l

[y
9

l/fljl/llllfllI/W MW
”8m 9'014
Michifl‘m 5“”

/ M
W , , »/ /
University

 

 

This is to certify that the

dissertation entitled
Transference Manifestations in Psychotherapy

presented by

Wendy Sabbath

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Psychology

 

 

   

Major professor

Date 12/3/86

MS U i: an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

 

MSU

LIBRARIES
m

V

 

 

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

 

 

 

 

 

TRANSFERENCE MANIFESTATIONS IN PSYCHOTHERAPY

By

‘Wendy June Sabbath

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Psychology

1987

ABSTRACT
TRANSFERENCE MANIFESTNTIONS IN PSYCHOTHERAPY

By
wendy June Sabbath

An initial consensus seems to have been reached across
practitioners of the major forms of psychotherapy regarding the
importance of the relationship between therapist and patient, as a
prominent factor in the process and eventual outcome of the therapy.
The phenomena of transference, or the re-creation within the
relationship with the therapist of earlier object relationships, has
always been the cornerstone of psychoanalytic theory and technique.
Although there has been lively interest within the psychoanalytic field
in the vicissitudes of the transference and perceptions of the

therapist, there have been relatively few empirical studies of

. transference itself.

The present study investigated the extent to which the therapist
became perceived as similar to the mother or the father in the course
of psychodynamic psychotherapy. Sixteen female clients undergoing six
to ten months of psychotherapy at the M.S.U. Counseling Center served
as subjects. The Rorschach was administered at both the beginning and
the end of therapy, the Interpersonal Check List (ICIJ was administered
at beginning, middle, and end of therapy, and the Helping Relationship

Questionnaire (HRQ) was given at the end of therapy.

Internal representations of the self and the mother changed
significantly (p _<_ .05) from early to mid-therapy. 'Ihe self-
representation also changed significantly from mid- to late therapy.
Adaptive psychological functioning was significantly correlated with
perceiving the therapist as similar to both mother (2 = .004) and
father (2 = .02) in mid-therapy. 'Ihe therapist seemed to become
perceived as more similar to mother or father in mid-therapy, and again
less similar to mother or father in late therapy, but not t0>a
significant extent. The total amount of change in representations of
self, therapist, and mother was significantly QEHS .05) correlated with
the amount of change in anxiety content on the Rorschach. Tbtal change
in representations of therapist, mother, and father was significantly
(p _<_ .05) correlated with the amount of change in hostility content on
the Rorschach. No increase was found in capacity for object relations,
as measured by the Developmental Analysis of the Concept of the Object
scale applied to Rorschach responses. Intensity of late therapy
representations was not significantly correlated with late therapy mean

object relations scores.

AWS

It's very difficult to do justice to the profound contributions
made by each member of my dissertation and guidance comnittees. Each
of these people has been an integral and deeply valued part of my life
for the past six years.

Any kind of recognition must begin with Dr. Bert Karon. From the
first lectures of his I heard in 1975 he has catalyzed my interest in
personality and psychoanalytic psychology. Over the past years, he has
been a staunch friend, advisor, and an inspiring supervisor. I
appreciate both the freedom he gave me to develop my own ideas, and the
careful attention and consideration he gave to them. Dr. Rabin gave
generously of his wisdom and concern. Working with him for the past
several years has been an honor, as well as a pleasure. He has
introduced me to numerous aspects of psychoanalytic thought, and his
interest has enriched my life in more personal ways as well. Dr.
Abeles was a supportive, informative, and very helpful supervisor and
committee member. Dr. Powell was cheerfully encouraging, both in terns
of research and in clinical issues.

I would like to thank the staff of the M.S.U. Counseling Center
for their cooperation with this project. I appreciate the attempts of
all the therapists who solicited the participation of their clients.
I'd like to thank the subjects themselves, who gave their time and were

willing to share their personal experiences and feelings with me.

iv

Richard was a dedicated and humorous tester and scorer. Carol, as
always, was a unique deeply appreciated consultant, test, and ally.
Suzy's technical and adndnistrative assistance has been consistently

over and above what was required, and is also appreciated.

TABLE OF CONTENTS

LI ST OF TABLES 0 O O O O O O O O O O O O O O O O O O O O

CHAPTERl

IWMICN. O O O O O O O O O O O O O O O O O O O O O

Transference...................
Einpirical Studies of Transference. . . . . . . . .
Developmental Aspects of Internal Representations.

Summary .
Theoretical Approaches to Object Relations and

Recent Applications of Projective
Interpersonal Theories and Research on
Stages of Psychotherapy. . . . . . . .

Summary of Literature. . . .
CHAPTER 2

MISC O O O O O O O O O O 0
CHAPTER 3

ms 0 O O O O O O O O O O O O

Sllbjxt O O O O O I O 0

Subjects and Therapist.

Premature Terminators .

Instruments . . . . . .

Prm$ureSO O O O O O O

Scoring Procedures. . .

CHAPTER 4

RESULTS .

Main Hypotheses. . . . . . .

CHAPTERS

Techniques . .
Psychotherapy

Rorschach.............
Helping Relationship Questionnaire.

DI$USSIWQ O 0 O O O O O O O O O O O O O O 0

vi

Page

viii

11
15

15
21
23
25

28

33

33
33
35
36
40
42

47
47

53
55

6O

@mral Issues 0 O O O O I O O O O O C O O O O O O O
The Relationship between Change in Representations
and Change in Rorschach Content. . . . . . . .

Hostility, Anxiety, and Level of Functioning. .
Change in the Capacity for Object Relations . .
Quality of the Transference . . . . . . . . . .

Maternal/Paternal Aspects and Temporal Changes

Later Therapy ICL and Object Relations Scores

APPEND ICES

APPENDIX A.

APPENDIX B.

APPENDIX C .

APPENDIX D.

ms.

in Transference. . . . . . . . . . . . . .

vii

60
65
66
67
68

71
72

73
75
76
77

78

 

LIST OF TABLES

Tables

1

T-tests of Differences between Prematurely Terminating
Clients and Clients Completing Therapy on Initial
variables. 0 O O O O 0 O O O O O O O O O O O O O O O O O

Rorschach Scores Used as Criteria for Level of
PersonalityFunctioning.................

Mean ICL Representation Scores for Self, Therapist,
Mother, and Father in Early, Middle, and Late Therapy. .

Mean Early and Late-therapy Rorschach HoStility,
Anxiety, Object Representation, and Criteria Scores. . .

Pearson Correlations between Total Amount of Change in
ICL Representations of Self, Therapist, Mother and
Father and Total Change from Early to Late Rorschach
Hostility, Anxiety, Mean Object Representation and
CriteriaScores.....................

T-tests Comparing Rorschach Object Representation Scores
in Early and in Late Therapy . . . . . . . . . . . . . .

Pearson Correlation Coefficients of Rorschach variables
between First and Second Administrations . . . . . . . .

Descriptive Statistics, Frequencies and Percentages for
RorSChaCh variable 0 O O O O O O O O O O O O O O O O O 0

Descriptive Statistics for the Helping Relationship
meStiomBir-e O O O O C O O O O O O O O O O O O O O O O 0

viii ‘

Page

37

43

48

51

51

52

54

56

58

CHAPTER 1

INTRODUCTION

The foundation of the psychotherapy literature has traditionally
rested on individual case histories and on clinicians' reports of
therapy's efficacy, but many of the factors influencing therapy process
and outcome have proved difficult to isolate and to quantify.
Achieving understanding of the many influences operating in this
process has been, according to Greenspan and Wieder (1984), one of the
most difficult challenges in the behavioral sciences. Graff and
Luborsky (I977) highlighted one difficult problem in quantitative
analysis of therapeutic process: discovering meaningful long-term
trends in therapy by means of both economical and reasonable methods.
Their hope was that solutions to this problem might provide
verification of clinical findings. It is one thing to assert, as
Smith, Glass, and Miller (1980) did, that psychotherapy is consistently
beneficial and that ”its efficacy has been demonstrated with near
monotonous regularity” (p. 183), but the question which remains, and
to Which this study will attempt to supply some answers, is just how
events in psychotherapy, and specifically aspects of the therapeutic
relationship and the transference itself, interact to produce
successful outcome in psychotherapy.

Some previous studies of the relationship between therapist and
client have offered observations about important features of this

1

 

relationship from both the therapists' and the clients' points of view.
Leary and Harvey (1956) described psychotherapy as an "inplausible
procedure offering to the individual the opportunity to learn those
things about himself which by definition he does not*wish.to know” (p.
123). In the face of such opposition, they felt that the continued
existence of psychotherapy was in itself sufficient testimony to its
effectiveness, a position reminiscent of Freud's (1937) frequently
cited definition of analysis as one of three ”impossible professions in
which one can be sure beforehand of achieving unsatisfying results" (p.
248). Janet Malcolm (1981) followed this observation with Limentani's
crunent that psychoanalysis ”is not only impossible, but also extremely
difficult" (Malcolm, 1981).

Iapkin and Lury (1983) took a developmental perspective toward
process, and suggested that psychotherapy should focus on repairing
previous omissions in the developmental building blocks of the
individual's life such as identifications, object constancy, and
reciprocity. Again, the emphasis was on the reparative possibilities
inherent in the relationship between therapist and client, and in the
ways in which this relationship was handled. They referred to
Guntrip's statement:

Psychotherapy involves that the patient must gr w out of

unrealistic positive and negative transference relations, in

which he is seeing his internal fantasied good and bad

objects projected in to his therapist, by means of

discovering what kind of actual relationship is given to him

by his therapist as a real person. This involves much more

than experienced psychoanalytical interpretation. That paves

the way, against the background of the kind of person the

analyst actually is, for the patient to grow'gradually to an

accurate perception of him as a real person in his own right.
(Guntrip, 1973, p. 66)

Lapkin and Lury concurred that this perspective toward the
therapeutic relationship transcended classical psychoanalytic technique
and constituted a 'personal relationship therapy' (Guntrip, 1973, p.
183) which included all the variants of psychodynamically oriented
psychotherapy. Stmpp (Goldfried, 1980) believed:

Psychotherapy works because (a) all human beings have a
strong tendency to 'transfer' patterns of interpersonal
relatedness from the past to the present, and (b) this
transference tendency can be utilized by the therapist to
bring about a 'corrective experience.‘ Thus the essence of
psychotherapy is interpersonal learning, a new significant
experience (ERLEENIS) that, if all goes well, modifies basic
aspects of the patient's patterns of relatedness in ways that
are called therapeutic....By patterns of interpersonal
relatedness I mean characteristic and often stereotyped ways
in which a person relates to significant others. Such
patterns become apparent particularly in situations that are
affectively charged and that activate engrained (overlearned)
tendencies of relating to care-givers and authority figures.
Patterns of relatedness are often sufosed by unconscious
fantasies and contradictory strivings. (p. 279)

Definitions such as the preceding ones provide for the emergence
of transference reactions in all forms of psychotherapy, but only in
psychodynamically oriented psychotherapy and psychoanalysis are
transference vicissitudes both anticipated and directly addressed as
crucial aspects of the treatment. The existing literature on
transference has been oriented toward theoretical speculations
regarding transference and clinical vignettes, but in recent years
there has been renewed interest in the therapeutic relationship, and
therapeutic alliance in general, as well as specific qualities of the

transference.

Transference

 

‘Waelder (1956) briefly defined transference as "an attempt of the

patient to revive and re-enact, in the analytic situation and in

relation to the analyst, situations and fantasies of his childhood" (p.
367). waelder mentioned bringing influence fer change to bear upon the
client in the form of new identifications in the superego or the ego,
and of therapy functioning as a corrective experience. He enumerated
three general approaches to the transference: (1) transference as an
obstacle to the treatment, (2) transference as possibly THE most
important vehicle of the treatment, and (3) transference as the
opportunity of direct influence, through identification or re-training.

Zetzel (1956) wrote of the development of traditional views of
transference moving from the structural approach to include greater
consideration of current, past, internal, and external object relations
and their effect both on the patient's phenomenological world and the
therapeutic relationship. She extensively discussed differing
psychoanalytic perspectives on transference, noting that crucial
aspects of the early relationship, and their repetition in the analytic
situation, were consensually validated events occurring with great
frequency.

Schlesingerévaccaro (1983) drew explicit parallels between
transference phenomena and Piaget's concept of conservation. She
pointed out that mental representations of persons and of situations
were conserved, and in the transference the patient revealed the inner
world as it had been conserved. The goal of therapy was seen as aiding
the patient to attempt to assimilate the therapist less to mother- and
father-schemata, and to work toward increased acconmodation with the
external environment.

Gill (1982) reiterated that transference clearly occurred in

therapy and ought to be taken seriously in all 'uncovering' modes of

therapy. He distinguished between two types of transference
interpretation: interpretations of resistance to the awareness of
transference, and interpretations of resistance to the resolution of
transference. Two types of resistance to resolution of transference
were enumerated. The first.were interpretations working with the
transference in the present by pointing out that the client's attitude
might not be as clearly determined by the in'nediate therapeutic
situation as he or she may claim. The second were the 'genetic
transference interpretations,‘ or interpretations of a similarity
between the transference attitude and the past.

Konig and Tischtau-Schroter's (1982) definition of transference

explicitly mentioned both its object-relations and interpersonal

aspects:

Transference is defined as a projection of inner objects and
parts of the self; the classical part of transference always
contains a nucleus of projection. The interactional part of
transference consists in verbal and nonverbal behavior with
the purpose of causing the other person to comply with the
transferred object or part of self: Transference
expectations may be unconscious and in conflict*with
conscious expectations. (p. 1191)

Schafer (1977) clarified the relationship between relived
experience and new experience in the interpersonal situation of the
transference. He noted that the positive aspects of the transference
('transference love') related to both the current life situation and to
the repressed past. A sequence of transference reactions was suggested
in which the analyst's consistent containing and tolerating of
aggressive impulses and feelings was seen as a pre-condition for

development of the client's capacity for loving. Having reached this

phase, the experience of having both positive and negative reactions

toward therapist was hypothesized to lead to both less splitting and
less idealization of the therapist, and to culminate in the patient's
increased ability to differentiate from both parents and analyst.

The analysand now tolerates his or her own experiencing of

relationships ambivalently, thereby demonstrating that new

and more inclusive conditions for loving have been developed.

This development bears as much on loving oneself as on loving

others. That is to say, tolerating the ambivalent mode is,

correlatively, a condition of healthy 'narcissism' and

'object love': the one implies the other rather than being

at its expense. (PP. 344-345)

Karme' (1982) proposed that the predominantly maternal or paternal
quality of the transference was predicted by the therapist's gender and
developmental qualities of both the client and the current depth of the
analysis itself. She suggested that maternal transferences occurred
toward male and female analysts, but existed only in preoedipal phases
for patients with male therapists. Similarly, she believed that a true
paternal transference neurosis to female analysts were quite rare, and
that during analysis of the oedipal phase transference became specific
to the gender of the analyst. Mogul (1982) extensively reviewed the
existing literature on the effect of the therapist's sex on therapy.
She found agreement among a number of clinical reports that gender of
the psychoanalyst had been related to the sequence in which
transference reactions occurred, and that this effect had been most
evident early in treatment. Her conclusions were that the specific
content and sequence of transference reactions were related to
therapist gender in neurotic patients, that more developmentally
regressed and fragmentary transference reactions, which were affected
by therapist gender, were more dramatic in developmentally regressed

patients, and that therapist gender had less impact in classical

psychoanalysis with.neurotic patients than in other forms of
psychotherapy with neurotic and other patient populations. Kulish's
1984 literature review confirmed these conclusions. Kulish
specifically commented upon the absence of a body of research to
support the significant number of clinical observations by such notable

figures as Freud, Glover, and Greenson, and that the phase of

 

PE
treatment, developmental fixation and maturation of the patient, and

therapist gender were all involved in the specific maternal or paternal

quality of the transference at a given moment in therapy. «.1

Empirical Studies of Transference

 

Luborsky, Crabtree, Curtis, Ruff, and Mintz (1975) attempted to
make a 'conceptual map' of dimensions of transference based on a factor
analysis of eight psychoanalysts' ratings of 23 clinically relevant
transference concepts from transcriptions of segments of a successful
analysis. The original definition of transference which guided the
study was influenced by object-relations theory: ”Transference is a
revival in a current object relationship, especially to the analyst, of
thought, feeling, and behavior derived from repressed fantasies
originating in significant conflictual childhood relationships” (p.
65). Four basic dimensions were found, the largest factor (accounting
for approximately one fifth of the variance) being 'expressed parental
transference.’ The other three factors were 'infantile distortion'
(material seen by the analyst as infantile and distorted), 'positive
transference,’ and 'sibling transference.‘ All transference
manifestations decreased over the course of the analysis, confirming

theoretical beliefs that successful outcome is marked by, and/or

8

somewhat the result of, lessening of rigid pathological patterns of
relating to others.
Mueller (1969b) conceptualized transference much along the lines
of Gill's 'genetic transference interpretations,‘ observing:
The pattern of a client's reactions to his therapist often
recapitulates the behavioral patterns that the client learned
as modes of coping with stress in earlier parent-child

interactions and that those repetitious behavioral patterns
are often indices of generic conflicts. (p. 2)

Mueller investigated the ways in which the pattern of client

 

reactions to the therapist converged throughout therapy with the
client's previous reactions to parents and other significant persons.
He found that, particularly when anxious, the client tended to behave
toward the therapist as he or she had reported behavioring toward the
parents. The therapist then tended to behave toward the client in a
similar fashion to the parents' reported behaviors. meller
interpreted these reciprocal interactions as clear evidence for the
development of transference relationships during therapy, and for
occurrence of predictable countertransference reactions.

Hoornaert and Pierloot (1976) referred to symbolic dimensions of
parental figures as "affective and mental schemata, formed by the
personal relation with the parents, enriched by the significant
additions awarded by the cultural world of the subject” (p. 261).
Their primary interest.was in the maternal and paternal aspects of
transference manifested in the attitudes and feelings of psychosomatic
patients toward their physicians. A.unique aspect of this study was
its documentation of the occurrence of transference in a setting

analogous, but not identical, to psychotherapy. The patient's image of

the doctor was found to have a predominance of paternal (autonomy,
dynamism, and action) qualities, and the correlation was higher between
images of the doctor and tie ideal fatter than that between the doctor
and actual fatter. The representation of the doctor showed almst no
similarity to eitrer the image of tie real or the ideal motler.

Blotcky (1984) suggested that an early examination of human
content and human movement responses in Rorschach protocols might
anticipate transference distortions which would emerge later in
psychotherapy. He described tie rapid development of a turbulent
transference, often including both devaluation and idealization, as
characteristic features of psychoanalytic psychotherapy with borderline
adolescents. He hoped that accurate anticipation of such transference
patterns might help prevent pranature termination, acting-out,
development of a therapeutic inpasse, or countertherapeutic behaviors
on tre part of the trerapist.

‘mo relatively recent studies have reported attempts to both
describe and treasure changes in the transference across the course of
psychoanalysis and psychotherapy. Although the investigators
conceptualized the role of the transference in slightly different ways,
the emergence of such studies highlights the growing trend in
psychodynamic research of recognizing the importance of tie therapeutic
relationship and of transference. Graff and Luborsky (1977) charted
analysts' quantitative ratings of transference and resistance for each
session of four patients' analyses in order to trace the analytic
process in the dynamic unfolding of transference and resistance. They
naintained a specifically object-relations conceptualization of

transference :

10

The degree to which the patient is dealing with material that

is overtly or covertly related to the analyst. This material

would be a manifestation of or a displacement from an early

inrxutant object relation. The previous object, however,

does not have to be mentioned; it may be inferred by the

rater because of inappropriateness, etc. (p. 476)

Distinct patterns of clinical change were found for the four
patients. The two whose posttherapeutic outcome was considered more
successful showed increasing levels of transference concomitant with
decreasing levels of resistance. comparing thiS‘with the‘widely held
belief that successful outcome is expected to be related to decreasing
resistance, the authors suggested that the pattern of change associated
with successful psychotherapy was movement from unconscious,
unrecognized, transference manifestations in affect and behavior to a
more conscious, and therefore more manageable, transference reaction.
This seems to echo, in quantitative fonm, Freud's original rule of
psychoanalysis 'to make the unconscious conscious'. Graff and
luborsky's opinion was that the patients were no more disturbed, but
that the transference theme remained evident in a context of increased
mastery. The two less successful cases showed fairly parallel
transference and resistance curves without the decrease in resistance
characteristic of the more successful analyses, suggesting that the
transference reactions could not be worked through as resistance to
analytic work in general renained so high.

Rawn (1981) made several comments on this study. First, he cited
his own 1958 paper as being the pioneer attempt to quantitatively
examine transference and resistance variables. More important, he

noted that his own original results were identical‘with those of Graff

and Luborsky. In both studies, the presence of heightened resistance

11

coupled with negative transference was inevitably found to lead to
unsuccessful therapeutic outcome.

Gill and Hoffman (1982) had judges code transcripts of audio-
recorded psychotherapy sessions for evidence of 'experience of the
(therapeutic) relationship', or transference manifestations and the
ways in which they were handled in therapy. These authors differed
somewhat from Graff and Luborsky in their explicit.concern‘with both
preconscious aspects of transference as well as the actual behavior of
the therapist expected to impact on the transference. This attention
to the significance of tie trerapists' actual statements and behaviors
brings this study conceptually close to a number of interpersonally-
oriented investigations (Crowder, 1972; Dietzel & Abeles, 1975:
Mueller, 1969a, b; Mueller & Dilling, 1968).

It is because of transference that a patient is sensitive to

certain facets of the 'reality' of the therapist as opposed

to others which he may ignore....The transference may

manifest itself in selective attention to some particular

aspect of the complex reality of the analyst's behavior, or

in some fixed mode of adaptation to that response. But it

need not and, indeed, usually does not distort reality in any

simple sense. (Gill & Hoffman, 1982, p. 14)

In this report of their ongoing work, Gill and Heffman claimed
impressive reliability data for ratings of transference and expressed

their joint certainty that a real clinical phenomenon, with crucial

implications for future research, had been measured.

Developmental Aspects of Internal Representations

 

Several studies report changes in identification patterns
occurring with successful psychotherapy. Groesbeck (1978) considered

this issue from a Jungian perspective, believing that the natural

12

evolution of interpretations of psychological types in the analysis of
the transference was related to later individuation. In support, he
cited Paulsen, "Because it transcends them, the term transference can
legitimately be distinguished from the term projection and used to
designate the successive stages of the individuation process as it
occurs in relation to the analyst" (Groesbeck, 1978, p. 44).

In clinical settings, more disturbed clinical populations have
been found to display identifications characteristic of earlier stages
of psychosocial development (Cava & Rausch, 1952), and quantitatively
less intense (Sopchak, 1952) than those of nonsymptomatic individuals.
Cava and Rausch observed conflict in areas related to identification in
psychoanalytic theory to be predictably related to twelfthrgrade boys'
perceptions of their fathers. Less conflicted boys tended to identify
more strongly with their fathers, and significant (p < .10) differences
discriminated between more— and less-conflicted boys in castration
anxiety, total identification, Oedipal intensity, and identification
differences. Sopchak asked 108 students to complete MMPIs for
themselves, and as they thought their father, their mother, and 'most
people' would respond. In this nonclinical sample, identification with
parents was correlated with good judgement, while abnormality (a
configuration of normatively deviant MMPI scores) correlated with lack
of identification with either parent or with people in general.
Better-adjusted male subjects tended to identify more strongly with
fathers than with mothers, but for women both positive identification
*with the mother and failure to identify with the father were positively
correlated with abnormality. Another MMPI study with a nonclinical

population (Beier & Ratzenburg, 1953) found significant gender

l3

differences in identifications with mother and father, with each gender
identifying more readily with.the parent of the same sex.

McDonald (1965) studied hostility and identifications in a sample
of single white pregnant.women in the seventh month of pregnancy. MMPI
scores were used as indexes of ego control patterns and the
Interpersonal Check List (ICIJ was used to measure representations of
the self, mother, and father. Significant differences were found
between members of four ego-control groups (expressors, repressors,
sensitizers, or expressor-sensitizers), and attribution of hostility to
parents was related to large differences between self-representation
and representations of parents: the less the subject identified with a
given parent, the more hostility was attributed to that parent.

Krieger and Wbrchel (1960) investigated patterns of parental
identifications in a "neurotic' university clinical population, a
homosexual clinical population, and a "normal" heterosexual non-
clinical pOpulation using a Orsort.technique to elicit representations
of the self, the ideal self, the mother, and the father. Normals
identified nearly equally with both parents, and neurotics tended to
identify significantly more with the opposite-sexed parent. Male
neurotics' ideal self was significantly more highly correlated with
mother than with father, while female neurotics' ideal self was more
like the father than the mother. Hbmosexual males' ideal selves
correlated more highly‘with the father than with the mother.

Mueller (1966) reported correlations between university students'
semantic differential ratings for parents and self and anxiety scores
from the Taylor Manifest Anxiety Scale. Highly anxious male subjects

described themselves and their parents as more passive, impotent, and

14

poorly evaluated than the low anxious group. Low anxious males
regarded their fathers significantly more highly than themselves, while
the highly anxious males regarded their fathers as much like
themselves. High anxious females attributed more passivity to their
fathers than did the low anxious females. Mueller speculated that
while low anxious individuals seemed to have discrete representations
of the father and themselves, more anxious individuals' self-
representations were patterned after the denigrated father. less
anxious females particularly depicted a psychological sense of distance
or differentiation between themselves and their fathers. Results also
suggested that females' responses carried better validity in studies of
psychotherapy than males' as females seemed less affected by
situational variables such as anxiety.

Schrier (1953) investigated the relationship between outcome of
short-term psychotherapy and aspects of the therapeutic relationship,
and in particular the patient's identification with the therapist. He
cited Fuchs, ”Each identification is therefore a monument of an object
relation, and our character, being itself composed of identifications,
contains the history of our object relations" (Schrier, 1953, p. 586).
Schrier felt this was similar to Freud's belief that identification, an
elementary capacity to form an emotional relationship, was a function
of the capacity for object relationships. His study of nine
hospitalized short-term psychotherapy patients yielded highly
significant (p < .001) relationships between variables measuring
identification, a positive therapeutic relationship, and positive

outcome.

15

weiner (1982) discussed theoretical issues relevant to
identification in psychotherapy. Beginning with Freud's concept that
identification might be the earliest expression.of an emotional
relationship, he suggested that growing identification with highly
valued persons might be one of the most important processes taking
place in psychotherapy, concluding that some degree of identification
‘with the therapist was necessary for eventual individuation in
psychological development.
MEX

The preceding studies indicate the importance of internal
representations of others. Such representations have been shown to
highlight pathological modes of regarding others, and change as a
result of successful psychotherapy. Given the assumption that
identification is the one of the earliest forms of an object
relationship, another group of studies involving object relations, a
number of which utilize the Rorschach, have relevance here in terms of
placing differentiation as a developmental process naturally succeeding
identification, and therefore representing movement towards integration

of internal representations and psychological development.

Theoretical Approaches to Object Relations

 

and Recent Applications of Projective Techniques

 

Recent developments (Blatt & Lerner, 1983) in psychoanalytic
theory have stressed a less ”mechanistic" and increasingly interactive
ViGW'Of personality development, in which "ongoing object relations are
internalized as part of the development of processes of representation

and thereby become part of the personality structure" (p. 8). The

16

process of internalizing significant interpersonal relationships is
believed to lead to the construction of cognitive-affective schemes
which thereafter govern the individual's construction of the world and,
most important, representations of the self in interaction with others.
In cases where these existing psychological constructions have
interfered with.adjustment, the therapeutic relationship has appeared
to be the natural vehicle for the development of more adaptive

representations of self and others.

If the internalization of object relations results in the
formation of psychic structures during normal development,
then the internalization of significant interactions between
the patient and the therapist must play an important role in
the therapeutic process. The therapist becomes available as
a new object by eliminating, step by step, the transference
distortions which interfere with the establishment of new
object relationships (Loewald, 1960). And it is the
internalization of new and relatively undistorted relations
‘with the therapist which leads to therapeutic change. But
the clinical examples presented in this paper suggest that
there are different types of distortions in the context and
structure of the representational world in various forms of
psychopathology. These differences should be expressed in
the therapeutic process in the nature of the transference and
countertransference. The quality of the object
representations prior to beginning therapy should provide
understanding about the potential transference issues that
may emerge in therapy. It should also provide information
about the types of changes that need to occur in the
individual's representational world if the patient is to
progress toward higher developmental levels. Different types
of changes may occur in different types of patients depending
upon the disruptions of their representational world when
they begin treatment. (p. 26)

Lerner (1983) wrote, "The development of representations is
regarded as the sine qua non of ego development and adaptation"
(p. 314). He mentioned the growing interest in studies of

interpersonal processes relating to internalized representations and

anticipated transference paradigms to counter criticisms of the ego

17

psychological model as being conceptually distant from the experiential
world. Lerner, among others, applied the Blatt, Brennais, SChimek, and
Glick (1976) scale to Rorschach protocols collected fromlclinical
populations. The scale was based on wernerian principles and was
designed for assessment of the concept of the object on the Rorschach.
In the original article, Blatt et al. reported that formal properties
of human responses on the Rorschach showed consistent changes with the
developmental level and degree of psychopathology of the patient.
Developmentally advanced responses,‘with higher levels of
differentiation, integration, and articulation, were found to occur
with increased psychological development, and to diminish in direct
proportion to psychological disturbance.

Spear and Lapidus (1981) found significant differences between
groups of obsessive/paranoid borderline, hysterical/impulsive
borderline, and undifferentiated schizophrenic patients in their
quality of object relations as measured by the Blatt et a1. scale.
Schwager and Spear (1981) compared pre- and post-therapy protocols of
ten chronic schizophrenic patients who had been treated in a long-term
psychiatric inpatient unit. Different directional shifts were found in
change criteria for paranoid and for nonparanoid undifferentiated
patients. The paranoid group showed a significant decrease in overall
levels of cognitive-structural differentiation while the nonparanoid
group showed the expected significant increase in capacity for object
differentiation. Lerner (1983) used this scale to illustrate clinical
change following intensive psychoanalytic psychotherapy of a sixteen-
year—old girl and demonstrated dramatic changes in her human

representations on the Rorschach which were consistent‘with progressive

18

transference manifestations. He attributed these changes to the
effectiveness of therapy in aiding the patient to integrate part-object
relations into whole object relations, promoting object constancy, and
achieving an integrated self-concept.

Ryan and Bell (1984) investigated changes in representations of
self and others over an extended period of recovery fromlpsychosis in
order to further test assumptions of object-relations views of
psychopathology. They used a scale which assessed object relations
through the patient's early memories, measuring concepts similar to
those of Blatt et a1. (1976). Level of object representation was found
to increase significantly from tie psychotic level at initial
hospitalization to the level at postrtreatment followup. PatientS'with
low object relations scores at followup were likely to be
rehospitalized at a rate almost twice as high as that of individuals
whose followup object relations scores were greater, supporting the
hypothesis that object relations scores were indicators of
psychopathology.

Tuber (1983) used both Rorschach structural and object relations
variables to predict later adjustment in children treated at a
residential facility. He found object relations scores to be
significantly related to incidence of hospitalization at followup, with
highest scorers most successful at avoiding later institutionalization.
Tuber suggested that this constitutes evidence that "these
individual's' early internalized representations of others were benign
enough to enhance the likelihood of generating equally benign

relationships later in life" (p. 383).

 

l9

Lerner and St. Peter (1984) looked at developmental patterns of
Object relations across a wide range of psychopathology, comparing
Rorschach percepts scored according to the Blatt et al. scale from
groups of patients falling within neurotic, borderline, and
schizophrenic categories. Further supporting the developmental basis
of this concept, they found "a steadily increasing relationship between
a person's quality of reality testing (defined by response accuracy),
developmental level of the concept of the object (developmental index)
and psychopathology" (p. 88). As predicted, the most impaired patients
presented the most poorly differentiated, articulated, and integrated
figures and the least impaired patients presented the most well
perceived responses.

Several underlying concepts link the above studies. First, it
seems that processes of identification and differentiation are crucial
to the development in capacity for object relationships, and may be
demonstrated in the resultant quality of object representations. The
level of development of these representations varies directlylwith
level of psychopathology and has been reliably assessed by means of
projective techniques, such as the Rorschach, The Blatt et al. scale
appears to be the most widely used, but similar scales (Krohn & Mayman,
1974; Urist, 1977; Urist & Shill, 1982) have been used‘with.success as
well. Athey, Fleischer, and Coyne (1980) accounted for the popularity
of the Blatt et al. scale by the fact that "it is the only empirical
attempt to address the relationship between thought organization and
object relations in any systematic sense, illustrating some degree of
correspondence between group thought disorder and developmental

qualities of object representation" (p. 276-277). Second, these

20

otherwise relatively hidden internal representations become apparent
when reenacted in the relationship with the therapist, both in
interpersonal behavior and in the perceptions and distortions of the
transference itself, and become demonstrably more developmentally
advanced at the conclusion of successful psychotherapy.

Classical psychoanalysis, the ego—psychology/object relations
perspective, and the interpersonal perspective have all made
contributions in trese areas, reaching remarkably similar conclusions.
Freud himself directly addressed the existence of both the reality-
based interpersonal relationship and transference proper in his 1937
discussion of "a certain man" and the man's analyst. Ernest Jpnes
later identified the analyst and analysand as Freud himself and
Ferenczi. Freud discussed the advantages and disadvantages of

interpreting the negative transference:

To activate it would certainly have required some unfriendly
piece of behavior in reality on the analyst's part.
Furthermore, he (the analyst) added, not every good relation
between an analyst and his subject during and after analysis
was to be regarded as a transference; there were also
friendly relations which were based on reality and which
proved to be viable. (p. 222)

Along similar lines, Gill (1982) wrote:

It may be that the recognition that all resistance
interpretations are interpersonal interpretations has been
obscured by tie fear that to do so would comnit oneself to an
interpersonal rather than an intrapsychic View of human
psychology. But one need not be driven into such a
dichotomy....the ultimate outcome of the interaction of
givens and environmental factors is the intrapsychic
formation which becomes interpersonally expressed in human
interaction. (p. 41)

21

Interpersonal Theories and Research on Psychotherapy

 

A.number of studies investigating clients' internal
representations of self and others, and changes in such representations
subsequent to therapy, have used the Interpersonal Check List (ICL).
Clark and Taulbee (1981) felt that although the ICL had been widely
used in the evaluation of psychotherapy outcome, its potential for
assessment of interpersonal perceptions and behavior had not been
tapped. Leary and Harvey (1956) used the ICL to study direction of
personality change in psychotherapy along the two major ICL parameters
of dominance-submission and hostility-affiliation. Men who had rated
themselves as 'hostile' or 'weak' before treatment were significantly
more likely to change than were women who rated themselves the same
way. HOwever, men who were 'conventional' and 'bland' before therapy
had a poorer prognosis than women rated that.way.

Mueller and Dilling (1968) scored the actual interactions of
nineteen therapist-client pairs according to ICL categories in an
investigation of reciprocal effects of therapist-client interactions,
those "attempts on the part of each person to establish an emotional
state in the interaction which tends to elicit a predictable response
from the other person" (p. 282). Ample evidence for the existence of
such interactions was found, particularly in competitive—hostile
therapist behavior which.was significantly correlated with competitive-
hostile client behavior as well as subtle passive-resistant behavior by
the client. Supportive-interpretative therapist behavior was
positively correlated with support-seeking client behavior. A second
paper (Mueller & Dilling, 1969) proposed using the same interpersonal

system to measure similarities between the client's interaction‘with

22

the therapist and his or her descriptions of other interpersonal
relationships, tracking interpersonal themes and observing in.what.ways
mention of such themes paralleled, preceded, or followed client-
therapist interactions. The authors stated that this concept was
basically an operational definition of transference and that "this
weaving together of past and present interactions, constantly using one
to explain the other, is the way that a therapeutic relationship
develops; so our system‘would seem to have clinical validity" (p. 51).

Lorr, McNair, Michaux, and Raskin (1962) used four subscales of
groupings of ICL items, and several other personality and symptom
measures to study the effect of the frequency of psychotherapy on
therapeutic outcome. ICL scores documented improvement after eight
months of treatment. One year after initiation of psychotherapy,
reduction in severity of distress was associated with general gains in
interpersonal relations as measured on the ICL. The pattern of changes
suggested to the authors that successful psychotherapy followed a
natural progression in which symptoms were initially reduced, followed
by subtler gains in the capacity for interpersonal relations.

Kiesler (1982) felt that interpersonal theorists had paid
insufficient attention of psychotherapy issues, and extensively
reviewed the various interpersonal measures with particular emphasis to
the concept of complementarity. He suggested that since
conplementarity stabilized and restricted the range of interpersonal
behavior, therapists could produce maximal change by providing
acomplementary responses to clients, thereby increasing the client's
repertoire of interpersonal experiences. He warned that the strong

anxiety engendered by consistently responding in an acomplementary

 

23

fashion might strain the limits of a fragile therapeutic alliance, and
suggested a two-stage approach to psychotherapy. In the first,
'hooked,‘ stage, the therapist would experience being virtually unable
to respond except in a complenentary fashion to a client who had many
years of experience in interpersonal manuevers designed (unconsciously)
to elicit just such responses in others. The second, 'disengagement',
phase would be marked by the therapist's gradual regaining of the
ability to respond freely in both.complementary and acomplementary
ways.

Stages of Psychotherapy

 

Several other theorists have proposed that successful
psychotherapy progresses in fairly discrete, predictable stages.
Dietzel and Abeles (1975) found a three-stage pattern of complementary
behaviors in their study of twenty psychotherapy cases. Successful and
unsuccessful therapist-client dyads were marked by high levels of
complementarity in the initial stages of therapy. More disturbed
clients were skilled at eliciting more complementarity at this stage,
but this was not related to eventual therapy outcome. In the middle
stage of therapy, successful dyads showed significantly lower levels of
complementarity than unsuccessful ones, indicating that "to facilitate
client change, the therapist must avoid a confirming, security-
enhancing stance to client elicitations, and therapists who become
caught up in, or ensnared by, the client's constricted elicitations
will reinforce these maladjusted behavioral modes" (p. 271).

Successful dyads continued to exhibit less complementarity in the late,

or termination, stage of therapy.

24

Crowder (1972) looked at the relationship between interpersonally
scored transference and countertransference in therapist-client dyads
and psychotherapy outcome in early, middle, and late stages of therapy.
countertransference in early interviews discriminated between
successful and unsuccessful dyads. Clients who improved were more
support-seeking in early and middle, but not in late, therapy. They
tended to be significantly more hostile-competitive and less passive-
resistant and supportive-interpretive early in therapy. In mid-
therapy, nonimproving clients continued to express more passive-
resistant behaviors. Crowder concluded that the quality of
interpersonally expressed transference and countertransference
reactions were indicative of the outcome of psychotherapy. Successful
dyads seemed to have established an alliance in which the client.was
willing to flexibly experiment with therapist-initiated new
interpersonal actions, while unsuccessful dyads appeared to become
bogged down in manuevers which maintained the status quo.

Tracey and Ray (1984) found that therapy moved progressively
through three stages of complementarity. The first stage continued
until an initial high level of complementarity was reached. Upon
reaching this point, the middle stage, marked by a low level of
complementarity, began. At this point, therapist and client appeared
to struggle to both maintain an alliance and to work through persistent
conflicts. In the last stage, complementarity reached a higher level,
indicating ”a new, more realistic consensus between the participants as
to what form the relationship should take" (p. 25).

Greenspan and‘Wieder (1984) described the stages of the

therapeutic relationship as:

25

Becoming invested in an emotional sense in the person of the

therapist (as well as the program of therapy), engaging in an

exchange of signals that are purposeful and organized (as a

minimum basis for embarking on the techniques generic to the

particular therapy): learning to tolerate even minimum

discomfort generic to the particular treatment and continuing

at higher levels of capacity to observe various behavioral

and emotional patterns in the context of a stable affective

therapeutic relationship and then engaging in.various ways of

altering trese patterns. (p. 6)

Eimmmuapgifliterature

The most immediate question regarding psychotherapy outcome
research was once whether or not psychotherapy could be proven to have
an effect which was demonstrably longer-lasting and more substantial
than other forms of social reinforcement such as friendship, advice-
giving, and benign attention. This question seems to have been
answered in the affirmative with enough consistency to satisfy most
critics. One major arena of current process and outcome studies has
been the client's general capacity to form relationships and the
qualities of such relationships, including elements of identifications
and projections. Silverman and Weinberger (1985) believed that the
activation of symbiotic fantasies accounted for much of the power of
therapy to promote change. Clinical researchers have been especially
interested in the influence of internal representations on
interpersonal behavior and on the transference itself as therapy
progresses, and in transfbrmations of representations and the quality
of the transference throughout the course of therapy.

Transference has frequently been conceptualized as the formation
of an internal representation of the therapist, composed of elements of
both fantasy and reality, the way in which the therapist is perceived

as similar to significant persons in the client's early life, and the

26

tendency of the client to attempt to recreate old patterns of behavior
with these significant persons in the therapeutic relationship. The
two conditions marking the presence of transference and distinguishing
it from other interpersonal behaviors are the repetitiveness and
inappropriateness of the behavior (Crowder & Feltoon, 1973). crowder
and Feltoon's extensive review of the literature yielded a number of
studies documenting significant perceived similarity between childrens'
perceptions of teachers and mothers, between 'the most important
parental figure' and the therapist for group therapy participants,
increasing similarity between parent and therapist with continuing
psychotherapy, and inconsistent similarity between the sex of the
therapist and that of the parent perceived as most similar to the
therapist. Although it seems generally accepted that the transference
may take on either a predominantly maternal or paternal quality, and
that during long-term psychoanalysis oscillation between these two
occurs, there is no consensus regarding immediate precipitants for the
therapist becoming perceived as predominantly like mother or father,
and little on the forms the quality of the transference takes in
shorter-term psychodynamic psychotherapy. As most research on
psychoanalytic concepts is carried out in psychoanalytic settings,
there is a need to test these same concepts in the more prevalent
setting of moderate-tenm psychodynamic or psychoanalytically-oriented
psychotherapy.

The present study will focus on two major aspects of the
therapeutic relationship and therapy outcome: changes in internal
representations of self and others throughout therapy with particular

attention to the relationship of such changes to initial, final, and

27

total amounts of change in personality functioning, and the quality of
the transference itself. Quality of the transference refers to
progressive distortions in the perception of the therapist in ways
which recapitulate earlier significant relationships, and the effect on
these distortions of both initial level of client personality

functioning and therapist gender.

CHAPTER 2

The purpose of this study was to illustrate changes in the
client's perception of the therapist across three stages of
psychotherapy, and to link transfermations in internal representations
of self, therapist, and parents with general growth in the capacity for
object relationships and more adaptive personality functioning.

l. Representations of the self, the therapist, the mother, and
the father will change significantly from early therapy to mid-therapy
and from mid-therapy to late therapy.

This hypothesis was tested by scoring the ICL rating of self,
therapist, mother, and father at each stage of therapy for dominance,
affiliation and intensity (distance from the midpoint of the ICL
circumplex). T-tests were then employed to determine whether a
significant difference existed between ratings at consecutive phases of
therapy.

2. In mid-therapy the representation.of the therapist will
become significantly more similar to the representation of one parent
than it had been in early therapy.

The theoretical basis for this prediction was that because both
client and therapist were aware of the time-limited nature of the

therapy, the transference would take on a predominantly maternal or

28

29

paternal quality, resulting from the particular developmental conflict
which the client was unconsciously re—enacting in the therapeutic
relationship, stemming fromlthe client's level of psychosocial
development. Although the transference is conceptualized as containing
residues of relationships‘with both mother and father, which
qualitatively shift from moment to moment, it was hypothesized that in
this mode of therapy the therapist was perceived as predominantly like
mother or like father, which was illustrated in shifts of ICL ratings
of the therapist from the more objective perceptions of the early
therapy stage to the more highly transferentially charged ratings of
the middle therapy stage.

Subjects were classified according to which parent the therapist
was perceived as more similar in mid-therapy. Tbtests were used to
determine whether the geometric distance between ICL scores for
therapist and that parent lessened significantly from early therapy to
mid-therapy.

3. Mid-therapy perception of the therapist as similar to one
parent will be related to the client's initial level of personality
functioning.

A. Psychological disturbance is predicted to be positively
correlated with perception of the therapist as similar to the mother.

B. Psychological disturbance is predicted to be negatively
correlated with perception of the therapist as similar to the father.

C. Higher-functioning clients are predicted to view the
therapist as more similar to the parent of the same sex as the

therapist.

30

D. Lower-functioning clients are expected to perceive the
therapist, regardless of gender, as more like the mother than the
father.

Level of personality functioning, or psychological disturbance,
was measured by subjects' number of Rorschach criteria scores at
initial testing. Subjects were then ranked on the number of criteria
and grouped into higher—functioning (fewer criteria) and lower-
functioning (greater number of criteria) subgroups. Hypotheses A and B
were tested by using Pearson correlation coefficients to determine the
relationship between number of criteria and distance between parents'
and therapist's ICL scores. C and D were tested by employing the phi
coefficient to measure the relationship between therapist gender and
gender of the parent to whom the therapist is viewed as more similar.

4. In late therapy, the representation of the therapist will
become significantly less similar than it had in mid-therapy to the
parental representation it had been more similar to in mid-therapy.

This increase in distance between therapist and parental
representations was believed to demonstrate a resolution of some
transferential issues and growing ability to view therapist, as well as
others, in a more objective fashion.

This hypothesis was tested by means of T-tests to determine
whether a significant difference existed between the distance between
the representation of the therapist and the mother at mid-therapy and
at late therapy for subjects who perceived the therapist as similar to
the mother in mid-therapy. Identical operations were performed for

therapist and father representations in mid-therapy and late therapy

31

for subjects perceiving the therapist as like the father in mid-
therapy.

The following hypotheses regarding the outcome of therapy were
also tested:

5. The total amount of movement in representations of self,
therapist, mother, and father - that is, the sum of the deviations
between ICL representations at early and mid-therapy and mid- and late
therapy will be positively correlated with changes in personality
functioning measured on the Rorschach from early to late therapy.

The total amount of movement in each representation was measured
by calculating the geometric distance between the representation at
each testing and summing these distances. This score was then
correlated using Pearson two-tailed correlations coefficients, with the
following measures of personality functioning: The difference between
hostility scores, anxiety scores, mean object representation scores,
and criteria for personality functioning scores between Rorschachs
administered early and late in therapy.

6. Significant differences on the Developmental Analysis of the
Concept of the Object scales will be found on the overall global scale
and each of the six subscales between initial and late Rorschach
protocols.

Paired T—tests were employed separately for the global and each of
the subscales to test this hypothesis.

7. The number of representations at “moderate" intensity on the
ICL in late therapy will be positively correlated with the Rorschach

overall mean object relations score in late therapy.

32

This hypothesis was tested with the Pearson correlation

coefficient.

 

CHAPTERB

Subjects

Sixteen women receiving individual psychotherapy services at the
Michigan State University Counseling Center during the 1984-85 academic
year served as subjects for this study. Therapists were senior staff,
interns, and practicum students at the Counseling Center to whom
clients had been assigned according to standard center procedures. The
length of therapy contact.was at least six months, but not more than
one academic year. All female students receiving psychotherapy at the
Olin branch of the Counseling Center, who were expected to remain in
therapy for at least six months, and were working with the present
therapist for the first time, were asked to participate.

Twenty-four subjects initially agreed to participate. TWo
subjects terminated therapy within the first three weeks, before any
data was collected. Three subjects terminated therapy before the sixth
session, so that only initial test data was available. Three more
subjects terminated prior to the sixteenth session, so that only
initial and mid-therapy data was available.

Subjects and Therapists

 

Subjects' ages ranged from twenty to thirty-four, with a mean age

at initiation of psychotherapy of twenty—five. Fourteen had never

33

34

married, and two were divorced. Three subjects, including the two who
had been divorced, had one or two children between the ages of two and
ten. Six subjects were currently enrolled in master's or doctoral
level graduate programs, and the other ten were undergraduates with
sophomore, junior, and senior status.

One subject had a history of substance abuse that included
treatment at a residential facility. Seven subjects had received
previous outpatient treatment of at least three months' duration. A
number of subjects had past histories of significant separations,
losses, and traumas. Family histories included two subjects whose
fathers were survivors of concentration camps. One subject's parents
had separated and later reconciled, but six subjects' parents had
divorced, with five of the divorces taking place in the subjects'
adolescences. One subject's father had permanently abandoned herself
and her mother within a short time of her birth.

Family psychiatric histories included two chronically mentally ill
family members who had been hospitalized over extended periods of time.
One subject's mother, and one's mother and sister had made suicidal
gestures and attempts, one had herself made a suicide attempt, and one
had an alcoholic parent. Past traumatic events for subjects included
two incidences of physical abuse, two cases of sexual abuse by the
father, two cases of childhood sexual abuse by non-family members, and
one case of date rape in young adulthood.

Presenting problems in fall 1984 nearly unanimously centered
around general dissatisfaction with significant relationships,
particularly'with parents and boyfriends. Most subjects complained of

symptomatic depression and anxiety, variously described as

 

35

uncontrollable mood swings, ”being manic-depressive" by self-
description rather than clinical diagnosis, and a number of different
descriptions of a sense of dramatically increasing tension, stress, and
hopelessness which had been unresponsive to the well—meaning effbrts at
intervention made by family and friends. A few identified academic
concerns as contributing to other problems, one had a psychologically
related gastrointestinal illness, and one pinpointed the process of
emotional separation from the family of origin as causing her distress.

A total of ten therapists worked with the sixteen clients. Four
therapists were female, and six were male, resulting in seven same-sex
therapist-client dyads and nine opposite-sex therapist-client dyads.
Three therapists (one female, two male) were in their first year of
practicum, one male was in his second year of practicum, three female
and two male therapists were Counseling Center interns, and one male
therapist was a senior member of the Counseling Center staff.

Premature Terminators

 

Post-hoc analyses were carried out to determine whether any
significant differences existed between those subjects who prematurely
terminated therapy and those who did not. Prematurely terminating
subjects' ages ranged from nineteen to twenty-five, with a mean age of
21.7. Five had never married, one was currently married, and none had
children. All were undergraduates, although one had already completed
a first bachelor's degree at a university in another country. One had
a past history of psychiatric hospitalization for a previous suicide
attempt, and three had histories of previous individual and family
therapy outpatient treatment. Two had made past suicide attempts

without subsequent hospitalization. Separations and losses for

36

premature terminators included one subject whose parents separated and
reconciled, and one whose father died during her adolescence. Family
psychiatric histories included one chronically alcoholic parent, one
sibling whose family was reacting with turmoil to his homosexuality,
and one father who, while married to the subject's mother, was
flagrantly involved with tie subject's high school classmate.
Presenting problems were identical to those of subjects who completed
therapy, with the addition of one subject who reported binge eating.

The subjects were seen by male therapists, and four saw female
therapists. All of the therapists also had clients who completed this
study. Two were in the first year of practicum, one was in the second
year, and three were interns. Table 1 gives descriptive statistics for
both premature terminators and completers on all initial testing
variables as well as the magnitude of the difference between mean
scores for the groups.

No significant differences were noted between initial ICL scores.
Premature terminators had significantly higher mean object relations
scores, and significantly more Rorschach FC scores, than completers.
There was also a trend toward terminators producing more reflection
responses and a lower proportion of good quality pure form responses.

Instruments

 

The central data for this study consisted of variables drawn from
Rorschach protocols administered according to the Exner (1974) system,
the Interpersonal Checklist (LaForge & Suczek, 1955; Leary, 1956), and
the Helping Relationship Questionnaire (Luborsky, 1984).

The dynamic processes involved in tle production of each Rorschach

response were elaborated by Rapaport, Gill & Schafer (1968) ,

Table 1

37

Thtests of Differences between PrematurelyaTerminatipg Clients and
Clients Completipg Therapy on Initial Vafiables

 

 

 

 

 

 

Clients Clients not

completing completing

therapy therapy

N=16 Nel6
variable Mean SD Mean SD '3
Early therapy ICL:
Self dominance 49.75 7.43 49.83 17.23 -.02
Self affiliation 51.81 6.96 53.50 14.35 -.38
Therapist dominance 61.38 4.91 62.50 8.02 -.40
Therapist affiliation 49.94 4.02 51.17 9.37 -.44
Mother dominance 57.38 6.62 63.00 12.46 -1.39
Mother affiliation 46.19 10.26 48.67 15.83 -.44
Father dominance 60.20 6.61 63.17 12.73 -.71
Father affiliate 37.33 7.30 40.00 9.76 -.69
Early therapy Rorschach:
Rorschach criteria 7.31 2.44 6.67 1.9 .66
Mean object representation 9.24 1.59 11.45 2.87 -2.32*
Hostility content 3.13 2.53 5.33 3.45 -1.66
Anxiety content 4.75 5.69 3.33 2.50 .81
R 24.81 10.20 26.67 8.21 -.44
Location:
W’ 11.69 7.24 15.83 7.19 -1.20
D 9.31 6.97 7.17 4.40 .49
Dd 3.56 3.48 3.67 2.16 -.08
S 2.31 2.21 4.17 2.93 .20
Determinants:
FM 3.06 2.91 2.83 2.04 .21
m 1.06 1.29 2.00 1.55 -1.44
M 5.44 3.48 5.67 1.63 -.21
FC 2.56 1.71 4.50 1.38 -2.74**
CF + C 2.12 2.03 2.67 2.94 -.49
FC' + C'F + C' .75 1.00 1.33 1.51 -1.06
FT + TF + F 1.19 1.80 2.17 2.64 -1.00
FV + VF +‘V .31 .48 .50 1.23 -.53
FY + YF + Y 3.31 3.42 3.00 3.69 .19
Fr + rF .31 .87 1.33 1.51 -2.00*
(2) 7.25 3.17 6.50 1.05 .83
FD 1.19 1.42 .50 .55 1.63
F 8.50 4.34 6.67 3.62 1.00

(table continues)

 

38

 

 

 

 

Clients Clients not

completing completing

therapy therapy

N=16 N:16
variable Mean SD Mean SD .E
Percentages & ratios:
E 9.44 5.55 11.00 3.08 -.83
es 9.75 6.02 11.33 5.50 -.59
D score -1.81 1.94 -1.83 .41 -.04
F+% .60 .20 .46 .12 2.00*
X+% .64 .13 .55 .09 1.74
X-% .05 .05 .04 .04 .59
3r + (2)/ R .34 .13 .38 .69 -.92
Lambda .64 .52 .39 .19 .26
S-constellation 3.34 1.93 3.50 1.38 -.17
Afr .65 .41 .52 .16 1.11
* p < .10.

** E < .05.

39

we see three prominent phases in the process of the coming about

of a Rorschach response: in the first phase, the salient

perceptual features of the blot initiate the associative process:
in the second, this process pushes beyond these partial perceptual
impressions and effects a more or less intensive organization and
elaboration of the inkblot; in the third, the perceptual
potentialities and limitations of the inkblot act as a regulating
reality for the associative process itself. Clearly, then, it
would not be correct to reason that the Rorschach response is to
be considered mainly either a perceptual product or one of free
association. Either view“would fail to reflect the cogwheeling of
the progress of perceptual organization with the associative

process. (p. 276)

The ICL, as summarized by Keisler (1983), is an interpersonal
measurement instrument. The subject is asked to indicate on,a written
form which, of a broad array of 128 attributes, are considered true for
the person being rated. A formula is then applied to the attributions
considered descriptive of the ratee, culminating in a single score
being assigned to the ratee which describes their position in "a two-
dimensional Euclidian space reflecting the joint action of two basic
interpersonal dimensions or motivations, almost universally designated
Control and Affiliation, These dimensions define, respectively, the
vertical and horizontal axes of the circle” (Kiesler, 1983, p. 186).
The final score, therefore, is a single numerical value which is the
intersection of standard scores on the two axis dimensions, reflecting
the rater's perception of the ratee. The degree of 'intensity'
(moderate versus extreme) is determined by the distance of this
intersection from the midpoint of the circumplex. Ratings falling
within one standard deviation (mean=50, SD=10) are considered moderate,
while ratings toward the periphery of the circumplex in any direction
are considered extreme, implying rigidity of interpersonal

behavior.

 

 

40

The Helping Relationship Questionnaire is a short, self-report
measure designed by Luborsky (1984) to access the client's view of his
or her experience in psychotl'erapy. This questionnaire consists of 11
statements answered on a Likert-style scale from ore (strongly
disagree) to six (strongly agree) describing the client's views of how
much change has occurred in tlerapy and the relationship with the
therapist, and includes two open-ended questions about areas in which
the client believes he or she has improved and not improved. Luborsky
has used this questionnaire as a part of his larger investigation into
both the degree to which supportive-expressive psychotherapy was
performed by the psychotherapist and to measure the patient's
experience of the helping alliance. All of these scales were presented

in his 1984 book, which was subtitled, A manual for supportive-

 

expressive treatment. Luborsky used the earliest forms of the manual
to train his psychotherapy supervisees in the Department of Psychiatry
at the University of Pennsylvania.

Procedures

 

Therapists were informed of the study and their participation was
requested at initial staff and practicum meetings in September. During
the first or second psychotlerapy session with clients meeting the
inclusion criteria, the therapist read a short description of the study
from the Therapist Information Form and asked the client if she would
be interested in participating. Clients expressing interest were given
the Initial Contact Sheet to complete. The therapist then returned
this form to my mailbox.

I called the client within two days and made an initial

appointment for testing. Testing during the early phase of therapy

41

took place as soon after the first therapy session as possible, but was
always completed prior to the fifth therapy session. I administered
the majority of the early, mid- and late therapy testing. To minimize
potential bias, an advanced graduate student, experienced in the
Rorschach, administered early and late therapy tests to my own clients.
Another advanced graduate student carried out three early therapy
testing sessions.

At the first testing, the study was described to the subject,
stressing that there would be three test sessions and that
participation had no bearing whatsoever on receiving counseling
services. The subject first signed the consent form, gave some basic
demographic infonmation, and was administered the Rorschach. When the
Rorschach had been completed, the subject was presented with the ICL
and offered the choice of completing it at once or completing it at
home and returning it in its self-addressed, stamped envelope. The
subject was instructed to read the list of statements, marking those
which she believed were true for her and leaving blank those which were
not true. After answering the statements for her perception of
herself, she was instructed to read through the list again, answering
them for her therapist, and to follow identical procedures for her
mother and for her father.

Participating therapists were periodically asked to supply the
actual dates of therapy sessions that had taken place with research
subjects. Mid-therapy testing was scheduled to take place between the
tenth and thirteenth session. The subject was called after the ninth

session, and a second appointment was made. During this session, the

42

ICL‘was administered in the same fashion as it has been given in the
previous session.

The final test administration was scheduled to take place as close
as possible to the final therapy session, and in each case occurred
‘within the last four sessions of psychotherapy. Subjects were called
and a final appointment.was made. The Rorschach was once again
administered, with the added instruction that the subject should report
what she saw, regardless of whether it.was the same or different than
what was seen at initial testing. The IC1.was administered next, in
standard fashion, followed by the Helping Relationship Questionnaire.

Scoring_Procedures

 

Rorschach. Each Rorschach protocol was scored for structural
variables according to Exner (1974). Eighteen of these variables and
ratios and two scores added later by Exner (Exner, 1986) were used to
construct a global measure of personality functioning. Using Exner's
1986 data drawn from 600 nonpatient adults (pp. 257-258) scores on 15
of these variables falling beyond one standard deviation from the mean
were determined. The five ratios that were included were based on
Exner's (1986) hypotheses regarding the meaning of directionality on
these ratios in terms of adaptive ego functioning. Table 2 gives the
variables and ranges of scores used.

Using these guidelines, the number of criteria - the number of
variables falling outside nonpatient normative performance - was
determined. A high number of criteria scores indicated greater
deviation from nonpatient norms, while a low number of criteria scores
indicated increased similarity to the nonpatient population. Subjects

were ranked according to this scale. Using early therapy criteria

43

Table 2

Rorschach Scores Used as Criteria for Level of Personality Functionipg

 

1. M : Sum C 5_2.00

2. chcrwc

3. D score # 0

4. M S_FM + m

5. M— > 0

6. FC' + C'F + C' > 2

7. FT + TF + T > 2

8. FY + YF + Y > 2

9. FV + VF +‘V > 2

10. FD > 2

11. P < 5 or > 8

12. Lambda < .31 or > .87
13. X+% < 71 or > 89

14. X-% > 11

15. F+% < 59 or > 93

16. Zd >.: 3.5

17. Afr < .47 or > .85
18. Egocentricity index < .28 or > .50
19. S-constellation Z 8

20. S > 2

 

44

scores, the eight subjects receiving the lowest scores on this scale
(low deviates from normative performance) were classified as higher-
functioning or having less impairment of ego functioning. The eight
receiving the highest scores (high deviation from normative performance

= greater impairment of personality functioning) were classified as

 

lower functioning.

An advanced graduate student was trained to score each protocol
for quality of object representation according to the Developmental
Analysis of the Concept of the Object on the Rorschach scale (Blatt, ‘ \
Brennais, Schimek & Glick, 1975). This cognitive-structural scoring
system is based on a theoretical integration of developmental theories
of Heinz werner and Jean Piaget with psychoanalytic ego psychology
(Blatt & Lerner, 1983). The scale is composed of six major subscales
on which Rorschach responses with human content are scored along a
developmental continuum. Subcategories in each subscale are weighted
according to amount of development indicated. For each protocol, the
total number of responses within each subcategory were tallied, the
weighted sum of the subcategories was computed for each subscale,
yielding a developmental index score for each subscale, and subscale
developmental index scores were summed and divided by the total number
of scorable responses, producing an overall mean object relations
score.

The same rater also scored each protocol for anxious and hostile
content on the Elizur (1949; Aronow & Reznikoff, 1976) scales. These
scales weight anxious and hostile content according to the intensity of
affect expressed whether by behavior attributed to figures in the

percept, symbolism in the percept, or the way in which the percept is

45

described. The weighted sum of scores for each protocol was then
computed.

[199. Subjects completed ICL ratings of four individuals - self,
therapist, mother, and father - a total of three times. Two subjects
said they had had so little contact with their biological father that
it.was impossible to describe him, and instead completed ratings for
their stepfathers, the men they considered to have been the father

figure in their lives. One subject had never had any contact with

either her biological father or any other male in a similar role, so
she was instructed to disregard father ratings.

Following standard procedures (Leary, 1956) the number of
statements scored as "true” for the ratee were added separately for the
16 subscales, and subscale scores were combined in standard formulas
which yielded raw scores on dominance and affiliation axes. The raw
score for each axis was then compared to a table of standard scores.

Once standard scores were obtained, the intersection of the two
scores on the axes was plotted on a diagnostic grid or circumplex
composed of eight discrete diagnostic octants. By noting which octant
a representation lay within, and the distance of that representation
from the midpoint, a diagnostic code was assigned, as well as whether
that representation was ”moderate" or "extreme.”

Measurement of the total amount of activity in representations of
self and others was done by applying the formula dx + dy , where dx
is the vertical discrepancy between representations and dy the
horizontal. The formula, which measures the geometric distance between
pairs of ratings, was applied to the distance between early and mid-

therapy representations and mid- and late therapy representations, and

 

46

the sum of the two distances was obtained separately for each

representation.

CHAPTER 4

RESUDTS

Main Hypotheses

 

ayppthesis l. Representations of the self, the therapist, the

 

mother, and the father will change significantly from early to mid-
therapy and mid— to late therapy.

Significant differences were found between early and mid-therapy
for the amount of affiliation in the self representation, the intensity
of the dominance in the self representation, and the amount of
dominance in the maternal representation. The self representation
became significantly less affiliative again from mid- to late therapy.
The intensity of self-dominance increased significantly in mid-therapy,
while the amount of dominance attributed to the mother decreased
sharply in mid-therapy. Table 3 presents the mean representation
scores at each stage of therapy.

Hypothesis 2. In mid-therapy the representation of the therapist

 

will become significantly more similar to the representation of one
parent than it had been in early therapy.

The deviation between parental and therapist representations
failed to decrease significantly between early and mid-therapy either

for the nine subjects perceiving the therapist as more similar to

47

48

 

 

 

.Ho. wrm.uo pancamacmam mmcaumu mmmumoulcfle can Manse consume monoHoMMflp u m

.mo. wlm.uo untowMflcmHm mmCHumu mmmuoca coca can ammumcunofle consume oocouomwwc u N

.mo. Iv. m an unmoflwcmam mmfiumu Eggnog can 3.8m smegma cocouowmac u H

genumnuauua

mN.mH mm.mH oo.mH vv.m Hm.h vm.h vv.¢ oo.m mo.m vv.v oo.m mo.m umvflmcmucH

oocmcaeoo

mH.NH mm.HH mm.NH vv.o ma.h mm.h mm.HH mm.¢ mm.HH Hm.o mwo.m mm.v uhnflmcmucH

no.hm mm.wm mm.hm mo.Hm Hm.m¢ ma.o¢ Hm.m¢ oo.mv vm.mv Nam.om Nmm.mv Hm.Hm COHumfiHHmm<

hm.mm ma.mm oN.oo mH.Nm Hmm.om mm.hm mm.Ho mm.am mm.Ho Hm.om Hm.om mh.mv OUCMCHEOQ
8.3 be; s28 83 92 s28 .83 2: 38d 83 e2 bums

End 850: umflmmumfi 2mm
mmmmum

 

acumen. 33 can .083: .bumm 5 Moscow com Josue: .umflmmuorfi. dam now mouoom coflumucomoumom .HUH coo:

m manna

49

mother or the six subjects perceiving the therapist as more similar to
the father. Hypothesis 2 was not supported.

Hypothesis 3. In mid-tterapy,

 

A. Psychological disturbance will be positively correlated with
perception of the therapist as similar to the mother.

B. Psychological disturbance will be negatively correlated with
perception of the therapist as similar to tte father.

C. Higher-functiom’ng clients will view tte tterapist as more
similar to the parent of the same sex as tte therapist.

D. lower-functioning clients will see tte tterapist as more like
the father than the mother, regardless of therapist gender.

Adaptive functioning was strongly correlated with perceiving both
mother and father as similar to the therapist. The correlation between
number of criteria for psychopathology and motter-therapist deviation
was .64 (p = .004) . Fetter-therapist deviation and Rorschach criteria
were correlated at a level of .52 (p = .02).

2 x 2 tables for C and D yielded significant results, with phi
coefficient values of .55 (df = 6) and .26 (df = 7) both significant at

p_<_ .01.

Hypothesis 4. In late therapy, the representation of the

 

therapist will become significantly more deviant than it had been in
mid-therapy to tte parental representation it had been more similar to
in mid-therapy.

Therapist and mother representations were more distant, but not to
a significant extent, at late therapy for subjects seeing the therapist
as more similar to mother than fatter in mid-therapy. A trend was

found for the distance between fatter and therapist representations to

50

increase sizably (pug .10) from mid- to late therapy for subjects
perceiving the therapist as more similar to father than to mother.

Hypothesis 5. The total amount of activity or movement in

 

representations of self, therapist, mother, and father will be
positively correlated with the difference in pre- and post-therapy
hostility content, anxiety content, mean overall object representation,
and level of functioning criterion Rorschach scores. Table 4 gives the
mean Rorschach for these variables at both stages of therapy, and Table
5 presents the correlations obtained between these Rorschach variables
and the ICL activity variables.

A significant relationship was noted between the total amount of
activity in ICL representations and changes in Rorschach content.
These content scores are actually difference scores, the difference
between hostility and anxiety upon entering and upon exiting
psychotherapy. The amount of anxiety and hostility content actually
increased rather than decreased throughout the course of therapy. No
significant correlations were obtained between the total amount of
movement and the pre-post differences in mean object representation or
number of criteria scores.

Hypothesis 6. There will be a significant difference between the

 

overall mean object relations scores and each of the six subscale
scores on the Developmental Analysis of the Concept of the Object scale
between pre- and post-therapy Rorschachs.

Table 6 present means, standard deviations, correlations, and E

values for the subscales and the overall scale.

 

Tl

51

Table 4

Mean Early and Late-therapy Rorschach Hostility, Anxiety, Object
Representation, and Criteria Scores

 

 

 

Early Late
Hostility 3.12 4.00
Anxiety 4.75 7.06
Object Representation 9.24 9.32
Criteria 7.31 7.37

 

Table 5

Pearson Correlations between Total Amount of Change in ICL
Representations of Self, TherapISt, Mother, and Father and Total Change
from Early to Late Rorschach Hostility, Anxiety, Mean Object
Representation and Criteria Scores

 

 

 

Rorschach Score

 

ICL Object

Representation Hostility Anxiety Representation Criteria
Self .15 .43* .25 .11
Therapist .52* .47* .21 .12
Mother .53* .45* .18 .19
Father .48* .14 -.20 -.07

 

* = p_< .05.

52

Table 6

T-tests Comparing Rorschach Object Representation Scores in Early and
in Late Therapy

 

 

Early to Late

 

 

Developmental Early Late Mean t
Index Mean SD Mean SD Difference value
Differentiation 23.19 10.82 24.31 13.22 1.13 -.51
Articulation 17.71 10.47 18.38 9.89 .63 -.21
Motivation

Object-action

Integration 9.38 6.51 9.81 7.95 .44 -.21
Nature of

Action 10.25 6.18 10.31 5.93 .06 -.04
content of

Interaction 6.56 4.08 6.44 3.86 .13 .11

Overall Mean

 

53

No significant differences between the overall score or subscale
scores were obtained. The majority of E values did, however, fall in
the predicted direction. For all but one subscale, small increases in
capacity for object relations were found in late therapy. Although
most differences between scores are so small as to prohibit assuming
clinical meaning, the fact that all but one lie in the expected
direction argues against random effects and offers some‘weak support
for the validity of this scale and for the occurrence of some slight
increase in capacity for object relations with therapy.

Hypothesis 7. The number of representations at moderate intensity

 

in late therapy will be positively correlated with the late therapy
Rorschach object relations score.

The correlation between these variables was .24 (p.= .19) and thus
this hypothesis was not confirmed.
Rorschach

Table 7 gives the correlation coefficients between first and
second administrations for each Rorschach variable. Testrretest
correlations were high for most variables, suggesting that perception
and response tendencies indicated on these protocols reflect
characteristic modes of functioning for these subjects. compared to
Exner's 1986 onedyear and three-year test-retest correlations for
nonpatient adults, however, the coefficients obtained in this study are
strikingly low. Exner established the range of .80 and above as
determining internal consistency, and his nonpatient samples produced
coefficients ranging from .70 to .91 for the majority of variables.
His 1978 study of 30 long-term psychotherapy cases retested after

approximately 180 days yielded coefficients comparable to those in this

 

54

Table 7

Pearson Correlation Coefficients of Rorschach variables between First
and Second Administrations

 

 

 

 

Present Exner
Study (1978)
N216 Ne30
Variable Description E p _r;
R NUmber of responses .83 .001 .64
W Number of whole responses .76 .001
D Number of usual detail responses .77 .001
Dd Number of unusual detail responses .46 .04
S Number of white space responses .48 .03
M Number of human movement responses .58 .009 .68
PM Number of animal movement responses .26 .17 .49
m Number of inanimate movement
responses .51 .02
FC Number of form dominant responses .72 .001 .39
CF + C Number of color dominant responses .27 .16 .47
FC' + C'F + C' Number of achromatic color responses .06 .41
FT + TF + T Number of texture responses .80 .001
FY + YF + Y Number of diffuse shading responses .73 .001
FV +‘VF + v NUmber of vista responses .56 .012
Fr + rF NUmber of reflection responses .67 .002
FD Number of form-based dimensional
responses .36 .08
F NUmber of pure form responses .67 .002 .51
Percentages and ratios:
es EXperienced stimulation
(PM + m + T + C' +‘V + Y) .62 .006 .61
D score Stress tolerance index
(Scaled EA-es difference score) .34 .10
3r + (2) / r Egocentricity index .72 .001 .73
Lambda Number of pure form/number of
non-pure form responses .54 .015 .48
F+% Good quality pure form .25 .19
X+% Extended good form .45 .04 .81
Xe% Poor form .34 .10
S-constellation .34 .10
EA Experience actual
(M + weighted sum C) .78 .001 .70
Afr Affective ratio .73 .001 .73

 

55

study. The similarity between this group and Exner's treatment group,
and tte difference between these groups and Ebmer's nonpatient sample
seem to demonstrate tte impact of six months of psychotherapy on
personality variables that would have most likely remained unchanged
without psychotherapy. The implication for this study is that
psychotherapy may be shown to have had a significant influence on
aspects of personality functioning, and that tte sample in this study
appears to be comparable with other samples.

Another comparison between normative groups and the sample in this
study is shown by Table 8, which presents descriptive statistics on
forty Rorschach variables and ratios for initial and second
administrations for this study and for Emer's 1986 data on nonpatient
adults and outpatient adults diagnosed as character problems. Initial
Rorschach means for subjects in this study seem to fall somewhere
between those of the two Exner groups, confirming that tte level of
psychotherapy for this group seems to be well within tte neurotic
range, neither quite as well cognitively and affectively organized as
the nonpatients, nor as disorganized as the character problems.

Helping Relationship Questionnaire

 

Table 9 shows the descriptive statistics for this instrument. The
majority of subjects agreed to each of tte items measuring such
qualitatively different aspects of general improvetent as feeling
helped, obtaining self-understanding, working jointly with tte
therapist, and increased autonomy. All clients indicated they believed
they had improved, with tte modal response being ”much improved.” A
post-hoc correlation between subjects' estimates of improvement and

level of personality functioning was performed. No relationship was

 

56

HamscHucoo «Hams.

 

m~ o oH AH.m ~m.a mH H a s~.m AH.o mH o m vH.m oo.o «H o mH en.o om.m a
s o c HH.H ac.o a o H so.H mH.H o o o mm.H so.H c o o ~4.H mH.H on
e o o vm.c Hv.o e o o me.o NH.o m o c a~.H ms.o n o o as. Hm.o mu+uu
H~ o o mv.~ mm.o AH H a mm.~ ve.m AH n s ss.n «a.» HH n m RH.m m~.s .m.
o c N HH.H nm.~ OH O n vo.~ oH.e HH H m oa.~ om.m cH n m RH.n ve.m :
a c o HH.H nH.H oH o o oo.H om.o HH H H v~.m ms.n SH a o ~v.n Hn.m ».m>.sm
n a o am.c m~.o m o o mm.o we.o H o o cm.o mH.o H c o o..o Hn.o >.m>.>u
m c o so.o vv.o m o H oo.o oH.H MH c c ce.n mH.~ m o c om.H aH.~ a.se.ea
m o a HH.H os.c cH o H c~.H Hm.H m c o am.H mo.H m o o oo.H ms.o .o+m.o..mo
n o o eo.~ me.H A o n vo.H oH.~ s c H Hs.H em.H o o H nc.~ ~H.~ o+mo
m a o mn.H sm.o a c m oc.~ sm.~ o o ,N Hm.~ om.~ o o m HA.H em.~ on
m o o HH.H no.H a o H eo.H m~.H m o o «m.H oo.H v o o o~.H mo.H a
A o H ne.H u~.~ a H . Hm.H Hm.n «H H H HH.H 4m.n HH o H Ho.“ oo.m 2m
mvgwug
s o H oo.H Hm.H a o H mo.H 4m.H s o o m¢.~ oo.m a o o H~.~ Hm.~ m
a c o sm.~ mm.~ vH o o es.~ ms.H mH o o Hm.e Ha.4 «H o o o¢.n om.n no
H~ a v Hm.v No.» v~ m 4H 45.. mm.~H mm o cc.n an.» ms.a H~ a m sm.o Hm.m o
mm H o ea.m mm.» o~ ~ a Go.~ cm.o RN m oo.m mm.m mo.oH mm m nH ¢~.A ae.HH :
coHuaooH
on oH AH Hv.m mm.sH an oH nw vm.m sm.- so a oo.HH HH.HH Hm.m~ av HH «H o~.oH Hm.v~ a
muamHum>
xa: :Hz moo: am can: xa: :Hz use: on cam: xa: cH: moo: am can: xm: 2H: one: am can:

oowuz mfiwmnoum

Hupuoumgu acmHHaduso

oowmz muasc‘ ucmwumacoz
Homon. umcxw

cofiumuumMCHect Honda

coHumHuchHEo< HuHuHcH
eHuz spasm Hammond

 

 

weBchS 209.089. no.“ mega—coupon can 8H3: 4moHumHH3w gHUAmHBB

u wanna

57'

 

 

 

m.mm HHH N.mH oHH m.NH N m.NH N N .v. 2 0H8
m.Hv 8 HH.H o c o n6 H m.H A 3953
m.mN HmH HH.N NH :6... a cém a mm. v H:
m.mN Hm n6 mm a v m.nv N l on. v :22 + an
HH.N v o.H HH c o c o o A 3 8333330 m
HH.N. on n.n ON n6 H n6 H H. A a I x
o.m¢ co H.oN NH .73 a o.nN HH ON. v o + m
mév mm o.oH ow oéo HH c.mN NH cN. v a + x
o.vm no c.HN oNH can a 99 N o m 0.80m o
m.mH an oén wHN m6 H n6 H 339.3388. N A 8.959
o.om NHH o.nN 9H m.Nn o 9.2 N 3:335:
HH.H: 3 TH”;
mJN Ho H6. HvN m6... a o.om a HgHuuvsoHucH. N A 9591.
o aucusvoum a aucmswopm a accusoohm a mucuswowm
o o v SH 84 m o c HH.N No.N a N N vH.H om.n N o m 34 qmé cooum
NHH HN.o mm... 36 Hm.o NH 36 on... «H... 36 m H H mm. SH 34 NN... Nw.o Hvé was but
H... o oo.H nv.H Hm.H mN.N 36 No... cN.o $5 oN.H o o mmé omé HH.N o o 36 $5 mos—.3
mm... o and 36 Ne... 36 no... Nmé HH.H. on... Hmé 86 8.0 86 :6 and 35 Nv.o 35 end H.223.”
mad 0 o mod mH.o 96 o 36 mo... ooé 26 o 0 mod 36 36 o c 3... 8.0 aux
oo.H on.c mN.° nH.o ON.c oo.H ov.c wa.o mo.o om.c H¢.o mm.o mm.o HH.o Nm.o mm.c om.o vo.o nH.o vo.o o+x
oo.H c o.H oHé HN.c oo.H c 8H :6 2.6 8H 35 cm... 26 $6 c.H NH.o $6 35 86 f...
N 7. o 8H 8.9: H n: o 3H 36 n m: c 3H NHH: v m: H: 3H 8.? 988 o
3 o v 3:.” $6 Hm m a $4 HN.m em N N :6 8.3 HN H 0 Ned mN.m mo
m.vH o N SN :4 N m.H a me «N... NN H N 2...... 36 RN m.N om.m mmfi 31m 5
moHumm H «mandamOHQm
3.5.88. 8:8: xifizooflomcmmz 33:89. 858: 35:30: 8 :8:

oomlz mEanoum

“Song £83.83

oouuz munzut ucwwumaccz

.wmmH. hmcxm

EHuahchHefi H2: Bumbag HaHuHcH

2.: exam anemone

 

58

Table 9

Descriptive Statistics for the Helping Relationship Questionnaire

 

 

l = strongly disagree 2 = disagree 3 = mildly disagree
4 = mildly agree 5 = agree 6 = strongly agree
Item Mean Mode SD Range
1. I believe that my therapist is helping me. 5.19 5 0.75 4-6
2. I believe that the treatment is helping me. 5.19 5 0.75 4-6
3. I have obtained some new understanding. 5.50 6 0.63 4-6
4. I have been feeling better recently. 4.44 4 1.09 2-6
5. I can already see that I will eventually

work out the problems I came to treatment

for. 4.75 4 1.34 1-6
6. I feel I can depend upon the therapist. 4.75 4 1.29 3-6
7. I feel the therapist understands me. 4.88 5 0.89 3-6
8. I feel the therapist wants me to achieve my

goals. 5.63 6 0.62 4-6
9. I feel I am working together with the

therapist in a joint effort. 5.06 6 0.85 4-6
10. I believe we have similar ideas about the

nature of of my problems. 4.88 5 0.62 4-6
11. I feel now that I can understand myself

and deal with myself on my own. 4.19 4 1.33 2-6
Total of items 1 - 11 54.44 59 7.16 40-65
Self-estimate of improvement
(l=not at all, 2=slightly, 3=moderately,
4=much, 5=very much) 3.56 4 1.03 2-5

 

59

found between the total HRQ score (total of items 1 - 11) and number of
Rorschach criteria on either initial or second administrations (£_= o
and .25 respectively), indicating that subjects' level of pathology was

unrelated to their subjective experience of improvement.

CHAPTER 5

DISCUSSION

The seven hypotheses address two general issues: (1) whether
there are distinct phases in therapy and the therapeutic relationship
in which self and others are experienced in subjectively different
ways, with particular attention to correlates of the quality of the
transference around the time it is hypothesized to be most intense, and
(2) how a number of personality features change across the course of
therapy.

General Issues

 

An immediate issue regarding the cumulative findings from this
study is that the low number of subjects prevents any definite
conclusions from being drawn. The low'N contributed to the difficulty
of attaining statistical significance. It also decreased the
likelihood that the findings of this study are true for the population,
increasing the likelihood that chance factors operated to confirm or
disprove hypotheses. ‘While the high possibility of random effects must
be taken seriously while evaluating these results, other factors also
merit consideration. Buckley, Conte, Plutchik, Wild and Karasu (1984)
addressed similar methodology issues and claimed that in exploratory
research, such as their study of therapy outcome and the present study,

Type I errors were more acceptable than Type II errors.

60

61

Auerbach mingled attitudes of scientific skepticism and pragmatism
in his 1983 paper on assessment of therapy outcome. He took the
position that,

Outcome should be viewed like the concept of a true score in

classical test theory; it is known only to God. Our

measurements of outcome are approximations, and because of

the nature of personality and of psychotherapy, they are a

good deal less precise than tte determination of IQ...We must

do that.which is feasible and seems reasonable while

remembering that the appearance of reasonableness may be

deceptive. (p. 538)

Auerbach articulated two primary attitudes toward psychotherapy
outcome research: 'lifelike complexity' and 'reliability,' each of
which had specific advantages and disadvantages. 'Lifelike complexity'
was described as more ambitious, usually involving explanation as well
as description, but also being significantly more difficult. On that
continuum, this study clearly falls on the side of complexity. It was
intended to offer in rich clinical description.what it lacked in

quantitative rigor and validity.

Length of Treatment. One strength of this study is its duration

 

of six to ten months of therapy. Buckley et al. (1984) pointed out
that the vast majority of psychotherapy studies have involved periods
of treatment of less than two months. The two most highly regarded and
oft-cited outcome studies, the Psychotherapy Project of the Menninger
Foundation and the Penn Psychotherapy Project, studied much longer
courses of therapy, from 44 to 800 hours of treatment. Between the two
extremes, little literature exists on relatively brief, but not
explicitly short-term, dynamic psychotherapy.

Shapiro and Shapiro's 1982 meta-analysis of therapy outcome
studies found relatively few studies using dynamic and humanistic

methods of treatment. Even.when all studies from these perspectives

 

62

were combined - a conceptually'weak assumption - the majority of such
studies used group rather than individual treatment, therapy lasted a
mean of eleven hours, was carried out by therapists with a mean 3.7
years of experience, and only one-third of the data was collected in
clinical settings.

Subject Characteristics. Demographic and other attributes of this

 

sample strengthen the relevance of this study to general clinical
situations. The mean age of subjects, the high proportion of graduate
students, the inclusion of subjects with children, and the severity of
some subjects' psychiatric histories argue that this sample bore more
resemblance to a general outpatient population than to that of a
nonpatient university undergraduate population. Inspection of
descriptive statistics for Rorschach variables for this group compared
to Exner's nonpatient and outpatient samples further supported this
conclusion.

Buckley et a1. (1984) found similar outpatient-nonpatient.
comparisons in their medical student population. Their therapy
patients also had measurably greater psychopathology than their
nonpatients, leading the authors to conclude that in this high-
functioning clinical sample the concerns leading subjects to seek
treatment were the result of psychopathology which would have been
manifested, and led them to seek treatment, in any setting. This has
relevance for this sample as well. Although they were necessarily
functioning well enough to remain in college, the kinds and severity of
presenting problems and history suggests that these were mild to

moderate neurotic and character problems which would have led them to

63

seek outpatient treatment whether they were enrolled in school or
currently employed.

Analyses of several hypotheses compared higher~ and lower-
functioning subgroups. Although the total sample divided neatly into
two groups of eight, differences in severity between the subgroups may
have been slight. One reason for this was subject selection. Clinical
needs took clear precedence over research, and in all cases in which
recruitment.was feared to interfere with the client's ability to feel
safe in therapy, with development of the therapeutic relationship, or
where the client appeared too agitated, no recruitment todk place.
This may well have eliminated many clients whose more severe
psychopathology caused them to present to the counseling Center in
crisis, and whose diminished trust and investment in others may have
led them to refuse to participate even if approached. The high
percentage of graduate students indicates an opposite effect. These
tended to be women who had personal experience‘with research, had
themselves undergone the tedious process of gathering data, several of
whom verbalized attitudes of pleasure in participating in the academic
community through such activities.

Two additional factors might account for some of the negative
findings in this study. The first factor is the therapists' level of
experience. With.the exception of one experienced faculty member, the
remaining therapists had no more than three years of prior experience
practicing psychotherapy. A substantial number of the cases studied
was the therapists' first cases of their careers. As stated above,

this level of training is commonly found in psychotherapy process and

—""1-1l

64

outcome studies, but suggests that tte impact of tte psychotterapeutic
process might well have been stronger‘with more seasoned clinicians.

The second factor is the adequacy of the measuring instruments.
The ICL and the Rorschach are standard instruments (particularly the
Rorschach), but are not often applied in exactly these ways.
conceptually, both measures seemed sufficient to measure the phenomena
in question. As they have not been previously applied in just this
way, however, it is entirely possible they simply failed to indicate
changes that did, in fact, occur.

One of the post-hoc analyses concerned differences between
subjects who completed the course of psychotherapy, and those who
terminated prematurely. There were no differences between these groups
on their representations of themselves, their therapists, their
fathers, or their mothers, although there was a nonsignificant tendency
for terminators to perceive both mother and father as more dominant and
more affiliative than completers. Significant differences appeared in
Rorschach variables between the two groups. Terminators had higher
mean object relations scores, a greater number of FC responses, a
higher number of reflection responses, and a lower percentage of good
quality pure form responses. EXner (1986) considered reflections to be
indicative of a degree of egocentricity or self-centeredness, and the
mean for terminators was significantly above that of Exner's
nonpatients. Terminators seemed to have a higher general level of
functioning in capacity to relate to others and to manage affect, but
to feel acutely uncomfortable, self-conscious, and perhaps withdrawn
from forming affective connections with others, possibly including the

therapist. Their higher level of general functioning may have led them

65

to feel less pressure to make the necessary commitment to continue in
psychotherapy, and their possible tendency to bend reality may have
aided them in distorting either their view of what occurred in therapy,
the realities of their current situation, or both. The terminators'
relative youth in comparison to completers may also have led them to
overestimate the situational nature of their difficulties, and to
underestimate the contribution of internal conflicts in bringing about
current distress.

The Relationship Between Change in Representations and Change in

Rorschach Content

 

Hypothesis 5 stated that the total amount of change (activity or
movement) in each representation would be positively correlated with
indicators of personality functioning. The underlying rationale for
this hypothesis was that at the conclusion of successful psychodynamic
psychotherapy the client would experience lessened symptomatology, as
measured by Rorschach content scores, increased capacity for object
relatedness, and more adaptive use of inner resources, as measured by
Rorschach criteria scores. These changes in functioning were believed
to be directly tied to the changes in internal representations of self
and others which were expected to occur in therapy through the
corrective emotional experience with the therapist, interpretation and
exploration of the transference itself, uncovering of earlier memories,
particularly those involving early interactions*with the parents, and
exploration of current interpersonal relationships.

The results of the correlations between ttese scores were mixed.
Significant results were obtained for hostility and anxiety content,

while the other two scores were unrelated to ICL activity. Self-report

[1.1.1.7.
-

66

and projective measures have often failed to be significantly related
even when they have been believed to be conceptually similar. A
gereral 1y accepted explanation has been tte proposition that ttese tap
different levels of depth in tte personality, whetter tte levels are
called unconscious and preconscious, or Ieary's (1956) Levels 11
(private self-description) and III (fantasy and underlying character
structure). As content scores changed significantly from early to late
therapy, while adaptive functioning (as measured by Rorschach criteria
scores) did not, this suggests that tte content scores may measure an
aspect of personality less fundamental than underlying personality
structure itself, but deeper than conscious knowledge of self and
others. It also suggests that symptomatic relief may substantially
precede structural personality change in psychotterapy.

Hostility, Anxiety, and Level of Functionirg

 

The relationship between hostility and anxiety change highlights
the relationship between level of functioning, anxiety, and hostility.
Post-hoc analyses found a very near significant (p = .059, .055)
negative correlation between initial level of functioning and initial
and final levels of hostility, and a significant (p = .04) positive
correlation between initial level of functioning and late therapy
anxiety. Better-functioning subjects had higher levels of hostility at
tte beginning and end of therapy. It was otherwise lower-functioning
clients whose hostility was expressed in more disguised forms, possibly
resulting in lower hostility scores. Early and late therapy hostility
and anxiety were negatively correlated (_r; = -.16, -.l9) but not to a
significant extent. While tte correlation between initial anxiety and

level of functioning was very slight (_r_ = .08, p = .39), subjects who

 

67

were better-functioning at the onset of therapy had significantly lower
levels of anxiety at the end of therapy than lower-functioning
subjects.

Rabinowitz (1975) proposed that hostility ratings were primarily a
measure of the individual's awareness of his or her inner self and
hailingness to accept hostile feelings and attitudes. He found that
the ability to fantasize was a major source of variation among several
hostility measures. A growing awareness or tolerance of one's own
hostility would explain the increase in hostility scores observed in
this study, but a post-hoe analysis failed to produce significant
correlations between initial hostility, late therapy hostility, the
difference between these scores, and the Rorschach M response, long
considered an index of capacity for fantasy and inner life.

Change in the Capacitygfor Object Relations

The increase in differentiation at the end of therapy’was so small
as to be statistically insignificant. The most likely explanation for
the small size of this increase is the form of therapy received by the
subject. Previous reports of this scale in the literature have cited
changes in differentiation following intensive, long-term
psychoanalytic psychotherapy. The modal form of therapy practiced at
the Counseling Center is a shorter-term, psychodynamic psychotherapy
typically involving one weekly session. The differences in length,
intensity, and goals of treatment probably accounted for the more
modest improvement attained in this study. The results of Hypotheses 5
and 6 together strongly suggest that in the six to ten months of
psychodynamic psychotherapy, changes in the handling and expression of

unpleasant affects took place, representing the uncovering and more

"31

fans. n‘n ‘

68

adaptive handling of previously repressed material, but only minimal
change in basic character and personality structure took place.

Quality of the Transference

 

This study was designed to measure perceptions of self and others
at three phases which were believed to reflect discrete stages in the
therapeutic relationship. The early or initial phase of therapy was
believed to be one in which the client's original, distorted
perceptions of self and others were as yet unchanged, and the
perception of the therapist was relatively neutral. By the midpoint of
therapy it was expected that the process of gradual exploration would
have brought many of these distorted perceptions into greater
awareness, and the cumulative effect of the experience with the
therapist would have resulted in a transference reaction with
predominantly paternal or maternal features. At the end of therapy,
with the termination process encouraging resolution of conflicts and
disengagement from the therapist, it.was predicted that the perception
of the therapist would be more neutral and less transferentially
colored than in mid-therapy. Because previous studies had indicated
that the developmental phase at which the analysis was being conducted
influenced the maternal or paternal quality of the transference,
differences between clients were expected in accord with level of
psychosocial development and probable psychosexual fixation.

A strong correlation was found between therapist-mother and
therapist-father deviation at mid-therapy and initial level of
functioning. Better-functioning clients tended to view therapists as
similar to both mother and father, while clients with greater initial

pathology saw the therapist as dissimilar to father and mother.

“I.

(.

 

69

Level of personality functioning seemed to be related to the
readiness of the client to perceive the therapist as similar to one or
both parents. Therapist gender was also clearly a factor. Higher-
functioning clients appeared to more readily view the therapist in ways
that recapitulated old object relationships, while lower-functioning
clients seemed somewhat less likely to do so.

The hesitancy of lower-functioning clients could be attributed to
a number of factors. It could signal a characteristic guardedness and
hesitancy to form relationships, evidenced in this case as a lower
level of transference after approxnmately ten therapy sessions.
Adternatively, it could indicate that these clients' experiences with

past significant others such as parents had been disappointing enough
that the therapist would be perceived in a somewhat idealized fashion,

dissimilar to either parent, both in emotionally accessibility and in
personal strength. Better-functioning clients seemed more capable of

forming a therapeutic relationship, while more poorlybfunctioning
clients appeared less open to the therapist.

These findings are in line with previous clinical reports of
therapy outcome from a number of sources. As Strupp (1980) phrased it,

By the same token, the more deficient, troubled, and
destructive the patient's experience with significant others
has been the greater will be the difficulties in achieving a
corrective experience in psychotherapy; This is the reason
therapy often faces an uphill battle and it defines the
limits of what can be achieved in a given period of time.
Patients with severely disturbed earlier relationships
frequently have great difficulty in forming a good
'therapeutic alliance' with the therapist, a sine qua non for
therapeutic change. (p. 291)

Schwager and Spear (1981) cited Appelbaum's analysis of the

results of the Menninger Foundation Psychotherapy Research Project,

 

70

which found results parallel to those in this study. In Appelbaum's
work, as well as in this study, this 'rich get richer' phenomenon
accounted for the greatest improvement subsequent to moderate or long-
term psychotherapy among those patients who had shown highest levels of
ego functioning at initial testing, and little structural change among
patients with more severe initial pathology. Harty, Cerney, Colson,
Coyne, Frieswyk, Johnson, and Mortimer also reported similar results in
their 1981 study and commented,

The strongest finding here thus seems to repeat a familiar
one in the field of psychotherapy research: that those who
begin a clinical process (here hospital treatment) with
greater ego strength, less disorganization, and more
accomplishments are very likely to be those who end in the
best condition as well...Do these findings, as some have
held, cast doubt on the effectiveness of the treatment? we
suggest that the answer is yes only when exaggerated or
inappropriately 'equalitarian' expectations are made of what
treatment should accomplish. In hospital work, as in
psychotherapy, psychological processes are to a great extent
both the object of the work and the medium in which the work
is carried out, i.e., both the material to be shaped and the
tools for shaping. we should not be dismayed to find that
the resources one brings to such a task determine, to a major
extent, what can be accomplished. (p. 221)

Some subjects' spontaneous comments on the Helping Relationship
Questionnaire confirmed this idea. One wrote, "It is important to note
that with any therapist it would probably take a long time for me to
really open up. My therapist did a good job, the result is a matter of
temporal limitations not quality.” The same client told her therapist
during termination that she believed it “would take a bulldozer“ to get
through to her and create a meaningful therapeutic relationship marked
by trust. Another client offered, 'I believe therapy has been helpful
and that my therapist has been trying to get me to open up, but it's

been really hard for me to allow myself to open up and express my

 

71
feeling which in turn is making therapy seem to move or proceed

slowly."
Maternal/Paternal Aspects and Temporal Changes in Transference

It was predicted that the transference would take a predominantly
maternal or paternal quality in mid-therapy which would be greatly
attenuated at late therapy, as the client worked through termination
issues. This would be indicated by the representation of the therapist
being significantly more like one parent than the other in mid-therapy,
and a significant difference occurring between perceived similarity
between the therapist and that parent in mid-therapy, and the therapist
and that parent in late therapy.

A predominantly maternal or paternal focus was observed in
subjects' perceptions of the therapist, although this pattern was not
robust enough to be quantitatively confirmed. Six clients saw the
therapist as paternal, and nine perceived the therapist in a maternal
fashion. No high-functioning clients had paternal transferences to
female therapists, but three high-functioning clients had maternal
transferences to male therapists. The remaining four high-functioning
clients' transferences were Specific to the gender of the therapist.
The occurrence of paternal transferences in three clients of female
therapists is an anomaly, according to most existing literature.

Mid-therapy transference quality took on added significance when
late therapy therapist—mother and therapist-father deviations were
examined. The significant difference between these deviations
disappeared in late therapy. Maternal transferrers in particular
seemed to view the therapist as nearly equally similar to each parent

in late therapy. Although the predominant maternal or paternal quality

72

seemed to diminish in late therapy, there was not a significant
difference between mid- and late therapy.

Iate Therapy ICL and Object Relations Scores

 

The conceptual rationale of Hypothesis 7 was that ”moderate" ICL
representations indicated a more highly differentiated vieW'of self and
others. It was believed that the phenomena of splitting, commonly

found in clients with lower levels of ego functioning, might be shown

 

on the ICL by representations of self and others at ”extreme” levels of
intensity. Therefore, the number of moderate late-therapy
representations was expected to correlate with the capacity for 4;
differentiation and mature object relationships, which was believed to
be represented in the overall mean Rorschach object relations score.
The correlation between these scores was positive but nonsignificant.
It seems that, as with Hypothesis 5, the differences in levels of
functioning assessed by Rorschach and ICl variables led to low

relationships between the two measures.

 

gum...

APPENDICES

 

APPENDIX A

Therapist Information Form

73

APPENDIX A
MEN'S COUNSELINS S‘IUDY

Wendy Sabbath, M.A.

Introduction This study is actually my doctoral dissertation. I hope
to collect the data here at the Counseling Center this year, and this
requires the help of the other staff members and interns, in terms of
convincing qualified clients to be subjects. All I am asking is that
you introduce the study to potential subjects (hopefully in an
enthusiastic manner). If they are interested, I will do all the
subsequent testing and make the necessary arrangements.

Subjects Any female client who is at least 18 years old, is a new
client to you, and whom you plan to see for at least six months is
eligible. The only other criteria are that the client be only mildly
to moderately disturbed, and not currently actively suicidal, or
otherwise severely impaired in her ability to make a therapeutic
relationship. If you are unsure whether a particular client is
eligible, please give her the benefit of the doubt, and later you and I
can discuss the issue.

Procedure During the first actual therapy hour, I would like you
to read or paraphrase as closely as possible the following statement:

This year an intern at the Counseling Center is undertaking a
study of some issues in women's emotional development, and
the ways in which counseling aids these processes. we are
asking women to participate in this study. Participation
would involve meeting three times with one of our counselors
to complete several measures. Ybur identity would be kept
strictly anonymous. Ybur decision.whether or not to
participate would have no bearing on your counseling, but we
would like to encourage you to take part. One of the
benefits you would receive would be the opportunity at the
end of the study to discuss*with the investigator what the
tests suggest about your counseling experience, and.what the
experience had been like for you.

If at this point the client is willing to participate, or wants
more information before deciding, please hand her an Initial Contact
Sheet to complete. These should be returned to my mailbox, and I will
take it from there. If she is unwilling, please thank her anyway. If
she is unsure whether she is interested, please try to persuade her to
fill out the Initial Contact Sheet so that I can talk to her directly
and respond to her specific concerns. It's hard to know in advance
what clients' particular questions or objections might be, but I'm very
willing to address these concerns and try to make this as interesting
and non-threatening an experience as possible. Most people find taking
the tests I am using in this study to be a revealing experience for
them, and as mentioned above, I am very willing, after all the testing

 

11

 

74

is completed, to go over each subject's tests with them and discuss in

general what their tests mean to them.

I think this is clear. If not, please let me know and I'll be
glad to try to clear things up. I can't do this study without your
help, so I'd like to thank you in advance for your assistance, and I'll
figure out a way to make my gratitude more tangible later.

 

 

 

APPENDIX B

Initial Contact Sheet

 

75

APPENDIX B

MICHIGAN STATE UNIVERSITY
Women's Counseling Study

INITIAL CIJNTACI‘ SHEET

1 am interested in getting some more information about the study

of women's development and perhaps participating in the study. My name a?"

is and my home phone number is

 

 

 

The best time to reach me is

(If there is another phone number at which you may be reached, please

 

indicate it here, along with the times you may be reached at that

number ). I understand that I

will be called within the next several days regarding this project.

 

 

APPENDIX C

Consent Form

 

 

l.

76

APPENDIX C

MICHIGAN STATE UNIVERSITY
Department of Psychology

DEPARTMENTAL RESEARCH.CONSENT FORM

I have freely consented to take part in a scientific study
being conducted by wendy Sabbath, M.A. under the supervision
of Professor Bertram Karon at the M.S.U. Counseling Center.
This study involves aspects of women's emotional growth and
development and investigating the ways in which counseling
aids that process. I understand that if I complete the study
I wall be contacted three times and that the total time
commitment will be less than ten hours over the course of six
to nine months.

The study has been explained to me and I understand the
explanation that has been given and what my participation
will involve.

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

I understand that the results of the study will be treated in
strict confidence and will remain anonymous. ‘Within these
restrictions, results of the study will be made available to
me at my request. I understand that I may choose to have
some of the information made available to my therapist.

I understand that my participation in the study does not
guarantee any beneficial results to me.

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

Signed:

 

Date:

 

 

APPENDIX D

Helping Relationship Questionnaire

 

77

APPENDIX D

dee #

 

The Helping Relationshipgguestionnaire

 

Below are listed a variety of ways that one person may feel or behave

in relation to another person.

Please consider each statement‘with

reference to your present relationship*with your therapist.

Mark each statement according to how strongly you feel that it is true,

or not true, in this relationship.

Please mark every one. ‘write in 6,

5, 4, 3, 2, or 1 to stand for the following answers.

6. Yes, I strongly feel that it is true.
5. Yes, I feel it is true.
4. Yes, I feel that it is probably true, or more true than untrue.

3. No,
2. NO,
1. NO,

l.
2.
3.

 

 

6.
7.

10.

11.

I feel that it is probably untrue, or more untrue than true.
I feel it is not true.
I strongly feel that it is not true.

I believe that my therapist is helping me.

I believe that the treatment is helping me.

I have obtained some new understanding.

I have been feeling better recently.

I can already see that I will eventually work out the
problems I came to treatment (counseling) for.

I feel I can depend upon the therapist.

I feel the therapist understands me.

I feel the therapist wants me to achieve my goals.

I feel I am working together with the therapist in a joint
effort.

I believe we have similar ideas about the nature of my
problems.

I feel now that I can understand myself and deal‘with myself
on my own (that is, even if the therapist and I were no
longer meeting for appointments).

Kinds of change:

I feel improved in the following ways:

 

 

 

 

I feel worse in the following ways:

 

 

 

Estimate of improvement so far:

not at all

(please circle one)

slightly moderately mush very mush

Any further comments:

REFERENCES

 

Aronow, E. & Reznikoff, M. (1976). Rorschach content interpretation.
New York: Grune and Stratton.

 

Athey, C., Fleischer, J., & Coyne, L. (1980). Rorschach object
representation as influenced by thought and affect organization.
In J. Kvawer, H. Ierner, P. Ierner, & A. Sugarman, Borderline
phenomena and the Rorschach test. NeW'YoIk: International
Universities Press.

 

Auerbach, A. (1983). Assessment of psychotherapy outcome from the
viewpoint of expert observer. In M. Lambert, E. Christensen, & S.
De Julio, The assessment of psychotherapy outcome. New'York:
wuley.

 

Beier, E. & Ratzeburg, F. (1953). The parental identifications of
male and female college students. JOurnal of Abnormal Psychology,
38, 569-572.

 

Blatt, S., Brennais, C., Schimek, J., & Glick, M. (1976). Normal
development and psychopathological impairment of the concept of
the object on the Rorschach. Journal of Abnormal Psychology, 85,
364-373.

 

Blatt, S. & Lerner, H. (1983). The psychological assessment of object
representation. JOurnal of Personality Assessment, 21, 7-28.

 

Blotcky, A. (1984). Early use of the Rorschach in anticipating the
transference of borderline adolescents. Dynamic Psychotherapy, 2,
157-165.

 

Buckley, P., Conte, H., Plutchik, R.,‘Wild, K., & Karasu, T. (1984).
Psychodynamic variables as predictors of psychotherapy outcome.
American JOurnal of Psychiatry, 141, 742-748.

 

Cava, E. & Raush, H. (1952). Identification and the adolescent boy's
perception of his father. JOurnal of Abnormal and Social
Psychology, 52, 855-856.

 

 

Clark, T. & Taulbee, E. (1981). A comprehensive and indexed
bibliography of the Interpersonal Check List. JOurnal of
Personality Assessment, 25, 505-525.

 

 

Crowder, J. (1972). Relationship between therapist and client
interpersonal behaviors and psychotherapy outcome. JOurnal of
Counseling Psychology, 19, 68-75.

 

 

78

79

Crowder, J. & Feltoon, H. (1973). Transference: Analysis of the
concept and review of the research. San Antonio: Tri-way
Printers and Mailers.

Dietzel, C. & Abeles, N. (1975). Client-therapist complementarity and
therapeutic outcome. JOurnal of CounselingPsychology, 22, 264-
272.

Elizur, A. (1949). Content analysis of the Rorschach with regard to
anxiety and hostility. Journal of Projective Techniques, l3, 247-
284.

Exner, J. (1974). The Rorschach: A comprehensive system (Vbl. 1).
New'York: JOhn Wiley & Sons.

Exner, J. (1978). The Rorschach: A comprehensive system (Vbl. 2).
New York: John Wiley & Sons.

Exner, J. (1986). The Rorschach: A comprehensive system (2nd ed.).
New York: Jehn Wiley & Sons.

Freud, S. (1937). Analysis terminable and interminable. (Standard
Edition). London: Hogarth Press, 1964, 216-253.

Gill, M. (1982). Analysis of transference (vol. 1). New Yerk:
International Universities Press.

 

Gill, M. & Hoffman, I. (1982). A method for studying the analysis of
aspects of the patient's experience of the relationship in
psychoanalysis and psychotherapy. JOurnal of the American
Psychoanalytic Association, 30, 137-167.

 

 

Goldfried, M. (1980). Some views on effective principles of
psychotherapy. Cognitive Therapy and Research, 4, 271-306.

Graff, H. & Luborsky, L. (1977). Long term trends in transference and
resistance: A report on a quantitative-analytic method applied to
four psychoanalyses. JOurnal of the American Psychoanalytic
Association, 25, 471-490.

 

Greenspan, S. & Wieder, S. (1984). Dimensions and levels of the
therapeutic process. Psychotherapy, gl, 5-23.

 

Groesbeck, C. (1978). Psychological types in the analysis of the
transference. JOurnal of Analytic Psychology, 23, 23-53.

Guntrip, H. (1973). Psychoanalytic theory, thergEYr and the self.
New York: Basic Books.

Harty, M., Cerney, M., Colson, D., Coyne, In, Frieswyk, S., Jehnson,
S., & Mortimer, R. (1981). Cerrelates of change and long-term
outcome. Bulletin of the Menninger Clinic, 45, 209-228.

80

Hoornaert, F. & Pierloot, R. (1976). Transference aspects of doctor-
patient relationship in psychosomatic patients. British JOurnal
of Medical Peyehology,‘49, 261-266.

 

 

Karme', L. (1982). El ana'lisis de un paciente varo'n par una
analista: el prdblema de a1 transferencia edipica negativa (The
analysis of a male patient by a female analyst: The prdblem of the
negative oedipal transference). Revista de Psicoana'lisis, 32,
883-896. (From Psychological Abstracts, 1984, ll, Abstract No.
21067).

 

 

Kiesler, D. (1983). The 1983 Interpersonal Circle: A taxonomy for
complementarity in human transactions. Psychological Review, 29,
185-214.

 

Konig, K. & Tischtau-Schroter, R. (1982). Der interktionelle Anteil
der Ubertragung bei Partnerwahl und Partnerveranderung (The
interactional part of transference by the choice of partners).
Zeitschrift fur Psychosomatische Medizin und Psychoanalyse, 28,
266-279. (From Psychological Abstracts, 1982, 62, Abstract No.
10838).

 

Krieger, M. & Worchel, P. (1960). A test of the psychoanalytic theory
of identification. Journal of Individual Psycholegy,‘l§, 56-63.

 

Krohn, A. & Mayman, M. (1974). Object representations in dreams and
projective tests. Bulletin of the Menninger Clinic, 38, 445-465.

Kulish, N. (1984). The effect of the sex of the analyst on
transference. Bulletin of the Menninger Clinic, 28, 95-110.

 

Lapkin, B. & Lury, J. (1983). Transference and countertransference
in the light of developmental considerations. American JOurnal of
Psychoanalysis, 43, 345-359.

 

 

Leary, T. (1956). Multilevel measurement of interpersonal behavior:
A manual for the use of the interpersonal system of personality.
Berkeley, CA.: Psychological Consultation Service.

Leary, T. & Harvey, J. (1956). A.methodology for measuring
personality changes in psychotherapy. Jburnal of Clinical

 

 

Lerner, H. (1983). An object representation approach to
psychostructural change: A clinical illustration. JOurnal of
PersonalityAssessment,.41, 314-323.

 

Lerner, H. & St. Peter, S. (1984). Patterns of object relations in
neurotic, borderline and schizophrenic patients. Psychiatry, 42,
77-92 0

 

 

81

Lorr, M., McNair, D., Michaux,‘w;, & Raskin, A. (1962). Frequency of
treatment and change in psychotherapy. JOurnal of Abnormal and
Social Psychology, 64, 281-292.

 

Luborsky, L. (1984). Principles of psychoanalytic psychotherepy. New
Yerk: Basic Books.

Luborsky, L., Crabtree, L., Curtis, H., Ruff, G., & Mintz, J. (1975).
The concept 'space' of transference for eight psychoanalysts.
British Journal of Medical Psychology, 48, 65-70.

 

McDonald, R. (1965). Ego-control patterns and attribution of
hostility to self and others. JOurnal of Personality and Social
Psychology, 2, 273-277.

 

Malcolm, J. (1981). Psychoanalysis: The impossibleprofession. New
York: Knopf.

Mogul, K. (1982). Overview: The sex of the therapist. American
JOurnal of Psychiatry, 139, 1-11.

 

Mueller, W. (1966). Anxiety level, inferred identification and
response tendencies on a semantic differential. JOurnal of
Counseling Psychology, l}, 144-152.

 

 

Mueller, W. (1969a). Patterns of behavior and their reciprocal impact
in the family and in psychotherapy (Monograph). JOurnal of
Counselipg Psychology, lg, (2, pt. 2).

 

 

Mueller, W; (1969, September). Mapping the psychotherapeutic process
through analyeis of interpersonal behavior. Symposium conducted
at the American Psychological Association,‘Washington, D.C.

Moeller,‘w. & Dilling, C. (1968). Therapist-client interview behavior
and personality characteristics of therapists. Journal of
Projective Techniques and Personality Assessment, 32, 281-288.

 

Mueller,‘W. 8 Dilling C. (1969). Studying interpersonal themes in
psychotherapy research. J0urnal of Counseling Psychology, lg, 50-
58.

Rabinowitz, A. (1975). Hostility measurement and its relationship to
fantasy capacity. JOurnal of Personality Assessment, 39, 50-54.

Rapaport, D., Gill, M., & Schafer, R. (1968). Diagnostic
(psychological testing (rev. ed.). New Yerk: International
Universities Press.

 

Rawn, M. (1981). A note on unwitting replication: Quantitative
studies of transference and resistance twenty years apart.
JOurnal of Clinical Psychology, 39, 782-784.

 

 

82

Ryan, E. & Bell, M. (1984). Changes in object relations from
psychosis to recovery. JOurnal of Abnormal Psychology, 23, 209-
215.

 

Schafer, R. (1977). The interpretation of transference and the
conditions for loving. JOurnal of the American Psychoanalytic
Association, 2;, 335-362.

 

Schlesinger-Vaccaro, R. (1983). A relationship between Piaget's
notion of conservation and the transference phenomenon: A
clinical example. The Psychoanalytic Review, 19, 83-94.

 

Schrier, H. (1953). The significance of identification in therapy.
American Journal of Orthopsychiatry, 23, 585-604.

Schwager, E. & Spear, W; (1981). New perspectives on psychological
tests as measures of change. Bulletin of the Menninger Clinic,
fl, 527-541 0

 

Shapiro, D. A. & Shapiro, D. (1982). Meta-analysis of comparative
therapy outcome studies: A replication and refinement.
Psychological Bulletin, 92, 581-604.

 

Silverman, In & weinberger, J. (1985). Mommy and I are one:
Implications for psychotherapy. American Psychologist, 49, 1296-
1308.

 

Smith, M., Glass, G., & Miller, T. (1980). The benefits of
psychotherapy. Baltimore: The Jehns prkins University Press.

 

 

Sopchak, A. (1952). Parental 'identification' and 'tendency toward
disorders’ as measured by the Minnesota MUltiphasic Personality
Inventory. JOurnal of Abnormal and Social Psychology, 41, 159-
165.

 

Spear, W; & Lapidus, L. (1981). Qualitative differences in manifest
object representations. JOurnal of Abnormal Psychology, 29, 157-
167.

 

Tracey, T. & Ray, P. (1984). Stages of successful time-limited
counseling: An interactional examination. JOurnal of Cbunselipg

Psycholegy, 3_1_, 13-27.

Tuber, S. (1983). Children's Rorschach scores as predictors of later
adjustment. Journal of Consulting and Clinical Psychology, §l,
379-385.

 

 

Urist, J. (1977). The Rorschach test and the assessment of object
relations. JOurnal of Personality Assessment, gl, 3-9.

 

 

 

83

Urist, J. & Shill, M. (1982) . Validity of the Rorschach Mutuality of
Autonomy Scale: A replication using excerpted responses. Journal
of Personality Assessment, 4g, 450-454.

 

Waelder, R. (1956) . Introduction to the discussion on problems of
transference. Internatioral Jounal of Psychoanalysis, g, 367-
368.

Weiner, M. (1982) . Identification in psychotherapy. American Journal
of Psychotherapy, g, 109-116.

 

Zetzel, E. (1956). Current concepts of transference. International
Journal of Psychoanalysis, _3_'_7_, 369-376.