EFFECTS OF EQQNéTWE-BEHA‘JEOR
MODEFICATSDN UPON QEPRESSED
OUTPATENTS

Bissertation for the Degree of Ph. D.
MiGHEGAfi STATE UMVERSW
VERNE GORDGN SCHMICKLEY

1978

 

will llllflll l l llljl ll l lllglll “1 l .

 

 

This is to certify that the
thesis entitled

EFFECTS OF COGNITIVE-BEHAVIOR MODIFICATION
UPON DEPRESSED OUTPATIENTS

presented by

Verne Gordon Schmickley

has been accepted towards fulfillment
of the requirements for

 

 

 

Date February 11, 1976

 

0-7 639

    
   
 

- A amuse Iv .
(MEI-80'8“
[BOOK HENDERY
5 UBRARY emvzns
it gnueropuicmu

 
  
  

  

 

OVERDUE FINES:
25¢ per day per iten'

.4 {{ “‘3. f murmurs Liamv mums:
\

v - 3. 9" Place in book return to remove
- "” . charge from circulation records

mm “A?! UNIVIJW

~ 1
V

 

 

   

 

 

 

 

 

.1 \Rw

ABSTRACT

EFFECTS OF COGNITIVE-BEHAVIOR MODIFICATION
UPON DEPRESSED OUTPATIENTS

BY

Verne Gordon Schmickley

The purpose of this study was to evaluate the
effects of cognitive-behavior modification upon depressed
individuals. These individuals were clients of an out-
patient community mental health center. The effects
of cognitive-behavioral intervention were assessed
using a client self-rated scale of depressive thoughts
and feelings, and by a selfhand-other-rated scale of
depressive behavior.

Four graduate students in counseling with train-
ing and experience in behavior-modification were trained
in the theory and practice of the cognitive modification
of depression. Specific training procedures for these
counselors are described. Three therapists were each
assigned to three subjects, and the fourth therapist
was assigned to two subjects. The 11 subjects, all

females, were selected from 39 clients of a community

Verne Gordon Schmickley

mental health center referred for possible inclusion in
this study by mental health staff persons.

Subjects were accepted for the study based upon
‘specified screening criteria including clinical signs,
observed client behavior, client history, presenting
problem(s), MMPI profile, Shipley—Hartford scores, and
score on the Beck Depression Inventory (BDI).

Following each subject's acceptance into this
study, she appeared for an interview with this investi-
gator. She was furnished a copy of the Beck Depression
Inventory and pre-dated copies of daily response cards.
The subject was requested to complete the BDI question-
naire each day throughout the study.

The subject was requested to bring two signifi-
cant others to the interview with this investigator.
The significant others were to serve as raters for her
daily behavior. The subject and the raters were fur-
nished copies of the Depression Behavior Rating Scale,
an abbreviation of the NOSIE-30, to complete each day
on a daily predated response card.

Subjects and their raters completed the daily
response cards through a baseline phase of 7, 10, or
14 days, a two-week treatment phase, and a two-week
reversal phase. Data are plotted for each subject for g
each experimental day and across four dependent

variables. The Beck Depression Inventory was one of

Verne Gordon Schmickley

the daily repeated measures. Items from the Depression
Behavior Rating Scale yielded the three remaining daily
repeated measures--Depression items, Psychomotor
Retardation factor, and Social Interest factor.

The treatment intervention consisted of four
sessions over a two-week period. During the first
session the therapist established rapport and structured
the treatment phase for the subject. The subject was
furnished a pamphlet, "Coping with Depression," and the
booklet was discussed. The therapist suggested that the
subject read the pamphlet once each day until the second
treatment session, and whenever else she experienced
feelings of depression.

During the second treatment session, the therapist
assisted the subject in identifying at least five depres-
sive cognitions and their corresponding challenges. The
subject was instructed to read the list of depressing
cognitions and challenges before engaging in high proba-
bility behaviors until the third session.

During the third session, the therapist assisted
the subject in generating a list of at least five
positive self-evaluatory statements. The subject was
to read the list before engaging in a high probability
behavior before the last treatment session.

In the fourth session, the therapist reviewed

the cognitive treatment of depression, reinforced

Verne Gordon Schmickley

subject learning, and scheduled the subject for a follow-
up appointment with this investigator two weeks later.

At that time the subject's current level of psy-
chological functioning was evaluated clinically, and
she was administered a second MMPI. Any concerns she
had regarding the treatment were discussed, and if the
subject desired, further mental health services at the
center were offered.

A multiple-baseline design was employed in the
study. Subjects self-reported level of depressive
symptomatology (EDI) and self-and-other-rated depressive
behaviors (BRS) were charted daily as repeated measures.
The treatment effects were contrasted across four depen-
dent variables and across individuals at the same rela-
tive points in the treatment schedule.

Based upon two assumptions made by Aaron Beck
concerning the effects of the cognitive modification of
depression, 12 research hypotheses were generated for
this study. For each of the four dependent measures
of depression, it was hypothesized that during the
treatment phase the subject's rated depression would
decline. The depression rating on each measure would
rise again during the reversal phase, a second hypothesis.
Finally it was hypothesized that depression scores
during the reversal phase would not rise to the level

of the baseline phase.

Verne Gordon Schmickley

The data from each subject are presented in an
intensive case study format. The subject's performance
on each dependent variable is presented graphically,
and significance tests have been performed for all 12
research hypotheses on the data from each subject.
Finally, supplementary data, such as pre-posttreatment
changes in the subject's MMPI profile, are presented
along with a discussion of the outcome of treatment
and her current therapeutic status.

The research hypotheses are supported by vary-
ing proportions of subject data. Specifically, it was
found that scores on the Beck Depression Inventory,
the Behavior Rating Scale Depression items, and the
Behavior Rating Scale Psychomotor Retardation factor
decreased during the treatment phase, and rebounded
after the termination of treatment, but remained lower
during the reversal phase than before the treatment
intervention. Scores for the Behavior Rating Scale
Social Interest factor for two subjects support the
hypothesis that subjects' performance during the treat-
ment phase was different than during baseline, but the
other two research hypotheses for Social Interest were
supported by data from four subjects.

Supplementary data presented for all subjects
included mean MMPI profiles for pre- and posttreatment

testing situations. The changes in the MMPI profile

Verne Gordon Schmickley

over the time of the experiment are discussed, and
results of statistical tests of the changes in profile
are presented. It was found that the posttreatment MMPI
mean profile was well within normal limits, and that
reductions on five of the ten clinical scales for the
group were statistically significant.

It was concluded that this cognitive-behavioral
treatment was to varying degrees effective in reducing
depressive symptomatology and behavior among clients of
an outpatient clinic. The research findings are dis-
cussed, as are implications of the present study for

practice and further research.

EFFECTS OF COGNITIVE-BEHAVIOR MODIFICATION

UPON DEPRESSED OUTPATIENTS

BY

Verne Gordon Schmickley

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

ACKNOWLEDGMENTS

I should like to express my appreciation to
those individuals who have greatly influenced or
assisted me in completing this dissertation. Mere words
seem inadequate to the task, but my gratitude goes to
these teachers, coworkers, friends, and loved ones for
their contributions.

Richard Johnson has been my advisor and doctoral
committee chairman over the past four years. Dick has
been a constant source of advice, feedback, and encour-
agement, and I have valued his friendship and thoughtful
comments. His expertise and interest in counseling
research saved me from becoming hopelessly mired in
the problems of this dissertation, and his prodding
helped me finish the study within a reasonable time
frame. I look forward to future associations with Dick,
as I have much more to learn from him.

The other members of my doctoral committee--
Herbert Burks, Mark Rilling, and Stephen Yelon--have
greatly influenced my thinking and counseling. Their

classes in counseling, learning theory, and educational

ii

psychology, respectively, were exceptionally well-
prepared and well-taught. Their teaching styles,
though different from one another, were excellent
models, and the classes they taught were my most
valuable courses of graduate school. I value their
input regarding my doctoral program and this disser-
tation, and I am particularly grateful to Herb for his
editorial expertise.

Doctoral committee members can be a source of
irritation and conflict for graduate students; my
committee was an exception to this rule. All meetings
were constructive, cooperative, and productive. The
members were excited about the educational process,
and worked hard to help me at various points along the
way. I thank Dick, Herb, Mark, and Steve for making
the committee meetings such positive experiences for me.

At this point I should like to thank two of the
leading authorities on the treatment of depression,
whose work greatly influenced my thinking, and in fact
whose approaches I tried to blend in my experimental
treatment. Their works are cited later, but Aaron T.
Beck, M.D., and Peter M. Lewinsohn, Ph.D., were
extremely generous in furnishing me with reprints
and treatment materials. Their comments and encour-

agement are greatly appreciated.

iii

This study was run on a low student budget, and
could not have been completed without the therapists who
donated their services to the treatment package. Carol
Best, Bill Brooks, Jan Skinner, and Tony Straseske are
to be commended for their hard work, professionalism,
and willingness to learn. Carol, Bill, Jan, and Tony
were a pleasure to work with, and their depressed clients
and I thank them. Nancy Voight helped train the coun-
selors for the treatment package, and was a real asset.
Because Nancy was involved in her own dissertation
research at the time, her participation is particularly
appreciated. I wish them all the best in their personal
lives and in their counseling careers.

I am grateful to the clients who participated in
this depression treatment program. Their willingness to
enter treatment for a painful emotional condition and
to make frequent observations of their performance was
I hope as helpful to them as it was to this study.
Similarly, the cooperation of the client behavior raters
in the data collection is most appreciated. For reasons
of confidentiality, the clients and their raters will
remain anonymous, but they have all been personally
thanked.

The staff of the mental health center has been
of great assistance to this study. Zoe Ann DaViS: Peggy

Goffin, Jo Ann Lowes, Gil Schmidt, Cher Swensen, and

iv

Tom Woodward, all mental health therapists, are thanked
for referring depressed clients for possible inclusion
into the study. Kathy Thompson, Nancy White, and Ruth
VanVleck are thanked for their typing various materials
for the study. Kathy and Nancy also scored the pre- and
post- psychological tests for clients. The encouragement
and constant support of the agency coordinator, Gil
Schmidt, have been appreciated during all phases of my
doctoral studies. The friendship and professional com-
petence of the staff at the center have been valued
during my four years as their coworker. Finally, I
thank the clinical director of the mental health board,
Gil DeRath, for his suggestions and approval of this
study.

Other individuals have made contributions less
direct but no less important to me. Ron Baker, Ellen
Betz, Steve Elson, Chuck Gasswint, John Grundvig, Tom
Linde, and Avi Scherman are instructors, supervisors,
or fellow graduate students who assisted me during
important points in my professional development. I
see these psychologists too rarely now, but I am
influenced daily by their teachings and encouragement.

Mention must be given to Marty North, who

typed the final drafts of my research proposal and

this dissertation. Her excellent typing and her knowl-
edge of the tasks necessary to prepare this final copy
are appreciated.

My family has contributed to this dissertation
in ways I'll probably never fully comprehend. My
parents, Harold and Margaret Schmickley, were teachers,
and education was highly valued in our home. They taught
me to love and respect others, and also taught me to
respect myself. They challenged me to learn, and think,
and work. They tolerated my reactions when they didn't
make things too easy for me. They said and showed their
love in many ways, although at the time I did not
appreciate their methods. Now I can look back with
the wisdom of old age (almost 30) and say, "Thank you,
Mother and Dad."

My brothers, Jim, Ron, and Dennis Schmickley,
have always encouraged me, and showed me how to avoid
some of the mistakes they made earlier. Especially
I thank them and their lovely wives for their interest,
encouragement, and advice during the final years of my
education.

My final words of appreciation go to Carol,
who has shared my frustrations and my joys over the
past three years. She has bolstered me when I was
discouraged, and kidded me back into reason when I

took myself too seriously. She has been a constant

vi

source of comfort and stability when my academic life
seemed ambiguous and threatening. She has been patient
with me, and when I was too busy to be with her, she
understood. To Carol: Thank you for making this

dissertation more possible and more tolerable.

vii

Chapter

I.

II.

TABLE OF CONTENTS

Magnitude of the Problem of Depression
Overview of the Problem of Depression
Behavioral Theory and Modification of

Depression . . . . . . . .
Learned Helplessness and Depression
Cognitive Theory and Modification of

Depression . . . . . . . .
Theoretical Conclusions . . . .

PROCEDURES 0 O O O O O O O 0

Population. . . . . . . . .
sample C O I O O O O I O 0
Measures . . . . . . . . .

Clinical Impressions. . . . .
MMPI O I I O O O O 0

INTRODUCTION AND REVIEW OF THE LITERATURE

Shipley-Hartford Institute of Living

scale 0 O I I O O O O I
Beck Depression Inventory . . .
Reinforcement Survey Schedule. .

Outpatient Depressive Behavior Rating

Scale . . . . . . . . .
Procedures. . . . . . . . .

Subjects. . . . . . .
Psychodiagnostic Instruments
Client Behavior Raters
Experimenters . . .
Treatment Intervention
Followup. . . . .

Design . . . . . . . . . .

Intensive Case Study Design--Review.

Summary of Design and Procedures.

Research Hypotheses. . . . . .
Data Analysis. . . . . . . .

viii

Page

13
21

22

22
22
23

23
25

28
29
31

33
35

35
36
39
40
43
45

47

47
50

51
54

Chapter

III.

IV.

RESULTS . . .

Single Subject

Subject
Subject
Subject
Subject
Subject
Subject
Subject
Subject
Subject
Subject
Subject

l—‘I—‘komxlOSU'l-waH

l—‘O

Results

Overall Results. . .

Hypothesis
Hypothesis
Hypothesis
Hypothesis
Hypothesis
Hypothesis
Hypothesis
Hypothesis
Hypothesis
Hypothesis
Hypothesis
Hypothesis

\ocoqmmthI—I

10.
11.
12.

Supplementary Findings

MMPI. O O

Interrater Reliabilities

DISCUSSION . .

Summary . .
Discussion .

Overall Results . .

MMPI. . .

Interrater Reliabilities

Implications.

ix

Page

57
57

S7
61
70
74
77
83
88
94
99
105
111

116

116
116
117
118
118
118
119
119
119
120
120
121

121

121
124

129

129
134

134
142
145

148

APPENDICES

APPENDIX

A. BDI QUESTIONNAIRE

B. BEHAVIOR RATING SCALE.

C. DAILY

D. COGNITIVE TARGETS

REFERENCES.

RATING CARDS.

Page

156
160
161
163

164

Table

1.

LIST OF TABLES

Binomial Tests of Phase Changes for Research
Hypotheses Involving Beck Depression

Inventory . . . . . . .

Binomial Tests of Phase Changes
Hypotheses Involving Behavior
Scale Depression Items. . .

Binomial Tests of Phase Changes
Hypotheses Involving Behavior
Scale Psychomotor Retardation

Binomial Tests of Phase Changes
Hypotheses Involving Behavior
Scale Social Interest Factor.

Results of Matched-Pairs T-tests Between
Pre- and Posttreatment Means (N = 10) on

MMPI Clinical Scales . . .

Interrater Reliability Coefficients for

for Research
Rating

for Research
Rating

Factor .

for Research
Rating

Behavior Rating Scale Variables.

Descriptions of Median Trends of Baseline
Performance on Four Dependent Measures

of Depression. . . . . .

xi

Page

60

66

67

68

125

127

143

Figure

l.

10.

ll.

12.

13.

14.

LIST OF FIGURES

Subject 1; Daily scores on four dependent
measures of depression . . . . . .

Subject 1; MMPI pretreatment profile . .

Subject 2; Daily scores on four dependent
measures of depression . . . . . .

Subject 2; MMPI pre- and posttreatment pro-

files 0 I O O O O O O I O 0

Subject 3; Daily scores on four dependent
measures of depression . . . . . .

Subject 3; MMPI pre- and posttreatment
profiles . . . . . . . . . .

Subject 4; Daily scores on four dependent
measures of depression . . . . . .

Subject 4; MMPI pre- and posttreatment
profiles . . . . . . . . . .

Subject 5; Daily scores on four dependent
measures of depression . . . . . .

Subject 5; MMPI pre- and posttreatment
profiles . . . . . . . . . .

Subject 6; Daily scores on four dependent
measures of depression . . . . . .

Subject 6; MMPI pre- and posttreatment
profiles . . . . . . . . . .

Subject 7; Daily scores on four dependent
measures of depression . . . . . .

Subject 7; MMPI pre- and posttreatment
profiles . . . . . . . . . .

xii

Page

58

62

64

69

71

73

76

78

80

82

84

87

89

93

Figure

15.

16.

17.

18.

19.

20.

21.

22.

23.

Subject 8;
measures

Subject 8;
profiles

Subject 9;
measures

Subject 9:
profiles

Subject 10;

measures

Subject 10;

profiles

Subject 11;

measures

Subject 11;

profiles

Group mean

Page
Daily scores on four dependent
of depression . . . . . . . . 95
MMPI pre— and posttreatment
O O O O O O I O O O O O 98
Daily scores on four dependent
of depression . . . . . . . . 101

MMPI pre- and posttreatment
O O I I O O O I O O O O 104

Daily scores on four dependent
of depression . . . . . . . . 107

MMPI pre- and posttreatment
O O O O I O O O O l O O 110

Daily scores on four dependent
of depression . . . . . . . . 112

MMPI pre- and posttreatment
0 O I O O O O O O O O O 115

pre- and posttreatment MMPI pro-

files over all subjects (N = 10). . . . . 123

xiii

CHAPTER I

INTRODUCTION AND REVIEW OF THE LITERATURE

Magnitude of the Problem
of Depression

 

 

Depression is a painful emotional condition which
has plagued mankind at least as long as history has been
recorded, and no doubt before that. Depression has been
described by Hippocrates, by Biblical scholars in the
Book of Job, and by Shakespeare in his plays. Known
throughout the ages as "the black bile," "melancholia,"
"manic-depression," or "depression," there have been as
many labels and types as there are theories of the causes
and treatment of the condition (Kushner, 1974).

Depression is no respector of social, economic,
or educational level, fame, or political power; such
individuals as Abraham Lincoln, Ernest Hemingway, and
Marilyn Monroe have suffered from depression (Kushner,
1974), as have Senator Thomas Eagleton and ex-Presidents
JOhnson and Nixon. Feelings of worthlessness, helpless-
ness, and pessimism characterize this most common form

of psychiatric disorder in the United States today

(Kushner, 1974); indeed, depression is often referred
to as "the common cold of mental disorders."

Grinker and Nunnally (1970) and associates
(Grinker, Miller, Sabshin, Nunn, & Nunnally, 1961)
became interested in the depressive syndrome because:

It seemed to have become the forgotten disease
although about half of the population in private
psychiatric hospitals were admitted and discharged
with this diagnosis. . . . The frequency, painful-
ness, and family disruption caused by depression
stimulated little work with the syndrome, and
little had been added to its understanding since
Kraepelin's and Bleuler's descriptions, which
were only somewhat better than the accounts left
by the ancients. (Grinker & Nunnally, 1970,

p. 392)

It is estimated that eight million people in the
United States are currently receiving treatment for
depression, and 600,000 of these are hospitalized because
their normal functioning is severely impaired (Kushner,
1974). In terms of numbers of individuals affected,
depression is beginning to rival schizophrenia as the
nation's number one mental health problem, and it is
estimated that 10% of the general population will suffer
a clinically significant depressive episode at some time
in their lives (NIMH, 1973).

Aaron T. Beck, M.D., probably the foremost con-
temporary authority on depression, writes that it is
"the most common psychiatric disorder treated in office

practice and outpatient clinics. The study of depression

is important, not only because of the human misery it

causes, but because of its byproduct, suicide" (Beck,
1974a, p. 61). Every year 20,000 Americans commit
suicide (Kushner, 1973), and Beck (1974) concludes that
depression is among the leading causes of death in this
country.

Overview of the Problem
of Depression

 

 

Many writers (e.g., Kushner, 1974) have described
depression as merely being on a continuum with other
unpleasant emotions: from "normal Monday-morning blues,"
to sadness, grief, disappointment, despondency, and
finally depression. An individual's "level" of depres-
sion would depend upon the frequency, intensity, and
duration of his unpleasant feelings. Other authorities
take an opposite view, such as Bernard J. Carroll, M.D.,

who views depression as "an illness which is categorically

 

different from normal mood lowering" (1974, p. 2). Peter

M. Lewinsohn, Ph.D., a behavior therapist, essentially

takes a moderate position, and assumes depression "to

be a continuous variable which can be conceptualized as

a state (which fluctuates over time) but also as a trait

(some people, 'depressives,‘ are more prone to becoming

depressed than others)" (1974, p. 65).. (Italics mine.)
Lewinsohn provides an excellent description of

depression:

The term "depression" is used to refer to the
syndrome of behaviors which have been identified
in descriptive studies of depressed individuals
(e.g., Grinker et al., 1961), including verbal
statements of dysphoria, self-depreciation, guilt,
material burden, social isolation, and somatic
complaints; it also involves a reduced rate for
many behaviors. (Italics mine.) (1974, p. 65)

 

Another behaviorist, Martin E. P. Seligman, Ph.D., simi-
larly describes the phenomenon:

Depression . . . does not have a necessary condition
that defines it. Rather, it is a convenient diag-
nostic label which denotes a constellation of

s toms [= a syndrome], no one of which is neces-
sary. (Italics mine.) (1974, p. 86)

 

Beck (1974a) delineates the signs and symptoms
which characterize depressed individuals:

Emotional: Sadness or apathy; crying spells, self-
dislike; loss of gratification; loss of feelings of
affection; loss of sense of humor.

Cognitive: Negative self-concept; negative expec-
tations; exaggerated view of problems; attribution
of blame to self.

Motivational: Increased dependency; loss of moti-
vation; avaidance, indecisiveness; suicidal wishes.
Physical and Vegetative: Loss of appetite; sleep
disturbance; fatigability; loss of sexual interest.
(p. 62)

 

 

Similarly, Seligman (1974) describes the mani-
festations of depression:

Passivity: the slower response initiation, retard-
ation, and lowered amplitude of behavior . . .
Negative expectations: the readiness with which
dépressed patients view their actions, even if they
succeed, as having failed or being futile . . .

The sense of helplessness, hopelessness, and power-
lessness which depressed patients frequently voice.
(Italics mine.) (p. 89)

 

 

 

Seligman sees clinical depression as related to his con-
cept of learned helplessness, which will be elaborated

upon in a later section. He writes:

The label "depression" denotes patients with
negative cognitive sets about the effects of
their own actions, who become depressed upon
the loss of an important source of gratifi-
cation--the perfect case for learned helpless-
ness to model. (Seligman, 1974, p. 85)

Once the symptoms of depression develop, they
generally increase in severity until the condition levels
off, and then the symptoms improve. The median duration
of the depressive episode is said to be three months.

In most individuals the intervals between depressive
episodes are symptom free; there is, unfortunately, a
tendency for depression to recur at some later time
(Beck, 1974a).

The goals of therapeutic strategies have been to
reduce the frequency, intensity, duration, and recur-
rence of depression. There have been therapeutic inter-
ventions devised to fit any of the numerous theories of
the cause of depression, e.g. psychoanalytic, ego-
psychological, and physiological/psychopharmacological.
A review of the various theories and their corresponding
therapeutic interventions is beyond the scope of this
dissertation; the interested reader is directed to the
excellent and comprehensive reviews by Beck (1972),
Becker (1974), and Friedman and Katz (1974). What will
be discussed in the following are theories and/or
therapies which offer great promise for the amelioration

of the problem of depression: behavior modification,

learned helplessness, and cognitive restructuring.

Behavioral Theory and Modification
of Depression

 

 

Although depression is one of the most frequently
occurring problems of mental health, social scientists in
general have neglected the problem until recently, and
a systematic behavioral theory of depression remains to
be developed (Becker, 1974). Lazarus (1968) attributes
much of this void to the difficulties in operationally
defining and measuring depression. Beck (1974a) notes
that their avoidance of the subjective components of
depression has subjected behaviorists to valid criticism
(e.g., American Psychiatric Association, 1973), and that
behavioral theory has thus far been applied only to
limited aspects of depression. Krasner (1971) states
that depression may be the last type of behavior to come
within the behavior therapy paradigm.

Fortunately, this trend is changing. Recently,
innovative theory and research on depression have begun
to flourish among certain social learning behaviorists
(Becker, 1974). These developments will be briefly
traced chronologically.

In his classic paper, Ferster (1965) was the
first to write that the state of clinical depression is
a consequence of decreased reinforcement. Ferster has
since (1974) expanded and elaborated his notions of

depression, but this basic assumption remains unchanged.

Lazarus (1968) similarly conceptualized depression as
a consequence of inadequate or insufficient reinforcement.

Lazarus (1968) prescribed three specific behavioral
techniques for the treatment of neurotic depression: time
projection with positive imaginal reinforcement; affective
expression of "repressed" anger; and behavioral depri-
vation and retraining to regain reinforcement. Burgess
(1969) has successfully applied contingency management
strategies to the extinction of depressive behaviors (by
shaping competing, nondepressive behaviors).

Jackson (1972) maintains that in the development
of depression, the person's behavior has become ineffec-
tive in securing positive reinforcers from his environ-
ment. This is due to alterations in reinforcement
schedules or changes in significant discriminative
stimuli under which behaviors were typically emitted.
Jackson has shown that self-reinforcement techniques
can serve to assist depressed individuals to regain
control over their sources of reinforcement. (Excellent
reviews of literature of behavioral self-control may be
seen in Cautela, 1969; Kanfer & Phillips, 1970; Elson,
1972; Goldfried & Merbaum, 1973; Johnson & Elson, 1974;
Mahoney, 1974; Mahoney & Thoresen, 1974; Thoresen &
Mahoney, 1974.)

The greatest strides toward a behavioral model

and therapy of depression have been made by Lewinsohn's

social learning group at Oregon. For the past six
years a comprehensive, behaviorally oriented theory
of depression has been developing. The thrusts of the
research have been:
. . . to test hypotheses about the socio-environmental
reinforcement conditions surrounding depression,
. . . to develop systematic and replicable treatment
strategies for depressed individuals, and
. . . to provide tools to assess and to evaluate
behavior change. (Lewinsohn, 1974, p. 65) '
The Lewinsohn group has been creative and prolific in its
study and description of behavioral applications to
depression (cf. Lewinsohn & Shaw, 1969; Lewinsohn &
Graf, 1973; Lewinsohn, 1974). The behavioral inter-
ventions successfully used appear to fall into five
categories (Lewinsohn, 1974):
1. Increasing the individual's activity level;
2. Identification of potential reinforcers for
the depressed individual;
3. Induction of affects incompatible with
depression;
4. Enhancing the individual's instrumental
skills; and
5. Increasing the drive level of the depressed
individual.
Lewinsohn's applications of behavior therapy successfully
meet the depressed individual's needs perceived by

Becker (1974) for increased positive reinforcement of

nondepressive behaviors.

The recent contributions of behavior theory to
depression are certainly exciting, both heuristically
and clinically. Becker concludes that:

It is too early to assess the contribution of
behavioral approaches to depression, but their
emphasis on testable formulations must inevitably
sharpen conceptualizations and expand the essential
data base for adequate theory building and rational
treatment programs. (1974, p. 119)

Learned Helplessness and
Depression

 

Rivaling the Lewinsohn group in the production of
research which holds exciting possibilities for a learning
theory and therapy of depression is the work of Seligman
and associates (Seligman, 1969, 1973, 1974, 1975;
Seligman & Hager, 1972; Seligman, Maier, & Solomon,
1971).

Seligman is an experimental psychologist at the
University of Pennsylvania who has extensively studied
the effects upon infrahuman subjects of unpredictable
and uncontrollable aversive events. A description of
his well-controlled and sophisticated methodology is
beyond the scope of this discussion; the reader is
referred to Seligman's own descriptions (Seligman, 1974,
1975; Seligman & Hager, 1972; Seligman et al., 1971)
or to excellent summaries by Becker (1974) or by
Thornton et a1. (1974).

Seligman subjected experimental dogs to massive

and sustained electrical shock, which they could not

10

predict nor escape from. When the dogs were later
included in standard escape-training trials, two-thirds
of the experimental animals failed to learn to escape
or avoid shock (all naive control animals learned to
escape or avoid shock):

The experimental dogs tended to manifest two
behavioral deficits. First, they attempted few
adaptive responses, and second, even when they
displayed an adaptive response, they failed to
learn from it. . . . They typically reverted to
the passive acceptance of 50 seconds of severe
pulsating shock. The latter behavioral phe-
nomena have been termed learned helplessness.

. . . (Becker, 1974, pp. 112-113)

Seligman describes the experimental phenomenon:

We use the term "learned helplessness" to describe
the interference with adaptive responding pro-

duced by inescapable shock and also as a shorthand
to describe the process that we believe underlies
the behavior. So learned helplessness is defined

by two behaviors: (1) dogs that have had experience
with uncontrollable shock fail to initiate responses
to escape shock or are slower to make responses than
naive dogs, and (2) if the dog does make a response
that turns off shock, it has more trouble learning
that responding is effective than a naive dog.
(1974, p. 86)

 

 

Seligman sees the behavioral deficits as having

a cognitive cause.

Learning that one cannot control important events

. . . that responding and reinforcement are inde-
pendent results in a cognitive set that has two
basic effects: fewer responses to control rein-
forcement are initiated, and associating successful
responding with reinforcement becomes difficult.
(Seligman, 1974, p. 97)

The notion of stimulus generalization is interwoven with
the learned helplessness cognitions. The more closely

subsequent traumatic experiences resemble the inescapable

11

and uncontrollable conditions under which passivity was
acquired, the more likely it is that passivity will
generalize to subsequent conditions, even though the
animal may actually be able to escape from the situation
(Becker, 1974).

Seligman (1974) suggests that the phenomenon of
learned helplessness provides a laboratory analogue of
human neurotic depression. More important, learned
helplessness provides a model for understanding the
development of depression in humans. Seligman admits
that his evidence is at this point tentative. But the
work of Seligman and his associates provides an exciting
and long-in-coming rapproachment between "pure" research
and clinical endeavors.

Bibring (1953), a psychoanalytic theorist,
viewed helplessness to achieve one's goals as a primary
cause of human depression (Beck, 1974a). Engel (1968)
described the "helplessness-hopelessness" emotion--the
patient's feeling that he cannot help himself, and that
he is unlikely to be helped by outside influences.
Grinker and Nunnally (1970) describe a factor charac-
teristic of depression: " . . . helplessness, hopeless-
ness, failure, sadness, unworthiness, guilt, and
internal-suffering. There is a self-concept of
'badness'" (p. 396).

Beck (l974d) sees strong relationships between

his own cognitive theory of human depression (to be

12

discussed in the next section), and Seligman's findings
on learned helplessness. The idea that one is helpless
is an irrational belief (Ellis, 1962) held by many
depressed individuals. He "finds evidence of his own
worthlessness in every experience and feels compelled
to blame himself for his faults" (Beck, 1974a, p. 63).
Beck has done extensive research on the content of dreams
of depressed individuals. In most of the dreams, the
depressed individual sees himself as "the loser,"
including airplanes being missed by a split-second,
exams being failed, slot machines giving the jackpot
to the very next gambler, and so forth (Kushner, 1974).
Beck's dream studies (1974a) concluded that the depressed
individual regards himself deprived of an important
source of gratification and defective in certain
attributes. The depressed individual has all-too-
successfully learned helplessness.

Seligman has found only one treatment which
"cures" helplessness in dogs. The animal has learned
that nothing he can do will terminate the shock, so his
negative expectations must be changed. The dog must
be forcibly dragged from side to side in the shuttle
box, so that eventually he learns that the trauma can
he escaped (1974). But successful therapy should be
preventive, in Seligman's (1974) view, arming the

patient against future depressions.

13

Obviously, depressed humans cannot be physically
forced into feeling better. Beck (l974d) reasons that
it must be cognitively demonstrated to the patient that
he is able to succeed:

"Paralysis of the will" may be considered as a
result of the patient's pessimism and hopeless-
ness; since he expects a negative outcome, he is
reluctant to commit himself to a goal or under-
taking. Conversely, when he is persuaded that he
can succeed at a particular task, he may be stimu—
lated to pursue it. (Beck, 1974a, p. 71)

How Beck does this is discussed next.

Cognitive Theory and Modification
of Depression

 

 

An interesting change is taking place in the
field of behavior therapy. Some of its most ardent
and "hardcore" practitioners have begun to see merit
in the theories and techniques of more psycho-dynami-
cally-oriented therapists (American Psychiatric Associ-
ation, 1973). The result has been the development of
an increasingly powerful and exciting psychology, in
which all manifestations, overt and covert, of human
behavior become active areas of investigation and
possible modification (Murray & Jacobson, 1971).
Even B. F. Skinner acknowledges the validity of
the resurgence of interest in internal events:
A small part of the universe is contained within
the skin of each of us. . . . We feel it and in
some sense observe it, and it would seem foolish
to neglect this source of information just because

no more than one person can make contact with the
inner world. (1974, p. 21)

14

Skinner views perceptions and reports of the inner world
as clues (a) to past behavior and the conditions affect-
ing it, (b) to current behavior and the conditions
affecting it, and (c) to conditions related to future
behavior.

Nevertheless, he cautions us to examine our own
behavior as we make contacts with the inner world.
Although attractive, the study of the world within the
skin has its potential pitfalls: "in teaching self-
knowledge (1) the verbal community must make do with
rather primitive nervous systems, and (2) it cannot
fully solve the problem of privacy" (Skinner, 1974,

p. 31). But behaviorism has progressed since Skinner
(1953) first wrote of "the world within one's skin"
more than 20 years ago. The recent advances in the
experimental analysis of behavior now allow us to study
internal events, and behaviorism "can now analyze, one
by one, the key terms in the mentalistic armamentarium"
(Skinner, 1974).

One such mentalistic term is "thinking," or
what an individual says to himself (Ellis, 1962). As
long as the individual's thinking is adaptive, there is
no problem. But, according to Ellis, once thinking
becomes maladaptive (irrational), problematic emotions
and behaviors result. Ellis' assumptions have received

much anecdotal and clinical support (Ellis, 1962, 1971,

15

1973), but only recently (cf. Elson, 1974) have the
theoretical assumptions been directly studied. Russell
and Brandsma (1974) studied:
. . . the assumptions of Albert Ellis' ABC theory
of psychopathology which relates implicit verbali-
zations to emotional arousal. . . . The ABC theory
was supported by the galvanic skin response on
three of the five hypotheses under investigation.
Rational-emotive psychotherapy (RET) has been

called both a type of and an essential component of

 

other techniques of behavior therapy (Beck, 1970; Ellis,
1971, 1973; Lazarus, 1971, 1974a, b). The value of RET
in changing behavior has gained considerable experimental
support through the efforts of Donald Meichenbaum and
associates at the University of Waterloo. Meichenbaum
(1973, 1974a, b) and Meichenbaum and Cameron (1974)
have successfully demonstrated that behavior modifi-
cation techniques can be used to modify what an indi-
vidual says to himself. Meichenbaum and Cameron (1974)
review several applications of self-instructional methods
used to teach "healthy talk" to various client groups,
including impulsive children, schizophrenics, and
neurotics. Clients may be taught to change their
evaluations of environmental consequences:

Our research on cognitive factors in behavior

modification has highlighted the fact that it

is not the environmental consequences per g3

which are of primary importance, but what S

says to himself about the consequences. How-

ever, what 3 says to himself, that is how he

evaluates afid interprets these events [Ellis'
(1962) irrational beliefs], is explicitly

16

modifiable by many of the behavior therapy tech-
niques which have been used to modify maladaptive
[overt] behaviors.’ (Meichenbaum & Cameron, 1974,
p. 103)
Thus, cognitive events become behaviorally modifiable.
Meichenbaum and associates did not originate
this notion. Jerome Frank (1961) noted that people tend
to change their attitudes to conform with what they
have been conditioned to say to themselves. Lloyd
Homme (1965) had argued that what one says to himself
influences his self-esteem, and that the self-esteem
can be altered by changing one's thoughts and feelings.
These internal events are called "coverants" (covert
operants) by Homme, and he assumes that they may be
controlled by manipulating reinforcing events based
upon Premack's principle (1959).
The common cognitive changes observed in
depressed individuals (low self-esteem, pessimism,
guilt, learned helplessness, etc.) have been reviewed.

Previous authorities have viewed these cognitions as

consequences of depression. Instead, Beck (1967,

 

1972) assigns primary significance to these cognitions
as the causes of depression (Lewinsohn, 1974). "By
proposing that the cognitive disorder precedes rather
than follows the depressed state, he contributes
alternative ways of thinking about etiology and

approaches to therapy" (Katz, 1974, p. 301).

17

Beck takes a radical stance, and views depres—

sion as a thought disorder:

 

Although thought disorders are traditionally
associated with schizophrenia and paranoia, recent
work has uncovered systematic distortions of
reality in depression. These findings cast doubt
on the classification of depression as a primary
mood disorder, and may also provide clues for the
treatment of the disease. (1963, pp. 62-63)
Beck (1967) defines a "depressive cognitive triad" of
a set of negative evaluations of the self, of the out-
side world, and of the future:
This triad is seen as the key to the consequences
of depression, such as the lack of motivation,
the affective state, and other internal and
behavioral manifestations. The depressed person's
cognitions lead to misinterpretations of exper-
iences, and hence many of the secondary responses
are logical consequences of such misinterpre-
tations. The depressed person is locked in an
insoluble situation, the result of which is future
despair. (Lewinsohn, 1974, p. 64)
The first component of Beck's "cognitive triad" is a
negative view of the self: The depressed individual
sees himself as deprived of gratification, inadequate,
and worthless. He believes that he is undesirable,
and perceives himself as defective (Beck, 1974a). He
believes he is responsible for irreversible and criti-
cal events, and he is beset with self-coercive
injunctions such as "should" and "must" (cf. Horney,
1950). No matter how well he performs, he is plagued
with the constant feeling that it is never good enough
(Becker, 1974). The second component of the cognitive

triad is a negative view of experiences. The patient

18

interprets all of his interactions with his environment
as overdemanding, blocking, depriving, depreciating,
and defeating (Beck, 1974a, Becker, 1974). Finally,

a "negative view of the future permeates his ideation.
As he looks ahead, the patient sees an indefinite con-
tinuation of his present difficulties and a life of
unremitting deprivation, frustration, and hardship"
(Beck, 1974a, p. 71).

The causes of the negative cognitive set are
traumatic life experiences. Once the cognitions are
established, however, events with cue properties simi-
lar to the original traumatic event may trigger
inappropriate depressive cognitions (Becker, 1974).
Depressed patients tend to distort their experiences,
and misinterpret and exaggerate aversive events (Beck,
1972). The patients' negative coqnitions "contribute
to . . . symptoms of depression such as sadness, pas-
sivity, self-blame, loss of pleasure responses, and
suicidal wishes" (Beck, 1974b, p. 2). Thus, due to
conditioning of depressive cognitions, the individual's
emotional response is disproportionate to the present
aversive situation--he has learned helplessness:

The predictions of depressed patients tend to be
overgeneralized and extreme. Since the patients
regard the future as an extension of the present,
they expect a deprivation or defeat to continue
permanently. If a patient feels miserable now,
he assumes he will always feel miserable. The

absolute, global pessimisms are expressed in
statements such as "Things won't ever work out

19

for me," "Life is meaningless," "It's never going

to be any different." The depressed patient

judges that, since he cannot reach a major goal

now, he will never be able to achieve it. Nor

can he see the possibility of substituting other

rewarding goals. Moreover, if a problem appears

insoluble now, he assumes that he will never find

a way of working it out. . . . (Beck, 1974c,
Becker (1974) notes that this stimulating and provocative
theory of depression "is highly related to clinical data."
From the striking similarity of Beck's above description
of the depressed individual with Seligman's (1974)
description of the learned helpless animal, it appears
furthermore that Beck's theory is substantially grounded
in the laboratory.

What has arisen, then, is a theory of human
depression based upon the work of Seligman and of Beck.
Depression would appear to be a condition resulting from
the conditioning of negative cognitions about one's self.
This position, as noted by Becker (1974), generates
several testable hypotheses, and results in implications
for the treatment of human depression.

In essence, the cognitive treatment of depression
involves assisting individuals in becoming aware of the
thoughts which mediate unpleasant emotions and maladaptive
behaviors, and in teaching them to think more effective
and adaptive self-statements (Meichenbaum & Cameron,

1974). This approach is compatible with what Skinner

(1974) calls an experimental analysis of behavior "which

20

goes directly to the antecedent causes in the [internal]
environment." Lewinsohn (1974) states that once

the individual is taught to correctly label his cognitions,
he will be in a better position to deal with his emotional
predicament. The underlying negative set is modified so
that the total syndrome of depression may be influenced.

This is the employment of cognitive modification (Beck,

 

1974b).

Beck's treatment focuses on the specific
depression-generating cognitions of the individual.
These maladaptive cognitions have been condensed and
compressed into a "shorthand," so that complicated
series of depressive thoughts may occur in an instant.
These "self-statements" or "internalized verbalizations"

(Ellis, 1962) have been called "automatic thoughts"

 

by Beck (1972). Beck's treatment interventions pinpoint
the specific conditioned cognitive distortions and demon-
strate their invalidity. Beck (1972, 1974b) assists the
individual to see himself as a "winner" rather than a
"loser," as adequate rather than helpless, by changing
the negative cognitive set to more Optimistic expec-
tations (similar treatments of depression are described
in Hauck, 1971a, b; Lazarus, 1971b, 1974a, b; Schmickley,

1973; and Gillette et al., 1974).

21

Theoretical Conclusions

 

Beck makes two conclusions which follow from
his theory and assumptions of the cognitive modification
of depression. If these conclusions are correct, Beck's
theory has tremendous potential for the understanding,
treatment, and prevention of neurotic depression in
humans. To this writer's knowledge, the conclusions
per fig have not been explicitly validated. The con—
clusions thus become the hypotheses of this proposed
study:
(1) By pinpointing the patient's specific cognitive
distortions and demonstrating their invalidity
through behavioral or cognitive techniques, the
psychotherapist can achieve a rapid improvement
in the symptoms of depression. (Beck, 1974c,
p. 19)
(2) Since each of the components of depression
[behaviora1, cognitive, emotional, motivational,
or physiological] contribute to the other compo-
nents, . . . an improvement in any one problem
area [e.g., cognitive] would lead to an improve-
ment in other areas [e.g., behavioral], which
would finally spread to include the entire syn-
drome of depression. (Beck, 1974c, p. 3)

The specific research hypotheses generated by Beck's

conclusions are presented in the next chapter.

CHAPTER II

PROCEDURES

Pppulation

 

Subjects were selected from a population of
adult new-referral clients seen at a community mental
health center. The center is located in a county seat
community in Michigan. The county is predominantly
rural but has some "bedroom" communities, and has a
population of approximately 70,000 people. The center
receives self-referred clients and individuals referred
by helping professionals, such as physicians and social

service caseworkers.

Sample

A sample of 11 clients was selected for the
study from the total population of new referrals at the
mental health center during the period April 1 through
October 10, 1975. Staff members referred 39 clients
who appeared to them to manifest clinical signs of
depression. The 11 subjects finally selected for the
study were classified "depressive neurotic" based upon
the clinical diagnostic procedures and psychological

test data described below.

22

23

Measures

Clinical Impressions

 

A pool of possible subjects was selected from
the client population at the mental health center.
Possible subjects were new referrals who appeared to
mental health staff persons to be suffering from neurotic
depression, as defined in DSM-II (American Psychiatric
Association, 1968). Although psychiatric diagnostic
accuracy has encountered increasing and recent criticism,
Beck (1967) defends the practice as being a necessary
subjective appraisal which is better than no judgment
at all, and which yields an inter-rater agreement of 63%
for the category "neurotic depressive reaction." The
psychodiagnostic judgments involved in this study were
merely precursors to further, more objective, selection
criteria.

Thirty-nine such possible subjects were next
interviewed by this investigator, who employed more
rigorous psychodiagnostic criteria previously used in
other studies:

Objective signs of depression in the facies,
speech, posture, and motor activity; a major
complaint of feeling depressed or sad; and at
least 11 of the following 14 signs and symptoms--
loss of appetite, weight loss, sleep disturbance,
loss of libido, fatigability, crying, pessimism,
suicidal wishes, indecisiveness, loss of sense

of humor, sense of boredom or apathy, overconcern

about health, excessive self-criticism, and loss
of initiative. (Beck, 1972, p. 229)

24

Additionally, individuals showing evidence of organic
brain syndrome, functional psychosis (including manic-
depressive psychosis and all of the schizophrenias), or
some other psychopathological state which was more promi-
nent than depression (Beck, 1972) were excluded from
further study. Available reports from family members,
summaries of other mental health contacts, and previous
psychological test data were reviewed by this investi-
gator before subjects were included in the study.

In treating depressed individuals, there is
always the risk of suicide, and there are ethical
issues involved when subjecting depressed individuals
to experimental techniques. To minimize the risks of
suicide, this investigator discussed their suicidal
intentions with clients during the evaluation interview.
All clients were required to agree not to attempt to
kill themselves during the course of the depression
treatment program. They were urged to call the clinic
at any time they might have suicidal intentions, and
they were reassured that they would receive immediate
supportive crisis intervention therapy. Any indi-
viduals who appeared to this investigator to be
potentially suicidal, and/or who would not make a
verbal contract not to attempt suicide, were referred
to other mental health therapists at the center and

‘were not involved in this study.

25

MMPI

The Minnesota Multiphasic Personality Inventory
(MMPI), developed by Hathaway and McKinley (1943), is
a widely used indicator of psychopathology. Although
the instrument has suffered increasing criticism by
theoreticians such as Mischel (1968) and clinicians
such as Beck (1974a), the tremendous influence of the
MMPI upon assessment research and psychodiagnostic
practice cannot be denied (Goldberg, 1974). In his
comprehensive review of the literature almost 30 years
ago, Ellis (1946) wrote that there was "no other person-
ality inventory on which so much theoretical and prac-
tical work" had been done. But more important than
the sheer quantity of attention focused on the MMPI,
Ellis concluded that no other psychodiagnostic inventory
was more valid (Horrocks, 1964).

The MMPI consists of 13 standard scales. Three
of the scales are validity indicators: L (lie),
F (validity), and K (correction). The remaining
clinical scales are: Scale 1 - Hypochondriasis (fig);
Scale 2 - Depression (2); Scale 3 - Hysteria (Ex):
Scale 4 - Psychopathic deviancy (2g); Scale 5 -
Masculinity-feminity (pg); Scale 6 - Paranoia (Pa);
Scale 7 - Psychasthenia (PE); Scale 8 - Schizophrenia
(Sp); Scale 9 - Hypomania (Ma); and Scale 0 - Social

introversion (Si) (Hathaway & McKinley, 1951).

26

The Depression scale "was established to measure
the degree or depth of the clinical symptom pattern of
depression. This mood state is characterized generally
by pessimism of outlook on life and the future, feelings
of hopelessness or worthlessness, slowing of thought and
action, and frequently by preoccupation with death and
suicidé'(Dahlstrom, Welsh & Dahlstrom, 1972). Dahlstrom
et al's picture of depression is essentially similar to
Hathaway and McKinley's (1956b) description of "sympto-
matic depression" measured by Scale 2. Scale 2 was
shown to differentiate to a significant degree clini-
cally depressed inpatients from normals (Hathaway &
McKinley, 1956b). Apparently Scale 2 is respectably
stable; test-retest reliability is reported as .77 t
.044 (Hathaway & McKinley, 1956b).

MMPI Scale 2 consists of 60 items which are
described elsewhere (e.g., Hathaway & McKinley, 1956b;
Dahlstrom et al., 1972). Dahlstrom et a1. (1972) note
that the items which best differentiate clinically
diagnosed depressed individuals from other groups are
"I feel weak all over much of the time," r = .78; "I
have a good appetite," r = .71; "My sleep is fitful and
disturbed," r = .66; "I'm afraid of losing my mind,"

r = .66; and "My memory seems to be all right," r = .65.

Although Scale 2 was devised with data from a

group of primarily psychotic inpatients, it has been

27

supported in the research "that the items reflected
depressive mood changes on a neurotic basis" (Dahlstrom
et al., 1972). The scale is reported to correlate .59
with an overall psychiatric rating of level of depression
(Dahlstrom et al., 1972). This is essentially what
Becker (1974) notes, that the MMPI Depression Scale does
reasonably well for predicting mean score differences
between nosological groups, but that it is less effective
for detecting individual differences within groups. With
all of its limitations, Scale 2 remains the most effec-
tive criterion-keyed "measure of the depth of the clini-
cally recognized symptom or symptom complex of depression"
(Horrocks, 1964, p. 546). Over the years variations and
abbreviations of Scale 2 have been reported (e.g.,

Rosen, 1962; Dempsey, 1964; Wiggins, 1966), but at this
point there is little evidence to Show any advantages
over Scale 2 in clinical practice (Dahlstrom et al.,
1972).

Also important in detecting depressive neurosis
is Scale 7 - Psychasthenia. The 48 items comprising this
scale are described elsewhere (e.g., Hathaway & McKinley,
1956a; Dahlstrom et al., 1972), and relate to "anxiety
symptoms, inability to resist, irrational fears, and
self-devaluation. This scale is a general measure of
anxiety and ruminative self-doubt. High scores tend
to be obsessionally worried, tense, indecisive, and

unable to concentrate" (Carson, 1969).

28

The importance of Scale 7 in detecting depression
is its frequent elevation in combination with Scale 2.
Dahlstrom et a1. (1972) note that the 27 combination
. . . is the most frequent high-point pair in
hospitalized psychiatric groups, male and female,
and is a predominant pattern among outpatient
psychiatric cases. . . . The predominant feature
of this group in presenting complaints is depres-
sion, with tenseness and nervousness as frequent
components.
A modal diagnosis of reactive depression is given for
the 27-72 pattern by numerous other investigators,
including Hathaway and Meehl (1951), Rosen (1958),
Silver and Sines (1961), and Donnelly and Murphy (1973).
In addition to the 27 combination providing
an indicator of neurotic depression, both scales appear
to be sensitive to changes in client functioning. In
fact, along with £9, Scales Q and 2; provide consistent
validity as indices of change (Bergin, 1971).

Shipley-Hartford Institute
of Living Scale

 

 

The Shipley-Hartford Institute of Living Scale
is a paper-and-pencil indicator of intellectual function-
ing. DevelOped by Shipley (1940), the scale was origi-
nally devised to measure intellectual deterioration due
to.organic brain dysfunction, but the test has been used
by later researchers as an estimate of intellectual per-

formance.

29

The Shipley contains 40 Vocabulary (synonym
matching) items and 20 Abstract (problem solving) items.
Reliability coefficients for the test are Vocabulary
(.87), Abstract-Thinking (.89), and Full scale (.92)
(Shipley, 1940). Shipley did not specify the type of
reliability reported.

The Shipley has been found to be a reasonable
estimate of intelligence, correlating .78 with the WAIS
Full Scale I.Q. score (Paulson & Lin, 1967). This
estimate is most accurate for average and higher I.Q.s;
accuracy declines below Shipley raw scores < 36 (esti-
mated WAIS Full Scale I.Q. = 92) as reported by Watson

and Klett (1968).

Beck Depression Inventory

 

Another measure employed in this study was the
Beck Depression Inventory (Beck, 1967, 1972; Beck et al.,
1961). The BDI was developed as an indicator of the
13131 of depression present in the patient. "The
implicit rationale underlying severity of depression
measures is that increased severity is associated with
a higher frequency and greater intensity of depressive
symptoms" (Becker, 1974). The scale purports to
measure the various psychological, physiological,
and behavioral manifestations of depression, and is
said to be useful in the diagnosis, treatment, and

research of depression (Beck, 1974a).

3O

Becker (1974) notes respectable reliability
data for the BDI, as indexed by internal consistency
and stability criteria: Split-half Spearman-Brown
corrected Pearson E'is said to be .93. All items sig-
nificantly related to the total score at the p < .001
level per item. And highly significant test—retest
correlations are noted.

In terms of concurrent validity, Beck (l974a)
cites several studies which report correlations of .62
to .73 between the BDI and psychiatrists' ratings of
depth of depression. Beck et a1. (1961) state the
inventory also is sensitive to changes in the severity
of depression over time intervals.

Becker (1974) agrees that the BDI is sensitive
to change, and that validity and reliability are more
thoroughly reported for the BDI than for any other
depression scale. He concludes that the BDI "is
probably the best developed and most widely used self-
report depression measure" (Becker, 1974).

The Beck Depression Inventory has also been

used as a screening device for depression (Beck &

 

Beamesderfer, 1974), based upon the correlations of
BDI scores with clinical ratings found by Beck et al.
(1961). A suggested cutting score for clinical neurotic

depression is BDI Z 17 (Beck, l974d).

31

For purposes of this study the BDI was relabelled
"BDI Questionnaire," and the factorial labels ("sadness,"
"pessimism," "sense of failure," etc.) above each cluster
of items have been deleted. Additionally, each cluster
of items has been numbered (1 through 21) and each item
within a cluster has been lettered (a through f). Other
than these minor changes in format, the instrument is
that devised by Beck et a1. (1961). The BDI Question-
naire is reproduced as Appendix A.

In this study the BDI will be entirely self-
administered, in contrast to the standard instructions
for the clinician-administration of the Beck Depression
Inventory (Beck, 1975). Other investigators (e.g.,

Bosse et al., 1975) have used the Beck as a self-
reported measure of depression, and claim this use

to be valid.

Reinforcement Survey Schedule

 

The Reinforcement Survey Schedule (Cautela &
Kastenbaum, 1967) is a paper-and-pencil instrument
"designed to assess the relative reinforcement values
of a variety of stimuli and stimulus situations" (Klein-
knecht et al., 1973). The scale was developed to find
possible reinforcers for use in operant conditioning
strategies for shaping new responses (Cautela & Kasten—

baum, 1967).

32

The RSS is divided into four major sections. In
the first three sections the respondent is asked to
rate [60] items on a five point scale representing
the degree to which the stimuli give joy or other
pleasurable feelings. The extreme points of the
scale are "not at all" and "very much." (Cautela

& Kastenbaum, 1967, pp. 1116-1117)

RSS section I consists of 10 items, comprising 33
decisions or subitems, and pertaining to objects.
Section II has 44 items--106 decisions--relating to
covert stimuli. Section III contains 6 items describing
social situations. In Section IV, the subject is asked
to list what he does or thinks about frequently (from
5 to 20 times per day) (Cautela & Kastenbaum, 1967).
The authors included Section IV so that Premack's
Principle (1959) might be incorporated in shaping
desired behavior:

The Premack Differential Probability Hypothesis
states that if two responses occur with different
probabilities, the response with the highest
probability can be used to reinforce the reSponse
with the lowest probability. . . . These responses
may include thoughts, activities . . . , aggressive
behavior, or derogatory remarks made to oneself.

. . . This section [IV] can be particularly valuable
in teaching the patient self-control responses,

and also in establishing therapeutic activities
which can occur in the absence of the researcher

or therapist. (Cautela & Kastenbaum, 1967,

pp. 1122-1123)

 

Additional applications of the Reinforcement Survey

Schedule may be found in Cautela and Wisocki (1971).

33

Outpatient Depressive Behavior
Rating Scale

 

A rating scale for measuring changes in overt
client behavior was devised, based upon the NOSIE-30
(Honigfeld, Gillis, & Klett, 1966). The NOSIE-30
(Nurse's Observation Scale for Inpatient EvaluatiOn)
is a ward behavior rating scale which has been used
extensively in V.A. and N.I.M.H. studies, particularly
psychopharmacological research. Contributing to the
instrument's "popularity are its: (a) brevity;

(b) easily rated behavioral items; (c) evaluation of
patient strengths and weaknesses; and (d) sensitivity
to patient change" (Honigfeld, 1974).

The NOSIE-30 consists of 30 behaviors typically
displayed by psychiatric inpatients. The behaviors are
rated by observers according to their judged frequency
of occurrence. The response categories are: (1) never,
(2) occasionally, (3) sometimes, (4) often, and
(5) always. Observers are asked to rate each behavior,
based upon direct observations over the previous three
days (Honigfeld et al., 1974).

For this study the NOSIE-30 was modified as the
Outpatient Depressive Behavior Rating Scale (BRS). The

response categories have been changed from judgments of

 

the frequency of behaviors to counts of the observed
frequency of behaviors. Raters were instructed to

count each behavior, and to indicate its observed

34

frequency of occurrence for that day only. The frequency
categories on the BRS are: (0) never, (1) once,
(2) twice, (3) three times, and (4) four times or more.
Additionally the BRS was shortened in length from
the NOSIE-30 based upon factorial analysis and visual
inspection of the items. Honigfeld (1974) reports that
26 of the NOSIE-30 items cluster into six factors:
social competence (COM); social interest (INT); per-
sonal neatness (NEA); irritability (IRR); manifest
psychosis (PSY); and (psychomotor) retardation (RET).
There is also a total assets (TOT) component sometimes
scored.
The social interest and (psychomotor) retardation
factors have apparent face validity for the assessment
of "depressive" behavior changes. The factors each
have respectable interrater reliability; INT = .95,
RET = .83 (Lentz, Paul, & Calhoun, 1971). The five
items making up INT and the three items comprising RET
have formed the basis of the 12-item depressed outpatient
client behavior scale devised by this investigator. The
items appear on the new scale (Appendix B) as items 2,
4, 6, 7, and 9 (INT), and as items 3, 10, and 12 (RET).
The four items of the NOSIE-30 which Honigfeld
does not report as loading heavily on any factor have

been included on the new behavior rating scale, because

35

these items appear to this investigator to tap depres-
sive behaviors. On the "behavior rating scale," these
behaviors appear as items 1, 5, 8, and 11.

The behavior rating scale thus contains 12
specific and observable behaviors which describe the
syndrome of depression defined earlier by Beck (l974a)
and Seligman (l974b). The frequency counts should
provide appropriate and valid data for a study of
behavior change (Thoresen & Anton, 1974) of depressed

individuals.

Procedures

 

Subjects
Potential subjects were referred to this inves-

tigator by staff persons of the community mental health
center. Those potential subjects were new clients who
were perceived by the staff members to manifest clinical
neurotic depression (American Psychiatric Association,
1968). This investigator interviewed all potential
subjects, and evaluated their psychiatric problem(s)
based upon the increasingly specific diagnostic cri-
teria described previously. These clinical and diag-
nostic procedures and criteria were:

. . . intended to yield a sample of "pure" depres-

sives, individuals in whom depression (state) is

present to a clinically significant degree, con-

stituting the major presenting psychopathology.
(Lewinsohn, 1974)

36

Beck (1975) has quite accurately referred to individuals

in this resulting sample population as primarily or

 

predominantly depressed.

 

Psychodiagnostic Instruments

 

Based upon the clinical diagnostic procedures,
all clients who appeared to be neurotically depressed
were requested to take three of the psychodiagnostic
instruments previously described. Clients were admin-
istered the MMPI prior to the therapeutic intervention
phase. The answer sheets were visually skimmed before
scoring to rule out gross response sets such as answer-
ing all items true, alternating true-false, etc. This
procedure was essentially that used by Mezzich et a1.
(1974).

Following the scoring of protocols all MMPI
profiles were visually inspected. Based upon other
relevant studies described by DiMascio et a1. (1968),
Dahlstrom et al. (1972), Donnelly and Murphy (1973,
1974), and Leonard (1974), clients with MMPI profiles
meeting the following criteria were included in the
study: Highest scores on Scales 2 and 7, with T scores >
70; Scale 8 < Scale 7; and Scale 9 T score < 70.

The MMPI was also administered to all subjects
following the termination of the treatment phase of

this study.

37

The Shipley-Hartford Institute of Living Scale
was administered to clients being considered for
inclusion in this study, following the clinical diag-
nostic procedure. The Shipley protocols were scored
according to the test manual (Boyle, 1967). Only
clients whose VOcabulary raw scores were less than 5
greater than their Abstract raw scores were included
in the study. A Vocabulary-Abstract difference > 5 is
said to be an indicator of a deficit in abstract
problem-solving ability due to psychotic thought dis-
order (Gasswint, 1971). Finally, WAIS Full Scale I.Q.s
were estimated from the Shipley Total raw scores (Paul-
son & Lin, 1967). Only clients with I.Q.s 1.90 were
included in the study.

Clients were administered the Beck Depression
Inventory ("BDI Questionnaire") as a final psycho-
diagnostic screening instrument. Only clients with
BDI total score 1 l7 (Beck, 1974d) were involved in
further study.

Eleven primarily neurotic depressed clients
were included in the study, based upon the above clinical
and psychometric diagnostic "cutting" procedures. None
of the psychometric information used in the subject
selection process was available to the therapist during

the course of treatment.

38

A11 eleven subjects were administered the Rein-
forcement Survey Schedule prior to their inclusion in
the treatment. The RSS was given at that time, so as
not to have a confounding effect upon the treatment at
some later date. The RSS was available to the therapist
for use in identifying potential reinforcers for sub-
jects during treatment.

Subjects were requested and required to complete
a copy of the BDI Questionnaire every day during the
course of the study. The BDI served as a repeated
measure of subjects' self-reported level of depressive
symptomatology.

Subjects also completed the BRS every day during
the course of the study. The Behavior Rating Scale
served as a repeated measure of client depressive
behaviors.

Subjects were furnished c0pies of the BDI and
the BRS, along with pre-dated, postal card-sized daily
response sheets for both instruments (Appendix C).

There were several methods for subjects to report their
daily BRS and BDI responses to this investigator. The
preferred method was for the investigator to telephone
the subject each evening, and for the subject to ver-
bally report the responses.

Unfortunately for the sake of experimental

consistency, not all subjects had telephones or even

39

regular access to a telephone. In such cases subjects
were instructed to fill out the response cards on a
daily basis, and to return them to the mental health

center upon their next appointment.

Client Behavior Raters

 

Subjects were requested to bring two "significant
others" (spouse, co-workers, friends, adult family mem-
bers, etc.) with them to the mental health center for a
brief interview with this investigator. Subjects were
asked to select individuals whom they saw frequently
(approximately daily), and whose judgment the subject
respected.

During the interview, the Behavior Rating Scale
was explained to them, and each rater was asked to com-
plete the scale at that time. Raters' questions about
scale completion procedures were answered at that time,
and the raters were requested to assist in the treatment
of the client's depression by independently rating the
client on the Behavior Rating Scale every day during the
course of the client's treatment.

Finally, all raters were furnished additional
copies of the scale and pre-dated, postal card-sized
daily response sheets for the BRS (Appendix C).

At a specified time each day this investigator
telephoned each rater to collect the data. It was pre-

dicted that telephone data collection procedures would

40

enhance reliability and completeness of data, and would
serve as a regular social reinforcement for the daily
rating of client depressive behaviors. When contacted
by telephone, raters would report counts of observed
behavioral frequencies verbally to the investigator.

On those occasions when raters were not available by
telephone, they were requested to mail or bring in their
completed rating cards to the mental health center as
soon as possible.

Although the nature of the treatment, and of the
client's psychological problem(s) was not explained to
them, they were given an explanation of the treatment
approach, following the course of treatment. Addi-
tionally, the importance of their assistance was stressed

and verbally reinforced.

Experimenters

 

Four experimenters were used in this study.
Therapists were graduate students in the Department of
Counseling, Personnel Services, and Educational Psy-
chology, Michigan State University, who volunteered to
assist in the study.

Therapists received training in cognitive-
behavior modification during the month of April, 1974.
This investigator was assisted in instructing the

training sessions by a second doctoral student (now

41

an assistant professor of counseling) with extensive
training and experience in rational-emotive, behavioral,
and sexual therapy.

Therapists' training consisted of three two-hour
sessions. The first of the training segments was a
didactic presentation of the cognitive treatment of
depression. Rational-emotive psychotherapy was sum-
marized, and excerpts were played from various audio
tapes made by Albert Ellis. Trainees had previously

read A_Guide pg Rational Living (Ellis & Harper, 1961).

 

 

Aaron Beck's treatment of depression was summarized,
and they were furnished copies of the training manual
developed by Beck and Rush (1974). Principles of
behavior modification were not stressed, as all trainees
had been experienced in using this treatment modality
as part of their graduate training in counseling at
Michigan State University.

The second training session began with a brief
review of the first session and a discussion of the
materials trainees had been furnished to study.
Specific questions and concerns were responded to and
problematic concepts were clarified. The treatment
interventions which trainees would introduce during
the study were then discussed (these interventions are
delineated in the next section). When it was believed

that all trainees understood what was expected of them

42

during each session of the treatment of depression, a
role-playing session was staged. With trainees observ-
ing, the instructors played a depressed client and
therapist, and the interventions of the depression
treatment were modeled.

The final training session was a rational-
emotive/cognitive group experience. The four trainees
and two instructors assisted each other in identifying
irrational and maladaptive assumptions underlying one
emotion which the participant identified as currently
problematic for himself.

Following the three training sessions, the
instructors concurred that all four trainees possessed
requisite theoretical understanding and adequate clini-
cal skills to administer cognitive/behavioral techniques
to the depressed subjects in this study. In addition
to his involvement in their training, this investigator
served as a consultant to therapists during the treat-
:ment phase of the study. Audio tapes and case notes
from each treatment session were reviewed to assure
that the therapist followed the treatment specifications
and that quality control of treatment was maintained.

Because of the therapists' training in cognitive
and behavioral treatments, they could not be assumed to
be "blind" to the treatment package or philosophy, and

by inference to the research hypotheses. However,

43

they did not have access during the course of treatment
to the screening or treatment measures used to assess
subjects' responses. Thus therapists were not double-

blind, but they were data blind during the course of

 

treatment.

Treatment Intervention
With any behavior modification technique, the

intervention depends a great deal upon the specific
problematic client behavior(s). With cognitive behavior
modification,

. . . the specific targets and techniques to be

selected depend on the therapist' judgment. . . .

It must be emphasized that the approaches must

be tailored to the individual patient. . . .

One or more cognitions or beliefs can be illumi-

nated and addressed as a target. (Beck & Rush,
1974, p. 7)

It is thus a dilemma to define a treatment which is
flexible enough to allow for generalization to other
clients, yet specific enough to permit replicability
and clear understanding. The following treatment
package was devised to meet these criteria:

1. In the first treatment session, the therapist
spent the first few minutes meeting the subject and
establishing rapport. The therapist then introduced
the cognitive-behavioral treatment of depression by
furnishing the subject with a copy of the pamphlet,
Coping with Depression, by Beck and Greenberg (1974)

to serve as a client treatment manual. The subject

44

read through the booklet during the session, and ther-
apist and subject discussed the reading. The therapist
emphasized and elaborated upon the section "The A.B.C.s
of Changing Feelings" (Beck & Greenberg, 1974, p. 12).
Finally, the therapist suggested that the subject
reread the booklet once each day, and whenever else

the subject experienced feelings of depression, until
the second treatment session.

2. During the second treatment session, the
therapist assisted the subject in identifying at least
five automatic and maladaptive depressive cognitions.
The therapist instructed the subject in challenging each
identified cognition. Therapists limited their verbal
therapeutic interventions to the "Cognitive Targets
Symptoms" delineated in Beck and Rush (1974, p. 9),
and which have been reproduced as Appendix D. The
therapist then assigned the subject to read the list
of depressive cognitions and corresponding challenges
before engaging in a high probability behavior, until
the next session. The high probability behavior was
selected by the therapist and the subject by reviewing
the Reinforcement Survey Schedule previously completed
by the subject. She was instructed to log the occur-
rence of reading the list of depressive cognitions and
challenges, and to report this frequency to the therapist

during the third session.

45

3. During the third session the therapist
assisted the subject in generating a list of at least
five positive self-evaluatory statements. The therapist
instructed the subject to read through the list before
engaging in a high probability behavior [such as light-
ing a cigarette (Seitz, 197la)], until the next session.
Therapist and subject selected a second high probability
behavior from the RSS, and the subject again logged the
frequency of reading the positive self-statements.

4. In the final session, the therapist reviewed
with the subject the techniques of cognitive-behavior
modification, and structured the termination of treat-
ment contacts. The therapist also scheduled a followup
session for two weeks after this fourth session.

Therapists thus saw each subject for four treat-
ment sessions of one hour maximum length over a two-week
period. Whenever possible during the treatment sessions,
therapists reviewed and reinforced client progress, so
that learning how to decrease depression was generalized

and maximized.

Followup

Subjects were requested and required to complete
the BDI questionnaire every day during the course of
treatment and for two weeks following their completion
of the treatment phase. At that time (two weeks later)

they had a second interview with this investigator.

46

Their current level of psychological functioning was
evaluated (clinically) and they were administered a
second MMPI. Any concerns they may have had regarding
their treatment were discussed, and they were offered
further mental health services at the center if they
desired. Three months following their completion of the
treatment phase, subjects were requested to complete the
BDI questionnaire and the MMPI.

Subjects and significant others were requested
and encouraged to complete a behavioral rating for the
client every day during the course of treatment and
for two weeks following the completion of the treatment
phase. At that time (two weeks later), the raters were
contacted by this investigator. The depression treat-
ment program was discussed with them, although specific
details of the client's problem(s) and functioning were,
of course, not divulged. Their assistance and cooper-
ation with the clients' treatment and with the depression
treatment program were again verbally reinforced. Three
months following the client's treatment program, the
significant others were requested to rate the client's

current behavior.

47

Design

Intensiye Case Study
Design-~Review

 

 

In recent years, much support has been generated
for the application of single-case, or "intensive,"
designs to clinical research settings. Such designs
have a long and respectable history in the physical and
biological sciences, and recently have gained credi-
bility in research in counseling and psychotherapy
(Thoresen, 1972). Contrariwise, comparative group
designs have drawn increasing criticism from applied
researchers as being irrelevant or ineffective in use
in counseling research:

Typically, client change during treatment is
obscured in group designs . . . and the relation-
ship between change and treatment for individual
clients may vary in fundamentally different ways
compared to the average change of a group of sub-
jects. . . . Exclusive concern with group means
and variabilities impedes understanding of the
treatment process, and may lead to erroneous
generalizations about treatment effects.
(Thoresen & Anton, 1972, p. 553)

Bergin (1971) likewise reviewed the criticisms
of comparative group designs and offered the "experi-
mental case—study approach" as a solution:

It involves planning specific interventions with
patients having specifiable disorders. Careful
measurement is done of the presenting problem
before, during, and after treatment, and an
equally precise reading of the nature of the
interventions is recorded. The course of change
is thus thoroughly recorded and the responses to
the intervention pinpointed. (p. 254)

48

This approach would appear to meet the need for measures
described by Zytowski and Betz (1972), "which can follow
the progress of the client in counseling, so that his
improvement can be charted as a physician does that of
his surgical patient" (p. 554). Bergin (1971) notes
that the experimental case study approach meets the need
for specificity in research, while not detracting from
the complexities of clinical practice. The approach
"nicely articulates a research strategy with the vicis-
situdes of clinical practice, and thus insures more
prompt transfer of findings to practice than is usually
the case" (p. 255).

Since Sidman published his classic works on
operant techniques and design (1960, 1962), more and
more behavioral researchers have found single case
designs effective in the study of deviant behavior
(Krasner, 1971). The single case approach would appear
to be well-suited for the study of the cognitive-
behavioral treatment of depression: (a) psychological
correlates of depression can be adequately isolated
(Beck, 1972); and (b) the effect(s) of specific treat-
ment modalities may be systematically observed.

There is one more advantage of the single-case
design in depression research. There is always the
risk of suicide when dealing with depressed individuals,

and there are ethical and even legal contraindications

49

for assigning potentially suicidal subjects to a control
group (Kushner, 1974). In the single-case design each
subject serves as his own control for events prior to
the treatment intervention (Thoresen & Anton, 1974),
and each subject may be involved in active treatment.
The client's progress is plotted on a repeated measures
basis, and the change in the slope of the data curve
over time is the concern. "If the counselor can demon-
strate comparable results with the same client each
time a technique is used . . . , then confidence is
increased that the counseling procedure is effective"
(Thoresen & Anton, 1974, p. 556).

Thoresen (1972) and Thoresen and Anton (1972)
review other advantages, applications, and variations
of single-case designs. One design which is widely
touted is the reversal, or "ABAB" design. Thoresen
and Anton note that the design may be inappropriate
"since the change in client behavior may not be
reversible" . . . or "the problem behavior may be
severe and should not be reinstated" (1974, p. 555).
No clinician (nor client!) would wish to see depressive
behavior reinstated, particularly with the risk of
suicide. As an alternative, Thoresen and Anton (1974) -
-propose the multiple-baseline design.

Hall et a1. (1970) present an excellent descrip-

tion of multiple-baseline procedures, and illustrate

50

three design variants--across individuals, across
behaviors, and across situations. Liberman and Smith
(1972) have effectively applied the design to behavior
therapy research. Mahoney and Thoresen (1974) describe .
the essential characteristics of the design:

(1) Data must be kept on three or more behaviors
(or individuals).

(2) Treatment is applied to target behavior one
until a change is shown in that behavior.

(3) Then the intervention is applied to additional
target behaviors.

(4) If the target behaviors change substantially
at the point of treatment intervention, then
a causal relationship may be inferred.

Summary of Design and
Procedures

 

In this study a multiple-baseline design was
employed. Individual subjects' self-reported level of
depressive symptomatology (EDI) and self- and other-
reported behavior ratings (BRS) were charted every day
on a repeated measures basis.

Subjects were divided into three groups, and
each group began treatment at a varying time relative
to the initiation of data collection. Specifically,
Group I subjects began treatment 7 days following
data collection, Group II subjects began treatment 10
days following data collection, and Group III subjects
began treatment after 14 days.

To control for the effect(s) of attention by a

counselor, Group II and III subjects were involved in

51

noncognitive, nonbehavioral, "placebo" counseling
sessions during the seventh and the seventh and tenth
days, respectively. These sessions consisted of client-
centered, empathic counseling.

Data were collected for each subject during the
7, 10, or 14 days preceding treatment, for 14 days
during the treatment phase, and for 14 days following
treatment. There were four daily repeated measures
plotted for each subject: his total score on the BDI
Questionnaire and his mean scores on each of the three
areas of the behavior rating scale (RET factor, INT
factor, and "depression" items).

The treatment effects were contrasted across four

 

behaviors and across eleven individuals at the same

 

 

relative points in the treatment schedule.

Research Hypotheses

 

Beck makes two conclusions which follow from his
theory and assumptions of the cognitive modification of
depression. Beck's theory has tremendous potential for
the understanding, treatment, and prevention of neurotic
depression in humans. To this writer's knowledge, the
conclusions per ee have not been explicitly validated.

Hypotheses of this proposed study are:

52

Conclusion 1:

 

By pinpointing the patient's specific cognitive
distortions and demonstrating their invalidity
through behavioral or cognitive techniques, the
psychotherapist can achieve a rapid improvement
in the symptoms of depression. (Beck, l974c, p. 19)

Hypothesis 1:

 

Specifically for this study, it was hypothesized
that there would be a reduction in the level of
depression, as measured daily by the Beck Depression
Inventory, immediately following the point of treat-
ment intervention. This reduction in level of
depression will be indicated by a reduction in the
median lepe trend.

Hypothesis 2:

 

The level of depression would increase during the
two-week reversal phase following treatment. This
increase in daily BDI scores will be indicated by
an increase in the median slope trend.

Hypothesis 3:

 

Although it was predicted that the level of depression
would rise during the reversal phase, it was hypothe-
sized that due to generalization of learning, the

BDI scores would remain lower than during the base-
line phase, as indicated by a reduction in the

median slope trends.

Conclusion 2:

 

Since each of the components of depression
(behavioral, cognitive, emotional, motivational,

or physiological] contribute to the other compo-
nents, . . . an improvement in any one problem

area [e.g., cognitive] would lead to an improvement
in other areas [e.g., behavioral], which would
finally spread to include the entire syndrome of
depression. (Beck, l974c, p. 3)

Hypothesis 4:

 

There would be a reduction in daily behavioral
Depression item (BRS) scores immediately following
the point of intervention, as indicated by a
reduction in the median slope trend.

53

Hypothesis 5:

 

There would be an increase in daily behavioral
Depression item (BRS) scores during the reversal
phase, as indicated by an increase in the median
slope trend.

Hypothesis 6:

 

During the reversal phase the daily behavioral
Depression item (BRS) would remain lower than
during the baseline phase, as indicated by a
reduction in the median slope trend.

Hypothesis 7:

 

There would be a reduction in daily behavioral
Psychomotor Retardation factor (BRS) scores
immediately following the point of intervention,
as indicated by a reduction in the median slope
trend.

Hypothesis 8:

 

There would be an increase in daily behavioral
Psychomotor Retardation factor (BRS) scores during
the reversal phase, as indicated by a decrease in
the median slope trend.

Hypothesis 9:

 

During the reversal phase the daily behavioral
Psychomotor Retardation factor (BRS) scores would
remain lower than during the baseline phase, as
indicated by a reduction in the median slope trend.

Hypothesis 10:

 

There would be an increase in daily behavioral
Social Interest factor (BRS) scores immediately
following the point of intervention, as indicated
by an increase in the median slope trend.

Hypothesis 11:

There would be a decrease in daily behavioral
Social Interest factor (BRS) scores during the
reversal phase, as indicated by a decrease in the
median slope trend.

54

Hypothesis 12:

 

During the reversal phase the daily behavioral
Social Interest factor (BRS) scores would remain
higher than during the baseline phase, as indi-
cated by an increase in the median slope trend.

Data Analysis

 

The data were analyzed by nonparametric trend
analysis statistics (Siegel, 1956). These statistics
gave tests for statistical significance of changes in
the slope of data curves at the point(s) of treatment
intervention(s). The statistical procedures involved
have been described by Campbell (1963) and by White
(1971, 1972).

According to the procedures outlined by White
(1971), a slope for the data in each phase of the study
for each subject was estimated. This estimated slope
is known as the "median trend." In order to compare
the data obtained during two phases, the median trend
of one phase may be extended into the other phase of
interest. According to probability theory, half of
the data will fall on one side of the extended median
trend, half on the other side, when the data in both
phases are from the same larger population of data.
With p = .5 that a bit of data will fall on a predicted
side of the extended median trend line, a binomial

x N-x

formula, Px = (g) p q , gives the probability for

an observed number of data points, x, falling in the

55

experimentally undesirable direction of the median
trend. If the binomial probability formula yields no
statistically significant differences between data from
two phases, then the phases cannot be said to be dif-
ferent in their effects upon the data (Siegel, 1956).

The data were analyzed across phase changes for
each of the subjects, and it was ascertained whether
there were statistically significant differences in
the hypothesized directions.

Presented below is a discussion of each subject
involved in this study. A brief description of the
subject's presenting problems and depressive symptoma-
tology will be followed by a summary of the subject's
performance on each of the four daily measures of
depression. Results of the statistical tests will be
noted, as will whether the subject's data support or
do not support each of the research hypotheses.
Supplementary data for each subject will next be pre-
sented, and to finalize the discussion of each subject,
the therapeutic outcome will be summarized.

To better summarize and organize the results
of all subjects, a section will follow in which the
performance of all subjects across all measures will
be described. Certain supplementary findings from the
data are then presented. The MMPI pre- and posttest

results for all subjects are analyzed and the results

56

described. Also presented will be inter-rater relia-
bility data obtained from the daily ratings on the

Behavior Rating Scale.

CHAPTER III

RESULTS

Single Subject Results

 

Subject 1

 

Subject 1 is a 27-year-old female who has a
six-year-old son, and who is attending a local community
college. Subject 1 had felt depressed for "several
years," and she has a history of suicidal gestures.

She had drifted from one unsuccessful male-female
relationship to the next, and in general was dissatis-
fied with her social life. She felt generally bored

in life, and reported symptoms of depression including
social withdrawal, crying easily, and feelings of worth-
lessness and hopelessness. Her speech was slow and
unvarying in tone, and she had a sad expression on her
face.

Subject 1's scores on the four daily measures of
depression are shown in Figure 1. On the first depen-
dent variable, the BDI scores, she stopped recording
four days into the treatment condition, and was not
re3ponsive to requests by this investigator or by her

therapist to complete the BDI on a daily basis. As

57

58

”£3“

 

 

30
m
msaxon zo
INVENTORY
10
o
5
u
as 3
mansion
mm 2
1
0
1o
9
e
7
wow 6
manner 5
"cm I:
3
2
1
o
no
15
ms
scam. 10
IN
ncron
5
o

 

“ll’lllll)liililhll'l‘I‘Iilhlil‘l'lllilthllilllll‘llllllll-

Figure 1. Subject 1; Daily scores on four dependent
measures of depression.

59

may be seen, there was negligible improvement in her

BDI scores, and statistical analysis of baseline-treated
phase differences between the BDI median trends was not
attempted (Table 1). Thus in the case of Subject 1,
Hypotheses l, 2, and 3 are untestable due to her failure
to respond to the BDI.

The data for the three daily measures of behavior
obtained from Subject 1's BRS scores are complete through
the treatment phase. As may be seen in Figure 1, her
daily scores on Depression items declined during the
study, but this tendency began during the baseline phase.
Changes between the median trends of the baseline and
treatment phases were nonsignificant (Table 1), and
Hypothesis 4 was therefore not supported. She did not
return for her fourth treatment session, even though
repeatedly contacted by this investigator, saying that
she was too busy "working" and "moving to a new apart-
ment." Because of her attrition from the study, there
are no BRS data for the reversal phase, and Hypotheses 5
and 6 remain untested for Subject 1.

There was a reduction in her Psychomotor Retard-
ation scores during treatment, and the decrease was
statistically significant (p=.001). In this case
Hypothesis 7 was supported at the a = .05 level, but
due to subject attrition, Hypotheses 8 and 9 were not

tested.

60

Table l

Binomial Tests of Phase Changes for Research Hypotheses
Involving Beck Depression Inventory

 

 

Baseline- Treatment- Baseline-
Subject Treatment Reversal Reversal
(H1) (H2) (H3)
1 Incomplete Incomplete Incomplete
Data Data Data
2 5 .001a 1.000 5 .001a
3 5 .001a .994 5 .001a
4 1.000 3 .001a 1.000
5 .788 .967 .967
6 .971 1.000 .0063
7 5 .001a ; .001a 5 .001a
8 .500 5 .001a 5 .001a
9 .019a 1 .001a < .001a
10 Incomplete Incomplete Incomplete
Data Data Data
11 .855 .145 1.000
Across All 6 of 9 4 of 9 6 of 9
Subjects Significant Significant Significant

 

aIndicates statistical significance at a = .05

61

Subject 1's scores on the Social Interest factor
increased throughout the study, but there was an upward
trend evident during the baseline phase. Consequently,
median trend differences were nonsignificant, and
Hypothesis 10 was not supported. Again due to incom-
plete data, Hypotheses 11 and 12 were not tested.

As a piece of supplementary data, her MMPI pre-
test profile is included as Figure 2. This profile is
typical of the pretest profiles of subjects in this
study. It may be seen that there are peaks on Scales 2,
7, and 0. This pattern would be expected for depressed,
anxious, and socially withdrawn and introverted indi-
viduals. The moderate elevation on Scale 4 and the
depression on Scale 5 are indicative of social dis-
satisfaction and a tendency to respond interpersonally
in a passive-dependent fashion. As was mentioned pre-
viously, she did not return for her final counseling

sessions; thus MMPI posttest data are not available.

Subject 2

 

Subject 2 is a 33-year-old married woman with
three children. She is a high school graduate, and is
a homemaker. Her husband is a skilled tradesman. She
was referred to the mental health center by her gynecolo-
gist for "nerves and depression." Subject 2 stated that
she had been crying a great deal, and that she became

anxious and shook uncontrollably. She felt inadequate

62

The Minnesota Multiphasic Personality Inventory
Starke R. Hathaway and I. Charnley McKinley

Sector's Initialn______

 

I a a 4 s 6 7 e 9 0 ram
Tor‘l'c ? L F K H:+.5K D Hy Pd+.4K Ml Pa PH-IK SC'HK Mal-x St TOITC Additional
11).. - . - _r ‘4” .,
“5 .' : ' . _. :.
female - : - .5; ”i “2
110-5 ' - 50_ ; - ;" .1-.- -1- -. 2"“0 ”w

‘0-

 

 

 

 

 

 

 

 

 

TorTc 7 L F K Hatsx D Hy Pd*.4K M! Pa PINK Sc’lKMaLZK
l 8 8 4 5 6 7 8 9

Tech:

em)

= pretest

Figure 2. Subject 1; MMPI pretreatment profile.

NOTE: Copyright 1948 by The Psychological
Corporation. All rights reserved as stated in the
manual and Catalog. The Psychological Corporation,
New York, N.Y.

63

as a mother and imperfect as a wife. She was extremely
sensitive to the criticism of others, and she felt like
a failure. Her sleep pattern was disturbed--she awakened
frequently, could not get a full night's sleep, woke up
tired, and slept much during the day. She had felt
depressed for three to four years, and had considered
committing suicide.

Subject 2's daily performance on the four depen-
dent measures of this study are presented as Figure 3.
As may be seen in the graph for the BDI, there was an
abrupt decrease in her self-reported level of depression
just after the point of intervention. This decrease in
the median trend is statistically significant (Table l),
supporting Hypothesis 1. There was not the predicted
upswing in BDI scores during the reversal phase. Treat-
ment and reversal phases had median trends which were
not significantly different, and Hypothesis 2 was not
supported. The lower median trend in the reversal phase
than during baseline is, however, statistically signifi-
cant, and Hypothesis 3 is thus supported. Overall,
Subject 2's scores on the BDI dropped appreciably after
treatment began, and continued to decline over the
period of the study.

Her performance on the three measures derived
from daily BRS ratings supports none of the Hypotheses

4-12. It may be observed that she "improved" during the

64

Pun-rs

io

 

OFNUF’UI

p

:3
8
OHNU#UOVO\OO

 

20
m 15
90cm.
name? to
um
5
o nus

 

w ;' I 1. ~ m I: ll .1. I- :n 3' '| ’n .:-~ .I 1,: : 'u w I- .' II In. |~
llHIHIHIHIHIHIHIHIHIHIHlHIWIHIHIV
I . 1

Figure 3. Subject 2; Daily scores on four dependent
measures of depression.

65

study--her scores on the Depression items decreased,

as did the scores on Psychomotor Retardation, but
Social Interest scores increased. However, in each

of the baseline phases for the three measures, median
trends are toward the desirable direction, and there
were no statistically significant differences between
baseline and treatment phases (Tables 2-4). Nor were
there statistically significant differences between the
median trends of the treatment and reversal phases.
Thus, although Subject 2's behavioral rating scores
became more desirable as the study progressed, they
were heading in this direction even during the baseline
condition, and positive changes cannot be said to be
due to the treatment.

Subject 2's MMPI pre- and posttest profiles are
presented as Figure 4. It may be seen that there were
drops of at least one standard deviation in her scores
on scales 2, 4, 6, 7, 8, and 0, and that her posttest
profile is well within normal limits.

Following completion of her treatment, Subject 2
felt happy--as she described it, "better than I've felt
in years." She felt more at ease with her husband and
with her children. Her crying had ceased, and she was
no longer sleeping during the day. She desired no
further counseling at the end of treatment, because
she wanted to see how she would do with her depression

on her own.

66

Table 2

Binomial Tests of Phase Changes for Research Hypotheses
Involving Behavior Rating Scale Depression Items

 

 

 

Baseline- Treatment- Baseline-
Subject Treatment Reversal Reversal
(H4) (H5) (H6)
1 .593 Incomplete Incomplete
Data Data
2 1.000 1.000 1.000
3 i .001a .685 ,3 .001a
4 1.000 3 .001a 1.000
s .971 _<__.001a 1.000
6 1.000 .994 1.000
7 3 .001a : .001a .033a
8 1.000 1 .001a 1.000
9 .019a 1.000 3 .001a
10 Incomplete Incomplete Incomplete
Data Data Data
11 1.000 1.000 1.000
Across All 3 of 10 4 of 9 3 of 9
Subjects Significant Significant Significant
aIndicates statistical significance at a = .05

69

The Minnesota Multiphasic Personality Inventory
Starke R. Hathaway and I. Charnley McKinley

Scout‘s Initiale_____.

 

     
   

 

 

 

 

 

 

 

 

 

 

n a a 4 s o 1 a 9 0 ram...
TorTc 7 L r K H:+.5K D HL PU+.4_K Ml Pa n+1: 5cm: Ham: 5. TaTc mamma-
m-_— , - -- ..7 ,1- r - ._-m r -
us -’ - ' - 3- 115
female - : - .5; ”i “2 i
110-: , so; ; —~~ ~~--.—- ~— -. _ ‘b- - E-no —---—
z ' ' ' ' 50": 50-. - : l
nos—_ 40- — so— , - , :105
loo-f : 45- 3 co- — - zs- , $5; ss-‘ ’ .7 {—100 ~
: - ‘ ' - - ' : ‘ 70- :
95 —i 35- I ‘5'. - ls- _ I : 35‘ : .- 95
‘ ' ' ° ' so- - 66-: '
93-1 5°- ‘ ‘ . "'2’-”
- l m€-
as -_ : 2‘ 35
g 30.. 55’:
so—_ l30- - :- -: -ao
- 15- - -
- so
75 f m— ‘ : " 75
E “0- ': :
.. o-— #
7° : ‘00- : : 70
Z w... : I
55 I m- E I 65
fi- -.
w“—: 70- ... ‘Tm
6°“ ‘ :
ss-Z 5°" ao-I 2-55
(0— - ; _
45-} 20': IS- 20% E-ts
4o-_ 5' 20‘- ' - 15-’ {-ao
‘ 15‘. ' :
as —: 0‘ 5‘- - : ,0- - - - = ; - no-: :- as
5‘: )0- : _ ‘5_ * IS: 0-: IO- : -
— — —- - -fl— *
3° - m- _ _—J——"- _""S" 3° .
o- - S_ - _ o— - - -
25 - - _ Io- - -zs
: 0- ‘ 50" 1 s- 5:. 3
20—7 - - - . _-zo
o-‘ 3—0 !
TorTc 9 L r K Hs*.5l( D Hy pdux Mt Pa me sco-IK Ma+1K s: TorTc
l 2 3 6 B 6 7 8 9 0

 

= pretest
————————————— = posttest

Figure 4. Subject 2; MMPI pre- and posttreatment pro-
files.

NOTE: Copyright 1948 by The Psychological
Corporation. All rights reserved as stated in the
manual and Catalog. The Psychological Corporation,
New York, N.Y.

70

Subject 3

 

Subject 3 is a 28-year-old married woman with
children. She complained of nervousness and depression,
crying easily, irritability, and being easily confused.
She had considered suicide, had withdrawn socially, and
appeared disheveled and unkempt. She was dissatisfied
about her marriage to a man who was alcoholic, and who
was irresponsible with the family finances.

Subject 3's performance on the four dependent
variables is shown as Figure 5. She showed a statisti-
cally significant (Table 1) reduction in her BDI scores
at the point of treatment intervention, supporting
Hypothesis 1. The data in the reversal phase did not
increase as hypothesized (Hypothesis 2), but the
reversal phase median trend was lower than during
baseline phase (p < .001), supporting Hypothesis 3.

Her scores on the BRS Depression items dropped
significantly (Table 2) during the treatment phase.
There was not a significant difference between the
treatment and reversal phases, but the reversal phase
scores remained lower than during baseline. For this
subject, then, Hypotheses 4 and 6 are supported, but
Hypothesis 5 is not supported by the data.

Subject 3's scores on the Psychomotor Retardation
factor show a similar pattern. A significant reduction

of the median trend during the treatment phase (Table 3)

BECK
DEPRESSION
INVENTORY

BRS
DEPRESSION
ITEMS

PSYCHOMOTOR
RETARDATION
FACTOR

BRS
SOCIAL
INTEREST
FACTOR

Figure 5.

71

POINTS
4O

30

_a

o—omw-hu-Imuoooo

 

20

 

u .' I I. ~ in I‘.‘ II ‘I1 m _'n 5’ ,‘I _v. 3» HI ._- 'I ~ -‘ -- 1. H " h DAYS

Subject 3; Daily scores on four dependent
measures of depression.

72

(p < .001), is followed in the reversal phase by a
median trend which did not increase (p = 1.000), but
which remained lower than during the baseline phase
(p < .001). Thus, the Psychomotor Retardation data
support Hypotheses 7 and 9 but do not support Hypothesis 8
for this subject.

On the Social Interest factor of the BRS,
Subject 3's scores increased significantly (Table 4)
(p < .001) during the treatment phase. There was a
decline in the median trend during the reversal phase
(p = .001), but the reversal median trend remained
higher than during baseline (p < .001). The data for
the Social Interest factor for her support Hypotheses 10,
11, and 12.

As indicated by her MMPI profiles (Figure 6),
Subject 3 showed only slight improvement between pre-
and posttest scores. The scales of interest, 2 and 7,
decreased .5 and 1 standard deviations, respectively.
Overall, there was no appreciable change in her "per-
sonality" functioning, as indicated by her MMPI profile.

Subject 3's MMPI posttest profile was discussed
with her during her session following the reversal con-
dition. Because of probable remaining problems in the
areas of depression (Scale 2), anxiety (Scale 7), marital
and interpersonal relationships (Scales 4, 5, and 6),

and with her social functioning (Scale 0), I urged her

73

The Minnesota Multiphasic Personality Inventory
Starks R. Hathaway and I . Charnley McKinley

Scorer's Initials______

 

 

 

 

 

 

 

 

 

 

 

 

TorTc 9 L r K H.151: 6 Hay 1:81.95 III P: PHI: scaxuazzx ; TorTc AZWSQH
120—; - -- 7 ~ —- -- - - --- ~71!) r
115-3 _ as; E-ns
sFemale .- 3°_ 3 ;
llO—_ ' - _ .-HO
.05.?
100—?
9.;
90-3
eé
80-:- 130—
75-3 120-
3 no-
NET-10
6.5-: :3: '
50-3 7°“
- w_
55-3 5°-
‘0-
so——-3o--—-
45-:
4o-
35-
30—
25—
m—
o_: _o l
TorTc 7 L F K HahSK D Hy Pd+AK M! Pa Ptth Sc'lKMatZK Si TorTc
l 8 8 4 5 6 7 O 9 O
= pretest
------------- = posttest
Figure 6. Subject 3; MMPI pre- and posttreatment pro-
files.
NOTE: Copyright 1948 by The Psychological
Corporation. All rights reserved as stated in the

manual and Catalog.

New York, N.Y.

The Psychological Corporation,

74

to continue coming for counseling at the mental health
center. Such counseling might have included group
assertion training and conjoint marital therapy.
Subject 3 declined the invitation for future
counseling. She claimed that she felt less depressed,
and that she had just obtained employment as a check-
out clerk in a department store. She promised that she
would return for further counseling if she needed it in
the future, but that she was satisfied with her present

psychological state of affairs.

Subject 4

 

Subject 4 is a 53-year-old female who claimed
to have been depressed "off and on" for as long as she
could remember. She is married, and lives with her
husband on their farm. She has children who are married,
and who live with their families near the farm place.
Subject 4 was hospitalized recently when there was a
financial setback on the farm, and she saw "everything
we'd ever worked for go up in smoke." She felt exhausted,
hopeless, disappointed, and frustrated. She could no
longer motivate herself to do housework. Her speech
was slow, and she appeared to be sad. Her sleep was
fitful--"I wake up in a cold sweat every morning"--
and her appetite was poor. She and her husband were
experiencing marital difficulties, and she was isolating

herself from friends.

75

Subject 4's performance on the four daily
measures is included as Figure 7. It can be seen that
her scores on the Beck Depression Inventory dropped
overall, although the median trend was downward even
during baseline. Differences between median trends
occurred only over one phase change (Table 1). Thus,
Hypothesis 2 is supported by the BDI data, but Hypothe-
ses l and 3 are not supported. The reduction of level
of depression cannot be said to be due to the treatment
intervention.

Similar results may be observed on her scores
on the BRS Depression items (Figure 7). Overall her
scores decreased, but there was also a decreasing median
trend during the baseline. The phase changes between
baseline and treatment, and baseline and reversal are
nonsignificant (Table 2), failing to support Hypotheses 4
and 6. Only Hypothesis 5 is supported.

Subject 4's BRS Psychomotor Retardation scores
also showed a declining median trend during the baseline.
The phase change into the treatment phase was not sta—
tistically significant (Table 3), nor was this phase
significantly different than reversal. Thus, Hypothe-
ses 7 and 9 were not supported. Only the difference
between the median trends of treatment and reversal
phases was statistically significant (p < .001),

supporting Hypothesis 8.

76

 

 

 

POINTS
40
30
BECK
DEPRESSION 20
INVENTORY
IO
0
5
4
BRS 3
DEPRESSION
ITEMS 2
I
0
IO
9
8
7
PSYCHOMOTOR 6
RETARDATION 5
FACTOR 4
3
2
I
0
20
BRS 15
SOCIAL
INTEREST IO
FACTOR
5
0 u g : ., s In l'.’ II N‘- IA tn :3 .'I I. .~ 'u x. ’I a» 1‘ I” DAYS

 

Figure 7. Subject 4; Daily scores on four dependent
measures of depression.

77

None of the phase changes on the Social Interest
factor were statistically significant (Table 4) in the
predicted direction, failing to support Hypotheses 10,
11, or 12.

Subject 4's pre- and posttest MMPI profiles are
shown as Figure 8. There were marked reductions in
scores on Scales 2 and 7. the clinical scales of
interest to this study. Scale 4 remained elevated,
although reduced, indicating social/marital discontent.
Scale 0, Social Introversion, was appreciably reduced.

Following termination of treatment, she reported
a lessening of her depression. She felt happy and more
self-confident, and appeared calmer and more composed.
Her marriage seemed improved, and she wondered to what
extent this was due to changes in her perceptions. She
was not desirous of further personal, marital, or group
counseling, stating that she wanted to try out her newly

learned anti-depression strategies.

Subject 5

 

Subject 5 is a 22-year-old mother of three
children who was referred to the mental health center
by her family physician. She had been depressed for
as long as she could remember, and was hypersensitive,
irritable, and nervous. She felt like crying much of
the time, but she found it difficult for the tears to

flow. She had a poor appetite, and she wanted to sleep

78

The Minnesota Multiphasic Personality Inventory
Starks R. Hathaway and I. Charnley McKinley

Scout's Initials

 

l 3 3 4 5 6 7 I 9 0 ram
TorTc 7 L F K HstSK D Hy Pot“ MI Pa PH‘IK ScflK Hat-IX Si TorTc AddanamlSmIes
lZl-_ - -- _]m . .
us -‘ : ' — “-
:Female - : - .5; ”f “2 5
110-—_ _ 50_ ; - -- 2 _ S—no - .-
- - - w— w: w-
105 - 4o— 50- _ — 105
100- : 45- 3 w- 25- 55_ 554 , :-Im
“ ' ' I T 70... I
95 - 35- ““ 15'- : 3“— 3 1. 95
so. ' 65- E
_ ~90
{- as
l-ao
'- 7s

  

 

 

.4HHHH.HH
8

 

 

 

 

 

 

 

: E as
: wiéfi .
Z i :- 4s
1 : ;_..
3 15‘. :
as -j 0‘ 5’ m_ ' - Io-E :- 35
: 5: [0- 45- IS-_ 0 ID- ‘_' - ‘
w— -———- - - -.— w
j - 1.0- - - -fi— w
. 0.. . 5- - 0.. E
25 - - - 10- ' - - 25
°‘ so- “‘ _' -
zo— - 5 —20
0-“ ‘._o
TorTr: 7 L F K Hst .SK D Hy Pd tAK MI Pa Pt 0 [K Sc ' IK MatZK St TorTc
l 2 3 4 5 6 7 8 9 0

 

= pretest
———————————— = posttest

Figure 8. Subject 4; MMPI pre- and posttreatment pro-
files.

NOTE: Copyright 1948 by The Psychological
Corporation. All rights reserved as stated in the
manual and Catalog. The Psychological Corporation,
New York, N.Y.

79

through much of the day. At night she had nightmares

of "frightening" and "morbid" subjects. The subject
had socially isolated herself, and thought she was
having a "mental breakdown." She had an extremely poor
and inaccurate self-concept, thinking she was stupid,
ugly, and totally unlovable. In addition to her depres-
sive symptomatology, she was troubled by her marriage.
Her major complaints were unsatisfactory communications,
nonorgasmic sex, and her husband's over-drinking.

Subject 5 entered willingly into the depression
treatment study. As may be seen in Figure 9, her scores
on the BDI decreased over time. There was a downward
median trend across all phases of the study. There were
no changes in phases which were statistically significant
(Table 1), so the reduction in depression level may not
be inferred to be caused by the intervention. Hypotheses
l, 2, and 3 are thus not supported in this case.

Subject 5 showed a reduction in her BRS Depres-
sion items scores. During the baseline phase there was
a downward median trend, so the difference between
baseline and treatment is nonsignificant. The median
trend of the reversal phase was in the desired direction,
and the difference between this phase and treatment was
statistically significant (Table 2). The difference

between the median trends of the baseline and reversal

80

 

POINTS
40
3O
BECK
DEPRESSION 20
NVENT NY
10
0
5
4
BRS 3
DEPRESSION
ITEMS 2
I
0
IO
9
8
BRS 7
PSYCHOMOTOR 6
RETARDATION 5
FACTOR 4
3
2
I
O
20
15
BRS
SOCIAL IO
INTEREST
FACTOR 5
O

 

_ . . . q..,...... . ....,. . m g . . - u . A .. DAYS

Figure 9. Subject 5; Daily scores on four dependent
measures of depression.

81

phases was nonsignificant, however. Thus, Hypothesis 5
is supported, but the data fail to support Hypotheses 4
and 6.

On the Psychomotor Retardation factor her median
trend for the treatment phase was lower than for the
baseline, and the difference was statistically signifi-
cant (Table 3). Likewise there was a statistically
significant difference between treatment and reversal
data, and between baseline and reversal conditions.

For the Psychomotor Retardation performance, Hypotheses 7,
8, and 9 were supported.

Because her scores on the BRS Social Interest
factor were increasing during the baseline phase, none
of the differences between median trends are statisti-
cally significant in the desired direction. Thus,
although her scores increased across time, the data
do not support Hypotheses 10, 11, and 12, and the data
cannot be said to be due to the treatment.

Figure 10 is a presentation of pre- and posttest
MMPI profiles for Subject 5. It may be seen that there
were large reductions on Scales 2 and 7. Although the
posttest profile is within normal limits, some depression
and anxiety remain, as do social and marital maladjust-
ments and social introversion.

At the end of the reversal phase, she felt some

lifting of her depression. She was still unsure of

82

The Minnesota Multiphasic Personality Inventory
Starks R. Hathaway and I. Charnley McKinley

Scorer's Initials

I 2 . S 6 7 8 9 0 ForReootdu-tq
TorTc 7 L F K Hs‘.5K D Hy Pd+.4K MI Pa “41;leth SI TorTc AddluanlSculu

I:_EI-"emale

105 —:

 

 

     

100—'

 

 

 

 

 

 

 

 

 

 

 

“
711-;
0- ".0
'I‘orTc " L F K Hs*.SK D Hy PthK Mi Pa PHIK Sc'lKMahZK S: TorTc
I 2 3 4 5 6 7 8 9 0
= pretest
——————————— = posttest

Figure 10. Subject 5; MMPI pre- and posttreatment
profiles.

NOTE: Copyright 1948 by The Psychological
Corporation. All rights reserved as stated in the
manual and Catalog. The Psychological Corporation,
New York, N.Y.

83

herself, and elected to remain in counseling. She has
been receiving individual rational-emotive therapy,

and recently her husband came in for conjoint marital
therapy. The subject is also involved in an assertion-

training group.

Subject 6

 

Subject 6 is a 25-year-old female who is married
and has two children. She was referred by her family
physician for anxiety. She was easily upset, expressed
difficulty in coping with her children, had been crying
easily, and had a very poor self-image. She had a poor
appetite, and had lost 15 pounds during the past year.
She felt as though she should be punished, and had
thought about killing herself. The subject had with-
drawn socially, and her motivation for successful treat-
ment had been questioned by the referring therapist.

Her performance on the four daily measures of
depression is presented as Figure 11. It may be
observed that over time her scores on the BDI decreased
slightly. There was no statistically significant dif-
ference between the median trends of the baseline and
treatment phases (Table 1), and thus Hypothesis 1 is
not supported. There was a decrease in the median trend
of the reversal phase as compared to the treatment phase,

failing to support Hypothesis 2. There was overall a

DEPRESSION
INVENTORY

BRS
DEPRESSION
ITEMS

PSYCHOMOTOR
RETARDATION
FACTOR

BRS
SOCIAL
INTEREST
FACTOR

Figure 11.

84

POINTS
40

3O

20

Nov-DU!

_s

o—tmwhmmxnmsoo

 

20

 

I 3 4 n H In I) u III III :1)‘ 2.‘ TI ‘16 :x In I: II 14 I! In I“ II I’. I~ an _ -l DAYS

Subject 6; Daily scores on four dependent
measures of depression.

85

decrease in the median trend from the baseline to the
treatment phase (p = .006). Thus, from the BDI data,
only Hypothesis 3 is supported.

None of the between—phase differences of median
trends for the BRS Depression items were statistically
significant in the desired direction (Table 2). As may
be observed in Figure 11, there was a decline in the
median trend during the baseline phase, an upturn during
treatment, and a second decline during reversal. All of
these changes were contrary to the hypothesized changes.
Thus, from the BRS Depression item scores, none of
Hypotheses 4, 5, and 6 may be supported.

Subject 6 showed little Psychomotor Retardation
during the baseline phase, and there was no statistically
significant difference (Table 3) between her baseline
and treatment performance. There was a slight drop in
her median trend during the reversal phase, which was
not statistically significantly different from the median
trend of the treatment phase data. Overall there was a
slight reduction in Psychomotor Retardation between the
baseline and reversal phases; and this decrease was sta-
tistically different (p = .029). For this factor the
data fail to support Hypotheses 7 and 8, but support
Hypothesis 9.

The BRS Social Interest factor performance

increased during the treatment phase, and the resulting

86

median trend (Figure 11) was significantly higher sta-
tistically than the trend for the baseline phase (p = .029)
(Table 4). There was no statistically significant
reduction of the data during the reversal phase as
compared to the treatment phase. The data from the
reversal phase remain higher than the median trend of

the baseline phase (p = .029). Subject 6's performance

on the BRS Social Interest factor fails to support
Hypothesis 11, but support Hypotheses 10 and 12.

Her pre- and posttest profiles from the MMPI are
presented as Figure 12. An overall reduction in the
clinical scale scores may be observed, although the
profile configuration remains relatively intact. At
the time of posttesting, peaks on Scales 2, 7, and 0
remain, with a depression on Scale 5. Like Subject 1,
this profile indicates a mildly depressed and anxious
individual who is socially introverted, and who is
passive and dependent on others.

Following the two-week reversal phase she was
interviewed by this investigator. She desired further
counseling, and treatment alternatives were discussed.
For two months, she was involved in individual RET-based
counseling, and presently is attending weekly group
assertion training sessions at the mental health center.
She continues to be easily irritated by her children,

and she is plagued by frequent neck aches. She says

87

The Minnesota Multiphasic Personality Inventory
Starks R. Hathaway and I . Charnley McKinley

Scorer's Initials

 

 

 

 

 

 

 

 

 

 

 

 

I 2 8 4 S 6 7 O 9 0 Form

TorTc 7 L F K Hs‘bK D Hy Pd+.4K Mi Pa Ptth $01K MahZK SI TorTc AddmonaISmIeg
I20—i ‘ 7'20
ns —’ - ' - '- 115

female - : ~ .5; ”i “2 5

I ' ' ' ‘ 33-; 60-_ - 3
105-7 40- - 50- - - _ _ _-IO$
100-;
954
g)-
35.;
w- 130-
75_ IZO-

_ 110-
m m-

- g)-
6.5- w_

w—
55- 5°"
40—

$——w-——
45':
40-
35-
w...
25'.
zo-l _
0- '_o
TorTc 7 L F K Hs'bK D Hy Pd*.4K MI Pa Ptth Sc°IK Ma+.2K SI TorTc

l 2 3 4 5 6 7 8 9 0

 

= pretest
———————————— = posttest

Figure 12. Subject 6; MMPI pre- and posttreatment
profiles.

NOTE: Copyright 1948 by The Psychological
Corporation. All rights reserved as stated in the

manual and Catalog. The Psychological Corporation,
New York, N.Y.

88

she feels better when she remembers "the ABC's of chang-
ing feelings," but she really doesn't work at it very
hard. She remains extremely dependent upon her family
doctor and her mental health therapists to solve her

emotional problems.

Subject 7

 

Subject 7 is a 36-year-old married woman with
three children. She and her husband originally were
referred to the mental health center by a friend because
of marital problems. It soon became evident to the
therapist that some of the marital difficulties were
due to her depression, and she was referred for screen-
ing into this study. She had been depressed for at
least three years. She was very critical of herself in
terms of her physical appearance, her intelligence, her
housekeeping, her maternal abilities, and her adequacy
as a wife. She cried easily, slept much of the day, and
felt nervous and irritable. She had lost interest in
sex, and she had been thinking of killing herself. It
was decided by the couple and by their therapist that
upon her completion of the study, conjoint marital
therapy would be reinstituted.

The subject's performance on the four dependent
variables is shown as Figure 13. During the baseline
phase, her scores on the Beck Depression Inventory were

high, and although there are some data missing, it may

89

 

POINTS
40
30
BECK
DEPRESSION 20
INVENTORY
IO
0
5
4
BRS 3
DEPRESSION
ITEMS 2
I
0
IO
9
a
7
PSYCHOMOTOR 6
RETARDATION 5
FACTOR 4
3
2
I
o
20
BRS ‘5
SOCIAL
INTEREST IO
FACTOR
5
o

 

I _' I I» i m I? ,4 :5 III '30 1.": ‘_‘l L‘II :m In .- )I A .- ' I': II II, I~ DAYS

F“igure 13. Subject 7; Daily scores on four dependent
measures of depression.

90

be noted that the median trend is upward. During the
treatment phase the BDI scores decreased and the median
trend reversed in the desirable (downward) direction.
The between-phase difference was statistically sig-
nificant (p < .001) (Table 1). During the reversal
phase the median trend was upward, and this difference
was statistically significant (also at p < .001). The
comparison between baseline and reversal phase median
trends was statistically significant in the desirable
direction; Subject 7's self-reported level of depression
lowered as time progressed. Thus, the BDI data support
research Hypotheses l, 2, and 3.

As may be seen in Table 2, all changes between
phases for the median trends of the BRS Depression item
scores were statistically significant in the desired
directions. A relatively horizontal median trend during
the baseline phase turned downward during the treatment
phase (p = .001), and leveled off during the reversal
phase, a change which was also statistically significant.
Her BRS Depression item performance is thus supportive
of Hypotheses 4, 5, and 6.

The BRS Psychomotor Retardation factor perfor-
mance yielded an upward median trend during the baseline
phase. This trend dropped and became horizontal during
treatment, a change which was statistically significant

(Table 3); the data thus support Hypothesis 7. The

91

low (almost zero) and horizontal median trend continues
into the reversal phase. The data do not indicate a
change in performance between treatment and reversal
conditions, and thus Hypothesis 8 is not supported.
Comparing baseline and reversal conditions, there was

a statistically significant difference (p < .001) in
the desired direction (Table 3). Thus, her scores on
the BRS Psychomotor Retardation factor decreased as
compared to the baseline period, and Hypothesis 9 is
supported.

Looking at the Social Interest factor (Figure 13),
there was an increasing median trend for the data during
the baseline condition. A similar trend continued into
the treatment phase, and the data during the two phases
are statistically not significantly different (Table 4).
There was a decreasing median trend, as predicted, during
the reversal phase, and the difference between the
treatment and reversal phases is thus statistically
significant (p < .001). Her performance during the
reversal phase was not higher to a statistically sig-
nificant degree than the baseline performance; overall,
her Social Interest cannot be said to have increased.
Thus, for this BRS factor, Subject 7's performance
supports Hypothesis 11, but fails to support Hypotheses

10 or 12.

92

An overall reduction in MMPI scores is indicated
in Figure 14. Her scores on Scales 2 and 7 considerably
decreased during the course of this study. Elevations
remain on Scales 4, 7, and 0, and there remains the
depression on Scale 5. The posttest MMPI profile is
indicative of an individual with mild anxiety and social
introversion and feelings of inadequacy. Marital and/or
social dissatisfactions remain, as does a tendency to be
overly dependent and submissive.

Following her treatment for depression, she
reported feeling much less depressed. Her crying had
ceased, she felt somewhat more confident around others,
she was less lethargic, and she felt more in control in
her home and with her children. Social and marital
concerns still plague her. She and her husband were
involved in conjoint marital therapy with male and
female co-therapists. Therapy emphasized development
of better communications skills, and clarifications
of each partner's values in their marriage. Two months
of such marital counseling were relatively fruitless,
and were frustrating to the couple and to the therapists.
It was perceived that she was dishonest in her marital
communications, and that she really wanted out of the
marriage. The couple terminated from marriage coun-
seling, and soon thereafter Subject 7 separated from

her husband.

93

The Minnesota Multiphasic Personality Inventory
Starks R. Hathaway and I. Charnley McKinley

Scout's Initials—

 

      
  

 

 
 
 

   

 

 

 

 

 

 

 

 

l 2 3 4 5 4 7 I 9 0 Paw
TorTc ? L r K HshSK D Hy PU+AK w Pa PINK sc+IK Mann: St TorTc mm
120-7w ~»_2 - .. ~—--— 2... 2 2 ,1 - - -3—m . —
mi - ' ins
female - : - .5; ”i “2 2 .
110-: ; 50-" ;— 2 .-- —«—-- -— -- -- ~ _ -— 1__._ ~rlIO ——~——
2 - - - - ' eo-_ so. ‘°'. : ‘
105 f 40— ' 50- - - - _ .- IOS .
IOO-_ , _—-—45-.--- : -OO-—— - —-- 25-- 55. -—ss- v -—~- —:-Im L—
- _ - - - 70.. .
95-: as— I ‘5'. - Is- _ I : 35' : .- 96 2
‘ ’ ' ' ' ' ' I i
m... _ MIX-m
as -_ - as
m—: 130—- E—Z-m -
n: m‘ ; ;}n
31w- : :5
.- --— fl
m:_fiF_”° 53- ::70
I m- ‘ I '. I
55‘: ao- .\ 3 : ‘5 :
- \ _ - ‘
50.3- 70‘ -37.“. Z+m .._- 1:---” I
: 60‘ : ' 1
so— - -
s- w- §_ m: s 1
45- 20', Is- , m—E 2-45 1
w; 3 - w}}w .
H: ' ::
35 -_ - IO-: :. 35 I
- W- :: .
22.; : ' = :-a ‘
V * :
m— n n V u }m
0— 2.0
Tor'l'c 7 L F K HshSK D Hy Pd+.4K MI Pa PH IK 5::th Ila-tn St TorTc
1 s s s s s 7 s 9 o

 

= pretest
————————————— = posttest

Figure 14. Subject 7; MMPI pre- and posttreatment
profiles.

NOTE: Copyright 1948 by The Psychological
Corporation. All rights reserved as stated in the

manual and Catalog. The Psychological Corporation,
New York, N.Y.

94

Subject 8

 

Subject 8 is a 39-year-old female who was
referred by her family physician. He described her
history of nervousness, depression, and psychosomatic
problems such as colitis. The physician said that she
was one of his more enjoyable patients--bright, verbal,
and articulate. She had become increasingly nervous,
depressed, and confused. She had been crying easily,
and had begun drinking in the afternoons to "relax."
She was confused as to how to reconcile her sometimes
conflicting roles of wife, mother, and school teacher.
She made herself feel guilty by constantly telling
herself that she did not measure up to "society's"
standards of how a woman "should" act. The subject
had become irritable toward her children and husband,
and her teaching performance had declined. She had con-
sidered killing herself, and she felt as though she was
being punished. She felt fatigued much of the time,
and she had lost interest in sex. A

Subject 8's daily scores on the Beck Depression
Inventory and on the three Behavior Rating Scale
measures are presented as Figure 15. On the BDI there
was a slightly downward median trend during the baseline
phase. During treatment the trend was even more down-
ward, but there is not a statistically significant

difference (Table 1) between the data for the two phases.

95

 

 

 

POINTS
40
30
BECK
DEPRESSION 20
INVENTORY
I.
O
5
4
BRS 3
DEPRESSION
ITEMS 2
I
O
10
9
8
7
PSYCHWOTOR 6
RETARDATION 5
FACTOR 4
3
2
I
0
20
15
BRS
SOCIAL IO
INTEREST
FACTOR 5
0

Figure 15.

 

II 1 1 t, K In I': II It. M '_'II ;_' ‘l :{L m I.» (7 .II 0- ‘I- I- |_‘ n u I- DAYS

Subject 8; Daily scores on four dependent
measures of depression.

96

During the reversal phase a similar downward trend is
apparent, but there is a statistically significant dif-
ference between the data of the treatment and reversal
phases (p < .001). Comparing baseline and reversal
conditions, there is a lowered level of depression
during reversal, and this change is statistically sig-
nificant. Subject 8's performance on the BDI fails

to support Hypothesis 1, but does support Hypotheses 2
and 3.

There was a declining median trend for BRS
Depression scores during the baseline phase. A declin-
ing trend during treatment was not statistically sig-
nificant (Table 2) in the desired direction from the
baseline data. A leveling of the median trend during
the reversal phase is indicative of data which were
statistically significant (p < .001) from the treatment
data. Because of the declining trend during baseline,
the reversal data were not statistically significantly
lower than baseline data. Thus for the Depression item
scores, Hypotheses 4 and 6 were not supported. Only
Hypothesis 5 is supported in this case.

A similar pattern may be observed in Subject 8's
performance on the BRS Psychomotor Retardation factor.
There was a downward median trend during baseline,
followed by another declining trend during treatment.

No statistically significant differences were found

97

(Table 3) between the data for the two phases. Hypothe-
sis 7 is thus not supported. The median trend during
the reversal phase is likewise downward, but is higher
than during treatment. This difference is statistically
significant (p < .001), and Hypothesis 8 is supported.
Again because of the declining median trend during base-
line, the data during the reversal phase are not statis-
tically significantly different in the desired direction,
and Hypothesis 9 is not supported.

The data for the BRS Social Interest factor are
described by a horizontal median trend during the base-
line phase. The data during the treatment phase are
statistically not significantly different (Table 4) from
data predicted by the baseline median trend. There is
a change to horizontal of the median trend during the
reversal phase. This change is in the desired direction,
and is statistically significant. The differences
loetween the baseline and reversal median trends are in
tihe desired (higher) direction, and are statistically
ssignificant. Thus, the Social Interest factor data

fail to support Hypothesis 10, but lend support to
Ifyqpotheses 11 and 12 for this subject.

Subject 8's pre- and posttest MMPI profiles
are included as Figure 16. It may be observed that on
t31€3 scales of interest, Scales 2 and 7, her posttest

Sc=<>res have declined at least one standard deviation.

98

The Minnesota Multiphasic Personality Inventory
Starks R. Hathaway and I. Charnlsy McKinley

Scorsr's Initials___.___._

 

  

I 3 3 4 8 6 7 I 9 0 Farm
TorTc ? I. r K Hs+.SK D IQ Pang MI Pa PINK Sc-HK new: St Tor‘l'c “awed.
me - 4120 .
“5L : ' “ _. 2.
female - : - ”i “2 2
“0-: . _ w; : -... ....... ___.____. _- 2___ - . . E—“o _.....
:. ' ‘ ' ‘ e0-: 60-: 40- :
105- 40- - 50- - _ 3105
100— -45-- 3 40- 25- 55; 355—1 2 - ~—:-lm -
' - ' : ' 70- t
95- 35" 45 IS- : : 35- E f 5
' w: - s3-
._ - _.—:—-w
I 50': :
: :' 85
30-

 
  
    

I
.~st

..............;.
8

i

 

 

 

 

 

 

 

$5
35
25; m? - $5
at w- as Eu
3 - ‘15—: {-40
15‘ ‘ : :
— - I0-: 5-35
- 10' : : ‘
o-
- -fi_--.-w ———
: ‘ = 525
5" ' :
0—“ 2.0
TorTc 7 L F K Hat-SK D Hy PthK Mi PI: PIth SC‘HK Ma+2K SI TorTc
l 8 8 4 5 6 7 s 9 0

 

= pretest
-------------- = posttest

I"'-'i-gure 16. Subject 8; MMPI pre- and posttreatment
profiles.

NOTE: Copyright 1948 by The Psychological
Corporation. All rights reserved as stated in the
mal'lual and Catalog. The Psychological Corporation,

ew York, N.Y.

99

Thus, she is apparently less depressed and anxious.
Also Scale 4 is down, indicating less marital and/or
social concerns. There remains an elevation on Scale 0,
social introversion, and a valley on Scale 5, an indi-
cation of dependency and subservience. There remains
an elevation on Scale 1, a measure of hypochondriachal
and/or psychosomatic tendencies.

Following treatment, Subject 8 was desirous of
further counseling. She states that she is much less
depressed, but that occasionally she has a "bad day."
She panics, and begins to worry if she will become
depressed again. Right now she wants only supportive
counseling. She wishes to avoid discussions about her
interpersonal and psychosomatic difficulties. Her
therapist is observing her wishes, for the time being,
but plans to initiate treatment for these other areas
in the future. At present the subject does not appear
depressed. She claims to be thinking clearer, and to
be less irritable and more confident. She no longer

considers suicide, and her drinking is under control.

Subject 9

 

Subject 9 is a 26-year-old female who was
referred by her family physician. She was described
by the physician as a married woman with two small
children, who since the recent birth of her baby had

become increasingly nervous and depressed. She was

100

sleeping a great deal, had withdrawn socially, and was
extremely irritable toward her children. She stated
that she had been depressed since she got married. She
felt as though she had given so much to others that
there was nothing left of herself. She is a college
graduate and a commercial artist, but marital and
maternal demands left little time for her career.
Subject 9 described a disturbed pattern of sleep,
crying easily, feelings of hopelessness and helpless-
ness, and a loss of interest in sex. She felt insecure
and unworthy of others' love, and she had isolated
herself from friends.

Subject 9's performance on the four daily
measures of depression is presented as Figure 17. On
her scores for the Beck Depression Inventory it may
be seen that there was a reduction in the median trend
during the treatment phase. This change was statisti-
cally significant (p = .019) (Table 1) and in the
desirable direction. There was an uplift of the data
during the reversal phase, also in the predicted
direction at a level of statistical significance.

The reversal phase data were lower than predicted by

the median trend of the baseline data; this difference
also was statistically significant. Thus, for Subject 9,
all changes between phases were statistically significant
in the desired directions. For the BDI data, Hypothe-

ses l, 2, and 3 are all supported.

101

POINTS

 

 

 

DEPRESSION

INVENTORY
5
4

BRS 3

DEPRESSION

ITEMS 2
I
0
IO
9
8

BRS 7

PSYCHOMOTOR 6

RETARDATION 5

FACTOR 4
3
2
I
O

SOCIAL

INTEREST

FACTOR

 

“ 2 I b: I‘ In I‘ II IS IN "I _; '.'I 26 IF .w r; 1| A ‘-~ H‘ I I In I~ .. I DAYS

Figure 17. Subject 9; Daily scores on four dependent
measures of depression.

102

Her performance on the BRS Depression items
yields an upward median trend during the baseline phase.
The median trend during the treatment phase is lower and
horizontal. This reduction is desirable, and there is
a statistically significant difference (Table 2) between
the data for the two phases. Hypothesis 4 is thus sup-
ported. Into the reversal phase, the data lowered
further (to almost zero), and the median trend remained
horizontal. The predicted upturn in the median trend
did not occur, and the data for the two phases are not
significantly different. Hypothesis 5 is not supported
in this case. The performance during the reversal
phase is, however, lower than that predicted by the
median trend of the baseline phase, and the difference
is statistically significant (p < .001). Hypothesis 6
is supported by Subject 9's BRS depression scores.

Looking at the data for the BRS Psychomotor
Retardation factor, there was a decreasing median trend
during the reversal phase, which was not statistically
significantly different from the baseline phase (Table 3).
Compared to the median trend for the treatment phase,
there was an upward trend in the data during the
reversal phase; this difference was statistically
significant. Finally, the data for the reversal phase
were lower than data predicted by the median trend of

the baseline phase. Thus, across time there was a

103

decrease in the scores on the Psychomotor Retardation
factor, and the data support Hypotheses 8 and 9, but
not Hypothesis 7.

The BRS Social Interest factor had an increasing
median trend during the baseline phase. There was an
increasing median trend during the treatment, but dif-
ferences between baseline and treatment phases were
not statistically significant (Table 4). During the
reversal phase, the data leveled off, and there was a
statistically significant (p < .001) difference between
treatment and reversal performance. Overall there was
an increase in Social Interest but the scores during
the reversal condition are not statistically signifi-
cantly higher than those predicted by the median trend
of the baseline phase. The data fail to support
Hypotheses 10 and 12, but do support Hypothesis 11.

Subject 9's MMPI profiles are presented as
Figure 18. Appreciable reductions in her scores on
Scales 2 and 7 may be observed. The posttest profile
is well within normal limits. The remaining peaks
on scales 3 and 6 indicate an individual who has a
tendency to react in an extreme and irrational fashion
to stressful situations, and who is somewhat overly
sensitive to views and perceived criticism from others.
She has a normal degree of depression and anxiety, and
she is far more pleased with her current level of social

functioning.

104

The Minnesota Multiphasic Personality Inventory
Starks R. Hathaway and I . Charnlsy McKinley

Scorer‘s Initials

 

    

 

 

 

 

 

 

 

 

 

 

I 2 3 4 I 6 7 I 9 9 Form
‘I‘orTc 7 L F K Hs+.SK D Hy Pd+.14KL MI Pa PI+IK Set-1K Mat-x St TorTc AddthmdSoaIee
I207. .- -. ., rm .

.: : T - L I. “5 '
female - : - .5; ”i “2 i i
110': 1 ’ ~ — ._ ...... w— __ .-z .... -...-. ..__._. ,__..__ ..._._.__ .... .. 1...-.. - ..._-. ..-..- . rllo _—.——

Z 3 - 50': 50-. “0-. “

105 - 40- 50- _ ¢ ,- 105

100- Z 45- 40- — — 25- 55; ~55:- --~ -- mE-Im - 2
- - - : - ,0- :

95- 35- “T 15- ' 3 35‘ 3 _- 95

- ' - - 50- 65-3 E
90-5 3 -—- ~ -:—-- 40- ~'—-»-: - --7 - “O_ - ~ --—-—--;-_-9

‘ - ' 40-. 35‘. : ' : _ : ...:g
85 - 30- 20— - :- 85

50— _ I - ”T 45- '5- 30- 55; ;
-- ... ,5; : .. r ~~ : - «3%:
75 - lm- T I - m- T T T w“: - 50-5 T. 75
- 25- 25' - ' 25- 4°: - - : : .
IIO- - - - - - 3 45": 3 ,
n - --
70 : W _ - : 25 — . 70 w

3 90- 10- - -_ ‘
“T; 80— - 20- 20- _ ‘5
...; "t : --:-eo — ~

w.—
55 — 50" ' 5— 15— 15- 30.: - 55 g
‘0' 5 ‘ ‘ : - I
w-—w—— — — - - — _._zs-:-:-w *I
_ ..... T - T “b‘fl E E
45 - 0_ 1° _ m_ _ - - _ - _ Is- \20: :45
_ 15- 15- - 40- - - \ - -
40— - - — ‘5': ~ 2 5T 20: 3 - - Is‘g--40
' - ' - ' 15': ' ' :
35 - 0’ 5T- : 10-3 I _ - 3 g - IO-‘ - 35

__ — - ---—- w
”- - - jo— - - - 2—5': 23°

T 0- 5- - OTT T . Z I
255 - _ IO— _ - _-

°‘ so— - 5" 5; s
20—' . ~- -- - ~— .-zO
0— “..o I
TorTc 7 L F K HshSK D Hy Pat-AK MI Pa PI+ 1K Sc‘th Ma+2K Si TorTc
l 2 s s s s 1 s 9 o
11 = pretest

_____________ = posttest

Figure 18. Subject 9; MMPI pre- and posttreatment
profiles.

NOTE: Copyright 1948 by The Psychological
Corporation. All rights reserved as stated in the

manual and Catalog. The Psychological Corporation,
New York, N.Y.

105

Following the posttesting, the MMPI results
were interpreted to her, and she was asked about her
current emotional status. Subject 9 reported that she
felt happier than she had in several years, and that
she felt more in control of her emotions. She did not
let "little things" bother her so much, and she felt
more adequate and satisfied as a wife and mother. She
had been able to do more of her domestic chores, and
because of this she was able to complete more commercial
art free-lance jobs, much to her delight. She declined
further counseling at the mental health center at this
time, stating that she wanted "to get used to feeling
good for a change." Her physician later contacted the
center, noting her progress emotionally and in her

child-rearing practices.

Subject 10

 

Subject 10 is a 23-year-old woman who came to
the mental health center for counseling following her
divorce. She was having an extremely difficult time
getting along without her ex-husband, and was unsure
whether she could ever find another man if she let go
of him. She felt extremely inadequate and unattractive,
and she made few attempts to make new friends, either
male or female. Subject 10 felt nervous and depressed,
and she cried over "just about anything." Her sleep

pattern was disturbed and restless, and her appetite

106

was poor. She had withdrawn socially, she lacked moti-
vation to do her housework or to work at her job as a
nurse's aide, and she felt guilty and disinterested
about sex.

Her performance on the four daily measures of
depression is presented as Figure 19. It may be
observed that the data cease about halfway through the
treatment phase. For various reasons she and her two
raters stopped recording data at approximately the same
times. Subject 10 apparently gave up on herself and the
depression treatment program near the time of the third
experimental session. Her therapist noted that she felt
very dejected that session, and that helping her to
generate five positive self-evaluatory statements was
"like pulling teeth." Each time a statement was decided
upon, she found several ways to negate the positive
attribute. She had practiced her homework for the
session only a few times, and in general seemed
unmotivated and uncooperative with the therapist.

This investigator repeatedly attempted to contact her.
She did not have a telephone, so messages were left
with a neighbor and with her mother to call and/or to
bring in her data-reporting cards to the center.
Subject 10 ignored these requests, stating that she

kept "forgetting" because she was "too busy."

BECK
DEPRESSION
INVENTORY

BRS
DEPRESSION
ITEMS

BRS
PSYCHOMOTOR
RETARDATION
FACTOR

BRS
SOCIAL
INTEREST
FACTOR

Figure 19.

107

POINTS
40

30

 

NU-hUI

 

—l

O—‘Nw-IQU'IO‘INQKDC

 

20

 

(I 3 I I . I. . III I. _‘I 3 I _. » | . 3: I‘ II I DAYS

Subject 10; Daily scores on four dependent
measures of depression.

108

One of her raters was a girlfriend. Sometime
early during the treatment phase, the friend was not
as supportive and sympathetic as the subject thought
she should be. Subject lO stopped associating with
the girlfriend and the girl notified this investigator
that she would not be filling out the daily Behavior
Rating Scale until their conflicts were resolved.
Apparently this resolution did not take place.

Her other rater was a co-worker at the nursing
home where the subject worked. The rater was assigned
to a different shift, also during the treatment phase.
The co-worker sent a letter to this investigator,
apologizing for not being able to rate Subject 10's
behavior any longer because of their infrequent meetings,
and offering to help in any additional fashion appro-
priate to the treatment. This offer was acknowledged,
but unfortunately the daily ratings relevant to this
study were terminated.

Thus, due to difficulties in obtaining daily
ratings on Subject 10's depressive feelings and
behaviors, data are complete only throughout the first
half of the treatment phase. Statistical analyses of
the data could not be performed, and this is indicated
where appropriate in Tables 1-4.

She did complete the posttesting following the

reversal phase, and her pre- and posttest MMPI profiles

109

are included as Figure 20. It may be seen that her
posttest scores on Scales 2 and 7 decreased, but mild
amounts of depression and anxiety remain. The lower
scores on Scales 4 and 0 indicate that she is more
satisfied with her social relationships and functioning.
The peaks remaining on Scales 6 and 8 are disturbing to
this investigator; in fact, the score on Scale 6 increased
as compared to the pretest score, and Scale 8 is essen-
tially unchanged. These elevated scales indicate an
oversensitivity to the opinions of others, if not out-
right suspiciousness and distrust. Her social with-
drawal and oversensitivity may be due to so-called
schizoid behavioral tendencies, if not underlying
psychotic or pre-psychotic emotional conflicts (Marks,
Seeman, & Haller, 1974).

Following her completion of the depression pro-
gram, Subject 10 was desirous of further counseling.
Due to apparent remaining emotional and interpersonal
problems, this investigator referred her to a therapist
at the mental health center who can be understanding
and supportive, yet who is putting upon her the
responsibility for solving her own problems. Subject 10
frequently does not care for this approach, but is

regularly attending her weekly counseling sessions.

110

The Minnosda Multiphasic Personality Inventory
immmoRkammmyathChunhyhkkmkw

Scout'- Initials

 

  
 

 

 

 

 

 

 

 

l 8 8 4 I 6 7 O 9 0 ram
TorTc 1' L F K HstSK D j} Pd+.4K Ml Pa HHK 8::th Mao-.2: St Tor'fc www.-
lm— "’ - -- ‘ -.. _., - - ‘

; ‘ _ '- ‘r- - _.= I!) r A - x
”5 .I : ' - L :. '

female - : . ..L ”f “2 i
110-? -- , _- 50-1 —; - ~ j-H —— ~—- —---- ---— :- ---—— 7110 -»---

1 ' ' ' ‘ 50-. 60': ‘0’. : 1
105 -: 40— ' 50- . - _ _ _- 105 1
mé , m, 3~ —w:- - s. f. '. mw—~§m n .-.

: - - 70'- :

95 -_ : I. 95

E 55.: Z
90- _ i _.._ 3.2.9

3 m; I

E 55.: I
ao-_ m—-~ ———-—3-;-ao —

E _. so;

75 t m I '- 7s

3 no— '
m-TIO Ec—7O -—-fi——

- - I

3 m. . -. - '
“f m- 3'“ f
(50-: 70- _ \ 35-— ' 2

. —w—-w e ,-

- w— \ - ,

50- - 1
55 - a: " 55 i
40— \- .
45 - zo-Z :- 45 I
0-1 - - 1591-40 ‘

_ ' I 1
x1 wééu '
ao ’ : E 3

- fi€.:-m nun—Tn—
25 .: Z :- x

3 !

20- —— 3—20 6

o_' Lo ,
TotTc ? L F K HMSK D Hy Pd+.4K MI Pa PINK Sc+lK Ma'hZK Si Tet‘l'c

I 8 8 4 3 6 7 I 9 O

 

= pretest
————————————— = posttest

Figure 20. Subject lO; MMPI pre- and posttreatment
profiles.

NOTE: Copyright 1948 by The Psychological
Corporation. All rights reserved as stated in the
manual and Catalog. The Psychological Corporation,
New York, N.Y.

111

Subject 11

Subject 11 is a 27-year-old divorced woman who
has a daughter. She was referred to the mental health
center by her fiance, who thought she "needed help."
She reported that she had been depressed for one and
one-half months since her boyfriend left her. She had
isolated herself socially, and she was frightened of
what she would do without him. She had difficulty
sleeping, and was crying almost all the time. She
found it difficult to take care of her housework, and
she was irritable toward her daughter. She had con-
sidered suicide, and threatened her fiance with slitting
her wrists if he didn't come back to her.

Her performance on the four daily measures of
depressive functioning are presented as Figure 21. It
may be seen that there was a declining median trend
on the Beck Depression Inventory even during the base-
line phase. A similar trend is apparent during the
treatment phase, and the data for the two phases are
statistically not significantly different (Table 1).
There was a leveling off of the data in the reversal
phase as predicted, but the difference between the
treatment and reversal phases is not statistically
significant at the a = .05 level (p = .145). Because
of the downward median trend of the baseline data,

there was not a statistically significant difference

1112

 

POINTS
40
30
BECK
DEPRESSION 20
INVENTORY
10
0
5
4
BRS 3
DEPRESSION
Uflfi 2
1
O
10
9
8
BRS 7
PSYCHOMOTOR 6
RETARDATION 5
FACTOR 4
3
2
1
O
20
15
BRS
SOCIAL IO
INTEREST
FACTOR 5
0

 

(b J I .. ‘14:“!!th :(n 'I - w u .5 :4 ~|u 1;” , DAYS

Figure 21. Subject 11; Daily scores on four dependent
measures of depression.

113

between the reversal data, and data predicted by the
baseline median trend. Thus, for this case, the BDI
data support none of Hypotheses l, 2, and 3.

Subject 11's performance on the BRS Depression
items during baseline also showed a downward median
trend. Consequently, the data during the treatment
phase were not lower than data predicted by the base-
line median trend, and statistical significance was not
achieved (Table 2). A bottoming-out of the median trend
in the reversal phase produced by the data was not
higher than that predicted by the median trend of the
treatment phase. The difference between the two phases
was not statistically significant in the desired
direction. Finally, her performance during the
reversal phase was not significantly lower statisti-
cally than that predicted by the median trend of the
baseline phase. The data do not support any of the
predictions regarding the BRS Depression items, that
is, Hypotheses 4, 5, or 6.

During the baseline period, there was likewise
a declining median trend of the BRS Psychomotor Retard-
ation scores. The median trend declined even more
during treatment, but the difference between the data
for the two phases was not statistically significant
(Table 3). The data, then, do not support Hypothesis 7.

During the reversal phase, there again is a bottoming-out

114

of the median trend, but these data are higher than the
data predicted by the median trend of the treatment
phase and this difference is statistically significant
(p = .004). Hypothesis 8 is thus supported. Although
there was a decrease in her scores on this factor as

the study progressed, the data during the reversal phase
are not lower than those data predicted by the median
trend of the baseline phase, and there is no statisti-
cally significant difference in the desired direction.
Hypothesis 9 is not supported.

A downward median trend during the baseline
phase of the BRS Social Interest factor predicted data
during the treatment phase which were not significantly
different statistically (Table 4) than actually observed.
Hypothesis 10 is not supported by this subject's data.
Likewise, the median trend of the treatment phase data
predicts data during the reversal phase which are lower
than the observed data. A statistically significant
difference in the desired direction was not obtained,
and Hypothesis 11 is not supported. Comparing baseline
and reversal phases, there was an increase in the scores
on Social Interest. This difference is statistically
significant, and the data support Hypothesis 12.

Subject ll's pre- and posttest MMPI profiles
are included as Figure 22. Considerable reductions in

her scores on Scales 2 and 7 may be observed. In fact,

115

The Minnesota Multiphasic Personality Inventory
Starke R. Hathaway and I. Charnley McKinley

Scout's Initials

 

 

 

 

 

 

 

 

 

 

 

 

I 2 3 ‘ 5 6 7 I 9 0 For Romtdmg
TorTc 7 L F K Hs'.5K D Hy Pd+.4K Ml Pu PHlK Sc+lK MahZK St TotTc Addmonalsmles
me - - :-|20
15 -‘ — 3- 115
female - ,5- ”i “2 i
“0—: 50'- : _ ' - - 40- E-IIO
: ‘ ’ ' ' 50-: 50-: - :
105 -—_ 40- - 50- _ - _ _ :- IOS
lm-_ _ 45": : ‘0- 25- 557 55:- - :-loo
- 70-
95 - 35— ‘Sf 15- : 3 35" 3 - 95
- ' 50- 6.5- -
90- ' i ' ‘ 50-: : - : 79°
40‘. 35.. : - ' : l 50‘: i
30' - m- 29- _ . m_ - 1' 85
c - - - - ss-
( - 35: -w
‘ 0- 50
m- - - - : - 75
25- _ 25- ‘0‘: — _ :
g— - - t -
— w ‘ -— — ou-m —
. - l 2 35.. 25 - : -
- - c I-
35- - ' ‘
20 ’ zs- ' ‘ - ‘ 3 55
- ‘ w- ' 35.: I
25:. 25- - , : - 4 : 32'. - - :- E50
'30- - |- - '_
- - lo .. .. -
IS- _ 3 20- _ Ex 25- - 3°: ; 55
- ... p - I ‘ o : :
PTATfk “-25-—;.‘Tv {3"750
\v I - - :
- - ' \ ' ' -
z)-
' v 15- 20- - 45
10- ~ \ - - 2
- 15- '5‘ {-_ ‘0 ‘ ‘ : -
«3.. 15 - '- 5 20 _ 15_ 7‘0
- 15': :
35 - 0‘ 5' 10— 3 3 - no— '- as
5" IO- ‘5_ 15f lo.- 10- I
— -—~-— - - U. —_
30 10- - . - -fij. w
0- 5; - o— . _ Z
25 — ~ - _ _ IO- _ - — 75
0- _ 50- : S- -
zo— - 5 -20
0- 2.0
TorTc " L F K Ha ' .SK D Hy Pd ! AK Ml Pa PH 1K Sc ' IK Ma 0.2K SI TorTc
I 3 8 4 5 6 7 I 9 0

 

= pretest
.............. = posttest

Figure 22. Subject ll; MMPI pre- and posttreatment
profiles.

NOTE: Copyright 1948 by The Psychological
Corporation. All rights reserved as stated in the

manual and Catalog. The Psychological Corporation,
New York, N.Y.

116

the clinical scale configuration has dropped to well
within normal limits. This was one of the more dra-
matic reductions in MMPI clinical scale scores of any
of this study's subjects, and it may be observed that
Subject ll's resulting posttest profile is similar in
appearance to Subject 9's (Figure 18).

Following the completion of the reversal phase,
Subject 11 felt no longer in need of counseling. She
was not depressed, and she was becoming more and more
socially active. She had demanded that her fiance come
to a decision regarding their relationship, and he had
rejoined her. She reported feeling more in control of
her situation and her emotions, and this pleased her a

great deal.

Overall Results

 

Hypothesis 1

 

Hypothesis 1 predicted a reduction in subjects'
level of depression, as measured daily by the Beck
Depression Inventory, following the treatment inter-
vention. It may be observed in Table 1 that this
hypothesis was supported by the data from 4 of the 9

subjects.

Hypothesis 2

 

It was hypothesized that the level of depression

as measured by the BDI would increase during the reversal

117

phase. As indicated in Table 1, data from 4 of 9
subjects support this hypothesis. For these subjects,
Hypothesis 2 is supported; there was an increase in
depression level during the reversal phase. This result
is particularly crucial to the study. When considered
along with Hypothesis 1, Hypothesis 2 lends support to
the notion that the treatment, and not extraneous
variables affected subjects' level of depression.

When the treatment was instituted, BDI scores decreased;

with treatment withdrawn (reversal) BDI scores increased.

Hypothesis 3

 

Although Hypothesis 2 predicted that BDI scores
would rise during the reversal phase, it was predicted
that due to generalization of learning which took p1ace
during treatment, BDI scores would be lower during
reversal than during the baseline phase. Again referring
to Table 1, it may be seen that this hypothesis was
supported by 6 of 9 subjects' data at a = .05.
Hypothesis 3 is thus supported by the data of six
subjects. Overall there was a reduction in level of
depression, and this reduction in BDI scores was resis-
tant to extinction, at least during the course of this

study.

118

Hypothesis 4

 

There was hypothesized a reduction in the
Behavior Rating Scale Depression item scores immediately
following the point of treatment intervention. It may
be observed in Table 2 that 3 of 10 subjects yielded
data which support the prediction at a level of statis-
tical significance. Thus for 3 of 10 subjects Hypothe-
sis 4 is supported; there was a reduction in the

Depression scores during treatment.

Hypothesis 5

 

It was predicted that there would be an increase
in the BRS Depression scores following the termination
of treatment, i.e., during the reversal phase. This
hypothesis was supported at the a = .05 level by the
data of 4 of 9 subjects (Table 2). As with Hypothesis 2,
Hypothesis 5 lends support to the notion that the treat-
ment itself affected the data, in this case behaviors

assumed to be related to depression.

Hypothesis 6

 

It was hypothesized that during the reversal
phase the daily BRS Depression item scores would remain
lower than during baseline. This hypothesis was sup-
ported by data from 3 out of 9 subjects at a = .05.

As measured by the BRS Depression items, learning which
took place during the treatment phase continued during

reversal.

119

Hypothesis 7

 

A reduction in BRS Psychomotor Retardation
scores was predicted following the treatment inter-
vention. The data from 4 out of 10 subjects (Table 3)
supported this hypothesis at the a = .05 level. For 4
subjects in this study, Hypothesis 7 was supported.
There was a decline in scores on the Psychomotor

Retardation factor during treatment.

Hypothesis 8

 

It was predicted that there would be an increase
in daily BRS Psychomotor Retardation scores during the

reversal phase. This hypothesis received support from

the data of 5 out of 9 subjects at a .05. Thus,
Hypothesis 8 is supported by the data of 5 subjects;
there was an increase in Psychomotor Retardation follow-
ing the termination of treatment. Hypotheses 7 and 8,
then, support a treatment, which while in effect, sig-

nificantly reduced subjects' scores on the BRS Psycho-

motor Retardation factor.

Hypothesis 9

 

It was predicted that the Psychomotor Retard-
ation scores would remain lower during the reversal
period than during baseline. Data from 5 out of 9

subjects supported this hypothesis at a = .05.

120

Hypothesis 9 was supported; the reduction in Psycho-
motor Retardation was resistant to extinction, con-

tinuing through the reversal phase.

Hypothesis 10

 

It was hypothesized that there would be an
increase in Social Interest as measured by the Behavior
Rating Scale immediately following the treatment inter-
vention. This hypothesis was supported by 2 out of 10
subjects at a = .05 (Table 4). There was an increase
in Social Interest scores during the treatment phase,
but this increase is not necessarily due to the treat-
ment. There were in the cases of seven subjects upward
median trends on the Social Interest factor during the
baseline phase. Such increasing median trends during
baseline would not have been significantly different
statistically from desirable data during the treatment

phase.

Hypothesis ll

 

Social Interest scores were expected to decline
during the reversal phase, as compared to data during
the treatment phase. To a statistically significant
degree (a = .05) this was the case for 4 out of 9 sub-
jects in this study. Thus, these data support Hypothe-
sis 11; there was a decline in Social Interest following

treatment. Even though there was a reversal effect,

121

the lack of support for Hypothesis 10 indicates that
variables other than the treatment may have been

Operating to increase performance on Social Interest.

Hypothesis 12

 

The Social Interest factor was predicted to be
higher during the reversal phase than during baseline
due to generalization of learning and resistance to
extinction. This hypothesis was supported by the data
of 4 out of 9 subjects (Table 4) at a = .05. Hypothe-
sis 12 is thus supported by the data of 4 subjects;
there was a general increase in scores on the BRS
Social Interest factor which persisted through the
reversal phase.

It has been observed that the 12 research
hypotheses are supported by the data of varying pro-
portions of subjects. The proportions range from 2 of
10 subjects' data supporting Hypothesis 10 to 6 of 9
subjects' data supporting Hypothesis 3. A discussion

of these results follows in the next chapter.

Supplementarnyindings

 

MMPI

As a supplementary procedure, all clients were
given the MMPI as part of their evaluation for possible
inclusion into this study. Upon completion of the

reversal phase of the study, subjects were requested

122

to complete a second MMPI. All subjects except Subject 1
completed a posttest, and the pre- and posttest MMPI
profiles have been included with the discussion of
each subject's performance during the study.

As a means of assessing overall changes in the
MMPI profiles, group mean pretest profiles and posttest
profiles have been computed for all subjects except
Subject 1. These profiles are presented as Figure 23.
The group mean pretest profile is the 2-7 type with
additional peaks on Scales 4 and 0. This profile is,
of course, representative of depressed, anxious indi-
viduals who are socially introverted and who have
social and/or marital dissatisfactions. This profile
is quite similar to the 2-7 code type profiles dis-
cussed by Marks et a1. (1974).

The group mean posttest profile may be seen to
have decreased considerably from the pretest profile.
The profile is completely within normal limits, and
reductions of at least two standard deviations each
may be seen for Scales 2 and 7. Scales 4 and 0 are
likewise lower (at least by one standard deviation
each), indicating decreased social dissatisfaction
and introversion. The posttreatment profile in
Figure 23 compares favorably with the mean discharge
profile for hospitalized 2-7 patients given in Marks

et a1. (1974). Thus, the decrease in MMPI clinical

123

The Minnesota Multiphasic Personality Inventory
Starks R. Hathaway and I. Charnlsy McKinley

Scorer‘s Initials

 

 

 

   

 

   

 

 

 

 

I 2 3 4 S 6 7 8 9 0 For Recording
TorTc " L F K H3'.5K D _Hy PdiAK M! Pa PHIK Sc'lK MotZK St TorTc Addtmlfxllg
‘33 "f - - --m
115 -‘ - ' - '- 115
female _ : . ,5; 3°? “2 E
110—:- _ 50_ ; _ - r _ 0 Z-no
I ' ‘ ' ‘ 50-: 601 I _ :
105 f 40— ' w’ _ - - - .- 105
:00; {—Im
95 -j 3 :- 95
90- ‘-:-so
85 7 3- 85
80-- I30- - :--m
15‘ z
75 ‘ 120— - ’ 75
HO- .
70 100— Z 2 7O
' 9°- ': :
65 : w_ t: :— 55
eo-j 7o— , 32: +60
6.0- ’ - 1
55 - 50" ao-I :- 55
40- I -
$0-—3G -—25.:'.-50 _—e_—
45 -_ 211-: .- 45
40- 15-° -—4o
35 -_ 10—: :- 35
.._- =— “w
30 T‘: : 3°
25 -: 1 f- 25
wé }m
0- 1.0
TorTc 7 L F K Hs t .SK D Hy Pd 0.4K Ml Pa Pt ‘ 1K Sc ’ IK MCI" 2K 5: TorTc

 

 

 

 

I 2 3 4 5 6 7 8 9 0

 

= pretest
-------------- = posttest

Figure 23. Group mean pre- and posttreatment MMPI pro-
files over all subjects (N = 10). (Time span
approximately six weeks.)

NOTE: C0pyright 1948 by The Psychological Cor-
poration. All rights reserved as stated in the manual
and Catalog. The Psychological Corporation, New York,
N.Y.

124

scale scores of subjects in this study is consistent
with previous studies of the treatments of depressed
individuals.

The pre- and posttest MMPI scores on each
clinical scale were compared using matched-pairs t-test
statistics. The results of the statistical tests are
given as Table 5. Probabilities for all significant
tests are given in Table 5. Because each probability
is two-tailed, reductions in MMPI scores may be analyzed
at'% = .025. Furthermore, since must be partitioned
due to repeated significance testing, each t-test must
be decided upon at (2727167'= .0025. It may be seen
that five of the 10 MMPI scales were reduced, and that
these decreases were significant at the (7TETTOY = .0025
level. The scales so significantly reduced include 2

and 7, MMPI scales of interest.

Interrater Reliabilities

 

Interrater reliabilities were calculated for
subjects' daily observations of themselves and by others
on their performance on items of the Behavior Rating
Scale. Reliability coefficients were computed for each
subject on each of the dependent variables obtained
from the BRS--Depression, Psychomotor Retardation, and
Social Interest.

For subjects whose data were obtained from

observations from three raters (self— and two

125

Table 5

Results of Matched-Pairs T-tests Between Pre- and
Posttreatment Means (N = 10) on MMPI
Clinical Scales. (Time span
approximately six weeks.)

 

Scale t-value d.f. p (two-tailed)
l-Hs 3.05 9 .014
2-9 5.61 9 1 .001a
3-Hy_ 4.58 9 '<_‘.001a
4-gg_ 5 15 9 1 .001a
s-gg €0.13 9 .900
6-23 2.44 9 .037
7—g5 8.37 9 1 .001a
8-Sc 5.88 9 _<_..001a
9-53 1.32 9 .218
o—g; 4.06 9 .003

 

aIndicates statistical significance at 2(10 = .0025.

126

other-raters) per day, Hoyt's analysis of variance
method (1941) was used to compute the interrater
reliabilities. For subjects whose data were typically
obtained from observation by two raters (self— and only
one other-rater) per day, a Pearson product-moment cor-
relation coefficient (Pearson, 1907) was computed.
Although this investigator repeatedly stressed to sub-
jects the desirability of having their behavior rated
daily by two others, in some cases this was not feasible.
There were three subjects whose daily scores were con-
sistently obtained from less than two raters. These
three subjects include individuals whose BRS data were
incomplete for the total time of the study; reliability
computations were not attempted for such data.

It should be noted that the Hoyt formula is
essentially a derivation of the Pearson 5, and inter-
rater reliabilities obtained by the two methods may
be considered equivalent statistics. Reliability coef-
ficients obtained from eight subjects for each of the
three BRS variables are presented as Table 6. All
computations were performed by computer with a widely
used program, the SPSS (Vogelback Computing Center,
1975).

It may be seen that there is much diversity
among the individual correlation coefficients in

Table 6. To summarize across subjects for each

127

Table 6

Interrater Reliability Coefficients for Behavior
Rating Scale Variables

 

 

Sweet (233535228 Depression 3323323352: Ifiiiiiit
2 Hoyt-ANOVA .43121 .30399 .73848
4 Hoyt-ANOVA Incomplete Incomplete .81133

Data Data

5 Hoyt-ANOVA .25649 -.02182 -.06886
6 Hoyt-ANOVA .28018 -.01389 -.85783
8 Hoyt-ANOVA .46302 .42000 .74383
3 Pearson-p .2910 .2219 .6405
7 Pearson-; -.0579 -.0616 .0921
9 Pearson-5 .6420 .3141 .6923

Across Zr trans- .344 .163 .399

Subjects formation

 

128

variable, individual correlation coefficients were
transformed into Zr scores (Fisher, 1921). The Zr
scores were summed and meaned, and transformed back
into mean pévalues for each variable. i—values are
presented in Table 6.

The resulting interrater reliability values for
the three BRS variables deserve further amplification,

and this will be discussed in the next chapter.

CHAPTER IV

DISCUSSION

Summary

The purpose of this study was to evaluate the
effects of cognitive-behavior modification upon depressed
individuals. These individuals were clients of an out-
patient community mental health center. The effects of
a cognitive-behavioral intervention were assessed using
a client self-rated scale of depressive thoughts and
feelings, and by a self- and other—rated scale of
depressive behavior.

Four graduate students in counseling with train-
ing and experience in behavior modification were trained
in the theory and practice of the cognitive modification
of depression. Specific training procedures for these
therapists are described. Three therapists were each
assigned three subjects, and the fourth therapist was
assigned two subjects. The 11 subjects were selected
from 39 clients of a community mental health center
referred for possible inclusion in this study by mental

health staff persons.

129

130

Subjects were accepted for the study based upon
specified screening criteria including clinical signs,
observed client behavior, client history, presenting
problem(s), MMPI profile, Shipley—Hartford scores, and
score on the Beck Depression Inventory.

Following each subject's acceptance into this
study, she appeared for an interview with this investi-
gator. The subject was furnished a copy of the Beck
Depression Inventory and pre-dated c0pies of daily
response cards, and she was requested to complete the
BDI questionnaire each day throughout the study.

The subject was requested to bring two signifi-
cant others to the interview with this investigator.
The significant others were to serve as raters for her
daily behavior. The subject and the raters were fur-
nished c0pies of the depression Behavior Rating Scale,
an abbreviation of the NOSIE—30, to complete each day
on a pre-dated response card.

Subjects and their raters completed the daily
response cards through a baseline phase of 7, 10, or
14 days, a two-week treatment phase, and a two-week
reversal phase. Data are plotted for each subject for
each experimental day and across four dependent variables.
The Beck Depression Inventory was one of the daily
repeated measures. Items from the depression Behavior

Rating Scale yielded the three remaining daily repeated

131

measures--Depression items, Psychomotor Retardation
factor, and Social Interest factor.

The treatment intervention consisted of four
sessions over a two-week period. During the first
session the therapist established rapport and structured
the treatment phase. The subject was furnished a
pamphlet, "Coping with Depression," and the booklet
was discussed. The therapist suggested that the sub-
ject read the pamphlet once each day until the second
treatment session, and whenever else she experienced
feelings of depression.

During the second treatment session, the therapist
assisted the subject in identifying at least five depres-
sive cognitions and their corresponding challenges. She
was instructed to read the list of depressing cognitions
and challenges before engaging in high probability
behaviors until the third session.

During the third session, the therapist assisted
the subject in generating a list of at least five posi-
tive self—evaluatory statements. She was to read the
list before engaging in a high probability behavior
before the last treatment session.

In the fourth session, the therapist reviewed
the cognitive treatment of depression, reinforced learn-
ing, and scheduled the subject for a followup appointment

with this investigator two weeks later.

132

At that time the subject's current level of
psychological functioning was evaluated clinically,
and she was administered a second MMPI. Any concerns
she had regarding the treatment were discussed, and if
the subject desired, further mental health services at
the center were offered.

A multiple-baseline design was employed in the
study. Self-reported level of depressive symptomatology
(BDI) and self- and other-rated depressive behaviors
(BRS) were charted daily as repeated measures. The
treatment effects were contrasted across four dependent
variables and across individuals at the same relative
points in the treatment schedule.

Based upon two assumptions made by Beck (l974a)
concerning the effects of the cognitive modification
of depression, 12 research hypotheses were generated for
this study. For each of the four dependent measures of
depression, it was hypothesized that during the treat-
ment phase the subject's rated depression would decline.
The depression rating on each measure would rise again
during the reversal phase, a second hypothesis.

Finally it was hypothesized that depression scores
during the reversal phase would not rise to the level
of the baseline phase.

The data for each subject are presented in an

intensive case study format. Her performance on each

133

dependent variable is presented graphically, and sig-
nificance tests have been performed for all 12 research
hypotheses on the data from each subject. Finally,
supplementary data, such as pre-posttreatment changes
in her MMPI profile, are presented along with a dis-
cussion of the outcome of treatment and the subject's
current therapeutic status.

The research hypotheses are supported by varying
'proportions of subject data. Specifically, it was found
that scores on the Beck Depression Inventory, the
Behavior Rating Scale Depression items, and the Behavior
Rating Scale Psychomotor Retardation factor decreased
during the treatment phase, and rebounded after the
termination of treatment, but remained lower during
the reversal phase than before the treatment inter-
vention. Scores for the Behavior Rating Scale Social
Interest factor for two subjects support the hypothesis
that subjects' performance during the treatment phase
was different than during baseline, but the other two
research hypotheses for Social Interest were supported
by data from four subjects.

Supplementary data presented for all subjects
include mean MMPI profiles for pre- and posttreatment
testing situations. The changes in the MMPI profile
over the time of the experiment are discussed, and

results of statistical tests of the changes in profile

134

are presented. It was found that the posttreatment
MMPI mean profile was well within normal limits, and
that reductions on five of the 10 clinical scales for
the group were statistically significant.

It was concluded that this cognitive behavioral
treatment was to varying degrees effective in reducing
depressive symptomatology and behavior among clients
of an outpatient clinic. The research findings are
discussed below, as are implications of the present

study for practice and further research.

Discussion

 

Overall Results

 

The data from the four daily repeated measures
of depression support the research hypotheses to vary-
ing degrees. Rated covert and overt indices of depres-
sion decreased during the treatment phase as predicted,
increased during the reversal phase as hypothesized,
but remained lower during the reversal phase than
during baseline, also as predicted.

Several conclusions may be drawn from these
findings. There was a treatment effect upon three
rated measures of depression. When the treatment was
instituted, there was an overall decline in depression
across all subjects on each of the variables BDI, BRS
Depression, and BRS Psychomotor Retardation. Further,

when the treatment was withdrawn, measures of depression

135

increased again. Thus, there was an effect upon
depression which cannot be explained by variables other
than treatment. As Thoresen (1972) has noted, the
intensive case-study design is a time-series design
described by Campbell and Stanley (1966). Such designs
control for all internal sources of invalidity which
might influence experimental data. Such extraneous
variables include history, maturation, testing, changes
in instrumentation, and regression effects. The only
variable left to explain a decline of depression during
treatment and an increase during reversal is the effect
of the treatment itself.

In the study of depression, control of extraneous
variables such as history and maturation is particularly
important because of possible cyclical changes in
emotions (Beck, l974a). Another possible confounding
variable in depressed individuals is the effect(s) of
psychotropic medications including anti-depressants.

At least five subjects (Subjects 4, 5, 6, 8, and 9)

had recently been prescribed tri-cyclic anti-depressants
by their physicians and were taking the medications as
prescribed to a varying degree. It had been a concern
that such medications might influence the outcome of
this study. If anything, psychiatric medications
decreased observed differences between subjects' per-

formance during baseline and treatment phases. As may

136

be seen in Tables 1-4, on any one dependent measure
only one of Subjects 4, 5, 6, 8, and 9 performed sig-
nificantly different (a = .05) during the treatment
condition than they had during baseline. These five
subjects who may have been taking anti-depressant medi-
cations did not contribute more than their share of
changes in performance in the hypothesized direction(s).
Thus a subject's medication status did not appear to
influence the results of this study in a positive
fashion; in fact, an individual's being on medication
to some degree may have camouflaged the effects of
cognitive-behavior modification. That is, there may
have been reductions in measured depression during the
baseline phase for a subject, due to medication, and
this reduction would have canceled out or decreased
further reductions in depression due to the treatment
intervention. Baseline-treatment differences would
have been lessened, and perhaps may have gone undetected.
Most subjects additionally self-medicated to varying
degrees with substances such as caffeine, nicotine,

and ethanol. To totally guard against the effects of
the so-called depression cycle and the effects of psy-
choactive drugs would have been virtually impossible,
and it was assumed that the intensive case-study design
would control for such confounding of the data. Evi-

dently the design did not control for such confounding

137

effects, but as had been mentioned, treatment effects
were not disproportionately inflated by possible
medication effects.

The treatment was shown to affect subjects'
performance on a variety of measures. That is, measures
consisted of self-reports of covert phenomena such as
thinking and feeling, and overt behaviors. Several
manifestations and symptoms of depression were influenced
by the cognitive-behavioral treatment used in this study.

Such conclusions are in agreement with, and lend
support to the basic premises underlying Beck's treat-
ment for depression. By using cognitive and/or
behavioral interventions, individuals' cognitive
distortions may be alleviated, and rapid decreases
in depressive symptomatology may be realized. Further,
such decreases in depression are interrelated, and an
improvement in depressive emotions would also be mani-
fested in behavioral improvements.

The foregoing discussion and conclusions hold
true for Hypotheses 1-9, and the three dependent vari-
ables BDI, BRS Depression, and BRS Psychomotor Retard-
ation. In contrast, data from observations of BRS
Social Interest support an overall reduction in depres-
sion (more accurately an overall increase in social
behaviors incompatible with feelings of depression)

(Hypothesis 12), but the change began during the

138

baseline phase, and cannot be said to have been due
gply to the effects of treatment. Behaviors tapped by
the Social Interest factor are associated with inter-
personal interactions. Possibly certain aspects of this
study which occurred before the treatment intervention
increased social interaction and therefore subjects'
scores on this BRS factor. Subjects were involved in
placebo counseling sessions during the time their
baseline data were being collected. Additionally,

they were being rated by significant others, and there
were probably social performance expectations perceived
by the subjects. Finally, just being rated (attended
to) by others was probably socially reinforcing for
subjects, and would be predicted to have increased
social interaction. It was hoped that the baseline
condition would control for such nonspecific-to-
treatment effects, and that a stable baseline on BRS
Social Interest would have been obtained, but this did
not occur during this study.

This investigator is uncertain what, if anything,
could have been done to stabilize the baseline on Social
Interest, and therefore control for nonspecific effects
upon the factor. Increasing the length of baseline
does not appear to have controlled for nonspecific
social effects. Only two subjects had median trends

during baseline which were significantly lower

139

statistically than treatment performance; these subjects
were Subject 3 and Subject 6 (Table 4). As may be seen
from the graphs of their data (Figures 5 and 11, respec-
tively) Subject 3 was part of the lO-day baseline group,
and Subject 6 was part of the two-week baseline group.
The remaining subjects were evenly divided into 7, 10,
or 14 day baseline lengths. Apparently, increasing
the length of the baseline phase did not contribute
to stabilizing the baseline, and thus to contrasting
baseline to treatment performance.

In retrospect such reactive effects of subjects
coming to a mental health center for counseling, even
of a nonspecific variety, and of having their performance
observed by others would be predicted to increase social
behaviors. Such increases would normally be desirable
for depressed individuals. It is unfortunate that the
effects of the cognitive-behavioral treatment of depres-
sion upon Social Interest are confounded with whatever
social effects were occurring during baseline.

Some support for the effects of this treatment
upon Social Interest may be drawn from the decrease in
scores on this factor during reversal (Table 4).

Hypothesis 11 was supported, and evidently something

 

was affecting social behaviors because there was a

decrease when treatment was terminated.

140

Subjects' increase in social functioning during
baseline, due to the aforementioned reasons, may have
been related to observed decreases on the three other
measures of depression during the baseline phase for
several subjects. The exact causal and correlational
relationships between the dependent variables of this
study are unknown, and deserve further investigation.

The effects of sex upon subjects' performance
on the dependent measures warrants discussion. It is
obvious that all subjects in this study were female.
Males were not specifically excluded during the study,
but there were no male clients who fit the selection
criteria specified for inclusion into the depression
treatment program. To some extent the generalizability
to males of results of this study is limited. There
may be systematic differences between the ways the
depressed females of this study responded to cognitive-
behavior modification and the ways depressed males
might respond. This investigator suSpects that there
would not be significant differences due to the sex of
subjects, but this is conjecture. If males could have
been found whose psychological characteristics had met
the selection criteria for this study, it is assumed
that they would have responded to the treatment similar

to the female subjects.

141

During one of the analyses of variance attempted
on the data from this study, sex of the therapist was
used as a covariate in the ANOVA. Although there were
not enough data to compute treatment effects, it was
shown that the sex of the therapist was not found to
have an effect upon the data. Thus, subjects seen by
the two male therapists responded to the treatment in
ways not unlike subjects seen by the two female therapists.

The effects of other possible covariates were
not computed, again due to incomplete data, and the
resulting small d.f. Such possible covariates include
I.Q. score, length of baseline, marital status, attained
level of education, and pretest scores on MMPI clinical
scales. All of these factors may have influenced subject
data, but to what extent cannot be ascertained from the
data for this study. Further investigations might shed
light upon what effects such covariates have upon the
cognitive-behavioral treatment of depression.

Concerning the effects of the length of the
baseline upon subjects' performance, over all subjects
the length of their baseline phase does not appear to
have affected their performance differentially.
Inspection of the subjects' data for the baseline
condition on each of the four daily measures yielded
no systematic differences between subjects in the 7-,

10-, or l4-day baseline groups. Descriptions of the

142

median trends for all subjects' baseline performance
are summarized as Table 7. As was mentioned, due to
insufficnent data, analysis of variance with length of
baseline as a covariate could not be run. Thus, the
data do not suggest systematic differences in subjects'
performance due to the length of the baseline condition
or due to any cyclical changes in depression which may

occur independent of treatment.

194121
Although the MMPI was not a measure of primary
research interest, it was useful as a screening instru-
ment. Decisions could be made with the MMPI providing
data in addition to clinical perceptions and client
self-report. The MMPI was an economical measure in
that it required little time of the investigator to
obtain each subject's profile. Administration time
was low, generally less than five minutes to direct
the subject in responding to the items and to answer
preliminary questions. Scoring each MMPI protocol
was performed by clerical personnel, taking about 10
minutes per protocol. The investigator could interpret
the resulting MMPI profile in a matter of a few seconds
and decide whether the subject's profile fit the study
criteria. The cost of a test administration was low,
approximately five cents per score sheet, and the MMPI

booklet (Form R) was reusable. Thus, in terms of

143

 

msfimmouomo

 

acwmmmuomo asfimmmnooo msflmmouooo HH

mawmmmuocw mcfimmmnocfl mawmmmuocfl maflmmmuomp oa

maflmwonocw Hansenwuon msammmuocw HousoNflHon m >monva

amusoNHHon HmuaoNHHon mawmmwuomo Hmuaonuog m

HmuaoNHHon acammmnooo mcfimmonowo acammouowp m

mafimmmuoaw maflmmmnoaw aspaONAHon mafimmmuosfl n

ocwmmmuomo mcflmmmnosfl mcflmmwnocw mawmmmnocfi m amouoa

maflmmwnoafl mcflmmwuomo mcflmmmuowp maflmmmuosw N

mcflmmmnosfi amuCONwHon mcflmmmuooo mcflmmonooo m

mcfimmwuocw mawmmmuooo mswmmmuomp mawmmmnomp w moors

mafimmmuocw Hmuq0uflnon mswmmmnoop amuGONauon H

.mwmwwm. ”wwwwwmwmwm ...mmm.... mmmwmwwmw. 18.11.. .mmwwmm.
mmm mmm xomm .

 

h OHQMB

scammonmma mo mmHDmmoz
ucoocmmmo Hsom so wosmeuomnom onwammmm mo mucosa swaps: mo msowumfinomwa

144

investigator time and expense, the use of the MMPI was
a real bargain. It does consume the subject's time to
take the MMPI, usually one to two hours, and this is a
consideration. With the exception of one, all sub-
jects completed two MMPI protocols during this study,
and none reported strong aversion to doing so. Some
subjects even reported that they enjoy taking tests,
and asked this investigator if they could complete
other tests.

In addition to the great economy of the MMPI,
and its utility as a screening device, the instrument
was sensitive in detecting pre-posttreatment differences
for individual subjects and for group mean performances.
As indicated previously, there were statistically sig-
nificant reductions in subjects' scores on Scales 2 and
7, the scales of most interest. Although there were
no changes in MMPI scores hypothesized, the reduced
profiles over the experimental period add support to
the notion that something desirable took place during
this time period with these depressed subjects. The
MMPI results do not indicate what specifically happened
to lower the posttest scores, but in addition to the
daily repeated measures indicate an overall reduction

in self-reported depression.

145

Interrater Reliabilities

 

The interrater reliabilities reported for the
BRS variables vary greatly between clients, and overall
are not as high as reported by other investigators and
not as high as might be desirable in this study. As
reported in Table 6, the mean interrater reliabilities
for the BRS variables are Depression, .344; Psychomotor
Retardation, .163; and Social Interest, .399. The
Depression items have to my knowledge never before had
interrater reliabilities computed for them. However
Lentz et al. (1971) reported interrater reliability
coefficients for Psychomotor Retardation of .83 and
for Social Interest of .95. Looking on the negative
side, there was considerably less agreement between
raters during the present than in previous studies.

This is probably due to small number of subjects, rela-
tive lack of sophistication of raters, perceptual biases
of clients rating their own performance, and of raters
rating the performance of friends or loved ones. Addi-
tionally, ratings took place over a variety of different
locations and situations.

On the positive side, ratings and raters could
not be highly controlled in this "real-world" study.
Raters were psychologically naive, relatively untrained,
unpaid, and sometimes emotionally involved with clients.

Ratings were made for a client by one rater in one

[Ill III.|.|I|I.II I: I II Iii.)

146

situation, and by a different rater in perhaps an
entirely different situation. Thus, high interrater
agreement could not be expected. It could be argued
that given the constraints of this study, the inter-
rater reliabilities obtained are respectable.

It may also be noted that the interrater relia-
bility coefficient obtained in this study for Depression,
.344, compares favorably to the reliability obtained for
Social Interest, .399. Because in previous studies
respectable interrater reliability was obtained for
Social Interest, perhaps in a more controlled rating
environment a more desirable coefficient would also be
obtained for the Depression variable.

Although the interrater reliabilities obtained
for the BRS measures are undesirably low, the relia-
bilities of the measures are not necessarily low. The
interrater reliabilities are obtained comparing ratings
across raters across situations. That there is not high
interrater agreement does not mean that the measures are
unreliable; highly reliable measures could yield poor
interrater-intersituation reliabilities. Honigfeld
(1974) does not specify test-retest stability coef-
ficients for his NOSIE-30 factors, but low reliabilities
for the three behavioral measures of this study may not

be inferred from the interrater reliabilities obtained.

ll‘ll.| II II I I I I i I I ‘ItllllIl.

147

Likewise, the validities of the measures are
not necessarily weakened by the low interrater relia-
bility findings. The measures could be extremely yalid
for purposes of rating depressive behaviors, and yet
across situations different scores may be obtained.

As a way of guarding against spurious results
obtained from data with low interrater agreement,
individuals' repeated behavioral measures represent
mean scores of ratings each day on BRS Depression items,
Psychomotor Retardation factor, and Social Interest
factor. Even though daily ratings on each of the
measures may differ for whatever reasons, the mean
daily score should be the best estimate of the true
daily score.

A final support that the mean reliabilities for
the BRS variables are within acceptable limits is the
overall results obtained from their use in this study.
Statistically significant changes in the predicted
directions were obtained for the research hypotheses
involving the three BRS variables. It could be argued
that unless the measures possessed adequate reliability,
changes between phases would not have been detected.
Data would have been randomly distributed across all
phases, and there would have been no statistically

significant differences.

148

Although the mean interrater reliability
coefficients for the BRS variables Depression, Psycho-
motor Retardation, and Social Interest are disappoint-
ingly low, there are coefficients for some subjects on
some measures which are promising. Perhaps further
-investigations under more controlled situations will
obtain more consistent and higher interrater reliability
values. Perhaps also, other investigators will develop
behavioral measures better suited for repeated self-
and other-ratings of depression in the natural environ-

ment.

Implications

 

The results of the present study suggest impli-
cations for the theory and treatment of depression and
for further refinements in research. It has been shown
that the data support 11 of 12 research hypotheses at
the a = .05 level. In turn these research hypotheses
lend support to Beck's assumptions regarding the treat-
ment of depression. Such support may be viewed as
verification which is additional to, and independent
of, Beck's own research. External verification of
assumptions is crucial to theory-building, and the
data of this study bolster and affirm Beck's assumptions
regarding a cognitive theory and treatment of depression.

Beck's two important assumptions regarding treatment are

149

supported: that a cognitive treatment will decrease
depressive emotions; and that behavioral correlates
of depressive emotions will likewise be reduced.

To a certain degree, the results of this study
have implications for Seligman's theory of learned help-
lessness. To the extent that learned helplessness in
humans is a cognitive phenomenon, and to the extent that
learned helplessness is equivalent to depression, a
cognitive treatment may be effective in ameliorating
learned helplessness in humans. That is, by directly
attacking the cognitions of learned helplessness, the
condition should be lessened.

Implications for the treatment of depression
are many. The present study was performed primarily
in an outpatient community mental health center. Sub-
jects were agency clients who met certain selection cri-
teria. They were a population of individuals diverse
in terms of education, marital status, family income,
age, and motivation level. Findings from this study
certainly may be generalized to similar depressed indi-
viduals who are clients of community agencies and clinics
elsewhere.

The procedures employed in this study may be
easily adapted for use in other settings by other
clinicians. Testing materials are inexpensive to

obtain, and time-economical to use. Training of

150

therapists was short-term, and did not require highly
experienced clinicians. Materials used in training
therapists are quite inexpensive and easily obtained.

The treatment intervention is straightforward
and well-received by most depressed clients. The
cognitive-behavior modification of depression "makes
sense" to clients, and many of them wonder why they have
never received this kind of help before. Upon reading
the pamphlet "Coping with Depression" (Beck & Greenberg,
1974) some clients exclaimed that the book described
their problems with amazing accuracy.

The rating procedures are applicable to field
settings. It is often not feasible or desirable to
follow clients around their natural environment observ-
ing behavior. For all the faults of the procedure,
enlisting significant others in rating client behaviors
was useful in assessing treatment effectiveness. As
the saying goes, "Some data are better than none at all!"
The rating materials were prepared inexpensively, and
data may be collected with a minimal expenditure of a
clinician's time. Clerical, volunteer or student per-
sonnel could be enlisted to do much of the data collection
and charting.

The N = l statistical approach is conceptually
straightforward, and the clinician can look at the data

and receive immediate feedback regarding client progress.

151

Although in this study the treatment package was
specified and constant, in an agency setting the treat-
ment may be modified according to client performance
indicated by the graphs of repeated measures.

The statistics used in this study are readily
applicable to small samples for which more traditional
statistical analyses may not be appropriate or possible.
Likewise, control groups of similar subjects are unneces-
sary, because in the intensive case design, each subject
becomes his own control. In addition to the practical
advantages of such a design, the ethical issues of which
clients get the real treatment and which are assigned
to a control condition are resolved; no client is assigned
to a control group. All clients are continually moni-
tored during baseline, and they may be placed in treat-
ment if necessary. By appropriately designing the
research phases, between-phase comparisons may be made
and inferences may be drawn as to the efficacy of treat-
ment.

As much as possible, this study was designed to
meet the demands and constraints placed upon clinicians
currently working in applied settings. Using therapists
who received only two sessions of training in the cog-
nitive treatment of depression, a specified treatment
was delivered. The treatment was administered during

four sessions over a two-week period to clients of an

152

outpatient clinic. Client ratings were obtained on
several measures of depression which are readily
available, relatively economical in cost, and efficient
of time to use. Given all these considerations, the
treatment was still shown to be effective in the allevi-
ation of depression in outpatient females. This
cognitive-behavior modification of depression was
designed to have applicability to the practices and
needs of mental health workers, and apparently that

goal was attained.

Although the clients who received treatment in
this study were adult female outpatients, there are no
apparent reasons why cognitive-behavior modification
would not be helpful to depressed clients in other set-
tings. For example, the treatment package is easily
applicable to a public school setting. In fact, two of
the therapists in this study report success in using the
treatment procedures with adolescent clients in junior
high school counseling settings. I have perceived that
depressed medical inpatients respond to cognitive-
behavior modifications, and in particular that depres-
sion which is secondary to medical conditions or physical
disabilities may be decreased.

A great number of medical outpatients are
depressed, and probably no group of human service

professionals sees more depressed individuals than do

153

physicians. A busy physician could be taught the
principles and practice of cognitive-behavior modifi-
cation of depression in a relatively short time, perhaps
during a one-day workshop. The physician could there-
after use cognitive-behavioral strategies with depressed
patients seen in his office. A good number of patients
would probably respond favorably, and psychiatric
referrals and/or the use of psychotropic agents could

be reserved for the more severe and intractable cases.

Cognitive-behavior modification is applicable
to institutionalized clients also. For instance, psy-
chiatric inpatients who are depressed could respond
favorably to this treatment so long as they are verbal,
at least normally intelligent, lucid, in touch with
reality, and motivated to help themselves. Of course,
similar constraints apply to most "talk therapies."
Clients in corrections institutions with these positive
attributes might also respond to cognitive-behavior
therapy.

Certainly research is in order which tests the
applicability of cognitive-behavior modification to
other individuals in a variety of settings. Cognitive-
behavior modification is certainly not a panacea. It
is but another treatment modality which shows promise
of success with some types of clients with some types of
problems. As with any promising technique, further

investigation and refinement are warranted.

154

Although this study has not been emphasized as
a study of behavioral self-management strategies, there
are implications which may be drawn. Subjects were
assisted in generating lists of rational (i.e., non-
depressing) cognitions in one session and lists of
positive self-evaluatory statements in the next, and in
making high probability behaviors contingent upon the
reading of the lists. Essentially, subjects were placed
in charge of their own treatment. As in the self-
management literature cited in Chapter I, this behavioral
technique was apparently successful in decreasing depres-
sive cognitions and the consequential emotions and
behaviors. Several clients reported using the behavioral
self-management techniques from time to time after the
termination of treatment, whenever they "had a bad day."
The two self-management techniques used, self-reinforce-
ment of reading anti-depression statements and self-
reinforcement of reading positive self-evaluatory state-
ments, were combined in this treatment package. Further
investigations might compare the two techniques to find
differential effects of the two strategies upon depression.

In this study cognitive-behavioral self-management
techniques were effective in decreasing depression. Two
other treatments described in Chapter I have shown much
success with depression; these techniques are the cog-

nitive treatment of Aaron Beck and the behavior

155

modification of Peter Lewinsohn. The cognitive-behavior
therapy employed in this study owes heavily to those
cognitive and behavioral treatments of depression, and
may be considered to be a logical hybrid of the two
different, but successful approaches. Additionally,

I have received materials, assistance, and encouragement
from Beck (1975) and from Lewinsohn (1975). An obvious
research study would compare the efficacies of the three
successful treatments of depression--cognitive,
behavioral, and cognitive-behavioral. From such
research, perhaps the most powerful components, or
combination of components, of the treatments would be
evident.

Regardless of which treatment, or treatments,
emerge as most successful in treating of depression,
mental health professionals may at last have treatment
strategies at their disposal which will assist clients
in decreasing depression and even preventing its recur-
rence. Such a promise is exciting to those who work
each day with depressed clients, as well as to theoreti-

cians and researchers.

APPENDICES

[[111

APPENDIX A

BDI QUESTIONNAIRE

NAME

APPENDIX A

BDI QUESTIONNAIRE

DATE

 

 

In each group of statements, please circle the letter
in front of the statement that best describes how you
feel right now.

 

If two or more statements are true, circle the latter
statement in the group.

Be sure to read all the statements in the group before
making your answer.

1. a
b
c

N
(D 00'!” 0900' 9’ (DD--

0000‘”

 

I do not feel sad.

I feel blue or sad.

I am blue or sad all the time and I can't snap
out of it.

I am so sad or unhappy that it is quite painful.
I am so sad or unhappy that I can't stand it.

I am not particularly pessimistic or discouraged
about the future.

I feel discouraged about the future.

I feel I have nothing to look forward to.

I feel that I won't ever get over my troubles.

I feel that the future is hopeless and that things
cannot improve.

I do not feel like a failure.

I feel I have failed more than the average person.
I feel I have accomplished very little that is
worthwhile or that means anything.

As I look back on my life all I can see is a lot
of failure.

I feel I am a complete failure as a person
(parent, husband, wife).

I am not particularly dissatisfied.

I feel bored most of the time.

I don't enjoy things the way I used to.

I don't get satisfaction out of anything anymore.
I am dissatisfied with everything.

156

10.

11.

12.

(TO! (I)

(DD-100‘” (DQJO

HHD CLO 6'91 910 0‘9!

040 0‘9! CLOUD!

0'9!

157

I don't feel particularly guilty.

I feel bad or unworthy a good part of the time.
I feel quite guilty.

I feel bad or unworthy practically all the time
now.

I feel as though I am very bad or worthless.

I don't feel I am being punished.

I have a feeling that something bad may happen
to me.

I feel I am being punished or will be punished.
I feel I deserve to be punished.

I want to be punished.

I don't feel disappointed in myself.
I am disappointed in myself.

I don't like myself.

I am disgusted with myself.

I hate myself.

I don't feel I am any worse than anybody else.
I am critical of myself for my weakness or
mistakes.

I blame myself for my faults.

I blame myself for everything bad that happens.

I don't have any thoughts of harming myself.

I have thoughts of harming myself but I would not
carry them out.

I feel I would be better off dead.

I feel my family would be better off if I were
dead.

I have definite plans about committing suicide.

I would kill myself if I could.

I don't cry any more than usual.

I cry more now than I used to.

I cry all the time now. I can't stop it.

I used to be able to cry but now I can't cry at
all even though I want to.

I am no more irritated now than I ever am.

I get annoyed or irritated more easily than I
used to.

I feel irritated all the time.

I don't get irritated at all the things that used
to irritate me.

I have not lost interest in other people.
I am less interested in other people now than
I used to be.

13.

14.

15.

16.

17.

18.

19.

20.

QJOO‘W CLOUD! 0400'!” O: 00"” (DD-:0 0‘91 OU‘W CLOUD! Q:

m

158

I have lost most of my interest in other people
and have little feeling for them.

I have lost all my interest in other people and
don't care about them at all.

make decisions about as well as ever.

try to put off making decisions.

have great difficulty in making decisions.
can't make any decisions at all anymore.

HHHH

don't feel I look any worse than I used to.

am worried that I am looking old or unattractive.
feel that there are permanent changes in my
appearance and they make me look unattractive.

I feel that I am ugly or repulsive looking.

HHH

I can work about as well as before.

It takes extra effort to get started at doing
something.

I don't work as well as I used to.

I have to push myself very hard to do anything.
I can't do any work at all.

can sleep as well as usual.

wake up more tired in the morning than I used to.
wake up 1-2 hours earlier than usual and find

't hard to get back to sleep.

wake up early every day and can't get more than
hours sleep.

I-‘HHH

UlH

don't get any more tired than usual.
get tired more easily than I used to.
get tired from doing anything.

get too tired to do anything.

HHHH

My appetite is no worse than usual.

My appetite is not as good as it used to be.
My appetite is much worse now.

I have no appetite at all anymore.

haven't lost much weight, if any, lately.
have lost more than 5 pounds.

have lost more than 10 pounds.

have lost more than 15 pounds.

HHHH

I am no more concerned about my health than
usual.

I am concerned about aches and pains or upset
stomach or constipation.

21.

159

I am so concerned with how I feel or what I
feel that it's hard to think of much else.
I am completely absorbed in what I feel.

I have not noticed any recent changes in my
interest in sex.

I am less interested in sex than I used to be.
I am much less interested in sex now.

I have lost interest in sex completely.

THANK YOU FOR YOUR COOPERATION!

APPENDIX B

BEHAVIOR RATING SCALE

APPENDIX B

BEHAVIOR RATING SCALE

CLIENT'S NAME y DATE
RATER'S NAME RELATIONSHIP TO CLIENT

 

DIRECTIONS

Please count this person's behavior as you observed it today
only. Indicate your rating by filling one block for each item:

(0
0.) CD
.5 8
u w-t Q)
U H
H 0) 0) O
o a o m u E
8 ° *4 5 8 ..
z 8 :3 s. s. o
0__ l____ 2___ 3__ 1+__ l. Cries.
O__.l_”,2__ ‘__ ‘__ 2. Enters into activities around him
which involve other people.
0__ l__'2_” 3__ 4__ 3. Sits. unless directed into activity.
0____ l___ 2_ 3__ 1&_ 1+. Tries to be friendly with others.
0___ l___ 2___ 3___ li___ 5. Refuses to speak.
0__ l__ 2__ 3_" 4__’ 6. Laughs or smiles at funny comments or
events.
0____ 1___ 2___ 3___ 4___ 7. Starts a conversation with others.
O___l___2__ 3__.h__ . Says he feels blue or depressed.
0___ 1___ 2__ 3___ 1+___ 9. Talks about his interests.
0__ l__.2___3_",h__. 10. Sleeps. unless directed into activity.
0__ l___21__3__,h__i ll. Says he is no good.
0 l 2 3 4 12. Is slow-moving or sluggish.

Thank you for your assistance.

160

APPENDIX C

DAILY RATING CARDS

APPENDIX C

DAILY RATING CARDS

_ _

_ _
4 4 4 4 4 4
_ _ _ _ n n
3 3 3 3 3 3
_ _ _ n _ _
2 2 2 2 2 2
_ _ _ _ _ _

_ _

1 1 1 I 1 I
u _ _ _ _ _
_ _ _ _
0 0 o o 0 0

m4<0m GZF<¢ m0.><:mm

10.

11.

 

CLIENT RELATIONSHIP DATE

161

162

BDI QUESTIONAI RE

INITIAL'S

DATE

 

Please fill in blank: with letter:

2___
3___

8___

9.—
10.—
11—
12___
13—
14—
15___
16.——

17___
18—

19
20
21

APPENDIX D

COGNITIVE TARGETS

.soammfifiumm >n pmusvoummm .m .m .vnma
.owce>H>mccom mo auwmuo>wca newsmaeocawnm .umfluomscsfi vmnmfiansacs .coflmmmuaoo
mo usefiueouu Heuow>mnonuo>wuacmoo you Assume uccfiucmuu one .h .nmsx a ..H .4 .xocm scum
.cowmwoop mo mooswsacmcoo mufisaouu

Immusomo on unawanoma :w« uonzsu .ucuuon Homu Haw: uommuoafia .ucaau
.hauucse o>aueauouacn .m we so>c cowmaoeo Sac onwxmz .m we axes H soamwoco 02 .m mmcao>fimfiompaH .m

mpasozm uo >sceu>a|

 

 

nuanced
.oou ha csam> maom mousmsoxn
.measoau. Aeneas- .maaanu coon ocaueemuounou amass» ounmucaom
scausnauuuemxon .v .uaom no one: 00» madam: .q can Ham nomseo on uo>owammu .e Emacsuauonwaow .v
.mowpoom bosons mcwuuwm uo:
.uowua>fluos sabeaOnsc one name so one: 00» ou.so>u
waunmfiam uo musmficawmm< Hau0a>enoml «muonuo sens Haw
.maou unwed mama sands» seen one no scanso
msaudomuomao ma mom on anemone max: .m ocwnuos >0nco >Haeeu 90% on: .m .mcanuaco wanna u.:mo H .m nfiwaueuo no xooq .m
osaecnoos .ua unn3.u a
. . .mmoao on
: ousm om 30m. uncommon soaumosoI 0m deem >23 00m m.uoq .AmGOmomu .mmwuasowmmao mo
.mCOHuoaomum o>wummmz u0\hoeaaem .cwumwfiaumoa >Hamowumwamouss Hcpwowsm cu unawefime 30w> ooueuommsxm
mo coausuumcoemo HeowufimEMI .N doom oaaooa commonest can: .~ unmwm use Rhos Hafiz asasuoz .~ nanosecoaoaom .~
msoauot m>wueaumvam .n
mamanoum
muco3
3oa> ou mamx m>auscuoua< .0
seasons 0>Huocuouaau so; use: use assuage sass- .uces H has: uwo u.:eo
u .uw onwoo ouowon “mmmapsm on Haws uHumuonuo
scan 0» scenes“: muwswusuom muss aaaeou we can m.u0H cu coousm “mamomcuoancuwmwe
cocoas>finfid mmoaxmu .H uoe manwmuo>ouua ma cowowsm .H 009 «mcw>aa ou usaoa oz .a magmas Hepaoasm .H
cmmacnoma causwmouons camcoflumm owusmmaumna opsumuu< Ecummxm

 

 

m>fiumspmamz mo mcomemm uncanny o>auasaoo .m

memwm<fi m>HBHZOOU

Q NHOzmmm<

1653

REFERENCES

REFERENCES

American Psychiatric Association. Diagnostic and statis-

 

tical manual‘of mental disordersTTan ed.).
("us -If‘). Washington, D.C.: Author, 1968.

 

American Psychiatric Association. Supplement:

Depression. American Journal of Psychiatry,
1968, 124. -

American Psychiatric Association. Report of task force

Beck,

Beck,

Beck,

Beck,

Beck,

Beck,

Beck,

 

on behavior therapy. Washington, D.C.: Author,
1973. ‘—T

 

A. T. Thinking and depression. Archives of
General Psychiatry, 1963, 9, 324-333.

 

 

A. T. Depression: Clinical, experimental, and
theoretical aspects. New York: Hoeber-Harper,
1967.

 

A. T. Cognitive therapy: Nature and relation to
behavior therapy. Behavior Therapy, 1970, l,

 

A. T. .Depression: Causes and treatmepp. Phila-
delphia: University of Pennsylvania Press, 1972.

 

A. T. Depressive neurosis. In S. Arieti & E. B.
Brody (Eds.), American handbook of_psychiatry,
second edition, volume 33 Adult clinibalgpsy-
chiatry. New York: Basic Books, 1974, 61-90. (a)

 

A. T. Cognitive modification in depressed, suicidal
patients. Paper presented at the Society for
Psychotherapy Research, Denver, 1974. (b)

’A. T. The development of depression: A cognitive

model. In R. J. Friedman & M. M. Katz (Eds.),
Theypsychology of depression: Contemporary
theory and_research. Washington, D.C.: V. H.
Winston & Sons, 1974, 3-20. (c)

 

 

164

165

Beck, A. T. Recent developments in cognitive therapy.
Paper presented at the Eighth Annual Convention
of the Association for the Advancement of
Behavior Therapy, Chicago, 1974. (d)

Beck, A. T. Critique of Verne Schmickley's dissertation
proposal: The effects of cognitive-behavior
modification upon depressed outpatients.
Personal communication. Philadelphia: Uni-
versity of Pennsylvania, 1975.

Beck, A. T., & Beamesderfer, A. Assessment of depression:
The depression inventory. In P. Pichot (Ed.),
Psychological measurements in psychopharmacolo .
Modernyproblems in_pharmacopsychiatry, Volume 7¥'
New York: S. Karger, 1974, 151-169.

 

 

 

 

Beck, A. T., & Greenberg, R. L. Copipg with depression.
Philadelphia: Aaron T. Beck, M.D., I974.

Beck, A. T., & Rush, J. The treatment manual for cogni-
tive-behavioral therapy of depression. Unpub-
lished manuscript. Philadelphia: University
of Pennsylvania, 1974.

Beck, A. T., & Stein, D. The self-concept in depression.
Unpublished study cited in A. T. Beck. Depres-
sion: Clinical, experimental, and theoretica
aspects. New York: Hoeber-Harper, 1967.

Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., &
Erbaugh, J. An inventory for measuring
depression. Archives of General Psychiatry,
1961, 4, 53-63.

 

Beck, A. T., Weissman, A., Lester, D., & Trexler, L.
The measurement of pessimism: The hopelessness
scale. Journal of Consulting and Clinical Psy-
chology, 1974, 42, 861-865.

Becker, J. Depression: Theory_and research. Washington,
D.C.: V. H. Winston & Sons, 1974.

Bergin, A. E. The evaluation of therapeutic outcome. In
A. E. Bergin & S. L. Garfield (Eds.), Handbook
of_psychotherapy and behavior change: An experi-

mental—analysis. New York: rain Wiley & Sons,
I - 7 e

 

166

Bibring, E. The mechanism of depression. In P. Greenacre
(Ed.), Affective disorders. New York: Inter-
nationaICUniversities Press, 1953, 13-48.

 

Bosse, J. J., Croghan, L. M., Greenstein, M. B., Katz,
N. W., Oliver, J. M., Powell, D. A., & Smith,
W. R. Frequency of depression in the freshman
year as measured in a random sample by a retro-
spective version of the Beck Depression Inventory.
Journal of Consulting and Clinical Psychology,
1975, 4245), 746-747.

Boyle, B. S. Manual for the Shipley-Institute of Living
Scale. Hartford: Neuropsychiatric Institute
of the Hartford Retreat, 1967.

Burgess, E. The modification of depressive behavior. In
R. Rubin & C. Franks (Eds.), Advances in behavior
therapy. New York: Academic Press, 1969.

 

Campbell, D. T. From description to experimentation:
Interpreting trends as quasi-experiments. In
C. W. Harris (Ed.), Problems in measuripg chan e.
Madison: University of Wisconsin Press, 1963,
212-242.

 

Campbell, D. T., & Stanley, J. S. Expgrimental and quasi-
experimental designs for research. Chicago:
RandiMcNaIly, 1966f

 

Carroll, B. J. Review of clinical research strategies in
affective illness. Paper presented to the American
Association for the Advancement of Science Sym-
posium. Implications of Some Recent Biological
Studies of Depressive Illness. San Francisco,
California, February 27, 1974.

Carson, R. C. Interpretive manual to the MMPI. In
J. N. Butcher (Ed.), MMPI: Research developments

and clinical applications. New York: McGraw-
Hl I I — e

Cautela, J. R. Behavior therapy and self-control:
Techniques and implications. In C. M. Franks
(Ed.), Behavior therapy: Appraisal and status.
New York: McGraw-Hill, 1969, 3234340.

 

 

Cautela, J. R., & Kastenbaum, R. A reinforcement survey
schedule for use in therapy, training, and
research. Psychological Reports, 1967, 20,
1115-1130.

167

Cautela, J. R., & Wisocki, P. The use of the Reinforce-
ment Survey Schedule in behavior modification.
In R. Rubin (Ed.), Advances in behavior therapy.
New York: Academic Press, 1971.

 

Chodoff, P. The depressive personality: A critical
review. In R. J. Friedman & M. M. Katz (Eds.),
The psychology of depression: Contemporary
theory and research. Washington, D.C.:
V. H. Winston & Sons, 1974.

 

Dahlstrom, W. G., Welsh, G. S., & Dahlstrom, L. G. Ag
MMPI handbook, volume I: Clinical interpretation.

Minneapolis: University of’Minnesota Press,
1972.

 

Dempsey, P. A unidimensional depression scale for the
MMPI. Journal of Consulting Psychology, 1964,
28, 364-371.

DiMascio, A., Meyer, R. E., & Stifler, L. Effects of
imipramine on individuals varying in level of
depression. American Journal of Psychiatry,
1968, I24, 55-58.

 

Donnelly, E. F., & Murphy, D. L. Primary affective dis-
order: MMPI differences between unipolar and
bipolar depressed subjects. Journal of Clinical
Psychology) 1973, 22, 303-306.

 

 

Donnelly, E. F., & Murphy, D. L. Primary affective dis-
order: Bender-Gestalt sequence of placement as
an indicator of impulse control. Perceptual and
Motor Skills, 1974, 38, 1079-1082.

 

 

Ellis, A. The validity of personality questionnaires.
Psychological Bulletin, 1946, 43, 385-440.

 

Ellis, A. Reason and emotion in psychotherapy. New
York: Lyle Stuart, 1962.

 

Ellis, A. Growth through reason: Verbatim cases in
rational emotive therapy. North Hollywood,
California: Wilshire,_I97l.

 

 

Ellis, A. Humanistic psychotherapy: Thepationalf
emotive approach. New York} The Julian Press,
1973.

 

 

Elson, S. E. Behavioral self-control: A survey of the
literature. Unpublished manuscript. East
Lansing, Michigan: Michigan State University,
1972.

168

Elson, S. E. Research evidence for the efficacy of
rational-emotive therapy. Unpublished manu-
script. East Lansing, Michigan: Michigan
State University, 1974.

Engel, G. A life setting conducive to illness: The
giving-up--given-up complex. Bulletin of the
Menninger Clinic, 1968, 22, 355-365.

 

 

Ferster, C. B. Classification of behavioral pathology.
In L. Krasner & L. P. Ullmann (Eds.), Research
in behavior modification. New York: Holt,
Rinehart, & Winston, 1965, 6-26.

 

Ferster, C. B. A functional analysis of depression.
American Psychologist, 1973, 22, 857-870.

 

Ferster, C. B. Behavioral approaches to depression.
In R. J. Friedman & M. M. Katz (Eds.), The psy-
chology of depression: Contemporary theory and
research. Washington, D.C.: V. H. Winston &

 

 

Fisher, R. A. On the "probable error" of a coefficient
of correlation deduced from a small sample.

Frank, J. D. Persuasion and healing. Baltimore: Johns
Hopkins Press, 1961.

 

Friedman, R. J., & Katz, M. M. (Eds.). The psychology
of depression: Contemporary theoryjand research.
Washington, D.C.: V. H. Winston & Sons, 1974.

 

Gasswint, C. D. Personal communication. Knoxville,
Iowa: Veterans Administration Hospital, 1971.

Gillette, P., Knoff, W. F., & Ellis, A. Camera three
presents dimensions of depression. New York:
Columbia Broadcasting System, 1974.

 

 

Goldberg, L. R. Objective diagnostic tests and measures.
In M. R. Rosenzweig and L. W. Porter (Eds.),
Annual review of psychology, volume 25. Palo
Alto, California: Annual Reviews Inc., 1974,
343-366.

 

Goldfried, M. R., & Merbaum, M. Behaviorchange through
self-control. New York: Holt, Rinehart, &
Winston, 1973.

 

169

Grinker, R. R., Miller, J., Sabshin, M., Nunn, R., &
Nunnally, J. C. Theyphenomena of depressions.
New York: Harper & Row, 1961.

 

Grinker, R. R., & Nunnally, J. C. The phenomena of
depressions. In W. S. Sahakian (Ed. ), Psycho-
pathology today: Ex xpe erimentation, theory, and
research. Itasca, Illinois: F. E. Peacock,

1976, 392-400.

Hall, R. V., Cristler, C., Cranston, S., & Tucker, B.
Teachers and parents as researchers using
multiple baseline designs. Journal of Applied
Behavior Analysis, 1970, 3, 2471253.

 

 

 

Hathaway, S. R., & McKinley, J. C. The Minnesota Multi-
phasic Personality Inventory (Rev. ed)).
Minneapolis, Minnesota: University of Minnesota
Press, 1943.

 

 

Hathaway, S. R., & McKinley, J. The Minnesota Multi-
phasic Personality Inventory manual: Revised.
New York: The Psychological Corporation, i9—1.

 

 

Hathaway, S. R., & McKinley, J. C. Scale 7 (psychas-
thenia). In G. 8. Welsh & W. G. Dahlstrom
(Eds.), Basic readings on the MMPI in psychology
and medicipg, Minneapolis: University of
Minnesota Press, 1956. (a)

 

Hathaway, 8. R., & McKinley, J. C. Scale 2 (depression).
In G. S. Welsh & W. G. Dahlstrom (Eds. ), Basic
readings on the MMPI in psychology and medicine.
Minneapolis: University of Minnesota Press,
1956. (b)

 

Hathaway, S. R., & Meehl, P. E. An atlas for the clinical
use of the MMPI. Minneapolis: University of’
Minnesota Press, 1951.

 

 

Hauck, P. A. An RET theory of depression. Rational

Hauck, P. A. A young woman with feelings of depression.
In A. Ellis (Ed.), Growth through reason:
Verbatim cases in rational-emotive therapy.
Hollywood: Wilshire, 1971, 123-I46. (bi

 

 

170

Hendricks, C. G., Thoresen, C. E., & Hubbard, D. R.
Effects of behavioral self-observation on ele-
mentary teachers and students. Stanford, Cali-
fornia: Stanford Center for Research and
Development in Teaching, Research and Development
Memorandum No. 121, 1974.

Homme, L. E. Perspectives in psychology XXIV: Control
of coverants, the operants of the mind. Psycho-
logical Record, 1965, 33, 501-511.

 

Honigfeld, G., Gillis, R. O., & Klett, C. J. NOSIE-30:
A treatment-sensitive ward behavior scale.
Psychological Reports, 1966, 33, 180-182.

 

Honigfeld, G. NOSIE-30: History and current status of
its use in psychopharmacological research. In
P. Pichot (Ed.), Psychological measurements in

psychopharmacology. Modern problems in phar-

maco s chiatr . Paris: Karger, Base , ,
volume 7, 235-263.

 

 

Horney, K. Neurosis and human growth. New York: W. W.
Norton, 1950.

 

Horrocks, J. E. Assessment of behavior: The methodology
and content of psychologiggl measurement.
Columbus, Ohio: Charles E. Merrill, 1964.

 

 

Hoyt, C. J. Test reliability estimated by analysis of
variance. Psychometrika, 1941, 3, 153-160.

 

Jackson, B. Treatment of depression by self-reinforce-
ment. Behavior Therapy, 1972, 3, 298-307.

 

Johnson, R. G., & Elson, S. E. The modification of
covert behavior: A survey of the literature.
Paper presented at the Annual Meeting of the
American Educational Research Association,
Chicago, 1974.

Kanfer, F. H., & Phillips, J. S. Learningfoundations
of behavior therapy. New York: John Wiley &
Sons, 1970.

 

 

Katz, M. M. Epilogue: On future directions. In R. J.
Friedman & M. M. Katz (Eds.), The psychology of
depression: 4ggntemporary theory and_research.
Washington, D.C.: V. H. Winston & Sons, 1974,
299-301.

171

Kazdin, A. E. Statistical analyses for single-case
experimental designs. In M. Hersen and D. H.
Barlow (Eds.), Single-case experimental desigps:
Strategies for studying behavioral change.
Oxford: Permagon Press, 1975.

Kleinknecht, R. A., McCormick, C. E., & Thorndike, R. M.
Stability of stated reinforcers as measured by
the Reinforcement Survey Schedule. Behavior
Therapy, 1973, 3, 407-413.

 

Krasner, L. The operant approach in behavior therapy.
In A. E. Bergin & S. L. Garfield (Eds.), Hand-
book of psychotherapy and behavior change: An
Experimental analysis. New York: John Wiley &
Sons, 1971, 612-652.

 

 

Kushner, R. Three-part series on depression. Chevy
Chase, Maryland: Montggmer County Journal,
December 27, l973-January l , 1974.

 

Lazarus, A. A. Learning theory and the treatment of
depression. Behavior Therapy and Research,
1968' 2, 83-860

 

Lazarus, A. A. Behavior therapy and beyond. New York:
McGraw-Hill, l971. (a)

 

Lazarus, A. A. New techniques for behavior change.
Rational Living, 1971, 3, 2-7. (b)

 

Lazarus, A. A. Desensitization and cognitive restructur-
ing. Psychotherapy: Theory, Research, and
Practice, 1974,11, 98-102. (a)

 

Lazarus, A. A. Multimodal behavioral treatment of
depression. Behavior Therapy, 1974, 3, 549-
554. (b)

 

Lazarus, A. A., & Davison, G. C. Clinical innovation
in research and practice. In A. E. Bergin &
S. L. Garfield (Eds.), Handbook of psychotherapy
and behaviorighange: An experimental analysis.
New York: John Wiley & Sons, 1971, 196-213.

 

Lentz, R. J., Paul, G. L., & Calhoun, J. F. Reliability
and validity of three measures of functioning
with "hard-core" chronic mental patients.
Journal of Abnormal Psychology, 1971, 13, 69-76.

 

172

Leonard, C. V. Depression and suicidality. Journal of
Consulting and Clinical Psychology: 1974, 43,
98-101.

Lewinsohn, P. M. Clinical and theoretical aspects of
depression. In K. S. Calhoun, H. E. Adams, &
K. M. Mitchell (Eds.), Innovative treatment

methods in s cho atholo . New York: John
Wiley & Sons, I974, 63-12g.

Lewinsohn, P. M. Personal communication. Eugene,
Oregon: University of Oregon, 1975.

 

 

Lewinsohn, P. M., & Graf, M. Pleasant activities and
depression. Journal of Consulting and Clinical
Psychology, 1973, g_1_, 261-268.

 

Lewinsohn, P. M., & Shaw, D. A. Feedback about inter-
personal behavior as an agent of behavior change:
A case study in the treatment of depression.
Psychotherapy and Psychosomatics, 1969, 11,
62-88.

Liberman, R. P., & Smith, V. A multiple baseline study
of systematic desensitization in a patient with
multiple phobias. Behavior Therapy: 1972, 3,
597-603.

Mahoney, M. J. Cognition and behavior modification.
Cambridge, MassaChusetts: Ballinger, 1974.

Mahoney, M. J., & Thoresen, C. E. Self-control: Power
to the person. Monterey, California: Brooks-
Cole, 1974.

 

Marks, P. A., Seeman, W., & Haller, D. L. The actuarial
use of the MMPI with adolescents and adults.
Baltimore: Williams & Wilkins, 1974.

Meichenbaum, D. Clinical implications of modifying What
clients say to themselves. Unpublished manu-
script. Waterloo, Ontario, Canada: University
of Waterloo, 1973.

Meichenbaum, D. Self-instructional methods. In F. H.
Kanfer & A. P. Goldstein (Eds.), Hel in eo 1e
change. New York: Permagon Press, E972. (a5

173

Meichenbaum, D. Self-instruction and the therapeutic
modification of self-statements or cognitive-
hyphen-behavior therapy. Paper presented at
the Eighth Annual Convention of the Association
for Advancement of Behavior Therapy, Chicago,
1974. (b)

Meichenbaum, D., & Cameron, R. The clinical potential
of modifying what clients say to themselves.
Psychotherapy: Theory, Research, and Practice,
l974,33(2),lO3-ll7.

Mischel, W. Personality and assessment. New York:
John Wiley & Sons, 1968. ‘—

 

Murray, E. J., & Jacobson, L. I. The nature of learning
in traditional and behavioral psychotherapy.
In A. E. Bergin & S. L. Garfield (Eds.), Handbook
of psychotherapy and behavior change: An experi-

mental analysis. New York: John Wiley &—§ons,
7 I -7 0

National Institute of Mental Health. Special report on
depression. Bethesda, Maryland: Author, 1973.

 

 

 

Mezzich, J. E., Damarin, F. L., & Erickson, J. R.
Comparative validity of strategies and indices
for differential diagnosis of depressive states
from other psychiatric conditions using the MMPI.
Journal of Consulting and Clinical Psychology,
1974, 33(5), 691-698.

 

Paulson, M. J., & Lin, T. Predicting WAIS I.Q. from
Shipley-Hartford scores. Journal of Clinical
Psychology, 1970, 33, 453-461.

 

 

Pearson, K. On further methods of determining correlation.
Dra er's Company Memoirs, Biometric Series, IV,
1 07.

 

Premack, D. Toward empirical behavior laws: I. Positive
reinforcement. Psyghological Review, 1959, 33,
219-233. '

 

Premack, D. Reinforcement theory. In D. Levine (Ed.),
Nebraska symposium on motivation,3965. Lincoln,
Nebraska: University of Nebraska Press, 1965.

 

Rosen, A. Differentiation of diagnostic groups by indi-
vidual MMPI scores. Journal of Consulting Psy-
chology, 1958, 33, 453-457.

 

174

Rosen, A. Development of MMPI scales based on a
reference group of psychiatric patients. Psy-
cholggical Monographs, 1962, 23(8, whole no.
527).

Russell, P. L., & Brandsma, J. M. A theoretical and
empirical integration of the rational-emotive
and classical conditioning theories. Journal of
Consultingyand Clinical Psychology, 1974, 13,
389-397.

Sahakian, W. S. (Ed.) Psychopathology today: Experi-
mentation, theory and research. Itasca,
IllinOis: F. E. Peacock, 1970.

Schmickley, V. G. Rational-emotive-behavioral counseling.
Unpublished manuscript. East Lansing, Michigan:
Michigan State University, 1973.

Seitz, F. C. A behavior modification approach to depres-
sion: A case study. Psychology, 1971, 3, 58-
63. (a)

Seitz, F. C. Behavior modification of depression.
Proceedings, 79th Annual Convention, American
Psychological Association, 1971, 425-426. (b)

 

Seligman, M. E. P. Depression and learned helplessness.
In R. J. Friedman & M. M. Katz (Eds.), The psy-
chology of depression: tContemporarytheory and
research} WaShington, D.C.: V. H. Winston &

Sons, 1974.

Seligman, M. E. P. Helplessness. San Francisco: W. H.
Freeman, 1975.

 

Seligman, M. E. P., & Hager, J. Biolggicalboundaries
of learning. New York: Appleton-Century-Crofts,
1972.

 

Seligman, M. E. P., Maier, S. F., & Solomon, R. L.
Unpredictable and uncontrollable aversive
events. In F. R. Brush (Ed.), Aversive con-

ditionin and learnin . New York: Academic
Press, 1371, 347-400.

Shipley, W. C. A self-administering scale for measuring
intellectual impairment and deterioration.
Journal of Psychology, 1940, 3, 371-377.

 

175

Sidman, M. Tactics of scientific research: Evaluating
experimsntal data in psychology. New York:
Basic Books, 1960.

Sidman, M. Operant techniques. In A. J. Bachrach (Ed.),
Experimental foundations of clinical psychology.
New York: Basic Books, 1962, 170-210.

Siegel, S. Non-parametric statistics for the behavioral
sciences. New York: McGraw-Hill, 1956.

Silver, R. J., & Sines, L. K. MMPI characteristics of a
state hospital population. Journal of Clinical
Psychology: 1961, 33, 142-146.

 

 

Skinner, B. F. Science and human behavior. New York:
Mammillan, 1953.

Skinner, B. F. Beyond freedom and dignity. New York:
Alfred A. Knopf, 1972.

Skinner, B. F. About behaviorism. New York: Alfred A.
Knopf, 1974.

 

Thoresen, C. E. Behavioral humanism. In C. E. Thoresen
(Ed.), Behavior modification in education.
Chicago: National Society for the Study of
Education, 1972, 385-421.

Thoresen, C. E. The intensive design: An intimate
approach to counseling research. Paper presented
at the Annual Meeting of the American Educational
Research Association, Chicago, 1972.

Thoresen, C. E., & Anton, J. L. Intensive experimental
research in counseling. Journal of Counseling
Psychology: 1974, 33, 553-559.

 

 

Thoresen, C. E., & Mahoney, M. J. Behavioral self-
control. New York: Holt, Rinehart, & Winston,
1974.

 

Thornton, J. W., Rieken, C., & Lindsay, K. J. Person-
ality characteristics of the learned helplessness
individual. Unpublished manuscript. San Angelo,
Texas: Angelo State University, 1974.

Vogelback Computing Center. SPSS: Statistical package
for the social sciences, Version 6.00. Evanston,
Illinois: Northwestern University, l975.

 

 

176

Watson, C., & Klett, W. Prediction of WAIS 1.0.8 from
the Shipley-Hartford, the Army General Classifi-
cation Test, and the Revised Beta Examination.

Journal of Clinical Psychology: 1968, 33, 338-
341.

Welsh, G. S., & Dahlstrom, W. G. (Eds.) Basic readings
on the MMPI in psychology and medicine. Minne-
apolis: University of Minnesota Press, 1956.

 

White, 0. R. The "split middle": A "quickie" method of
trend estimation. Working paper no. 1. Eugene,

Oregon: Regional Resource Center for Handicapped
Children, 1971.

White, 0. R. A manual for the calculation and use of the
median slope--A technique of progress estimation
and prediction in the single case. Working paper
no. 16. Eugene, Oregon: Regional Resource
Center for Handicapped Children, 1972.

Wiggins, J. S. Substantive dimensions of self-report in

the MMPI item pool. Psychological Monographs,
1966, 32(22, whole no. 630).

Zytowski, D. G., & Betz, E. L. Measurement in counseling

research: A review. The Counseling Psychologist,
1972, 3, 72-86.

"‘iIWWW“