry I just uh my stomach
is getting all uh. I'm, all I'm just getting all up tight
and I couldn't do it. I'm sorry. I'm sorry.

Male (young, red sweater): So what are you looking at.
I'm not scared of you. You can't do anything to me. You
piece of shit.

Black male and female: Laughing at you.

Female (old, green blouse): I know you don't want me
around here. I know you just don't want me here anymore.
You don't need me. Alright. You don't want me here, I'll
leave.

Female (young): You did pretty well, uh. I think you have
a great deal of potential. Um. I think if you keep working
at it. And keep trying. Um. You'll get much better.

Male (young, with moustache): You know what. I really like
being with you. You really make the day special you know.
I just really think you're neat. You really are.

Male (young, with moustache): I want you. I want to feel
you. I want to lie next to you. I want to touch you. And
I want you to touch me.

Male (young, with moustache): I thought we had something.
I thought that you really cared. I thought you felt the
way I did. Why did you let me make a fool of myself?

109

110

Female (young, low cut blouse): This is really hard for
me to say. You've been very nice to me. No one ever has
been that nice to me. I just wanted to say I appreciate
it.

Male (black): I tell you man I can't take much more. Get
the fuck off my back. I'm going to have to kick you ass.
I'm going to have to kick your fucking ass.

APPENDIX E

Procedures for Operating the Polygraph

 

Turn main power switch on at least one-half hour before
machine will be used.

Turn to "stand-by" the power switches of the two ampli-
fier units to be used at least one half-hour before
machine will be used.

After leads have been attached to the subject turn power
switch on cardiac amplifier to "on."

Adjust driver sensitivity and base position if necessary.
Base line should be at the mid-line and sensitivity
should be such that -100 mv will move the.needle 2 cm.
Turn polarity switch to "use" (up).

Tachograph scale should be set at 60-180.

Turn tackograph to calibrate - center, +2 cm., -2 cm, and
mark on the record.

Turn tackograph to AC low. Adjust threshold so that the
record is clear. If this cannot be done turn tackograph
to AC Fast. Re-adjust threshold.

Turn amplifier on.

Adjust driver sensitivity and base position if necessary.
Base line should be at the mid-line and sensitivity should
be such that -100 mv will move the needle 2 cm.

Turn polarity switch to "use" (down).

Turn sensitivity to .01 mv/cm.

Bring needle back to midline using "zero adjust" on the
skin conductance bridge.

Turn sensitivity down to l mv/cm.

111

10.

11.

12.

112

Switch subject in on skin conductance bridge.

Bring needle to approximately midline position by bal-
ancing micromhos on the skin conductance bridge. Record
the number of micromhos needed to balance on the record.

Turn timer on.

Run the machine at speed of 100 mm/minute.

 

 

BIBLIOGRAPHY

BIBLI OGRAPHY

Altrochi, J., Parsons, O.A. & Dickoff, H. Changes in self-
ideal discrepancy in repressers and sensitizers.
Journal of Abnormal and Social Psychology, 1960, 61,
67-720

Archer, J., Fiester, T., Kagan, N., Rate, L., Spieling, T.
& Van Noord, R. New method for education, treatment,
and research in human interaction. Journal of Counsel-
ing Psychology, 1972, 19, 275-281.

 

Aristotle, On the Soul (Translated by W.D. Ross) In R.M.
Hutchins (ed.) Great Books of the Western World (vol.
8) Chicago: Encyclopaedia Britannica, 1952 (c. 340-
322 B.C.).

Ax, A.F. The physiological difference between fear and
anger in humans. Psychosomatic Medicine, 1953, 15,
433-442.

 

Bahrick, H.P. Incidental learning under two incentive
conditions, Journal of Experimental Psychology, 1954,
11, 170-172.

Benjamin, L.S. Statistical treatment of the law of ini-
tial values (LIV) in autonomic research: A review and
recommendation. Psychosomatic Medicine, 1963, g;, 556-
566.

Berlyne, D.F. Conflict and the orientation reaction. Journal
of Experimental Psychology, 1961, 63, 476-783.

Berlyne, D.E. Curiosity and exploration. Science, 1966, 153,
25-33.

Berlyne, D.F., Craw, M.A., Salapatek, P.H. & Lewis, J.L.
Novelty, complexity, incongruity, extrinsic motivation,
and the GSR. Journal of Experimental Psychology, 1963,
66, 560-567. .

Bernstein, A.S. Phasic electrodermal orienting response in
chronic schizophrenia: II. Response to auditory sig-
nals of varying intensity. Journal of Abnormal Psycho-
logy, 1970, 15, 146-156.

113

114

Bittker, T.E., Buchsbaum, M.S., Williams, R.B. & Wynne, L.C.
Cardiovascular and neurOphysiologic correlates of sen-
sory intake and rejection II. Interview behavior.
Psychophysiology, 1975, 13, 434-438.

 

Block, J. Ego control scales in the MMPI. In W. Dahlstrom &
G. Welsh (eds) An MMPI handbook: A guide to use in
clinical practice and research. Minneapolis: Univer-
sity of Minnesota Press, 1960.

 

 

Bull, K. & Lang, P.J. Intensity judgments and physiological
response amplitude. Psychophysiology, 1972, 9, 428-436.

 

Buss, A.H. Psychopathology. New York: Wiley, 1966.

 

Byrne, D. The repression-sensitization scale-rationale,
reliability, and validity. Journal of Personality,

 

Byrne, D. Repression-sensitization as a dimension of person-
ality. In B.A. Maher (ed.) Progress in Experimental
Personality Research, vol. 1, Academic Press: New York,
1964, 170-220.

 

 

Byrne, D., Barry, J. & Nelson, D. Relation of the revised
repression-sensitization scale to measures of self-
description. Psychological Reports, 1963, 1;, 323-334.

 

Byrne, D., Golightly, C. & Sheffield, J. The repression-
sensitization scale as a measure of adjustment: rela-
tionship with CPI. Journal of Counsulting Psychology,
1965, g2, 586-589.

Byrne, D. & Sheffield, J. Responses to sexually arousing
stimuli as a function of repressing and sensitizing
defenses. Journal of Abnormal and Social Psychology,
1965, 19, 114-118.

Byrne, D., Steinberg, M.A. & Schwartz, M.S. Relationship
between repression-sensitization and physical illness.
Journal of Abnormal Psychology, 1968, 1;, 154-155.

Callaway, N.E. & Dembo, D. Narrowed attention. A psycholog-
ical phenomenon that accompanies a certain physiological
change. Archives of Neurology and Psychiatry, 1958,

79, 74-90.

Campos, J.J. & Johnson, H.J. The effects of verbalization
instructions and visual attention on heart rate and
skin conductance. Psychophysiology, 1966, 3, 305-310.

 

115

Cannon, W.B. The James-Lange theory of emotion: A critical
examination and an alternative theory. American Jour-
nal of Psychology, 1927, 99, 106-124.

 

Carrier, N.A. & Orton, K.D. Skin conductance trends during
learning by bright, normal, and retarded children.
Journal of Comparative and Physiological Psychology,
1964, 99, 315-317.

Chase, M. Stress: The role of defensive predisposition
(repression-sensitization) in GSR reactivity, defen-
sive behavior and stress reduction. Unpublished
doctoral dissertation, University of Pittsburgh, 1967.

Crider, A. & Augenbraun, C.B. Auditory vigilance correlates
of electrodermal response habituation Speed. Psycho-

Danish, S.J. & Kagan, N. Emotional stimulation in counseling
and psychotherapy. Psychotherapy: Theogy, Research
and Practice, 1969, 9, 261-263.

 

Darrow, C.W. Electrical and circulatory responses to brief
sensory and ideational stimuli. Journal of Experimen-
tal Psychology, 1929, 99, 267-300.

 

 

Darrow, C.W. The rationale for treating the change in gal-
vanic skin response as a change in conductance.
PsychophysiolOgy, 1964, 9, 31-38.

 

Davis, R.C. Response patterns. Transactions of the New
York Academy of Science, 1957, 19, 731-739.

 

 

Davison, L.A. Adaptation of a threatening stimulus. Unpub-
lished doctoral dissertation. University of California,
Berkeley, California, 1963.

Deane, G.E. Human heart-rate responses during experimentally
induced anxiety: A follow-up with controlled reSpira-
tion. Journal of Experimental Psychology, 1964, 91,
193-195.

Deane, G.E. Human reart rate responses during experimentally
induced anxiety: effects of instructions on acquisi-
tion. Journal of Experimental Psychology, 1966, 1;,
772-73.

Depue, R.A. & Fowles, D.C. Electrodermal activity as an
index of arousal in schizophrenia. Psycholggical
Bulletin, 1973, 19, 233-238.

 

116

Descartes, R. Meditations. In L. Bair (transJ Essential
Works of Descartes, New York: Bantam, 1961 (1649).

 

 

Docter, R.F. & Freidman, L.F. Thirty-day stability of Spon-
taneous galvanic skin responses in man. Psychophysi-
ology, 1966, 9, 311-315.

 

Dublin, J.E. Perception of and reaction to ambiguity by
repressers and sensitizers: A construct-validity
study. Journal of Consulting and Clinical Psychology,
1968, 99, 198-205.

 

Dykman, R.A., Reese, W.G., Galbrecht, C.R., Ackerman, P.T. &
Sunderman, R.S. Autonomic responses in psychiatric

patients. Annals of the New York Academy of Sciences,
1968, 147, 237-303.

 

Easterbrook, J.A. The effect of emotion on cue utilization
and the organization of behavior. Psychological_
Review, 1959, 99, 183-201.

 

Edelberg, R. Electrical properties of the skin. In C.
Brown (ed.) Methods in Psychopysiology. Baltimore:
Williams & Williams, 1967.

 

Edelberg, R. The information content of the recovery limb
of the electrodermal response. Psychgphysiology,
1970, 9, 527-539.

 

Edelberg, R. Electrical activity of the skin. In N.S.
Greenfield, & R.A. Sternbach (eds.) Handbook of

Ps cho h siolo , Holt, Rinehart & Winston, Inc.:
New York, 1972, pp. 367-418.

Edelberg, R. Electrodermal recovery rate, goal-orientation,
and aversion. Psychophysiology, 1972, 9, 512-520.

 

Edelberg, R. Mechanisms of electrodermal adaptations for
locomotion, manipulation or defense. In E. Stellar
& J.M. Sprague, J.M. (eds.) Progress in Physiological
Ps cholo , vol. 5, Academic Press, New YorE, 19737
155-2I0.

Edelberg, R. 5 Wright, D.J. Two galvanic skin response
effector organs and their stimulus specificity.
Psychophysiology, 1964, 9, 39-47.

Edwards, P. & Pap, A. (eds.) A Modern Introduction to
Philosophy, The Free Press: New York, 1965.

 

Elliott, R. The significance of heart rate for behavior: A
critique of Lacey's hypothesis. Journal of Personalipy
and Social Psychology, 1972, 99, 398-409.

117

Elliott, R., Bankout, B. & Light, T. Differences in the
motivational significance of heart rate and palmar
conductance: Two tests of a hypothesis. Journal
of Personality and Social Psychology, 1970, 99, 166-
172.

Epstein, S., Bourdreau, L. & Kling, S. Magnitude of the
heart rate and electrodermal response as a function
of stimulus input, motor output, and their inter-
action. PsychophysiolOgy, 1975, 99, 15-24.

 

Fechner, G.T. Elements of Psychgphysics (vol. 1) (Transla-
ted by H. Aller: ed. by E. Boring and D. Howes) New
York: Holt, Rinehart & Winston, 1966.

Feder C.Z. Relationship of repression-sensitization to
adjustment status, social desirability, and acquies-
cence response set. Journal of Consulting Psychology,
1967, 99, 401-406.

Feder, C.Z. Relationship between self-acceptance and adjust-
ment, repression-sensitization, and social competence.
Journal of Abnormal and Social Psychology, 1968, 19,
317-322.

Fenichel, O. The Psychoanalytic Theory of Neurosis. W.W.
Norton & Co., Inc.: New York, 1945.

Fenz, W.D. & Epstein, S. Manifest anxiety: unifactorial or
multifactorial composition? Perceptual and Motor Skills,
1965, 99, 773-780.

 

Féré, C. Note sur les modifications de la tension électrique
dans 1e corps humain. Comptes rendus des Séances de
la Société de Biologie, 1888, 99, 23533.

 

Flanagan, J. Galvanic skin response: emotion or attention?
Proceedings of the 75th Annual Convention of the Amer-
ican Psychological Association, Washington, D.C., 1967,
7-9.

Freud, S. The Dynamics of Transference In P. Rieff (ed.)
Therapy and Technique, New York: Collier Books, 1963,
(1912), 105-116.

Freud, S. A General Introduction to Psychoanalysis (Trans-
lated by Joan Riviere) Simon & Schuster: New York,
1971 (1924).

Gang, M.J. & Teft, L. Individual differences in heart rate
responses to affective sound. Psychophysiolqu: 1975,
99, 423-426.

 

118

Gellhorn, E. Central nervous system tuning and its impli-
cations for neuropsychiatry. Journal of Nervous and
Mental Disease, 1968, 147, 148-162.

 

 

Gellhorn, E. Further studies on the physiology and path-
ophysiology of the tuning of the central nervous
system. Psychosomatics, 1969, 99, 94-104.

 

Gellhorn, E. The emotions and the ergotropic and tropho-
tropic systems. Psychologische Forschung, 1970, 99,
48-94.

 

Gendlin, E.T. & Berlin, J.I. Galvanic skin response correl-
ates of different modes of experiencing. Journal of
Clinical Psychology, 1961, 91, 73-77.

 

Germana, J. & Klein, 8.8. The cardiac component of the
orienting response. Psychophysiology, 1968, 9, 324-
328.

Golin, S., Herron, E.W., Lakota, R. & Reineck, L. Factor
analytic study of the manifest anxiety, extraversion,
and repression-sensitization scales. Journal of Con-
sulting Psychology, 1967, 99, 564-569.

 

Goodenough, D.R., Witkin, B.A., Koulack, D. & Choen, H.
The effects of stress films on dream affect and on
reSpiration and eye-movement activity during rapid-
eye-movement sleep. Psychpphysiology, 1970, 99,
313-320. .

 

Gordon, A. & Glass, D.C. Choice ambiguity, dissonance and
defensiveness. Journal of Personality, 1970, 99,
264-272.

 

Gordon, J.E., Martin, B. & Lundy, R.M. GSRs during repression,
suppression, and verbalization. Journal of Consultisg
Psychology, 1959, 99, 243-251.

 

Graham, F.K. & Clifton, R.K. Heart rate change as a com-
ponent of the orienting response. Psychological
Bulletin, 1966, 99, 305-320.

 

Gray, A.L. Autonomic correlates of chronic schizophrenia:
A reaction time paradigm. Journal of Abnormal Psy-
chology, 1975, 99, 189-196.

Gruzelier, J.H. & venables, P.H. Evidence of high and low
levels of physiological arousal in schizophrenics.
Psychophysiology, 1975, 99, 66-73.

119

Haggard, E.A. On the application of analysis of variance
to GSR data: I. The selection of an appropriate
measure. Journal of Expsrimental Psychology, 1949,
99, 378-392.

Hare, R.D. Denial of threat and emotional response to im-
pending painful stimulation. Journal of Consulting
Psychology, 1966, 99, 359-361.

 

 

Hare, R.D. Orienting and defensive reSponses to visual
stimuli. PsychOphysiology, 1973, 99, 453-464.

Hare, R.D. & Blevings, G. Conditioned orienting and de-
fensive responses. Psychophysiology, 1975, 99, 289-
297.

 

Hebb, D.O. Drives and the CNS (Conceptual Nervous System).
Psychological Review, 1955, 99, 243-254.

Hobbes, T. Leviathan. In R.M. Hutchins (ed.) Great Books
of the Western World, vol. 23, Chicago: Encyclopaedia
Bfitannica, 1953 (I651).

 

Hodges, W.F. & Speilberger, C.D. Effect of threat of shock
on heart rate for subjects who differ in manifest
anxiety and fear of shock. Psychgphysiologyy 1966,
9, 287-294.

 

Hoffman, H. Use of avoidance and vigilance by repressers
and sensitizers. Journal of Consulting and Clinical
PSYChOlQHXI 1970, 99, 91-96?

 

Hord, D.J., Johnson, L.C. & Lubin, A. Differential effect
of the law of initial values (LIV) on autonomic var-
iables. Psychophysiology, 1964, 9, 79-87.

 

Horney, K. The Neurotic Personality of Our Time. W.W.
Norton & Co.: New York, 1937.

 

Howe, E.S. GSR conditioning in anxiety states, normals and
chronic functional schizophrenic subjects. Journal of
Abnormal and Social Psychology, 1958, 99, 183-189.

 

Hudson, L. Contrary imagination: aspsychological study of
the young student. .Schocken Books: New York, 1966.

 

Israel, N.R. Leveling-sharpening and anticipatory cardiac
response. Psychosomatic Medicine, 1969, 99, 499-509.

 

James, W. What is an emotion? Mind, 1884, 9, 188-205.

Johnson, L.C. Some attributes of spontaneous autonomic acti-
vity. Journal of Comparative and Physiological Psychol-
an, 1963' _5_6-’ 415-D20

120

Johnson, L.C. & Lubin, A. Spontaneous electrodermal acti-
vity during sleep and waking. Psychophysiology, 1966,
‘9, 8-17.

 

Jones, H.E. The study of patterns of emotional expression.
In M.L. Reymert (ed.) Feelings and emotions: The
Mossehart symposium. McGraw-Hill: New York, 1950.

 

 

Kagan, N. & Millhouse, J. Physiological responses of IPR-
trained and non-IPR trained teachers. Unpublished
manuscript. Michigan State University, 1975.

Kagan, N. & Schauble, P. Affect simulation in interper-
sonal process recall. Journal of Counsleipg_Psycho-
logy, 1969, 99, 303-313.

 

Kagan, N., Schauble, P., Resnikoff, A., Danish, 8., Krathwohl,
D. Interpersonal Process Recall. The Journal of Ner-
vous and Mental Disease, 1969, 148, 365-374.

 

 

Kaplan, F. Effects of anxiety and defense in a therapy-
like situation. Journal of Abnormal Psychology; 1966,
19, 449-458.

 

Kaplan, H.E. Social interaction and GSR activity during
group psychotherapy. Psychosomatic Medicine, 1963,
99, 449-458.

 

Kaplan, M.F. Interview interaction of repressers and sensi-
tizers. Journal of Consulting Psychology, 1967, 99,
513-516.

 

Katkin, E.S. & McCubbin, R.J. Habituation of the orienting
response as a function of individual differences in
anxiety and autonomic lability. Journal of Abnormal
Psychology, 1969, 99, 54-60.

 

 

Kintsch, W. Habituation of the GSR component of the orient-
ing reflex during paired-associate learning before and
after learning has taken place. Journal of Mathemat-
ical Psychology, 1965, 9, 330-341.

 

 

Kleinsmith, L.J. & Kaplan, S. Paired-associate learning
as a formation of arousal and interpolated interval.

Journal of Experimental Psychology, 1963, 65, 190-193.

Klorman, R., Wiensenfeld, A.R., & Austin, M. Autonomic
reSponses to affective visual stimuli. Psychophysiol-

Koepke, J.E. & Pribram, K.H. Habituation of GSR as a function
of stimulus duration and spontaneous activity. Journal
of Comparative and Physiological Psychology. 1966. 99,
442-448.

121

Lacey, B.C. & Lacey, J.I. Studies of heart rate and other
bodily processes in sensorimotor behavior. In P.A.
Obrist, A.H. Black,.J.Brener, & L.V. DiCara (eds.)
Cardiovascular PsychoPhysiology, Chicago: Aldine-
Athenton, 1974, 538-564.

Lacey, J.I. The evaluation of autonomic responses: Toward
a general solution. Annals of the New York Academy
of Sciences, 1956, 91, 123-164.

 

 

Lacey, J.I. PsychOphysiological Approaches to the evalua-
tion of psychotherapeutic process and outcome. Re-
search in psychotherapy, vol. 1. Proceedings of the
American Psychological Association Conference, I958,
Washington, D.C. 1959.

 

Lacey, J.I. Somatic response patterning and stress: Some
revisions of activation theory. In M.H. Appley & R.
Trumbull (eds.) Psychological Stress: Issues in Re-
search, Appleton-Century-Crofts: New York, 1967, 14-42.

 

Lacey, J.I., Kagan, J., Lacey, B.C., & Moss, H.A. The vis-
ceral level: Situational determinants and behavioral
correlates of autonomic response patterns. In P.H.
Knapp (ed.) Expressions of the Emotions of Man, Inter-
national Universities Press: New YorE, 1968, 161-196.

Lacey, J.I. & Lacey, B.C. Some autonomic-central nervous
system interrelationships. In P. Black (ed.) Physi-
ological Correlates of Emotion. Academic Press: New
York, 1970, 205-227.

Lacey, O.L. An analysis of the appropriate unit for use in
the measurement of level of galvanic skin resistance.
Journal of Experimental Psychology, 1947, 91, 449-457.

Lacey, O.L. & Siegel, P.S. An analysis of the unit of measure-
ment of the galvanic skin response. Journal of Exper-
imental Psychology, 1949, 99, 122-127.

 

 

Lader, M.H. Palmar skin conductance measures in anxiety and
phobic states. Journal of Psychosomatic Research,
1967, 99, 271-281.

Lang, P.J. The application of psychophysiological methods to
the study of psychotherapy and behavior modification.
In A.E. Bergin and S.L. Garfield (eds.) Handbook of
Psychotherapy and Behavior Changs, Wiley & Sons: New
- York, 1971, 75-125.

 

 

 

Lang, P.J. & Buss, A.H. Psychological deficit in schizo-
phrenia: II. Interference an activation. Journal
of Abnormal Psychology, 1965, 19, 77-106.

 

122

Lazarus, R.S. & Alfert, E. The short circuiting of threat
by experimentally altering cognitive appraisal. Jour-
nal of Abnormal and Social Psychology, 1964, 99, 195-
205.

Lazarus, R.S., Speisman, J.C., Mordkoff, A.M. & Davison,
L.A. A laboratory study of psychological stress
produced by a motion picture film. Psychological
Monographs, 1962, 19, 1-35.

 

 

Lazarus, R.S., Speisman, J.C. & Mordkoff, A.M. The relation-
ship between autonomic indicators of psychological
stress: heart rate and skin conductance. Psychoso-
matic Medicine, 1963, 99, 19-30.

 

 

Learmonth, G.J., Ackerly, W., & Kaplan, M. Relationship be-
tween palmar skin potential during stress and person-
ality variables. Psychosomatic Medicine, 1959, 99,
150-157.

 

Lefcourt, R.M. Repression-sensitization: A measure of the
evaluation of emotional expression. Journal of Con-
sulting Psychology, 1966, 99, 444-449.

 

 

Lindsley, D.B. Emotions. In 8.8. Stevens (ed.) Handbook of
Experimental Psychology. Wiley: New York, 1951.

 

 

Lomont, J.F. The repression-sensitization dimension in rela-
tion to anxiety responses. ggurnal of Consultigg
Psychology, 1965, 99, 84-86.

 

 

Lomont, J.F. Repressers and sensitizers as described by
themselves and their pairs. Journal of Personality,
1966, 99, 224-240.

 

Lovallo, W. The cold pressor test and autonomic function:
A review and integration. Psychophysiology, 1975, 99,
268-282.

 

MacKinnon, P.C. & Harrison, J. The influence of hormone
associated with the pituitary-adrenal and sexual
cycle activity on palmar sweating. Journal of Endo-
crinology, 1961, 99, 217-225..

 

McCurdy, H.G. Consciousness and the galvanometer. Psycho-
lqgical Review, 1950, 91, 322-327.

 

McReynolds, P., Acker, M. & Brackbill, G. On the assessment
of anxiety IV. By measures of basal conductance and
palmar sweat. Psychological Rsports, 1966, 99, 347-
356.

123

Malmo, R.B. & Davis, J.F. Physiological gradients as in-
dicants of "arousal" in mirror tracing. Canadian
Journal of Psychology, 1956, 99, 231-238.

Malmo, R.B., Shagass, C. & Smith, A.A. Responsiveness in
chronic schiZOphrenia. Journal of Personality, 1951,
99, 359-375.

 

Martin, R.D. & Edelberg, R. The relationship of skin resis-
tance changes to receptivity. Journal of Psychoso-
matic Research, 1963, 1, 173-179.

 

 

Merbaum, M. & Badia, P. Tolerance of repressers and sensi-
tizers to noxious stimulation. Journal of Abnormal
Psychology, 1967, 19, 349-353.

 

 

Merbaum, M. & Kazaoka, K. Reports of emotional experience
by sensitizers and repressers during an interview
transaction. Journal of Abnormal and Social Psycho-
logy, 1967, 19, 1013105.

Obrist, P.A. Some autonomic correlates of serial learning.
Journal of Verbal Learning and verbal Behavior, 1962,
9, 100-104.

Obrist, P.A. Cardiovascular differentiation of sensory
stimuli. Psychosomatic Medicine, 1963, 99, 450-459.

Obrist, P.A., Black, A.H., Brener, J. & DiCara, L. (Eds.)
Cardiovascular psychOphysiology: Current issues in
response mechanisms, biofeedback and methodoIEgy,
Chicago: Aldine-Atherton, 1974.

Obrist, P.A., Webb, R.A. & Sutterer, J.R. Heart rate and
somatic changes during aversive conditioning and a
simple reaction time task. Psychpphysiology, 1969,
9, 696-723.

Obrist, P.A., Webb, R.A. & Stutter, J.R. Cardiac deceler-
ation and reaction time: An evaluation of two hypoth-
eses. Psychopsysiology, 1970, 9, 695-706.

Obrist, P.A., Wood, D.M. & Perez-Reyes, M. Heart rate during
conditioning in humans: Effects of UCS intensity,
vagal blockage, and adrenergic block of vasomotor
activity. Journal of Experimental Psychology, 1965,
70, 32-40.

Oken, D. The psychOphysiology and psychoendocrinology of
stress and emotion. In M.H. Appley & R. Trombull (eds.)
Psycholo ical Stress, Appleton-Century-Crofts: New
York, 19 7.

 

124

Oken, D., Grinker, R.R., Heath, H.A., Herz, M., Korchin,
S.J., Sabshin, M. & Schwartz, N.B. Relation of phy-
siological reSponse to affect expression. Archives
of General Psychiatry, 1962, 9, 336-351.

 

Orne-Johnson, D.W. Autonomic stability and transcendental
meditation. Psychosomatic Medicine, 1973, 99, 341-
349.

Panek, D.M. & Martin, B. The relationship between GSR and
Speech disturbances in psychotherapy. Journal of
Abnormal and Social Psychology, 1959, 99, 402-405.

 

Petzel, T.P. & Gynther, M.D. Task performance of repress-
ers and sensitizers under ego-oriented versus task-
oriented instructions. Journal of Consulting and
Clinical Psychology, 1968, 99, 486-487.

 

Plato. Dialogues (Translated by B. Jowett) In R.M. Hutch-
ings (ed.) Great Books of the Western World, vol. 7.
Chicago: EncycIBpaedia Britannica, 1952 (c390-348B.C.).

Porges, S.W. & Raskin, D.C. Respiratory and heart rate
components of attention. Journal of Experimental
Psychology, 1969, 99, 497-503.

 

 

Raskin, D.C. Attention and arousal. In W.F. Prokasy, &
D.C. Raskin (eds.) Electrodermal Activity in Psygho-
lo ical Research, New York: Academic Press, 1973,
12 -156.

 

 

Raskin, D.C., Kotses, H. & Bever, J. Autonomic indicators
of orienting and defensive reflexes. Journal of
Experimental Psychology, 1969, 99, 423-939.

 

 

Reich, W. Character Analysis (translated by V. Carfagno)
New York: Farrar, Straus & Gironx, 1972 (1933).

 

Roessler, R., Alexander, A.A. & Greenfield, N.S. Ego strength
and physiological responsivity: I. The relationship
of the barron ES scale to skin resistance, finger blood
volume, heart rate and muscle potential responses to
sound. Archives of General Psychiatry, 1963, 9, 142-
154.

Roessler, R., Burch, N.R. & Childers, H.E. Personality and
arousal correlates of specific galvanic skin responses.
Psychophysiology, 1966, 9, 115-130.

Schacter, S. The interaction of cognitive and physiological
determinants of emotional state. In L. Berkowitz (ed.)
Advances in Experimental Social Psychology, vol. 1,
New York: Academic Press, 1964.

125

Schacter, S. & Singer, J.E. Cognitive social, and physio-
logical determinants of emotional state. Psycholog-
ical Review, 1962, 99, 121-128.

 

 

Schacter, S. & Wheeler, L. Epinephrine, chlorpromazine, and
amusement. Journal of Abnormal and Social Psychology,

Schill, T., Emanuel, G., Pedersen, V., Schneider, L. &
Wachowiak, D. Sexual responsivity of defensive and
non-defensive sensitizers and repressers. Journal
of Consulting and Clinical Psychology, 1970, 99, 34-47.

Simal, P.J. & Herr, V.V. Autonomic responses to threatening
stimuli in relation to the repression-sensitization
dimension. Journal of Abnormal Psychology, 1970, 19,
106-109.

Smith, Jonathan, C. Meditation as Psyghotherapy. Unpublished
doctoral dissertation, MiEhiéan State University, 1975.

 

Sokolov, E.N. Neuronal models and the orienting reflex. In
M.A.B. Brazier (ed.) The Central Nervous System and
Behavior. New York: Josiah Macy, Jr. Foundation,
1960.

 

 

Sokolov, E.N. Higher nervous functions: the orienting
reflex. Annual Review of Physiology, 1963,.99, 545-
580.

 

Sokolov, E.N. Perception and the conditioned reflex. New
York: Macmillan, 1963.

Stennett, R.G. The relationship of performance level to
level of arousal. Journal of Experimental Psycholggy,
1957, 99, 54-61. 49'

Speisman, J.C., Lazarus, R.S., Mordkoff, A. & Davison, L.
Experimental reduction of stress based on ego defense
theory. Journal of Abnormal and Social Psychology,
1964, 99, 367-386}

 

Speisman, J.C., Osborn, J. & Laxarus, R.S. Cluster analysis
of skin resistance and heart rate at rest and under
stress. Psychosomatic Medicine, 1961, 99, 323-343.

 

Spinoza, B. Ethics. (Translated by W.H. White, Revised by
A.H. Striling) In, R.M. Hutchins (ed.) Great Books of
the Western World, vol. 31, Chicago: Encyc10paedia
Britannica, 19977 (1677).

 

126

Spohn, H.C., Thetford, P.E. & Cancro, R. The effects of
phenothiazine medication on skin conductance and
heart rates in schiZOphrenic patients. Journal of
Nervous and Mental Disease, 1971, 999, 129-199.

 

 

Stern, J.A. Stability-lability of physiological response
systems. Annals of the New York Academy of Sciences,
1966, 134, 1019-1027.

 

Stern, J.A. & McDonald, D.G. Physiological correlates of
mental disease. Annual Review of Psychology, 1965,
-]I._6_’ 225-2650

 

Sternbach, R.A. & Tursky, B. Ethnic differences among
housewives in psychOphysical and skin potential
responses to electric shock. Psychophysiology, 1965,
9, 241-246.

 

Syz, H.C. & Kinder, E.F. Electrical skin resistance in
normal and psychotic subjects. Archives of Neurology
and Psychiatry, 1928, 99, 1026-1035.

 

 

Tan, B.K. Physiological correlates of anxiety: a prelimin-
ary investigation of the orienting reflex. Journal of
the Canadian Psychiatric Association, 1964, 9, 63-71.

 

 

Tempone, V.J. Extension of the repression-sensitization
hypothesis to success and failure experience. Psy-
chological Reports, 1964, 99, 39-45.

 

Tempone, V.J. Some clinical correlates of repression-
sensitization. Journal of Clinical Psychology, 1964,
99, 440-442.

 

Tempone, V.J. & Lambi, W. Repression-sensitization and its
relation to measures of adjustment and conflict.
Journal of Consulting Psychology, 1967, a, 131-136.

 

Thayer, J., & Silber, D.E. Relationship between levels of
arousal and responsiveness among schizophrenic and
normal subjects. Journal of Abnormal Psychology, 1971,
11, 162-173.

Thelen, M.H. Repression-Sensitization: Its relation to
adjustment and seeking psychotherapy among college
students. Journal of Consulting and Clinical Psychol-
ggy, 1969, 99, 161-165.

 

Thetford, P.E., Spohn, H.E., Everds, J.P. Psychophysiological
patterns of response in schizophrenia. Journal of Ner-
vous and‘Mental Disease, 1974, 158, 104-111}

 

127

Ullmann, L.P. An empirically derived MMPI scale which mea-
sures facilitation-inhibition of recognition of threat-
ening stimuli. Journal of Clinical Psychology, 1962,
‘99, 127-132.

 

Uno, T. & Grings, W.W. Autonomic components of orienting
behavior. Psychophysiology, 1965, 9, 311-321.

 

Van Der Valk, J.M & Green, J. Electrical resistance of the
skin during induced emotional stress. Psychosomatic
Medicine, 1950, 99, 303-314.

 

Venables, P.H. Psychophysiological aspects of schiZOphrenia,
British Journal of Medical Psychology, 1966, 39, 289-
295. —

 

Venables, P.H. & Christie, M.J. Mechanisms, instrumentation,
recording techniques, and quantification of responses
in W.F. Prokasy and D.C. Raskin Electrodermal Activity
in Psychological Research, New York?' Academic Press,
1973, 2-124.

 

 

Vigouroux, R. Sur 1e r61e de la resistance électrique des
tissus dans l'électrodiagnostic. 99mptes rendus des
Séance de la Societé de Biologie, 1879, 99, 336-339.

 

 

Walker, E. Action decrement and its relation to learning.
Psychological Review, 1958, 99, 129-150.

 

 

Watson, R.I. The Great Psychologists. Philadelphia: J.B.
Lippincott Co., 1968. .

Wenger, M.A., Clemens, T.L., Coleman, D.R., Cullen, T.D. &
Engel, E.T. Autonomic reSponse Specificity. Psycho-
somatic Medicine, 1961, 99, 185-193.

 

Westcott, M.R. & Huttenlocher, J. Cardiac conditioning:
the effects and implications of controlled and uncon-
trolled respiration. Journal of Experimental Psychol-
ggy, 1961, 99, 353-359.

 

Wiener, M., Carpenter, B. & Carperter, J.T. Determination
of defense mechanisms for conflict areas from verbal
laterial. Journal of Consulting Psychology, 1956,
99, 215-219.

 

Wilder, J. Stimulus response the law of initial value.
Bristol: John Waight, 1967.

 

Wood, D.M. & Obrist, P.A. Effects of controlled and uncon-
trolled reSpiration on the conditioned heart rate
response in humans. ggurnal of Experimental Psychology,
1964, 99, 221-229.

 

128

Zahn, T.P., Rosenthal, D. & Lawlor, W.G. Electrodermal and
heart rate orienting reactions in chronic schizophrenia.
Journal of Psychiatric Research, 1968, 9, 117-134.

 

HICHIGR

1 111

129

T

ATE UNIV. LIBRQRIES
lllllllllflllllWlllllVlllllllllllIIIWHI
3103215053