MEAWRING PRIMARY AND SECONDARY
PROCESS THINKING IN SCHIZOPHRENICS
AND NQRMALS BY MEANS
OF THE RORSCHACH

Thai: for II“ Degree of pH. D.
MICHIGAN STATE UNIVERSITY

Eugene Zukowsky
1961

 

This is to certify that the

thesis entitled

Measuring Primary and Secondary Process Thinking
In Schizophrenics and Normals by Means of
The Rorschach

presented by
Eugene Zukowsky

has been accepted towards fulfillment
of the requirements for

Ph. D. degree in Psychology

/%Mj/L~—mn -

Major professor
Date g/y/g / [/41~/

0-169

LIBRARY

Michigan State
University

 

 

ABSTRACT

MEASURING PRIMARY AND SECONDARY PROCESS THINKING
IN SCHIZOPHRENICS AND NORMALS BY MEANS OF
THE RORSCHACH

by Eugene Zukowsky

The present study was undertaken for the purpose of testing the
psychoanalytic constructs of primary process and adaptive regression
as they apply to schizoPhrenia. Measures of primary process and
adaptive regression were provided by the Holt scoring system for the
gauging of primary process on the Rorschach, and the results provided
a test of these measures. Four hypotheses, based on psychoanalytic
theory regarding ego functioning, were proposed.

The subjects of this study consisted of twenty normal, twenty good
premorbid schizophrenics and twenty poor premorbid schizophrenics.
The Rorschach test was individually administered to all subjects by the
author, following the procedure outlined by Klopfer. All subjects were
reasonably matched on age, last school grade, and occupational level.

The Rorschach scoring followed the manual provided by Holt.
This manual contains specific directions for evaluating the degree to
which primary process is manifest in Rorschach responses and the extent
of controls over primary process productions. To provide a measure
of reliability for scoring the protocols, the author and another scorer
independently scored 10 records, using the overall measure of adaptive
regression for this correlation. The inter-scorer reliability was . 944.
Intra-subject reliability for the same measure was also demonstrated,
using the split-half method. The Split-half reliability coefficient was

. 648, significant beyond the . 01 level.

Abstract ‘ Eugene Zukowsky

A primary process response was defined as any response containing
a scorable content and/or formal deviation. The operational measure of

adaptive regression is given by the formula

ARS=DDx(FL+DC1+DC2+ DCn)

where ARS is the adaptive regression score, DD is the defense demand
for each response, FL is the form level, and DC is the defensive contribu-
tion for each reSponse.

Measures of primary process were also obtained by calculating the
number of content .and formal deviations, and number of Level 1 re3ponses,
for the purpose of determining if the manner in which primary process
was expressed was different for the three groups.

The data were analyzed by using analysis of variance first, to
determine if differences existed. If differences were found, comparisons
of group means were obtained with the t -ratio using the studentized range
distribution. This procedure was followed to reduce the risk of error
in making several comparisons between means.

Four hypotheses were tested. The first two were concerned with
the number of primary process responses. It was hypothesized that the
schizoPhrenic groups would have significantly greater numbers of primary
process responses than normals, and that the poor premorbid group would
have more primary process responses than the good premorbid group.
These hypotheses were not confirmed. It was found, however, that the
manner in which the primary process is expressed differentiated the
groups. The schizophrenic groups, regardless of premorbid status,
were found to give re3ponses which were closer to the primary process
pole, as measured by content and formal deviations and the blatancy of
reaponses. It was proposed that the primary process is revealed more
clearly when expressed as the manner of responding, i. e. , the more

qualitative expressions of primary process, than by absolute number of

Abstract Eugene Zukowsky

primary process responses. When interpreted in this way, the data were
found to be in support of psychoanalytic theory.

Two hypotheses dealt with the factor of adaptive regression. It was
predicted that the normals would have a higher mean adaptive regression
score (ARS) than schiZOphrenics, and that the good premorbid group
would have higher ARS scores than the poor premorbid group. Although
neither hypothesis received statistical substantiation, the data were
strongly suggestive that these hypotheses would be confirmed with further
testing on larger samples. Further analysis of the data led to the finding
that normals and good premorbid schizophrenics achieved higher form
level scores on primary process reSponses than the poor premorbid group.
There were no differences among the groups on total form level score.
These findings were interpreted to mean that normals and good premorbid
schiz0phrenics have similar levels of ego strength, as measured by
degree of perceptual accuracy, than poor premorbid schizOphrenics.
However, the two schizophrenic groups were judged to manifest primary
process in similar ways in that they operated at levels closer to the primary
process than did normal subjects.

It was concluded that the Holt method for the assessment of primary
process on the Rorschach is a valid one. However, it was suggested
that the overall measures of primary process responses and, to some
extent the adaptive regression score, are too gross. The method was felt
to be a potentially valid tool in both future research and clinical diagnosis,

and suggestions for future research along these lines were made.

Approved .fl‘é/a fit/1%— / “WM“...

7

Committee Chairman

Date $12,], /{ ://,/

 

I

MEASURING PRIMARY AND SECONDARY PROCESS THINKING
IN SCHIZOPHRENICS AND NORMALS BY MEANS OF
THE RORSCHACH

BY

Eugene Zukowsky

A THESIS

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

DOCTOR OF PHILOS OPHY

Department of Psychology

1961

ACKNOWLEDGMENTS

There are many people who have contributed to this project,
far too many to mention. I would like to express my appreciation to
the staff at both the Ann Arbor and Fort Custer V. A. Hospitals for
their cooperation and suggestions. Many thanks are due to Gerald
Casey and David Pryor for giving up their valuable time to serve as
judges. Special appreciation goes to David Pryor for having worked
so closely with me in teaching me the Holt scoring system.

I am indebted to the help given to me by my doctoral committee,
for their constructive criticisms and serious interest in this study.
They deserve special mention here.

Dr. Albert 1. Rabin, Chairman
Dr. Bill L. Kell

Dr. Gerald F. King

Dr. Terrence M. Allen

In particular, to Dr. Albert 1. Rabin who has served as my
advisor throughout my program of studies, I owe a special word of
gratitude and appreciation.

Thanks are also due to Dr. Robert R. Holt of the Research
Center for Mental Health, New York University, and to Dr. Martin
Mayman of the Menninger Foundation, Topeka, Kansas, for providing
the manuals used in this study.

Most importantly, I would like to dedicate this work to my
wife Zita, who has, from the start, given so much understanding and
support. This thesis could not have been accomplished without her.

)5: 3:: >1< >:: 9,: 3,2 95 3k 3:: )1: air 3::

ii

In Memory

To Dr. Gerald F. King, who died
September 6, 1961, just prior to the
completion of this thesis, and whose
death is a profound loss not only to
those who knew him, but to the field

of psychology.

iii

TABLE OF CONTENTS

CHAPTER

I.

II.

III .

IV.

VI.

VII .

INT RODU CTION .

PRIMARY AND SECONDARY PROCESS THINKING.

Freud's Formulations .............
Subsequent Elaborations: Ego Psychology . .

Attempts at

SCHIZOPHRENIA

Measurement ...........

AND DEVIATIONS IN THOUGHT

PROCESSES ..................

Dichotomies in SchiZOphrenia .........
Thinking in Schizophrenia ...........
Thinking in Schizophrenics as Reflected in
the Rorschach ..............
Primary and Secondary Process as Reflected

in the

Rorschach ............

STATEMENT OF HYPOTHESES ...........

METHOD .....

Characteristics of the Sample .........
Methodology ..................

Procedure .

T reatment of Data ...............

RESULTS . . . .

Analysis of Results ...........
Summary of Results ...........

DISCUSSION . . .

Summary and Integration of Findings
Implications for Future Research

SUMMARY . . . .

BIBLIOGRAPHY .

APPENDICES

iv

Page

1

ll

12
16

18
22
Z4
Z6
Z6
Z9
36
37

39

39
50

54

62
64

66

69

76

TABLE

10.

ll

12.

13.

LIST OF TA BLES

. Comparison of the Groups

and Occupational Level .

. Comparison of the Groups

Rorschach ReSpons es (R)

.. Comparison of the Groups

Process Responses . . .

. Comparison of the Groups
Process ReSponses . . . .

. Comparison of the Groups
Deviations (CD) ......

. Comparison of the Groups

Deviations (FD) . .

. Comparison of the Groups
Deviations .........

. Comparison of the Groups
Deviations .........

. Comparison of the Groups
Score (ARS) . .......

Comparison of the Groups

(T F-L) ..........

.. Comparison of the Groups
Level Scores (PP F-L) . .

Page

on Age, Last School Grade,
............... 28
on Total Number of
............... 39
on Number of Primary
............... 40
on Total Level 1 Primary
............... 42
on Total Number of Content

. . . . ........... 43
on Total Number of Formal
............... 44
on Level 1 Content
............... 45
on Level 1 Formal
............... 45

on Total Adaptive Regression

OOOOOOOOOOOO O O O

on Total Form Level Score

on Primary Process Form

Summary of Results (Significance tests) ........

Summary of Results (Means and Standard Deviations) .

47

48

49

52

53

LIST OF A PP ENDIC ES

APPENDIX Page
A. General Information Sheet .............. 77
B. The Phillips Scale . . . . . ........ . . . . . 78
C. Primary Process categories used for scoring the
Rorschach protocols, and the Defense Demand (DD)
associated with each . .............. . . 80
D. Control and Defense categories used to score the
Rorschach protocols, and the Defense Contribution
(DC) associated with each . . .......... . . 83
E. Form Level scores (FL) and examples for scoring . 85

vi

INTRODUCTION

Schizophrenia is the most serious form of mental illness in this
culture and the most widely studied. We are still, however, far from
a full understanding of this disorder. Nevertheless, the multitude of
observations and empirical findings, and the varieties of frames of
reference from which schizoPhrenia has been viewed have added
important bits and pieces to the puzzle which will hopefully culminate
in our understanding of this disorder.

We have come to view schiz0phrenia as the result of a conglomer-
ation of possible factors, many of which are inextricably interrelated.
This very knowledge, albeit incomplete, is both an aid to our further
understanding and a hindrance to precise research. Most often research
has to be carried out with a relatively restricted scope, often at the
neglect of apparently important factors. This would seem to be an
inevitable result of our incomplete picture of schiz0phrenia, as well as
the difficulties inherent in "in vivo" research. Nevertheless, greater
methodological precision has accompanied our increased knowledge.

The various points of view from which the problem of schizophrenia
has been attacked often seem unrelated. But it seems safe to say that
the study of this disorder is truly an interdisciplinary one. The psycho-
logical approach, and particularly the area of disturbances in thinking
in schizophrenia has been a widely studied area, and one which has
added to our understanding of the syndrome. Disordered thinking was
one of the earliest observations made about schizophrenics, and has
been widely held by many as being a crucial fact of the disorder.
However, the theoretical position of psychoanalysis relative to the con-

Structs of primary and secondary process thinking in schiz0phrenia

have, as yet, not been subjected to systematic testing. Considering
that this theory is one of the most elaborated, and offers excellent
opportunities for objectification, this gap in our knowledge becomes
important to close.

The major purpose of this research will be to study the broad area
of thought disturbances in schiz0phrenics. We will particularly concern
ourselves with the theoretical speculations of psychoanalysis as it per-
tains to the constructs of primary and secondary process thinking and
to the relationship between primary and secondary process modes of

thought to schizophrenics and normals.

CHAPTER I

PRIMARY AND SECONDARY PROCESS THINKING

The theoretical position used as a base in this study will be that
of Freud's schematic elaborations of primary and secondary process
thinking as well as later modifications and additions offered by "Ego
Psychology. " We will here be concerned with Freud‘s development
of the concepts of primary and secondary processes, the later notions
of adaptive regression, the relation between these concepts to schizo-
phrenia, and their measurement, i. e. , their "operational definition, "

via the Rorschach.

Freud's Formulations

 

Freud developed and elaborated his ideas on primary and secondary
processes in three major works (1911, 1915, 1938). He has dealt most

extensively with this topic in The Interpretation of Dreams, Chapter VII

 

(1938), originally published in 1900.

In essence, the theoretical elaboration of primary and secondary
process is intimately related to two major concepts. First, the tOpo-
graphical "divisions" of unconscious (ucs), preconscious (pcs), and
conscious (cs), particularly as they pertain to the "System ch, " are
most relevant. Secondly, from a "genetic" point of view, the primary—
secondary process distinction is intimately connected with ego develop-
ment.

The postulating of unconscious mental life is both basic to psycho-

analytic theory in general and to the concept of primary process in

particular. Freud (1912.) distinguished the difference between "conscious"
and "unconscious" in terms of conceptions which were present in aware-
ness, and "latent conceptions. " These latent conceptions, however, have
the potentiality of entering awareness. One case in point here is that
of memory. However, with the exploration of dreams, Freud came to
understand unconscious mental life not as a particular mental act, but
rather as "a system of mental activity. " He therefore designated mental
activity belonging to this system as The System Unconscious (ch).
It now became possible to investigate the properties of this system as they
differ from the properties of The System Conscious (Cs). This investi-
gation began with probings into the meaning of hypnotic phenomena and
culminated in the analysis and understanding of dream-processes.

Freud (1915) conceptualizes unconscious mental life in two ways.
"The unconscious comprises, on the one hand, processes which are
merely latent, temporarily unconscious, but which differ in no other
respect from conscious ones and, on the other hand, processes such as
those which have undergone repression, which if they came into
consciousness must stand out in the crudest contrast to the rest of the
conscious mind" (p. 104). The former, latent processes, are those
aspects of the preconscious which, given sufficient intensity and attention,
could become conscious. In regards to the latter processes, he is
directly alluding to the peculiar properties of the primary process. The
"abnormal processes" are a result of the characteristics of the system
ch. Since this system consists of "instinct-presentations" whose aim
is to discharge their cathexis under the laws governing the pleasure
principle, the processes of discharge are also subsumable under the
laws of the pleasure principle. Freud (1916) outlines four major proper-
ties of these processes.

1. Ideas may become concentrated in a single conceptual unit.

That is, ideas are condensed.

 

2. So as to facilitate discharge, intermediary or compromise

 

ideas are established.
3. These ideas which transfer their intensities to one another

are loosely associated and not bound by conventional logic.

 

4. Contradictory ideas can exist side by side.

 

In further elaboration of the qualities of unconscious activity (1915) he
sees the processes of the ch as timeless, unburdened by considerations
of reality, where there exists no negation, dubiety or varying degree of
certainty nor mutual contradiction. "In the ch there are only contents
more or less strongly cathected. "

In Freud's system, since these processes follow the pleasure
principle and, since infantile life is governed by this principle, the
psychic systems can be expressed in a temporal order. Thus, Freud is
able to speak of the primary process as temporally older, "the residues
of a phase where they were the only kinds of mental processes. ”

The postulation of psychic systems as older, more primitive, as
opposed to recent or "mature, " raises several points worth-while
elaborating on. For one, it allows for the concept of "regression. "
Freud (1938) speaks of a "threefold species of regression. " There is a
topological regression, a regression within the psychic systems of Cs,
Pcs and ch. He also notes a Temporal regression, which has already
been described, and lastly, a Formal regression to more primitive modes
of representation and expression. The characteristic of this latter type
of regression is clearly seen in dreams where "the idea is changed back
into the visual image from which it once originated. " Secondly, in an
earlier quote (p. 4), Freud alluded to the idea that only repressed ideas
are subject to the primary process. He has stated this idea more
explicitly. "Such an abnormal psychic elaboration of a normal train of
thought takes place only when the latter has been used for the transference

of an unconscious wish which dates from the infantile life and is in a state

of repression" (Freud, 1938, p. 532). Thus it will be that thoughts which
have been repressed, i. e. , are cathected by the unconscious, will
succumb to the primary processes and will reveal themselves according
to the formal qualities related with this primitive mode of representation.
From a developmental point of view the relation befween primary
and secondary processes and ego functions are important and help to throw
light more fully on the nature of the primary processes. We have stated
earlier that the ch follows under the regulation of the pleasure principle.
Under this state of affairs, where the purpose is to avoid excitation, the
only channel open to discharge was through motility. With the accumu-
lation of excitation, which is felt as "pain, " the system is set in motion
to relieve this excitation, the relief of which is experienced as pleasure.
Freud calls this striving for pleasure a "wish" and "the first occurrence
of wishing may well have taken the form of a hallucinatory cathexis of
the memory of gratification” (1916, p. 533). Thus, in early infantile life,
whatever was wished was hallucinated (imagined). Under repeated frus-
tration, the child learns that these hallucinations are incapable of bringing
satisfaction, and are eventually abandoned. Therefore, the activity of
a second system was required which "had to decide to form a conception
of the real circumstances in the outer world and to exert itself to alter
them" (1911, p. 4). This introduced a new principle of mental function-
ing, that of the reality principle. Whereas in the primary process energy
is loosely bound so as to facilitate discharge, under the requirements of
reality, restraint of motor activity was now necessary. This newly
"bound" energy replaced automatic discharge with, out of economic con-
siderations, small amounts of cathexes manifesting in "experimental
thought. " Thus, thinking, under the sway of the reality principle, becomes
an experimental way of acting. But in order for this to achieve the
required satisfaction, thinking itself had to follow the rules of reality.

Thus, primitive mental activity is modified "by bitter practical experience"

into a secondary and more appropriate activity, that of the reality
principle. Thinking under this new system is logical, tied to consider-
ations of reality and is economic. These are the modes of the latter
secondary processes. Freud generalized to these notions after under-
standing dream processes. He noted that two processes participate in
dream formation. "One forms perfectly correct and fitting dream-
thoughts, equivalent to the results of normal thinking, while the other
deals with these thoughts in a most astonishing and, as it seems, in-
correct way" (1938, p. 532). The former corresponds to the secondary
elaboration of dreams and is analogous to the work of the secondary
processes. The latter consists of the dream-work proper and is subject

to the laws of the primary process.

Subsequent Elaborations: Ego Psychology

 

The characteristics of the primary and secondary processes have
been further elaborated by several authors (Fenichel, 1945; Rapaport,
1950; Schafer, 1954, 1958; and Bellak, 1958). In a paper on "regression
in the service of the ego, " Schafer (1958) integrates the notions of Freud,
Hartmann (1951) and Kris (1951, 1952) as to the adaptive use of the primary
process. Hartmann (1951), in postulating a "conflict-free" sphere of
ego functions, observed that the ego operates with varying degrees of
neutralized energy, the more neutral the less encumbered by conflict and
the greater degree of autonomy of the ego. The degree of neutralization
corresponds to Freud's conception of sublimation, but goes beyond it in
the sense that there is a recognition of ego functions not tied down to
instinctual life (Hartmann, 1955). In delineating the differences between
primary and secondary processes with the inclusion of the later ideas
from ego psychology, Schafer (1958) has said: "Primary process, which
is genetically and formally the more primitive, operates with unneutralized

drive energies" whereas in the secondary process "energies are

relatively neutralized, i. e. , relatively bound in motives and structures
of a highly socialized nature. " Thus, in primary process thinking there
is a sexualization or aggressivization of thought, being under the domi-
nation of drives. However, in Kris' (1951, 1952) conceptions it was
observed that the primary process could be utilized for constructive,
creative purposes. Freud (1911, 1938) had already pointed the way here
by observing that unconscious strivings can be directed and guided into
superior aims (1938) and that the artist allows full play to erotic etc.
wishes but finds a way back from fantasy to create a new reality (1911).
Kris (1951) has termed the utilization of primary process for creative
purposes "regression in the service of the ego. " This process implies
central controlling functions in the ego which may, however, suspend
other functions, such as defensive and logical functions. Thus, this
concept refers " . . . to the ego's permitting relatively free play to the
primary process in order to accomplish its adaptive task" (Schafer,
1958, p. 125). '

That this process involves ego control and is, therefore, an active
process, has implications for psychopathology. Generally, the encroach-
ment of the primary process into normal, waking life, when not used for
adaptive purposes, is a passive, relatively overwhelming process. It
requires ego weakness rather than ego strength as is required for its
adaptive use. It follows that in states of relative ego weakness, e. g. ,
schizophrenia, primary process will be more abundantly exhibited.

Freud (1924) has conceptualized schizophrenia as a conflict between
the ego and the environment such that libido is withdrawn from external
objects back into the ego resulting in a regression to primary narcissism.
This regression also accounts for the observation ". . . that in schizo-
phrenia a great deal (of primary process) is consciously expressed" (1915).
Freud's ideas have been reviewed and summarized by Fenichel (1945),

but a theoretical paper by Hartmann (1953) deserves our close attention

here. Using Freud's speculations as his premise, Hartmann elaborates
on the theme of ego functions in schiz0phrenia. Taking as a starting
point that in schizophrenia the ego is weak, he speculates as to why this
might be so. In terms of specific ego functions he notes that the ego's
ability to act as a mediator between drives and reality is impaired,

either because of deficiencies in ego defensive mechanisms or that those
ego functions which maintain contact with reality may be incompletely
developed or weakened. Thus, since repression is either lacking or
faulty in the schizophenic, due to the ego's inability to maintain sufficient
countercathexes, this leads to a tendency toward conflict, and, at the
same time, the ego is incapable of dealing with it. Therefore, he con-
cludes, "the ego may be the weak link in the psychological setup before
the outbreak of psychosis. " More specifically, it is the ego's impaired
ability for neutralization which is "a fundamental character of the ego
disorder in schizophrenia. " Hartmann postulates that the degree to which
instinctual, libidinal energies have been neutralized may coincide with
the degree to which the primary process has been replaced by the
secondary process. Non-neutralized ego functions are those which have
been sexualized or aggressivized. In essence then, Hartmann hypothe-
sizes a deficiency in primary autonomous factors in the ego which contribute
to the vulnerability of defense and of the capacity for neutralization in
schizophrenia. Further, given the ego's inability to neutralize instinctual
energy, it can be concluded that there will be a greater reliance on
primary process in schiZOphrenics.

This view has been reinforced by Bellak (1958) who considers
schiz0phrenia a result of, for a variety of reasons, poor ego functioning.
He holds, further, that thought disturbances are the most sensitive
indicators of ego disturbances and that "the formal characteristics of
schiz0phrenic thinking are identical with the formal descriptive character-

istics of the primary process" (p. 19). He postulates that any event

10

sufficient to lead to a primary developmental ego weakness or to

secondary weakening of the ego may lead to the schizophrenic syndrome.

Attempts at Measurement

 

In terms of the measurement of primary process thinking in the
Rorschach, this has generally been implicitly done. Most Rorschach
workers utilize some means of designating the signs indicative of a
thought disorder, but there is little systematization among workers.

In some scoring schemes (Rapaport, Gill, and Schafer, 1946) there are
ways of classifying deviant responses, e.g. , fabulizations, peculiar,
queer, etc. , whose presence may or may not be pathognomonic. But by
and large the "measurement" i. e. , objectifying, of degree of thought
disorder has remained wanting.

Holt (1956) has attempted both to describe as well as objectify
primary process manifestations. He starts out with the assumption that
the more "primary" the thinking the more it is organized and compelled
by drives. Further, as a result of being compelled by drives, thinking
will then have certain formal characteristics, such as autistic logic,
loose and inappropriate associative links and various kinds of reality
distortions. Essentially by a process of classifying responses as to their
instinctual drive derivatives, formal qualities, and defensive attributes,
he has developed a scheme by which primary process can be detected
and measured, i. e. , operationalized. In Chapter IV. Holt's system will

be dealt with more elaborately.

CHAPTER H

SCHIZOPHRENIA AND DEVIATIONS IN
THOUGHT PROCESSES

Introduction

 

Early investigations in "dementia praecox" (an earlier term for
schiz0phrenia) were focused on attempts to discover a causative agent
for the disorder. In the late 19th century and early in the 20th century
the belief was widely held that dementia praecox was either due to a
specific disease agent or was a somatic disorder, perhaps inherited.
As a result of these views energy was directed towards finding the
etiological factor responsible for the disorder. Despite shifts in per-
spective due to more current information about schizophrenia, much of
contemporary research is based on long held views concerning organic
etiology. There is a good deal of highly sophisticated and important
work being done, for example, in the area of physiological research
with schiz0phrenics. This has evolved to the point where differences
within the general group of "schizophrenics" have been noted, differences
which cut across the classical diagnostic subcategories. In some
instances new typologies have arisen which are based on autonomic dif-
ferences (Meadow and Funkenstein, 1951), and important distinctions
are now being made which are based on the past history of the patient.
A good deal of research and speculation has been generated concerning
the distinction of "process-reactive" schizophrenia, and there are
implied etiological differences in this distinction (Brackbill, 1956).

The present study is not concerned with possible somatic differences
within schizophrenics, nor, except possibly in an indirect way, with

etiological factors. We are mainly concerned with behavioral correlates

11

12

and psychological processes in schizophrenia, particularly as they relate
to disturbances in thought. The research to be reviewed in the following
pages will, therefore, largely be confined to studies of thinking in
schizophrenia, with emphasis on Rorschach findings. It is to be noted
that, regardless of the point of view taken with respect to schizophrenia,
most if not all workers have taken cognizance of deviations in thinking.
Thus, as a psychological variable, disturbances in thinking play a part
in the phenomenology of schiZOphrenia, without necessarily owing
allegiance to any particular point of view. We will, however, explore
some works: pertaining to dichotomies in schizophrenia, but only for the

purpose of integrating the notions of disordered thinking with this view.

Dichotomies in Schizophrenia

 

It is not a recent observation that there are differences among
schizophrenics with regard to symptomotology and prognosis. ' Much
effort has been spent attempting to separate schizophrenics into two
groups, this separation being based on a combination of somatic and
psychological observations. We will review some of the literature in
this area and show its methodological importance for contemporary
research.

Bleuler (1950) was one of the first to take issue with Kraepelin
insofar as the course of dementia Praecox leading to inevitable
deterioration was concerned. Bleuler noted that patients with acute
onsets and clouded sensorium generally had a good prognosis. Meyer
(1921), noted that one of the more frequent prepsychotic indices of poor
prognosis was the presence of introversion or "shut-in" personality,
and'the observations of Hoch (1921) tend to support this belief.

In a study of 25 recovered cases of dementia praecox, Strecker
(1929) pointed out that a crucial prognostic sign was the presence of a

precipitating cause. Sullivan (1929), studying the outcome of 100

13

schizophrenics, found that recovery was twice as good in patients where
there was an acute onset than in those cases with insidious onset.
Further, as partial corroboration of Meyer's observations, the "shut=in"
personality was characteristic of most of the poor prognostic group,

but absent from all of the recovered cases. However, another study
(Hunt and Appel, 1936) observed that a common factor in both a recovered
and unrecovered group of 30 patients "lying midway between schizophrenia
and manic-depressive psychosis" was the presence of early onset and a
serious precipitating factor. What differentiated the recovered group,

in comparison to the unrecovered cases, was relief from the precipitat-
ing stress situation.

Langfeldt (1937), following a group of 100 patients for 6-10 years,
found higher improvement rates in previously identified atypical cases
than in typical cases. Those patients with poor recovery rates were
characterized by early, insidious onset, gradual personality change, and
the typical symptoms described by Kraepelin. Stressing the need for
further differentiation within the schiz0phrenic group, he suggests
(Langfeldt, 1951) the distinction of "Typical SchizOphrenia" for those
patients coinciding with the original formulations of Kraepelin, and
"Schizophreniform, " for atypical groups with good prognosis.

A large number of studies reported in the literature (Stalker, 1939;
Paskind and Brown, 1940; Kant, 1940, 1942; Meduna and McCullough,
1945; Clow, 1953; and Schofield e_t a_l. , 1954) indicate that there seem
to be essentially two types or extremes along the schizophrenic dimension.
Darrah (1940), has suggested that the term dementia praecox be reserved
for those cases with early insidious onset and development, and where
there is a steady course towards deterioration. Those patients showing
no particularly predisposed prespychotic personality, with acute onset
following an obvious environmental stress, and where there is good prog-
nosis should be termed schizophrenic. This dichotomization is also

pr0posed by Bellak (1948, 1949).

14

The studies of Wittman (1941, 1948), and Wittman and Steinberg
(1943) tend to corroborate already existing evidence of the existence of
two extremes within the category of schizophrenia. ' Phillips (1953)
devised a scale for rating the premorbid adjustment of schiZOphrenics
and found that maturity in the premorbid period, particularly social and
sexual adequacy, was related to good prognostic potential. This scale
was used effectively in a group of studies reported on by Rodnick and
Garmazy (1957), who emphasize its methodological importance in terms
of reducing the heterogeneity in schiZOphrenic samples.

The first study utilizing the distinction process and reactive schizo—
phrenic was that of Kantor, e_t a_l. (1953). They dichotomized a sample
of schiZOphrenics into process and reactive on the basis of Wittman's
classification. Rorschachs on each patient were independently rated as
to the presence, or absence of a psychosis. They found that the
Rorschachs of the process schizophrenics were more frequently judged
psychotic than those of the reactives; the latter protocols tended to more
closely resemble the records of normals. An earlier study by Stotsky
(1952) had also found that WAIS and Rorschach records of remitting
patients tend to resemble those of neurotics, while nonremitting patients
looked more psychotic on these tests. The author concluded that emotional
and intellectual functioning are less impaired in remitting than in non-
remitting patients.

In a critical review of the literature on physiological studies in
schizophrenia, Brackbill (1956) speculates that ". . . The evidence is
sufficient to postulate the existence of organic brain disease in some,
but not all, schizophrenic patients. Those who show the classical signs

of formal thinking disorder . . . may have some kind of structural brain

 

damage. On the other hand, those who show . . . no thought disorder,
may not have central nervous system pathology" (p. 223). Using this

distinction to test the hypothesis that some schizophrenics are suffering

15

from some kind of organic involvement, Brackbill and Fine (1956) used
known organics with which to compare process and reactive schizo-
phrenics. They hypothesized that the process group would not differ
from known organics on the incidence of Piotrowski's "organic signs, "

but both will differ from reactives. Their hypotheses were confirmed.

Summary

Although the initial discriminating feature between patients who
tended towards recovery and those who tended towards deterioration
was the distinction acute or chronic, this has subsequently been ampli-
fied to include a wide variety of factors. The most common, and prob-
ably the most reliable of these factors is that of early, insidious de-
velopment, with a generally withdrawn, schizoid personality as con-
trasted with an acute, obviously precipitated onset in a relatively benign
or“ well-adjusted prepsychotic personality. Thus, it is important both
conceptually and methodologically, to estimate the level of prepsychotic
adjustment (King, 1954). Many have speculated that with chronic,
"typical" (in the Kraepelinian sense) schizOphrenics we may be dealing
with essentially a psychosis of organic origin, while 'schizophrenics
with good prognosis may be of psychogenic etiology. The tenability of
this view is still Open to serious question. So far, it would seem that
Brackbill's notions that there is organic involvement in what might be
called "process" schizoPhrenics is not untenable. However, he himself
cautions that, at this stage, due to the general lack of normal controls
in most of this research, the findings of disordered physiology in schizo-
phrenics can as easily be interpreted as a consequence of the psychosis

rather than as an etiological factor.

16

Thinking in Schizophr enia

 

This section will be devoted primarily to studies of thought disorders
in schizophrenics, with particular emphasis on Rorschach findings.

Deviations from conventional thinking in schiz0phrenia were noted
early, but it wasn't until Bleuler (1950) that they were described
accurately. In fact, as opposed to all other symptoms, he relegated the
thought disturbances to the role of primary symptoms. In other words,

the presence of a disturbance in thinking was the sine qua non of schizo-

 

phrenia according to Bleuler.

A good deal of investigation has taken place in the area of concept
formation and abstract ability. Much of this gained impetus from the
work of Vigotsky (1934) and Goldstein (1939, 1943). In general, most of
the studies tend to show that schizophrenic thinking is impaired in three
ways: They perform poorly on tasks requiring the ability to make
generalizations, the handling of abstract relationships, and the analysis
and synthesis of conceptual wholes.

The studies of Hanfmann and Kasanin (1938, 1942), utilizing Vigotsky's
concept formation test with schizophrenics, generally corroborate the
idea of a loss of abstract ability in schiz0phrenics. They note, in com-
paring schizophrenics with organics, that although many cases of
schizophrenia often bear a close resemblance to cases of irreversible
brain disease, they differ in having a greater variability and "a certain
imaginative quality" to their productions, lacking in organic cases.

In relating impaired abstract ability to the type of clinical picture, these
authors found that those schizophrenics who appeared to resemble severe
neurotics, and those who generally presented an acute, overactive,
schizoaffective clinical picture, showed no impairment in thinking as
measured by the Vigotsky test. However, there was severe impairment

with the older, chronic patients.

17

The relationship between abstract ability and prognosis has also
been investigated. Meadow and Funkenstein (1951), have shown a
relationship between the loss of abstract ability and autonomic reactivity.
Both of these have been related to poor prognosis. Where no loss of
abstract ability is found the prognosis remains good.

In a study attempting to relate the observations of "looseness of
association" and "impairment in abstraction" in schizophrenics, Meadow
and associates (1953) note that both have been considered by various
writers as primary symptoms and related to prognosis. Taking a sample
of 41 chronic schizophrenics who seemed to be a ". . . true schizophrenic
group as it was conceived by Kraepelin, " they correlated several tests
measuring looseness of association and impaired ability for abstractions
and found that all measures were correlated significantly with each
other. As the sample consisted only of chronic patients, these measures
were therefore also correlated with poor prognosis. Because no control
group was used it is impossible to make any comparisons or generali-
zations as to the characteristic patterns exhibited by these chronic
patients in relation to other groups, particularly patients with good prog-
nosis.

The bulk of the studies on thought in schizophrenics deal mainly
with conceptual thinking. By and large, the results indicate a loss or
deficit in conceptual thinking of schiZOphrenics. However, many of these
studies, particularly the earlier ones, tended to suffer from methodological
inadequacies in the sense that the schizophrenic samples used were
usually heterogeneous groups and were not adequately described. It would
seem that the loss of the abstract attitude is insufficient to explain .
impairment in schiz0phrenic thinking, as this concerns itself with only

one aspect of the thinking process.

18

Thinking in Schizophrenics as Reflected
in the Rorschach

 

 

The comments and observations of Benjamin (1944) pertaining to
his findings on thought disturbances as measured by his Proverbs test
and the Rorschach are pertinent. He reports similar findings between
these two tests. He notes that almost all the "typically" schiz0phrenic
formal Rorschach characteristics which permit a blind diagnosis are in
the field of thinking or closely related to it (Succession, F+%, DdW and
DW, formally absurd Dd’s, and discrepancies between these and other
factors). However, his preliminary Rorschach results reveal three
important findings: 1) not all schizophrenics show signs of a formal
thinking disorder, 2) when apparent, these signs of thinking disorder
are relatively independent of the stage of the illness and the clinical
picture, and 3) there is a "definite prognostic significance of these signs. "
Benjamin hypothesizes from these observations that ". . . only those
cases with formal thinking disorders tend to "deteriorate, " and that the
severer these disorders are, the greater are the chances for, and the
quicker the course of, the deterioration. " He goes on to speculate that
if further research corroborates this hypothesis, i. e. , that there are
schiz0phrenics with and without formal thinking disorders, "this would
perhaps constitute the best criterion for distinguishing between process
and reaction. " This is similar to the hypothesis of Brackbill (1956).

The trend in Rorschach research along the line of thought disorders
has followed several theoretical orientations. Some stem from Rorschach
"theory, " others from developmental theory, and still others from psycho-
analytic theory. Despite the differing orientations they all deal with the
measurement of deviations in thinking.

In 1952, Watkins and Stauffacher published "An index of pathological
thinking on the Rorschach. " This index, stemming from theoretical

considerations about the formal characteristics of the Rorschach which

19

reflect thought patterns, arrived at an "index of deviation" in thinking
between an unselected group of psychotics, neurotics, and normals.
These formal characteristics were classified under such headings as
vagueness, confusion, autistic logic, confabulation, etc. , and each
incidence was given a specific scoring weight. They found that this
index of deviation was able to reliably discriminate among psychotics,
and normals and neurotics. However, it could not distinguish normals
from neurotics. Rieman (1953) found that several Rorschach factors
showed significant differences between neurotics and ambulatory schizo-
phrenics, these factors being essentially concerned with form level

and pathological verbalizations .

Developmental Theory and Rorschach Findings

A group of related research studies stemming from Werner’s
(1948) deve10pmenta1 theories as they pertain to perceptual development
and regression are pertinent here. First, however, it‘may be well to
summarize briefly Werner's thinking on this t0pic.

Werner conceives of the process of development as going through
various stages of differentiation and integration. Mental life proceeds
from a primitive, "syncratic" stage, in which there is little differentiation
of motor, sensory, and affective functions to a mature stage where there
is "discrete" differentiation of functions. ‘One of the characteristics of
the child's thinking, for example, is its concrete, "pars pro toto" quality.
To put this in. more explicit terms so that the relation between his
theorizing and its application to perceptual processes as measured by
the Rorschach becomes evident, Werner holds that development proceeds
from a "diffuse, " homogeneous level, to an articulated level where the
whole is divided up into related parts.

Based on the theoretical notions of Werner, Friedman (1953)

developed a scoring system for the Rorschach which was designed to

20

reflect degree of perceptual differentiation. Essentially, he used a break-
down in formulevel scoring as applied to wholes and details. Using
samples of undifferentiated schizophrenics, normal adults, and normal
children, he found, as hypothesized, that the schizophrenics, in the
structural aspects of their perception, function at a genetically lower
level, similar to, but not identical with, that of young children. Thus,

he feels it is valid that there is a perceptual regression in schizophrenics
although by no means a complete one.

Using the scoring developed by Friedman, Hemmendinger (1953,
1960) directly tested Werner's theory in a study of normal children from
3-10 years, and a group of normal adults. He found a developmental
hierarchy in perceptual integrations from global, amorphous whole per-
cepts, through mediocre details, to well integrated, form dominant
perceptions.

Siegal (1953) applied this scoring schema to various diagnostic
subgroups of schiz0phrenics. He found that the perception of the paranoid
corresponds to the more differentiated, but little integrated perception
of children between 6-10 years, while the hebephrenic and catatonic
resemble the global, amorphous perceptions of the 3-5 year olds.
However, all groups show combinations of genetically early and late
characteristics. In some early studies by Cameron (1938, 1938a, 1939),
using indices of thinking other than the Rorschach, the concept of regres-
sion was questioned. Cameron reports that he could not find the thinking
of schizophrenics comparable to the thinking of normal children and
senile patients. He concluded that the concepts of regression and
deterioration in schiz0phrenia was faulty. He felt that schizophrenics
showed a disintegration in their ability to use symbols, but that this was
a reversible process.

Several studies have attempted to apply Werner's theory of genetic

development as reflected in the system devised by Friedman in the

21

process-reactive dichotomy. Two such studies by Becker (1955,, 1956)
found that when the scoring is interpreted as revealing degree of thought
disorder, process schiz0phrenics revealed a greater degree of thought
disturbance than reactive schizophrenics (Becker, 1955). Interpreting
the scores as revealing a "genetic-level, " Becker (1956) found that
process schiz0phrenics had a lower mean "genetic-level" score than
reactives. Becker interpreted these results to indicate both a perceptual
and thought regression in process schizOphrenics. However, since both
studies lacked a normal control group, no comparisons could be made
between, e. g. , reactives and normals. Considering the fact that the
Kantor gt il.(l953) study found no differences between Rorschachs of
reactives and normals, this omission is to be criticized.

In what is essentially a repetition of Becker's later study (also
without a normal control group), Fine and Zimet (1959) found the same
differences between process and reactive schizophrenics in that the
process schiz0phrenics reveal genetically "lower" type perceptions.
However, the authors draw the analogy between genetically high and low
perceptions to primary and secondary process thinking. They speculate
that what might constitute the difference between process and reactive
schiZOphrenia may be that of a particular kind of thinking disorder.
Benjamin (1944) however, speculated that, since some schizophrenics
reveal disturbances in thinking and some do not, that the difference
between schizophrenics was a quantitative one rather than a qualitative
difference. Thus, Benjamin conceives the differences to be similar in
kind, but not in degree, while Fine and Zimet think of the differences as
being different in kind. It would seem both more parsimonious and more
well founded both clinically and theoretically, to hypothesize along the
lines Benjamin has directed. Nonetheless, the analogy that Fine and
Zimet make between genetically early and late to primary and secondary
process modes of thought are important and enables us to bridge the

gap between genetic and psychoanalytic theory.

22

Primary and Secondary Process as Reflected
in the Rorschach

 

 

The constructs of primary and secondary process thinking have
been, as far as is known to this writer, applied to an understanding of
the schiz0phrenic disorder only along theoretical, speculative lines.

To date, there has been no systematic attempt to tie these constructs
.down by means of experimental investigation. Yet, as was seen in the
previous chapter, these constructs are, according to psychoanalytic
theory, of basic importance in conceptualizing the psychotic process.

Some mention has been made (Rapaport, Gill, and Schafer, 1946)
of attempts to categorize deviant Rorschach responses. Although these
systems are gross, they allow certain inferences to be made concerning
the thinking processes, and particularly the manifestations of the
primary process. However, there has not been any attempt made to
empirically relate these deviant classes of reSponses to primary process
thinking.

In terms of differentiating schizophrenics from normals by means
of the Rorschach, most of the research has generally been inconclusive.
However, it is possible to infer from some Rorschach findings that
primary process thinking is more characteristic of schiZOphrenics than
normals. Some studies mentioned previously (Watkins and Stauffacher,
1952; Rieman, 1953) have been able to differentiate between schizophrenics
and normals on certain test patterns. The best discriminators among
Rorschach signs have tended to be those of Form Level and pathological
verbalizations. There has been some success in using the "affective
ratio" to separate a schizophrenic from a normal group (Taulbee, 1955).

In an attempt to "operationalize" the psychoanalytic concepts of
primary and secondary process thinking, Holt (1956, 1960) has devised a
highly elaborate system for scoring the Rorschach. This system will

be dealt with more Specifically in Chapter IV. Several studies

23

(Goldberger and Holt, 1958; Pine and Holt, 1959, and Cohen, 1960) have
indicated its reliability and construct validity as well as its potential
usefulness as a research tool.

The Rorschach test provides an excellent means by which to view
thought and perceptual processes. The availability of a means for
minutely analyzing these processes affords us the opportunity to test the
psychoanalytic constructs of primary process thinking as it pertains to

schizophrenia.

CHAPTER III

S TA T EMENT OF HY POTHES ES

Following the theoretical reasoning of Freud and Hartmann, it is
assumed that, given the relatively weak ego of the schizophrenic, the
ability to neutralize instinctual energy is impaired. The degree to which
this libidinal energy is neutralized coincides, according to Hartmann
(1953), with the degree to which the primary process has been replaced
by the secondary process. It follows then, that in disorders where the
ego is weakened, e. g. , schizophrenia, the ability to neutralize instinctual
energy will be impaired. The greater the impairment of neutralization,
the greater the reliance on the primary process. Given, according to
Bellak (1958), that thought disturbances are the most sensitive indicators
of ego disturbances, and that the characteristics of schizophrenic thinking
are identical with the characteristics of the primary process, the follow—
ing hypothesis can be made:

Hypothesis 1. Schizophrenics will reveal a higher mean number

of primary process responses to the Rorschach (to be defined in
Chapter IV) than normals.

 

 

 

 

The above hypothesis refers to only one measure of primary
process, total number of primary process responses. There are other
means by which the primary process can be manifest in Rorschach
responses. These would be: 1) the blatancy of the response, 2) the
existence of content and formal thought deviations, and 3) departures
from reality-oriented percepts as measured by form level. Although we
will make no specific hypotheses regarding differences among the groups

on these variables, they will be explored with the purpose of determining

24

25

the relationship among the groups and these indices of primary process
manif e stations .
A second hypothesis can be made regarding expected differences
in levels of ego functioning between the two schiZOphrenic groups. The
research mentioned in the previous chapter is suggestive that schizo-
phrenics who have relatively good premorbid histories have achieved
higher, more mature, levels of ego integration than patients with poor
premorbid backgrounds. The second hypothesis can be stated as follows:
‘Hypothesis II. Good premorbid schiZOphrenics will reveal a

significantly lower mean number of primary process responses
to the Rorschach than poor premorbid schizophrenics.

 

 

 

 

A third hypothesis can be made relative tothe effectiveness of
controls over primary process. The reasoning for this hypothesis is
that the degree to which the ego is strong determines the ability to utilize
and integrate the primary process for adaptive purposes, i. e. , to enhance
the secondary process. Hypothesis III can thus be stated:

Hypothesis III. Normals will achieve a significantly higher mean

Adaptive Regression Score (to be defined in Chapter IV) than
schiz0phrenics.

 

 

 

 

It is presumed that poor premorbid schizophrenics differ from
good premorbid schizophrenics in not having developed sufficient ego con-
trols over the primary process. This expected difference in ability to
effectively utilize the primary process can be stated as follows:
Hypothesis IV. Good premorbid schiz0phrenics will. achieve a

significantly higher mean Adaptive Regression Score than poor
premorbid schizophrenics.

 

 

 

 

CHAPTER IV

METHOD

Characteristics of the Samples

 

The present study used three samples: normals, and two schizo-
phrenic groups (n = 40) consisting of "Good Premorbids" and "Poor
Premorbids, " dichotomized according to the scale devised by Phillips
(1953, see Appendix B). The schizophrenic subjects, twenty in each
group, were obtained from two sources. One of these was the Fort Custer
Veterans' Administration HOSpital, a neur0psychiatric hospital. The other
source was the Ann Arbor Veterans' Administration HOSpital, a general
medical and surgical hospital with three psychiatric wards.

The normal sample (n = 20) was obtained from two sources.

Fifteen subjects were obtained from the local United States Army Reserve
Unit in Lansing, and five subjects came from the Veterans' Administration
counseling service at Michigan State University. By necessity, all of

the normal subjects were volunteers. Only white, male veterans were
used in both groups, the limiting of subjects to white males being consistent
with the sample on which the Phillips Scale was originally used. The
criterion for being classed as a veteran was to have served a minimum of
two years of active duty. The criterion for considering the control group
as "normal" was that they never have been referred for or received any
psychiatric care.

An attempt was made to match the three groups used in this study
(Normals, Good, and Poor premorbid schiz0phrenics) on the variables of
age, last school grade, and occupational level (Dictionary of Occupational

Titles, 1949). Table 1 reports the means and standard deviations on these

26

27

variables for the three groups. Inspection of this table shows that, for
the control variables of last school grade and occupational level, the
means for the three groups are quite similar. On the factor of age, it
was possible to match the Normal and Poor premorbid groups quite
closely. However, the mean age for the Good premorbid group was higher
than for the other two groups. The range in ages for the three groups

was similar, the Normal group ranging from 25 to 41 years, the Good
premorbid group ranged from 26 to 41, and the Poor premorbid group
ranged in age from 23 to 42 years. It was felt that, although differences
in means existed between the Normal and Poor premorbid group, and the
Good Premorbid group, these differences were not sufficient to cause the
results to deviate markedly due to this factor, particularly since the mean
age for the two schiZOphrenic groups combined was not significantly dif-
ferent from the mean age for the Normal group.

The selection of subjects for the two schizophrenic groups varied
somewhat depending upon the hospital from which they were obtained.

At the Ann Arbor V. A. Hospital, a study was in progress using the
Phillips Scale. All the clearly diagnosed schizophrenics who had already
been rated as Good premorbid on the Phillips Scale, and whose age, last
school grade and occupational level were within limits that the other groups
could be expected to fit into, were used in the sample.

At the Fort Custer V. A. Hospital, records from several wards
were screened for clearly diagnosed schizoPhrenics, who matched the
sample from the Ann Arbor Hospital on the variables of age, last school
grade and occupational level.

The two schiz0phrenic groups were obtained without regard for type
of ward, Open or closed. As it happened, the two groups came out with
similar numbers of Open and closed ward patients. The Good premorbid
group had thirteen patients from open wards and seven from locked wards,

while the Poor premorbid group had eleven patients from open wards and

28

Table 1. Comparison of the Groups on Age, Last School Grade, and
Occupational Level

 

 

Oc cupational
Age Grade Level

55 30.40 11.45 1.35
Normal

SD 4.64 3.758 1.24

55 36.25 11.15 1.85
Good

SD 2.88 3.31 .745

i 30.8 10.25 .85
Poor

SD 5.278 1.65 .671

SE 33.525 10.70 1.35
Total
Schizophrenic

SD 10.30 1.7 1.01

 

nine from closed wards. However, this does not mean to imply that the
wards, whether Open or closed, can be considered comparable. The
characteristics of the wards depended on the hospital from which the
patients came. This is due to the nature of the two. hospitals. The

Ann Arbor Hospital is a general-medical installation with three psychiatric
wards, an Open, intermediate, and closed ward. They tend to select
patients with relatively good prognoses, and their quota of patients is
limited. This is further emphasized by the fact that more than two-thirds
of the schizophrenic patients were rated as Good premorbids on the basis
of the Phillips Scale as part of a study carried out at the Ann Arbor
hospital. The Fort Custer installation, on the other hand, is a neuro-

psychiatric hospital, primarily custodial in nature. The types of wards

29

from which subjects were Obtained in this hospital could be described as
intermediary Open wards, where patients are housed with ground
privileges, an "acute ward" for the more labile, overtly disturbed
patients, and a "chronic ward" for the more withdrawn, regressed
patients. The latter two were closed wards.

The two schizophrenic groups differed considerably in terms Of
length Of hospitalization. ‘ The Poor premorbid group had a median time
of 8 years in the hospital, while the Good premorbid group's median length
of hospitalization was 2. months, 13 days. This degree of difference might
be expected given the dichotomization of the groups according to their
premorbid histories.

NO data were Obtained relative to the traditional diagnostic sub-
categories represented in the schizOphrenic samples. Although studies
have indicated that the problem of reliably establishing a diagnosis Of
"schizophrenia" is not serious, it is difficult to arrive at a reliable
diagnoses Of the schizOphrenic subclassifications (Ash, 1949; Mehlman,
1952). The schizophrenic subjects used for this study were selected only
when there was sufficient evidence that the diagnosis of schizophrenia
was reliably made. Further, evidence exists that there is little relation-
ship between the patients premorbid background and his diagnostic sub-
classification (Langfeldt, 1937; Schofield, 1954; Stalker, 1939; Wittman,
1948), and that it is not unusual for diagnoses to vary depending on the

hospital making the diagnosis.

Methodology

 

Measuring Premorbid Adjustment

 

The original scale devised by Phillips (1953) was used to predict
outcome of shock treatment. However, subsequent studies (Rodnick and

Garmezy, 1957; Seidel, 1960; and Farina, 1960) have utilized this scale

30

with broader application and have found it to be a reliable discriminator
between schizOphrenics with good and poor premorbid adjustment.

The scale in its entirety consists of three areas: 1) premorbid
history (Area I), 2) possible precipitating factors (Area II), 3) signs
of the disorder (Area .111). Although all areas contributed to the scale's
overall significance in the outcome study by Phillips, Area I was more
highly correlated with outcome than were scores from either of the other
two areas. As a result, the studies by Rodnick and Garmezy (1957),
Seidel (1960) and Farina (1960) have only used Area I. This study will
follow suit.

Area I of the Phillips scale concerns itself with relatively recent
and historical indices of premorbid adjustment. It consists of five sub-
sections to be described below (see Appendix B for the complete listing
of all categories).

A. Recent Sexual Adjustment. This section concerns itself with

 

the subjects recent sexual adjustment and heterosexual or homosexual
experiences. Its major interest is in how close a relationship has been
formed just prior to hospitalization.

B. Social Aspects of Sexual Life During Adolescence and

 

Immediately Beyond. This area is interested in determining the subjects

 

interest in relating to members of the Opposite sex and with peers.

C and D. ‘ Social Aspects of Recent Sexual Life. ' Only one Of these

 

areas is rated depending on the age of the subject. If he is over thirty
years Old, section C is rated, if younger, section D is rated. Both sections
are concerned with similar problems as are covered in section B, only

for more recent experiences.

E. Personal Relations: History. This section is concerned with

 

the subjects interest and ability to form close friendships.

F. Recent Premorbid Adjustment in Personal Relations. Essentially

 

the same as Area B, only for the period just prior to hospitalization.

31

In general, it would be fair tO say that in great measure, the scale
is concerned with an individual's interpersonal relationships in terms
Of forming close and intimate contacts with people or remaining aloof,
seclusive, and withdrawn.

The method of scoring the scale is as follows. For each of the five
areas a rating Of from O to 6 is assigned.' Ratings at the lower end Of
the scale indicate good adjustment, while the higher ratings suggest poor
adjustment. The possible extreme ratings would therefore be from "O"
to"'30" for the five subscales for anyone subject. - Rodnick and Garmezy
(1957) reliably established cutoff scores for poor and good adjustment at
the midpoint of the scale; scores of 16 or above for poor adjustment and
15 or below for good premorbid adjustment. . These cutoff scores were
also used by Seidel (1960) and Farina (1960), and will also be used for
this research. Thus, this study considered those S's who received
scores of 16 or above on the Phillips scale to have poor premorbid
adjustment and those who obtained scores of 15 or below to have estab-
lished good premorbid adjustment.

A recent study at Duke (Garmezy _e_t a_l. , 1960) has established the
reliability for rating the Phillips scale by direct interview with the
patient. Their correlation coefficients were felt to be sufficiently high
to justify the use Of the structured interview as a reliable means Of gain-
ing information about the patient's history, rather than using case history
material from hospital records.

The schizophrenic subjects from Fort Custer were rated as to
their premorbid adjustment, the ratings being established by interview
according to the scale devised by Phillips (1953). The reliability of these
ratings was established by having the interview rated by two independent
raters, the author being one rater. The rank order reliability coefficient
was . 95, indicating that the ratings were made in a highly consistent

manner by the two raters. Of the ten subjects rated, five received

32

identical ratings, and no rating differed by more than three points. There
were no category reversals, that is, no subject rated as Poor premorbid
by one rater was rated as Good premorbid by the other rater, or vice
versa. These reliability results are consistent with those generally
obtained by other workers using the Phillips Scale (Farina, 1960; Garmezy,
1960; Rodnick and Garmezy, 1957).

Measuring Primag Proc es s

 

The Rorschach scoring procedure used in this study was devised by
Holt (1960). This system has as its essential purpose the measurement
Of the incursion of primary process modes Of thought in Rorschach
responses, the determination of defenses employed against its incursion,
and a rating Of the subject's ability or lack of ability to utilize primary
process manifestations for adaptive purposes.

A brief description of the Holt system will' be given here. The
reader is referred to his manual for a more detailed examination Of the
system (Holt, 1960), or to his chapter in Rickers-Ovsienkina's volume

(1960a).

Form Level

 

Each response is first scored for form-level following the system
outlined by Mayman (1959). Essentially, this system breaks down form
perceptions into eight categories: F+ (well elaborated and integrated
perceptions), FO (popular or near popular forms), Fw+ (plausible but not
convincing), Fw- (forms that bear only a slight resemblance to the blot),
F- (arbitrary forms), Fv (vague forms, e. g. , clouds), Fa (amorphous
responses, e.g. , sky, night), Fs (spoiled form responses) (see

Appendix E).

33

‘Primarj Proc es s

 

Not all responses are scorable for primary process. Primary
process may become manifest in a response in two ways, either in the
content Of the response or in its formal qualities (see Appendix C).

There are twenty-five content categories which are divided up into
ideational and affective drive derivatives. Under ideational drive
representations are included libidinal (oral, anal, sexual, exhibitionistic-
voyeuristic, and homosexual), aggressive, and ”anxiety, " which is a
residual group presumed to be a reaction to instinctual threat. Each of
these (ideational or affective) are further subdivided into Level 1 or
Level 2, these referring to their position on the primary-secondary
process continuum. This is established by the "primitive vs. civilized"
dimension and refers to its appropriateness as a social communication
in a professional setting. Thus, the socialized responses are scored
Level 2; responses that are more "raw, direct, intense, or blatant, ”
therefore closer to primary process, are scored Level 1.

Some examples of this are as follows:

Level 2 Oral: ”Two people kissing"

 

Level 1 Oral: ("A wide open mouth"

 

Level 2 Sexual: "Two people getting married"

 

Level 1 Sexual: "Reminds me of sexual intercourse"
(See appendix C for a full list Of the content categories.)

The formal categories, of which there are thirty-four, attempt to
measure deviations from logical, orderly thinking. The following are
examples: autistic logic, fusion of percepts, internal-external view,
impossible combinations (see Appendix C for a complete listing of the

formal categories).

34

Control and Defense (CD)

 

The third aspect of the scoring has tO do with controls and defense.
Without going into the 40 categories under this heading, these are designed

to determine, for those responses scorable for primary process, the

 

nature of the controls or their absence. Some examples of the categories

of Control and Defense are, l) Remoteness of the percept, 2) the Context

 

Of the response (Cultural, Esthetic, etc.), 3) Sequence (whether the
response follows or is preceded by a, scorable primary process response).

Appendix D gives the complete list of CD categories.

Over-all Scores From the System

 

There are four over-all scores derivable from this system: defense
demand, defensive contribution, total adaptive regression score, and
number of primary process responses.

1. Defense Demand (DD). Based on the "primitive vs. civilized"

 

dimension, the manual sets the demand for each category scored, the DD
ranging from 1-6. The higher the DD, the more blatant and extreme the
response.

2. Defensive Contribution (DC). On the assumption that certain

 

defensive Operations are more effective in controlling drive representations
than others, Holt has assigned numerical weights to each Control and
Defense category. For example, a response with a human percept has a
Defensive Contribution (DC) Of 0, while animals have a DC of +1, plants

+1. 5, and inanimate objects have a DC of +2. 0. Each response will have
one or more Control and Defense (CD) categories appropriate, and each

CD will have an assigned Defensive Contribution. The DC score is the
algebraic sum of each weight assigned to the Control and Defense aspects

of the reSponse.

35

3. Total Adaptive Regression Score (ARS). The ARS for each re-

 

sponse is the product of the total Defense Demand (DD) for each response,
plus the Defensive Contribution (DC) assigned to that reSponse, plus the
Form' Level rating (FL). Thus, ARS 2 DD x (DCl + DCZ + . . . DCn + FL),

 

and the total ARS is the sum of the right hand portion of the formula for

all scorable primary process responses. High positive ARS scores are
indicative Of adaptive regression, and constitutes its Operational definition,
while high negative ARS scores are suggestive of maladaptive regression.

4.‘ Primary Process Re8ponses. All responses to which there can

 

be assigned either a content or formal category, or both, will be defined
as a response containing primary process elements.

Two hypotheses of this study are concerned with an over-all
measure Of primary process responses, which are expressed as the total
number Of scorable primary process responses. However, we will also
be concerned with other indices of primary process manifestations such
as the number of content and formal deviations, and the blatancy of
primary process responses.

Two reliability measures were obtained, one pertaining to the
reliability of scoring the Rorschach protocols according to the Holt method,
the other concerning reliability of the adaptive regression score. The
total ARS score was used to Obtain information regarding the reliability Of
scoring the protocols. Ten records were randomly selected for scoring
by an independent scorer, and the rank order reliability coefficient
between the author and the independent scorer for ARS was . 944. This
suggests that the protocols were scored in a highly consistent manner.
This degree of reliability is similar to that found in a study by Cohen
(1960), and is further corroborating evidence that the Holt system can be
reliably scored. Indirectly, the number Of primary process responses
can also be considered to be reliably scored, since an adaptive regression

score can only be assigned to a scorable primary process response.

36

In the determination Of the reliability of the ARS score itself, we
were confronted with the general problem of the reliability of Rorschach
scores. There are many difficulties inherent in the determination Of the
reliability of these scores, and there are flaws in any method (Klopfer,
et al., 1954, pp. 441-461; Holzberg, 1960). However, it would seem
that the split-half method, using Odd-even cards, presents the least
margin for error in establishing the reliability of Rorschach scores, and
this. method was used to determine the reliability of the ARS score.

The product-moment correlation coefficient between the ARS score from
odd numbered cards and those from even numbered cards is .648, which
is significant beyond the . 01 level. We may thus assume that the ARS

score is a reliable measure.

Procedure

 

The procedure essentially consisted of individually administering
the Rorschach test to each subject. Since at the Ann Arbor V. A.hospital
the subjects had already been rated as to their premorbid status, a brief
interview Of a "rapport" building nature, and the gathering Of some
general information such as age, last school grade, occupation, marital
status, and previous hospitalizations, preceded the administering Of the
Rorschach (see Appendix A for the General Information Sheet).

At the Fort Custer V. A. hospital, possible subjects were screened
via their case histories, and each potential subject was interviewed so
as to be able to rate the Phillips scale. For the first ten interviews, a
graduate intern in clinical psychology sat in on the sessions so as to be
able to establish the reliability of ratings. Following this, the author
rated all other potential subjects himself until a total of twenty subjects
rated as Poor Premorbid were Obtained. Once a subject had satisfied

the criterion of being rated as Poor Premorbid, and was matched on the

37

three control variables, he was then given a Rorschach. ' All Rorschachs
were given according to the instructions for administration as outlined
by KlOpfer, (it al. (1954). After each subject had been rated and tested,
a code number was assigned to him by a person other than the author,
and this code number was the only means Of identifying any Rorschach

protocol.

T reatment of Data

 

This section will describe the method of analyzing the data so that
the reader can get a clear idea of the procedures followed and their
rationale. The procedures to be described below apply to both the testing
of the hypotheses as well as further explorations Of the data.

For each variable analyzed, the means of the three groups (Normal,
Good, and Poor) were tested by analysis Of variance. If the F-test among
these means was not significant, no further analysis of that variable
was undertaken. However, if the analysis Of variance showed differences
among the groups, pairs Of means were tested to determine where these
differences lie. Since there are a number of possible comparisons of
means via t-tests, the risk that some might turn out to be significant at
the . 05 level by chance alone is greater than . 05. Therefore, the pro-
cedure using the "studentized range" distribution was used (Dixon and
Massey, 1957). This procedure allows all possible comparisons, with
a . 05 risk of any of them being statistically significant if no true differences
exist. For three groups of 20 sugjects each, the t-ratio required for
significance at the . 05 level using the studentized range distribution is
2.43, while for the standard t-test a value of 2.025 is required. Although
this procedure assumes homogeneity Of variance, and it was found that
for some of the variables explored the variances were not homogeneous,

it was felt that since there were no extreme departures from homogeneity

38

the added risk of bias was not sufficient to cause the results to be seriously
in error. However, where the variances are heterogeneous a more con-

servative interpretation of significance levels is required.

CHAPTER V

RESULTS

Analysis of Results

 

Before testing the hypotheses it was necessary to determine whether
the groups differed on the total number of reSponses to the Rorschach (R).
The non-significant F-ratio gives no indication Of differences in total
responses (see Table 2). It can thus be assumed that any differences

between the groups cannot be attributed to the factor Of responsivity.

Table 2. Comparison of the Groups on Total Number of Rorschach
Responses (R)

 

Group Mean SD F
Normal 23. 9 9.136

*Good 28.6 11.887 1.116 N.S.
Poor 28.15 11.604

 

3::
"Good" and "poor" refer to good and poor premorbid schiZOphrenics.

Primary Proces 5 Responses

 

Hypothesis I stated that the Total schizophrenic group would reveal
a significantly higher mean number of primary process reSponses to the
Rorschach than the Normal group. The non-significant F-ratio gives no
indication that the groups differ (see Table 3). AlthOugh the schizophrenic

groups had higher means than the Normals, the failure to find statistically

39

40

significant results gives no support to the hypothesis as stated. As a
result Of this analysis Hypothesis II also is not supported; Good pre-
morbid schiz0phrenics would have fewer primary process responses
than Poor premorbids. In fact, the Good premorbid group had a

slightly higher mean number Of primary process responses than the Poor

premorbid group (see also Table 3).

Table 3. Comparison of the Groups on Number of Primary Process

 

 

Responses
Group Mean SD F
Normal 11.95 4.879
Good 16.4 9.417 1.866 N.S.
Poor 14. 8 7.128

 

Since the over-all number of primary process responses did not
discriminate among the groups, other indices of primary process mani-

festations were explored.

Level 1 Primary Process Re8ponses

 

For each scorable primary process response the system provides
for its position on the primary-secondary process continuum by assigning
either a Level 1 or a Level 2 weight. A Level 1 weight indicates that the
response is closer to the primary process pole, while a Level 2 weight
places the response closer to the secondary process pole. As will be re-
called (see Chapter IV), the determination of the place along the primary-
secondary process continuum for each reSponse is dependent on how
blatant, raw, or intense the response. By taking an over-all measure of

the total number of Level 1 responses for each group it is possible to

41

determine whether or not the groups differ on the variableof "blatancy. "
The total number of Level 1 responses was Obtained by counting all
primary process responses containing a Level 1 weight, regardless of
whether or not the reSponse had a content or a formal deviation associated
with it. The occurrence of at least one score with a Level 1 weight was
counted as one Level 1 response. The F-ratio was found to be significant
at the . 01 level, revealing that the groups differ with reSpect to number
of Level 1 primary process responses (see Table 4). The comparisons
of the groups indicate that the Normal group differs significantly from

the Total schizOphrenic group in the expected direction, Normal having
fewer Level 1 responses than the two schizophrenic groups combined.
Although there were no statistically significant differences between the
Normal and the Good Premorbid groups, the data are highly suggestive
of differences, the means for the two groups differing in the expected
direction. There are clear differences between the Normal and the Poor
Premorbid groups, also in the expected direction. In general then,
Normals in this sample had significantly fewer Level 1 responses than
the Total schizophrenic group and the Poor Premorbid group, with strong
indications that the Normals might differ from the Good Premorbid
schizophrenics given a larger sampling Of the population. Thus, in
relation to Hypotheses I and II, on the variable Of the intensity of primary
process expression, the results indicate that schizophrenics manifest
primary process more blatantly than Normals, although they do not differ

on absolute numbers Of primary process responses.

Content Deviations

 

As described earlier (see Chapter IV), Holt has divided primary
process responses into two categories, Content Deviations and Formal
Deviations. To be scorable as a primary process response, a response
must contain either a content deviation and/or a formal deviation. The
question can be raised as to whether or not the three groups differ on

number of content or formal deviations. The purpose of these analyses

42

Table 4. Comparison of the Groups on Total Level 1 Primary Process

 

 

 

 

Responses
Grows

Normal Good Poor Total
Mean 1.8 4.3 5.8 5.05
SD 1.824 4.0 3.887 5.684
F 7. 11**
Comparison t_
Normal vs Total 3. 03%<
Normal vs Good 2. 336
Normal vs Poor 3. 74*
Good vs Poor 1.40

 

>3
Significant at the . 05 level

*4:
Significant at the . 01 level

was to determine, by an over-all measure, if the manner in which the
primary process is expressed is different for the three groups.

To get an over-all measure of the number of content deviations,
all responses containing at least one content deviation, regardless of
whether or not is was also associated with a formal deviation, were
tabulated. Any response containing at least one content deviation was
counted as one such deviation. All such deviations were counted regard-
less of whether they were Level 1 or Level 2 on the primary-secondary
continuum. The results are reported in Table 5. The analysis, plus
inspection of the means for the three groups, Offers no evidence that
the groups are different with regard to over-all numbers of content

deviations .

43

Table 5. Comparison of the Groups on Total Nutnber of Content
Deviations (CD)

 

 

 

CD
Group Mean SD F
Normal 10.25 4.216 .565 N.S.
Good 11.4 6.668
Poor 9.65 4.671

 

Formal Deviations

 

Following the same procedure as with the content deviations, the
over-all number Of formal deviations was also obtained. The F-ratio
which was significant at the .01 level (see Table 6), indicated real dif-
ferences between the groups on this factor. The comparisons of group
means is also reported in Table 6. They give significant evidence that
the Normal group differs from each of the schizOphrenic subgroups on
number of formal deviations in the expected direction, but no significant
differences between the two schizOphrenic groups on this variable were
found. Thus, the Normal sample gave fewer numbers Of formal devi-
ations, while the two schizophrenic groups, regardless of premorbid

status, gave no evidence Of differing on this factor.

Level 1 Content and Formal Deviations

 

Further analyses Of Level 1 responses were made to see if there
were any differences in Level 1 content or Level 1 formal deviations be-
tween the groups, to determine if there are differences in the manner in
which these more blatant primary process reSponses are expressed.

The occurrence of at least one Level 1 content deviation, even though it

44

Table 6. Comparison of the Groups on Total Number of Formal
Deviations (FD)

 

 

 

 

Groups

Normal Good Poor Total
Mean 3.85 8.85 8.6 8.725
SD 2.96]L 7.24 9.232 9.234
F 5. 066==<==<
Comparison t_
Normal vs Total 2. 75*
Normal vs Good 2. 82*
Normal vs Poor 2. 68>:<
Good vs Poor .141

 

lVariances are heterogeneous
**Significant at . 05 level
Significant at . 01 level

might be associated with a formal deviation, was counted as one Level 1
content deviation for that response. The same procedure was used for
Obtaining a measure of Level 1 formal deviations, each occurrence being
summed to give a total sum of Level 1 content or formal deviations.

The F-ratios for both the Level 1 content deviations and formal
deviations were significant. at the . 01 level, indicating real differences
between the groups (see Table 7 and 8). Comparisons of the group means
on these variables indicated significant differences between the Normal
and Total schizophrenic groups, and the Normal and Poor Premorbid
groups, both in the expected direction. There was no suggestion of any
differences between the two schizophrenic subgroups, pointing to a lack
of relationship between blatant content and formal deviations, and pre-

morbid status. Despite there being no statistically significant differences

45

Table 7. Comparison of the Groups on Level 1 Content Deviations

 

 

 

 

 

Groups

Normal Good Poor Total
Mean .4 1.25 2.4 1.825
SD .598T 1.618 2.521 3.11
F 6. 479""?
Comparison 3.
Normal vs Total 2. 54*
Normal vs Good 1. 52
Normal vs Poor 3. 58*
Good vs Poor 2. 06

 

T

*Variances are heterogeneous
)“(Significant at the . 05 level
Significant at the . 01 level

 

 

 

 

Table 8. Comparison of the Groups on Level 1 Formal Deviations
Groups

Normal Good Poor Total
Mean 1.5 3.65 5.1 4.375
SD 1.468T 3.99 3.998 5.746
F 5. 777M<
Comparison t_
Normal vs Total 2 69*
Normal vs Good 2.018
Normal vs Poor 3 38*
Good vs Poor 1. 36

 

lVariances are heterogeneous
M‘Significant at the . 05 level
Significant at the . 01 level

46

between the Normals and the Good Premorbid groups on these two vari-
ables, the data are highly suggestive that differences between these two
groups would likely have been found with larger sample sizes. Thus, in
general, the schizophrenics in this sample tend to give more blatant

content and formal deviations than Normal subjects.

Adaptive Regression Score

 

The third hypothesis of this study stated that Normals would achieve
a significantly higher mean adaptive regression score than the Total
schizophrenic group. Table 9 reports the means, standard deviations,
and F for the three groups. The significant F-ratio indicates that there
are differences between the groups. Although the differences between
the means for the Normal and Total schizophrenic groups were in the
expected direction, the difference for these two groups was not statistically
significant (see Table 9). Despite this lack Of support for Hypothesis III,
further comparisons between the groups reveal differences. The comparison
between the Normal and the Poor premorbid groups on ARS indicates that
the Normals have higher mean ARS scores than the Poor Premorbid group.
The difference between the Normals and the Good premorbid group was
not significant, and there is no evidence from inspection of the means for
those two groups that differences exist. Concerning Hypothesis IV,
although there were no statistically significant differences found between
the Good and the Poor premorbid groups, the t value missed significance
at the . 05 level by a narrow margin, and observation Of the means for
these two groups is strongly suggestive that differences exist. Despite
the fact that there were no statistically significant differences between the
two schiZOphrenic groups, on adaptive regression scores, the nearly
significant results raises the question of whether or not differences in
length Of hospitalization between the two groups may have contributed to

these results. In other words, we cannot be certain that the results are

47

Table 9. Comparison Of the Groups on Total Adaptive Regression
Score (ARS)

 

 

 

 

Groups

Normal Good Poor Total
Mean 56.88 55.95 26.41 41.18
SD 37.87 46.63 27.13 56.67
F 3. 973*
Comparison _t_
Normal vs Total 1. 277
Normal vs Good .075
Normal vs Poor 2.479*
Good vs Poor 2.403

 

>1:
Significant at . 05 level

due to actual differences in ARS, or to prolonged hospitalization. To test
this, an analysis of covariance on ARS, with length of hospitalization
covaried out was made. The F-ratio of 10. 56 was significant beyond the
. 01 level giving support to the contention that the differences between
the two schizOphrenic groups was a result Of differences in ARS rather
than length of hospitalization.

To summarize, Hypothesis III found partially supporting evidence

from the over-all measure of ARS, and there is some suggestion that,

 

perhaps with larger sample sizes the hypothesis might be fully supported.
The means for Normals and Good premorbid schizophrenics in this
sample were almost the same on ARS, but differences between the
Normals and the Poor premorbid groups were statistically significant in
the expected direction. There is a strong suggestion that, with further
testing on larger samples, differences would be found between the Good

and Poor premorbid groups on ARS, giving support to Hypothesis IV.

48

Total Form Level

 

Since previous studies (Becker, 1955; Friedman, 1953; Hemmen-
dinger, 1953; and Seigal, 1953) have shown that the ability to perceptually
integrate responses to the Rorschach play a part in discriminating schizo-
phrenics from other groups, an analysis Of the total form level score was
made. Form‘ Level and ARS are related in that they both are measures
of ego control. In this sense the analysis of form level scores will have
a bearing on Hypotheses III and IV. The total form level score was
obtained by taking the form level on all responses, regardless of whether
or not they had any primary process elements. The non-significant
F-ratio gives no evidence that there are any differences between the
groups on this factor, although there are differences between the means

in the expected direction (see Table 10).

Table 10. Comparison of the Groups on Total Form Level Score (T F-L)

_—
—_

 

 

T F-L
Group Mean SD F
Normal 15.0 12.63 .585 N.S.
Good 11.55 15.316
Poor 9.8 17.978

 

Primary Process Form Level

 

Since no differences were evident for total form level, the question
is raised as to whether or not form level scores are different for the
three groups on only those responses scorable for primary process. The
F-ratio comparing the three groups on primary process form level was

significant beyond the .01 level Of significance, giving clear evidence of

49

differences among the three groups (see Table 11). The comparisons
between the groups, also reported in Table 11, showed that the Normal and
Good premorbid groups, although not differing significantly between each
other, both differed significantly from the Poor premorbid group. Thus,
while the Normals and the Good premorbid schizophrenics in this sample
were about the same on their level of perceptual differentiation on primary
process responses, the Poor premorbid group had a significantly lower
level Of perceptual differentiation than either the Normal or Good pre-
morbid groups. Since form level also measures the degree of ego con-
trols over primary process productions, these results are in support of

Hypothesis IV .

Table 11. Comparison of the Groups on Primary Process Form Level
Scores (PP F-L)

 

u

 

 

 

Groups
Normal Good Poor Total

Mean 5.9 5.6 .65 3.125
‘SD 3.267T 7.387 1.268 6.965

F 7. 801**

Comparison 1

Normal vs Total 1. 858

Normal vs Good . 20

Normal vs Poor 3. 516*

Good vs Poor 3. 315*

 

Variances are heterogeneous
:I‘9‘Significant at . 05 level
Significant at .01 level

50

Summary of Results

 

The two hypotheses concerning differences in number Of primary
process responses to the Rorschach between normals and the combined
schizophrenic groups (Hypothesis 1), and between the two schiZOphrenic
groups (Hypothesis II), were not directly supported. There was no
evidence that these hypotheses would be supported even with larger
samples. We can say then, that this study gives no indication that the
Over-all number of primary process responses is related to level of ego
integration.

' With regards to Hypotheses III and IV, which were concerned with
the adaptive use of the primary process by the three groups, we found

no statistically supporting evidence for the ARS score. However, the data

 

is strongly suggestive that with further testing on larger samples both
these hypotheses would likely be supported. Thus, we can say that the
data Of this study revealed differences in the predicted direction on
adaptive regression scores between the normal and combined schizo-
phrenic group, and between the good and poor premorbid groups.

Further explorations of the data lead to the finding of differences
between the groups on several important variables.

The data suggests that, in general, the schiZOphrenics in this
sample give more over-all blatant responses, as measured along the
primary-secondary process continuum, than the normal subjects. The
manner in which these more blatant responses are expressed show no
differences between the two schizophrenic groups, that is, they both tend
to give more blatant content and formal deviations than normal subjects,
but do not themselves differ on these variables.

In terms Of the over-all manner in which the primary process is
expressed, via content or formal deviations, the data gave no evidence

Of differences on total number Of content deviations for the three groups.

51

However, there were clear differences in the number of formal deviations,
the normals having significantly fewer than the schizophrenic subjects.
There was no relationship on this factor to premorbid status, as there
was no evidence of any differences between the two schizOphrenic groups.
Thus, the schizophrenics in this sample gave, regardless of premorbid
status, greater numbers Of formal thought deviations than normals.

In terms Of level of perceptual differentiation, as measured by the
form level score, though there were no differences between the three
groups on total form level score, significant differences became apparent
when only the form level for primary process responses were measured.
In this instance, there was clear evidence of a relationship between pre-
morbid status and degree of perceptual differentiation, the normal and
the good premorbid groups having significantly higher primary process
form level scores than the poor premorbid group.

For the convenience of the reader, the results of this study are

summarized in Tables 12 and 13.

52

22,3 8. a... 88556

 

 

 

s32 mo. 6.6 28585....
0...

3.264 Euoh mmmoounm anagram n mam

H0464 Euoh 1308 .1. AMP

ouoom cofimmvummm o>flmmp< u mimic.

mcoflmgon Ransom 0 Oh

983.34,.va «GOOCOO .1. GO

momcommom mmmoounm knumgwunw. Wu naunm

*2. .m 00 .N . 2v; noonm .05 p000

sum .m .wwm .m *3. .m noonm m> EEHOZ

om . mm .H 0mm .N @000 m> HmEuOZ

ow .H *vm .N .35 .m 1308 m> Hmauoz

mumvulu

X.w.> .35va cm. 43...: .n cow; .m

17m Q0 Q0 nmunm nHunH

mum H #364 H H0464

 

Amumoh oucmOfifismmmv mfidmom mo >ud§§dm .NH 3nt

53

H964 Euoh mmOOOHnH >HMEHHnH .1. Ah nmunH

 

 

 

 

 

 

 

H96..— Euoh H.308 n Aha.
mOmcommOm mmOOOHnH .meHnnH n nHunH
mGOHHmH>OQ HmEnoh 0 Oh
9838189 36300 .1. Q0
mead m>.m woodm owed mm.o HH .m o.>H mod 9%.: Om
H.308
mmH .m 2.0.2 wH.H¢ mhmé mwé me mm.oH mo.m Aw..mH H>H
>N.H 36.: HVHKN womé mm.a de pod. www.mu MHJ. nHm
.HOOOH
mo. w.o HHVON H.m @.w Him mod w.m m3: H2
bwmé. mm.mH MOOHV ~36 HVNH. NoH wooO ozv Nata Om
HOOOO
o.m mm.HH mo.mm mod mad mNH HiHH. m6 “v.5: H)H
oom.m mo.N.H >w§m SLH 0min. mam. OHNQ wme wwHV Qm
HMEHOZ
o.m o.mH wwOm m.H mw.m H6. mm.oH w.H mo.HH 2
Ah AME mm< Oh QnH Q0 Q0 nHunH munH
mnnH H954 H HO>O..H H Ho>O1H
3:33ng pumpsmum paw mcwozv mHHdmOm HO raumgasm .mH 03.68

CHAPTER VI

DISCUSSION

The purpose Of this study was twofold. On the one hand, we were
concerned with the psychoanalytic constructs of primary process and
adaptive regression, as they relate to schizophrenic ideation and
thought. Secondly, we were interested in the measurement of these
constructs via the Holt method for the gauging of primary process
responses on the Rorschach. The ensuing discussion of the results will
therefore pertain to both the theory and the method underlying this study.
We shall also attempt to point to follow-up research along these lines.

The four hypotheses Of this study are based on psychoanalytic
theorizing concerning levels of ego functioning. Two of the hypotheses
relate to previous research findings Of differences between schizophrenics
as a function of premorbid adjustment and, therefore, of levels of ego
integration. ' It was‘presumed that, given lower levels of ego integration,
as in schizophrenia, there would be a greater reliance on more primitive,
i. e. , genetically earlier, mechanisms, particularly the mechanisms of
the primary process. It was thus hypothesized that there would be
greater "amounts" of primary process manifestations on the Rorschach
in schizophrenics than normals (Hypothesis 1). Since we can assume
differences in ego status between schizophrenics based on their premorbid
adjustment, we hypothesized that there would be differences in amounts
of primary process expression related to degree of ego maturity.

We therefore predicted that good premorbid schiZOphrenics would reveal
fewer primary process reSponses on the Rorschach than poor premorbid

schizOphrenics (Hypothesis II). The data of this study Offers no support

54

55

to either of these hypotheses. However, subsequent support may be

found below.

Level 1 Primary Process Responses

 

Since there was no evidence Of differences among the groups on
number, i. e. , degree, Of primary process incursions, we investigated
the possibility that the primary process would be revealed in the
manner in which it is expressed. Thus, the primary process may be
revealed by its intensity, deviations in thought and/or drive-dominated
content. The "manner" refers to the more qualitative aspects of primary
process expression.

In the recognition that in every response process containing a
scorable primary process element there are always aspects of ego
functioning, and, therefore, secondary process elements, Holt has in»
cluded the factor of the blatancy Of the reSponse. How blatant a response
is depends upon its acceptability as a social communication in a profes-
sional setting, both in terms of content, i. e. , how directly the impulse
is expressed, and formal qualities, i. e. , the extent and degree the
response deviates from logical considerations. It is assumed that the
more direct and blatant the response the closer it is to the primary
process. Thus, a Level 2 reSponse is closer to the secondary process,
and assumes greater ego control and flexibility over drive-laden material.
A response that is closer to the primary process end Of the continuum
may come about in two ways, following the theorizing of Hartmann (1953).
It may be as a result of weak counter-cathexes, or the relative inability
of the ego to have neutralized the drive. Both are interrelated in that
the weakness Of the ego defense results in its inability to neutralize the
impulse, and the ego's ineffective neutralization Of the drive may result
in its inability to form strong counter-cathexes. (Hartmann has posited
a possible somatic deficiency in the ego's inability to neutralize impulses

in schizophrenia.)

56

In our over-all measure of Level 1 responses we have found that,
in general, the schizophrenic groups, regardless of premorbid back-
ground, give more blatant responses to the Rorschach than do normals.
Thus, the schizophrenics tested gave greater numbers of responses to
the Rorschach which can be said to be closer to the primary process
end of the continuum than normals. We can say then, that our samples
of schizophrenics have shown that their ability to neutralize drives is

impaired, resulting in their more blatant expression.

Content Deviations

 

Before embarking on the task of interpreting the results Of this
section, it will be necessary to elucidate the meaning of a content devi—
ation. A response containing a scorable content element is one in which
there is an observable impulse derivative. That is, any such response
is assumed to be either a sexualized or aggressivized response, or both.
In this sense, it is a relatively direct expression of the underlying drive,
and can be so labeled (Oral, Anal, Sexual, Aggressive, etc.). It is
from this sexualization or aggressivization of a response that we can,
in part, infer the operation of the primary process.

The finding that the groups did not differ on number of over-all
content deviations would seem to carry the following meaning. On the
surface, it implies that the expression of libidinal drives on the Rorschach
is independent of the status of the ego. The groups seem to express
drives directly, via content deviations, in the same manner. However,
it must be kept in mind that this was a measure of total number of con-
tent deviations without regard for their position on the primary—secondary
process continuum. To get a fuller answer to the meaning of this finding
it is necessary to explore the variable of blatancy, which refers to the

position of the response along the primary-secondary continuum.

57

Although there were no differences among the groups on over-all
numbers of content deviations, the results indicate that both schizo-
phrenic groups, particularly the poor premorbid group, tended to give
more blatant content deviations than the normals. Thus, the schizo-
phrenics in this sample, in giving a response scorable for primary
process, will more likely have given a raw, intense, or blatant expression
of an impulse. It is now possible to more accurately interpret the
absence of differences on the over-all measure of content deviations.

It does not seem tenable to state that the lack of any differences on the
over-all measure is independent of the level of ego integration, as was
previously mentioned. Rather, the reverse would now seem to be true.
Although, when expressed in total numbers, both groups gave similar
numbers of content deviations, it can be seen that the content deviations
of the schizOphrenic groups were closer to the primary process pole
than those given by the normal subjects. There is, therefore, a distinct
relationship between the status of the ego and the giving of content
deviations, the healthier the ego, the more these content deviations will

be expressed in reSponses closer to the secondary process pole.

Formal Deviations

 

The libidinization Of a response via content deviations is only one
way the primary process can manifest itself. We can also view the
Operations of the primary process by means of departures from logical,
reality-oriented thinking. This latter means Of inferring the work of
the primary process by the Holt method is by the occurrence of formal
thought deviations. A scanning of the scorable categories of formal
deviations reveals that their mechanisms are similar, and closely related,
to dream processes as Freud described them (Freud, 1938). In the
sense that there is no categorizable drive associated with a formal

deviation, these deviations may be thought of as relatively indirect

58

expressions of drives and/or an attempt at their concealment. Thus,
the existence of a formal deviation is an instance of a noncommuni-
cative response, both in the sense that it does not reveal the underlying
drive, and that it calls for a relatively gross distortion of reality and
the logic that reality assumes. With this in mind, we can proceed to
interpret the obtained results.

The findings indicated that both schizophrenic groups, regardless
of premorbid status, gave greater numbers of formal deviations than
normals. Taking the meaning of a formal thought deviation as suggesting
the presence of a drive, but the distortion inherent in the response as
precluding its direct knowledge, it then becomes clear that both schizo-
phrenic groups have given greater numbers of responses that were, by
their nature, noncommunicative. In terms of these responses being
similar to dream processes, it is also appropriate to speak about a
regression to genetically earlier psychic mechanisms as a means of
dealing with reality. Given that primary process thinking is a more
primitive, or genetically earlier mechanism, and that the formal thought
deviations as measured here are the observable referents to these earlier
processes, then it follows that both schizophrenic groups have revealed
a greater reliance on these more primitive modes than do normals.

The finding that both schiZOphrenic groups have also given a greater
abundance of blatant formal thought deviations adds to the interpretive
significance of this variable, and strengthens its meaning. Since the two
schizophrenic subgroups did not differ significantly on number of blatant
formal deviations this is of importance with regards to previous
speculations on differences between schizophrenics relative to premorbid
background. Our findings indicate that, regardless of premorbid status,
this sample of schizophrenics rely more heavily on non-communicative,

genetically earlier thought processes.

59

'Summary of Findings on Primary Process

 

Since the first two hypotheses as stated in Chapter III received
no support, investigation was made as to whether other means of
gauging primary process manifestations on the Rorschach would dis-
criminate between the three groups. The two hypotheses concerning
primary process were tested by evaluating the over-all amount, or
number of primary process reSponses. It is evident, from this study,
that this Over-all measure of primary process is too gross and, perhaps,
irrelevant. The data are strongly suggestive that the manner in which
the primary process is expressed is a more important factor than the
absolute number of such responses. There were clear differences when
the manner in which primary process material manifests itself was
calculated. The schizophrenic groups, regardless Of premorbid back»
ground, utilized more verbalizations that were indicative of the break-
through of relatively blatant primary process material, either of a
libidinized or illogical nature. Thus both pathological groups tested
operated at a level closer to the primary process pole, while the normal
subjects, not differing in the degree (amount) of impulse expression,
functioned at a point closer to the secondary process pole.

It is with these results in mind that we can answer the question
as to whether or not our data are in support Of psychoanalytic theory
regarding schizophrenia. The answer would have to be in the affirmative.
The schizophrenic groups, regardless of premorbid status, revealed,
in the manner in which they responded to the Rorschach, characteristics
of ideation and thought that were closer to genetically earlier mechanisms,
and a general inability of the ego to bind impulse eXpression, or to
prevent its occurrence. Thus, we may Speak of the manner in which
the primary process is expressed in schizOphrenia as a result of faulty
repressions, or the ego's inability to maintain strong counter-cathexes

over impulse breakthrough. The manner in which the normal subjects

60

have revealed the work of the primary process has been in the

direction of a greater degree of ego autonomy, and a more socialized
expression of drives, indicating that drives have been relatively
neutralized. Therefore, our data indicate that schizophrenics do not
differ in having regressed to the utilization of more primitive modes

of representing reality, in the degree that their reSponses are non-
communicative, nor, therefore, in the manner in which they deal with
reality. In this respect, the data of this study are in support of psycho-

analytic theorizing concerning schizophrenia.

Adaptive Regres sion

 

The assumption underlying the ARS score related to the degree
to which ego controls are maintained over primary process material.
The higher the ARS, the higher the level of ego functioning. Following
theory, it was predicted that a group of people with known ego weakness
would score lower than a comparable group with presumedly higher ego
strength (Hypothesis III). This prediction received partial support, and
the data supported the prediction that there is a relationship between
this variable and premorbid status (Hypothesis IV). The finding that
schizophrenics with good premorbid histories do not differ from normals,
and that both these groups tend to differ from the sample of schizOphrenics
with poor premorbid backgrounds has some empirical precedent.
Several studies (Kantor, e_t a;l. , 1953; Becker, 1955, 1956; and Fine
and Zimet, 1959) have, if not directly then by inference, shown the
relationship between level of ego functioning and premorbid status.
Since the ARS score is presumed to be a measure of the ego's ability to
maintain realistic controls while under the sway of primary process
material, the data of this study suggest that both the Normals and Good
premorbid schizophrenics operate at similar levels of ego functioning,

these levels being higher than the level of ego functioning displayed by

61
the poor premorbid group. These results appear to lend further weight
to the findings of the investigators mentioned above, as they related

these variables to the process-reactive continuum.

Form Level

 

As pointed out earlier (Rieman, 1953) form level has been used
successfully to descriminate between normals and schizophrenics, and
this Rorschach factor has long been held to be a reliable estimate of ego
functioning (Schafer, 1954; Korchin, 1960). The degree to which good
form quality can be maintained on the Rorschach is said to be an indication
of the ego‘s ability, through the perceptual mode, to maintain reality
contact. More specifically, a response to the Rorschach containing a
primary process element can be considered to be, in large measure,

a "good" response if the form level of that response is accurate. In this
sense, form level is closely related, theoretically, to the concept of
adaptive regression and its Operational definition, the ARS score in Holt's
system.

To be considered adaptive, the ego must maintain a degree of
control over primary process incursions, and there are a variety of
ways that the ego can achieve this control. One such way is by means
of perceiving accurately. The results of this study indicate that there
is a relationship between the ability to perceive accurately, as measured
by the form level score, and premorbid adjustment. There were no dif-
ferences between the groups on total form level score. This would be
explained in terms of the lack of threat to responses without primary
process elements. ‘ Where little threat exists, the ego can function
smoothly. However, on responses scorable for primary process, there
is a sharp relationship between ego status and accurate perceptions.

The finding that normals and schizophrenics with good premorbid adjust-

ment achieve higher levels of ego functioning as measured by form level

62

than poor premorbid schiZOphrenics, lends supporting weight to previous
research using similar dichotomizations. This also gives support to

the notion that schizophrenics can be classed according to their ego
strength. However, it is necessary to restrict the use of the term

"ego" in this instance, as we are only viewing one aspect of ego function-

ing, the perceptual, when referring to form level.

Summary and Integration of Findings

 

The results of this study seem to present what may appear to be a
contradiction. On the one hand, we have shown that our samples of
schizophrenics have less of an ability to maintain control over primary
process incursions, and give more blatant expression of primary process
material. The schizOphrenic subjects tested tend to Operate at a level
closer to the primary process pole, and these relationships hold regard-
less of premorbid status. On the other hand, we have found that normals
and good premorbid schizOphrenics achieve higher levels of ego function-
ing, as measured by ARS score and form level, than poor premorbid
subjects. In short, we have found that in one instance, the level of ego
functioning was high for some schizOphrenic subjects, while, for these
same subjects, other responses indicated poor ego functioning.

Since the data indicated that, for ARS, the effects of length of
hospitalization were minimal, this contradiction will have to be explained
by other means. It will be prOposed that ARS and form level, as opposed
to other indices of primary process manifestations, measure different
aspects of ego functioning. A previous study using the Holt system
(Cohen, 1960) has found that form level is the major contributing factor
to the ARS score. We can therefore speak of these factors together when
discussing their relationship to ego functioning.

As Schafer (1954) has pointed out, the processes inherent in the

giving of a Rorschach response involve differing "levels" of ego functioning,

63

and these levels are not mutually exclusive. That is, in one response
process it is possible to view several facets of ego organization simul-
taneously. The relationship between ego functioning and the perceiving
of accurate forms relates to a level different from that of the content
and/or elaboration of that reSponse. The former reflects the ego's
ability to maintain reality contact and can therefore be taken as a
measure of the strength of the ego. However, a response may contain
a clear perception of the form of the object, but may also present a
distorted idea or thought of that object. Or, as in the case of many
complex percepts, the objects seen may be accurate, but their relation-
ship to each other may be distorted. It is in this latter sense that the
relationship between ego integration and primary process can be seen.
In other words, what a person "sees, " and how he elaborates his percept,
what he does with impulse derivatives, and how he handles the relation-
ship between his percept and reality, offers an excellent Opportunity to
view these differing facets of the ego in operation.

The findings of this study indicate that normal subjects and good
premorbid schiZOphrenics function at similar levels of ego integration,
and may therefore be said to have a similar degree of ego strength.

In this respect, the findings of the present study are not at variance with
previous research. However, both good and poor premorbid schizo-
phrenic subjects differ from normals in the E that they deal with their
percepts, schizOphrenics relying more on primitive, regressive, non-
communicative modes of representing reality than normal subjects.

Thus, in the psychological aspects of the Rorschach response process,

the schizophrenics exhibit a similar phenomenology, but different from
normals, even though some schizophrenics have achieved a good deal

of ego strength. The findings then, are in direct support of psychoanalytic

theory.

64

At this point a word is in order regarding the methodology of
this research, the Holt system for the measurement of primary process
in the Rorschach. We have found that this system can be reliably scored,
and that the ARS score is reliable, even with a psychotic population.
This alone lends heavy weight to its potential as a research tool.
However, the major criterion of its efficacy as a research instrument
lies in its ability to test theory. One of the major purposes of this study
was to attempt to give some construct validity to the Holt system. If the
results of this study are valid, then the scoring system devised by Holt
is strengthened in its validity. In fact, the results of this study suggest
that the scoring system may be more sensitive than anticipated. Although
it would appear that the over—all measures of numbers of primary process
responses, and, to some extent, the total ARS score are too gross and
ill defined, the categories which reveal the manner in which primary
process is expressed, the content and formal deviations, and the Level 1,

Level 2 distinction, may be valuable for further clinical research.

Implications for Future Research

 

Research attempts utilizing this instrument for measuring primary
process can follow several paths. A most urgent need for future work
with this system would be in the area of clinical diagnosis. The separat-
ing of pathological groups according to the traditional methods of symp-
tomatology are far frOm adequate. Perhaps a differentiation can be made
in terms of ego functioning, particularly as it pertains to the handling of
drives and the need for distortions of reality as can be seen via the
operations of the primary process and the manner in which it is handled.

It would appear to this writer that the Holt system can be readily
applied in the area of personality research. It would be enlightening to see if

any relationships existed between personality types (character structure)

65

and the manner of utilizing the primary process. It might be predicted
that, for example, compulsive characters would utilize the primary
process to a lesser degree than hysterical personalities.

Another possible area for future work with this method is in the
field of psychotherapy research. For example, it would be of im-
portance to learn if there were any differences in the manner in which
the primary process was dealt with by patients before and after psycho-
therapy, and whether or not successful and unsuccessful patients would
be differentiated by this method. It might also prove fruitful to see if
any relationships existed between the patient's and the therapist's mode
of utilizing the primary process.

Lastly, but not of least importance, would be a replication of
the present study with larger sample sizes and more finely discriminat-

ing hypothes es .

CHAPTER VII

SUMMARY

The present study was undertaken for the purpose of testing the
psychoanalytic constructs of primary process and adaptive regression
as they apply to schizOphrenia. Measures of primary process and
adaptive regression were provided by the Holt (1960) scoring system
for responses to the Rorschach, and the results provided a test‘of these
measures. Four hypotheses, based on psychoanalytic theory regarding
ego functioning, were proposed.

The subjects of this study consisted of twenty normal, twenty
good premorbid schizophrenics and twenty poor premorbid schizophrenics.
The Rorschach test was individually administered to all subjects by the
author, following the procedure outlined by Klopfer (Klopfer, it a_._l,
1954). All subjects were reasonably matched on age, last grade Of
school, and occupational level.

The Rorschach scoring followed the manual provided by Holt
(1960). This manual contains specific directions for evaluating the
degree to which primary process is manifest in Rorschach responses
and the extent of controls over primary process productions. To provide
a measure of reliability for scoring the protocols, the author and another
scorer independently scored 10 records, using the overall measure of
adaptive regression for this correlation. The inter-scorer reliability
was . 944. Intra-subject reliability for the same measure was also
demonstrated, using the split—half method. The split-half reliability

coefficient was . 648, significant beyond the . 01 level.

66

67

'A primary process response was defined as any response contain-
ing. a scorable content and/or formal deviation. The operational

measure of adaptive regression is given by the formula

ARS = DDx(FL+ DC1+ Dc2 ....DCn)

where ARS is the adaptive regression score, DD is the defense demand
for each response, FL is the form level, and DC is the defensive
contribution for each response.

Measures of primary process were also obtained by calculating
the number of content and formal deviations, and number of Level 1
responses, for the purpose of determining if the manner in which primary
process was eXpressed was different for the three groups.

The data were analyzed by using analysis Of variance first, to
determine if differences existed. If differences were found, compari~
sons of group means were Obtained with the t-ratio using the studentized
range distribution. This procedure was followed to reduce the risk of
error in making several comparisons between means.

Four hypotheses were tested. The first two were concerned with
number of primary process reSponses. It was hypothesized that the
schizophrenic groups would have significantly greater numbers of
primary process responses than normals, and that the poor premorbid
group would have more primary process responses than the good pre-
morbid group. These hypotheses were not confirmed. It was found,
however, that the manner in which the primary process is expressed
differentiated the groups. The schiz0phrenic groups, regardless of
premorbid status, were found to give responses which were closer to
the primary process pole, as measured by content and formal deviations
and the blatancy of responses. It was proposed that the primary process
is revealed more clearly when expressed as the manner of responding,

i. e. , the more qualitative expressions of primary process, than by

68

absolute number of primary process responses. When interpreted in
this way, the data were found to be in support of psychoanalytic theory.

Two hypotheses dealt with the factor of adaptive regression.

It was predicted that normals would have a higher mean adaptive regres-
sion score (ARS) than schizophrenics, and that the good premorbid group
would have higher ARS scores than the poor premorbid group. Although
neither hypothesis received statistical substantiation, the data were
strongly suggestive that these hypotheses would be confirmed with
further testing on larger sample sizes. Further analysis of the data lead
to the finding that normals and good premorbid schizophrenics achieved
higher form level scores on primary process responses than the poor
premorbid group. There were no differences among the groups on total
form level score. These findings were interpreted to mean that normals
and good premorbid schizophrenics have similar levels of ego strength,
as measured by degree of perceptual accuracy, than poor premorbid
schizophrenics. However, the two schiZOphrenic groups were judged to
manifest primary process in similar ways in that they Operated at levels
closer to the primary process than did normal subjects.

It was concluded that the Holt method for the assessment of primary
process on the Rorschach is a valid one. However, it was suggested
that the over-all measures of number of primary process responses
and, to some extent, the adaptive regression score, are too gross.

The method was felt to be a potentially valid tool in both future research
and clinical diagnosis, and suggestions for future research along these

lines were made.

BIBLIOGRAPHY

Ash, P. The reliability of psychiatric diagnosis. J. abn. soc. psychol.,
1949, 44, 272-277.

 

Becker, W. C. The relation of severity of thinking disorder to the

process-reactive concept of schizophrenia. Unp. Doct. Diss. ,
Stanford, 1955.

 

A genetic approach to the interpretation and evaluation of the

process-reactive distinction in schizophrenia. J. abn. soc. psychol.,
1956, 53, 229-236.

 

Bellak, L. Dementia praecox. New York: Grune and Stratton, 1948.

 

A multiple factor psychosomatic theory of schizophrenia.
Psychiat. quart. , 1949, 23, 738-755.

 

(Ed.) Schizophrenia. New York: Logos Press, 1958.

 

Benjamin, J. D. A method for distinguishing and evaluating thinking
disorders in schizophrenia. In Kasanin, J. S. (ed). Language and
Thought in Schizophrenia. Berkeley: University of California
Press, 1944.

 

 

Bleuler, E. Dementia praecox or the group of schizophrenias. New York:
International University Press, 1950.

 

Brackbill, G. A. Studies of brain dysfunction in schizophrenia.
Psychol. bull., 1956, 53, 210-226.

 

, and Fine, H. J. Schizophrenia and central nervous system
pathology. J. abn. soc. psychol., 1956, 52, 310-313.

 

Cameron, N. Reasoning, regression and communication in schizo-
phrenics. Psychol. Mono., 1938, 50, 1-34.

 

A study of thinking in senile deterioration and schizOphrenic
disorganization. Amer. J. psychol., 1938, 51, 650-664 (a).

 

. Deterioration and regression in schizOphrenic thinking.
J. abn. soc. psychol., 1939, 34, 265-270.

 

69

70

Cohen, I. Adaptive regression, dogmatism and creativity. Unp. Doct.
Diss., Michigan State University, 1960.

 

Clow, H. E. The use of a prognostic index of capacity for social adjust-
ment in psychiatric disorders. In Hoch, P. and Zubin, J. (Ed's).
Current problems in psychiatric diagnosis. New York: Grune and
Stratton, 1953.

 

Darrah, L. W. Shall we differentiate between schiz0phrenia and
dementia praecox? J. nerv. ment. dis., 1940, 91, 323-328.

 

Dictionary of Occupational Titles, occupational classification and industry
index. Vol. 11, U. S. Department of Labor, Washington, D. C.,
1949.

 

Dixon, W. J., and Massey, F. J. Jr. Introduction to statistical analysis.

 

McGraw-Hill Book Company, Inc. , New York, 1957.

Farina, A. Patterns of role dominance and conflict in parents of schizo-
phrenic patients. J. abn. soc. psychol., 1960, 61, 31-38.

 

Fenichel, O. The psychoanalytic theory of neurosis. New York:
W. W. Norton and Co., Inc., 1945.

 

Fine, H. J. and Zimet, C. N. Process-reactive schiz0phrenia and
genetic levels of perceptions. J. abn. soc. psychol., 1959, 59,
83"“860

 

Freud, S. Formulations regarding the two principles in mental function-
ing. Collected Papers, Vol. IV, New York: Basic Books, 1959,
(1911).

 

A note on the unconscious in psycho-analysis. Collected
Papers, Vol. IV, New York: Basic Books, 1959, (1912).

. The unconscious. Collected Papers, Vol. IV, New York:
Basic Books, 1959 (1915).'

 

. Metapsychological supplement to the theory of dreams.
Collected Papers, Vol. IV, New York: Basic Books, 1959 (1916).

 

. Neurosis and psychosis. Collected Papers, Vol. II,
New York: Basic Books, 1959 (1924).

 

The interpretation of dreams. In Brill, A. A. , The basic
writings of Sigmund Freud. New York: The Modern Library, 1938.

 

 

71

Friedman, M. Perceptual regression in schizophrenia. An hypothesis
suggested by the use of the Rorschach Test. J. proj. tech. ,
1953, 17, 171-185.

 

Garmazy, N., Farina, A., and Rodnick, E. H. Reliability studies of
the Phillips scale. Unpublished schizophrenia research project.
Dept. of Psychol. , Duke University, 1960.

Goldberger, L. and Holt, R. R. Experimental interference with reality
contact: individual differences. New York: Research Center for
Mental Health, 1958 (dittoed).

Goldstein, K. The significance Of special mental tests for diagnosis and
prognosis in schizOphrenia. Am. J. Psychiat., 1939, 96, 575-587.

 

The significance of psychological research in schizophrenia.
J. nerv. ment. dis., 1943, 97, 261-278.

 

Hanfmann, E. , and Kasanin, J. S. Disturbances in concept formation in
schizophrenia. Arch. neurol. psychiat., Chicago, 1938, 40,
1276-1282.

 

. Conceptual thinking in schizophrenia. New York: Nervous
and Mental Disease Publishing Co. , 1942.

 

Hartmann, H. Ego psychology and the problem of adaptation. In,
Rapaport, D. Organization and pathology of thought. New York:
Columbia University Press, 1951.

 

. . Contribution to the metapsychology of schizOphrenia.
Psychoanal. stud. child, 1953, Vol. VIII, 177-198.

 

. Notes on the theory of sublimation. Psychoanal. stud. child,
1955, V01. X, 9'29.

 

Hemmendinger, L. Perceptual organization and development as re-
flected in the structure of Rorschach Test responses. J. Proj. tech. ,
1953, 17, 162-170.

 

. Developmental theory and the Rorschach. In Rickers-
Ovsienkina, M. A. (Ed). Rorschach psychology. New York: John
Wiley and Sons, Inc. , 1960.

 

Hoch, T. A. Acute psychoses with symptoms resembling dementia praecox.
Am. J. Psychiat., 1921, 78, 364-372.

 

72

Holt, R. R. Gauging primary and secondary processes in Rorschach
responses. J. proj. tech., 1956, 20, 14-25.

 

Manual for the scoring of primary process manifestations
in Rorschach responses. New York: Research Center for Mental
Health, 1960.

, and Havel, J. A method for assessing primary and secondary
process in the Rorschach. In Rickers-Ovsienkina, M. A. (Ed.),
Rorschach psychology. New York: John Wiley and Sons, Inc. ,
1960 (a).

 

Holzberg, J. D. Reliability re-examined. In Rickers-Ovsienkina,
M. A. (Ed). Rorschach psychology. New York: John Wiley and
Sons, Inc. , 1960.

 

Hunt, R. C. , and Appel, K. E. Prognosis in the psychoses lying midway
between schizOphrenia and manic-depressive psychoses. Am. J.
Psychiat. , 1936, 93, 313-329.

Kant, 0. Differential diagnosis of schiz0phrenia in the light Of concepts
of personality stratification. Am. J. Psychiat., 1940, 97, 342-357.

 

. Problem of psychogenic precipitation in schiZOphrenia.
Psychiat. quart. , 1942, 16, 341-350.

 

Kantor, R. E., Wallner, J. M., and Winder, C. L. Process and
reactive schiZOphrenia. J. consult. psychol., 1953, 17, 157-162.

 

King, G. F. Research with neuropsychiatric samples. J. psychol.,
1954, 38, 383-387.

 

Klopfer, B., Ainsworth, M. D., KlOpfer, W. G., and Holt, R. R.
Developments in the Rorschach Technique. Vol. 1, Yonkers,
New York: World Book, 1954.

 

Korchin, S. J. Form perception and ego functioning. In Rickers-
Ovsienkina, M. A. (Ed). Rorschach psychology. New York:
John Wiley and Sons, Inc. , 1960.

 

Kris, E. On preconscious mental processes. In Rapaport, D.
Organization and pathology of thought. New York: Columbia
University Press, 1951.

 

. Psychoanalytic explorations in art. New York: International
Universities Press, 1952.

 

73

Langfeldt, G. Prognosis in schizophrenia and factors influencing course
of disease: Catamnestic study, including individual re-examinations
in 1936 with some considerations regarding diagnosis, pathogenesis
and therapy. Acta psychiatrica and neurologia supplementum.

1937, 13, 1-228. '

 

The diagnosis of schiz0phrenia. Am. J. Psychiat. , 1951,
108, 123-125.

 

Mayman, M. Rorschach form level manual-Menninger Clinic, 1959
(dittoed).

 

Meadow A. , and Funkenstein, D. H. The relationship of abstract think-
ing to the autonomic nervous system in schizOphrenia. In Hoch, P.
and Zubin, J. (Eds.) Relation of psychological tests to psychiatry.
New York: Grune and Stratton, 1951.

 

, Greenblatt,’M., and Solomon, H. C. 'Looseness of association

and impairment in abstraction in schizOphrenia. J. nerv. ment.
dis., 1953, 118, 27-35.

 

Meduna, L. J., and McCullough, W. S. The modern concept of schizo-
phrenia. Med. clin. N. Amer., 1945, Chicago number, 147-164.

 

Mehlman, B. The reliability of psychiatric diagnosis. J. abn. soc.
psychol. 1952, 47, 577-578.

 

Meyer, A. Constructive formulation of schizOphrenia. Am. J. Psychiat.,
1921, 78, 355-364.

 

Paskind, H. A., and Brown, M. Psychoses resembling schizOphrenia
occurring with emotional stress and ending imrecovery.
Am. J. Psychiat., 1940, 96, 1379-1388.

 

Phillips, L. Case history data and prognosis in schizophrenia. J. nerv.
ment. dis., 1953, 117, 515-525.

 

Pine, F. , and Holt, R. R. Creativity and primary process: a study of
adaptive regression. New York: Research Center for Mental
Health, 1959, (dittoed).

 

 

Rapaport, D. On the psychoanalytic theory of thinking. Int. J. Psycho-
analysis, 1950, 31, 1-10.

 

Rapaport, D., Gill, M., and Schafer, R. Diagnostic psychological
testing. Vol. 11, Chicago: Year Book Publishers, 1946.

 

Rieman, G. W. The effectiveness of Rorschach elements in the dis-
crimination between neurotic and ambulatory schizophrenic
subjects. J. consult. psychol., 1953, 17, 25-31.

 

Rodnick, E. H. , and Garmezy, N. An experimental approach to the
study of motivation in schizOphrenia. Nebraska Symposium on
Motivation, 1957, 109-184.

 

 

Schafer, R. Psychoanalytic interpretation of Rorschach testing.
New York: Grune and Stratton, 1954.

 

Regression in the service of the ego: The relevance of a
psychoanalytic concept for personality assessment. In Lindzey,
G. (Ed. ). Assessment of human motives. New York: Rinehart
and Co. , Inc. , 1958.

 

Schofield, W., Hathaway, S. R., Hastings, D. W., and Bell, D. M.
Prognostic factors in schizOphrenia. J. consult. psychol., 1954,
18, 155-166.

 

Seidel, C. The relationship between KlOpfer's Rorschach Prognostic
Rating Scale and Phillips' case history prognostic rating scale.
J. consult. psychol., 1960, 24, 46--49.

 

Siegel, E. L. Genetic parallels of perceptual structuralization in
paranoid schizophrenia: an analysis by means of the Rorschach
technique. J. proj. tech., 1953, 17, 151-161.

 

Stalker, H. Prognosis in schizophrenia based on follow-up study of
129 cases treated by ordinary methods. J. ment. sci. , 1939,
85, 1224-1240.

 

Stotsky, B. A. A comparison of remitting and nonremitting schizo-
phrenics on psychological tests. J. abn. soc. psychol., 1952,
47, 489-4960

 

Strecker, E. A. Prognosis in schizOphrenia. In Schiz0phrenia (Dementia
Praecox). Assoc. for research in nerv. ment. dis., Vol. X,
Baltimore: The Williams and Wilkins Co. , 1929.

 

Sullivan, H. S. Relation of onset to outcome in schizOphrenia. In Schizo-
phrenia (Dementia praecox). Assoc. for research in nerv. ment.
dis. , Vol. X, Baltimore: The Williams and Wilkins Co. , 1929.

 

75

Taulbee, E. S. and Sisson, B. D. Rorschach pattern analysis in
schizophrenia: A cross validation study. J. clin. psychol.,
1954, 10, 80.

 

Vigotsky, L. S. Thought in schizophrenia. Arch. neurol. and psychiat.,
1934, 31, 1063-1077.

 

Watkins, J. G., and Stauffacher, J. C. An index of pathological think-
ing in the Rorschach. J. proj. tech., 1952, 16, 276-286.

 

Werner, H. The comparative psychology of mental development:
Revised Ed. Chicago: Follet Publishing Co. , 1948.

 

Wittman, P. Scale for measuring prognosis in schizOphrenic patients.
Elgin State Hospital Papemrfi, 1941, 4, 20-33.

 

. Diagnostic and prognostic significance of the shut-in
personality type as a prodromal factor in schizOphrenia.
J. clin. psychol., 1948, 4, 211-214.

 

_ , and Steinberg, D. A study of prodromal factors in mental
illness with special reference to schizophrenia. Am. J. psychiat. ,
1943, 100, 811-816.

 

APPENDICES

76

APPENDIX A

INFORMA TION SHEET

 

 

 

 

 

 

 

 

Code:
Name: , Age:_____ Mos:_____ Year:
Last School Grade:
Occupation:
Job Description:
Job Level (rate)
3. Managerial-professional
2. Skilled (D. O.T. , l, 4 and 5)
1. Semiskilled (D. O. T. 6 and 7)
0. Unskilled (D. O. T. 8 and 9)
Marital Status: Married: Separated: Divorced:
How long married: Previous Marriages:____Chi1dren:

 

Previous Hospitalizations:

Where?

 

How Long ?

 

Previous Outpatient Treatment:

 

 

Current Admission Date:

 

Date Tested:

 

Diagnosis:

 

Rating:

 

Remarks:

 

 

 

 

77

Patient' 5 Name

A. Recent Sexual Adjustment

1-0
2-1
3-2
4-3(a)
(b)

5-4(a)
(b)

6-5(a)
(b)

7-6

(b)
6-5(a)

(b)
7-6

A PPENDIX B

Date Rater

 

 

 

Phillips Scale
Section 1 Premorbid History

Stable marriage

Marriage but unable to establish home
Marriage broken by permanent separation
Marriage but with low sex drive

Deep heterosexual relation, but unable to
develop it into marriage

Casual but continued heterosexual "affairs"
Homosexual contacts, with lack or failure in
heteros ex

Occasional casual homo- or heterosex
Solitary masturbation, no attempt at homosex or
heterosex experiences

No sexual interest in men or women

Social Aspects of Sexual Life During Adolescence
and Immediately Beyond

Always healthy interest in girls, with steady
girl during adolescence

Started dating regularly in adolescence
Always mixed closely with boys and girls
Consistent deep interest in male attachments,
restricted or no interest in girls

Casual male attachments, with inadequate attempts
at dating girls

Casual contacts with boys and girls

Casual contacts, lack interest girls
Occasional contacts with girls

NO desire to be with boys and girls; never
went out with girls

Social Aspects of Recent Sexual Life;
30 years of age and above

Married with children, living as family unit
Married and children but unable to establish

or maintain a family home

Has been married and children but perm. septd.
Married but considerable marital discord
Single, has had engagement or deep heterosex
relationship but unable to develop into marriage

78

5-4
6-5(a)

(b)

7-6(a)
(b)

(a)

(b)

www-
1
rbwv—I

U'Irk
ll
O‘UW

79

Single, with emotionally shallow "affairs"

Single, has dated a few girls but a generally dis-
continuous interest in girls

Single, consistent deep interest in male attachments
but no interest in women

Single, occasional male contacts, no interest in women
Single, interested in neither men nor women

Social Aspects of Recent Sexual Life; below
30 years of age

Married, living as a family, with or without kids
Married, with or without kids, but unable to establish
or maintain a family home

Single, engaged or in a deep relationship which is pre-
sumably leading to marriage

Single, has had engagement or deep relationship but
unable to deveIOp into marriage

Single, consistent deep interest in male attachments with
restricted or no interest in women

Single, casual male relationships with restricted or

no interest in women

Single, has dated a few girls casually but generally
discontinuous interest in women

Single, never interested in or associated with men or
women

Antisocial
Personal Relations: History

Always had a number of close friends

From adolescence on had a few close friends
From adolescence on had a few casual friends
From adolescence on stOpped having friends
No intimate friends after childhood

Casual but never any deep intimate friends
Never worried about boys or girls; no desire
to be with boys and girls

Recent Premorbid Adjustment in Personal Relations

Habitually mixed with others (though not a leader)
Mixed with only a close friend or group of friends
No close friends; very few friends; had friends but
never quite accepted by them
Quiet, aloof, seclusive, preferred to be by self
Antisocial
Total:
Disagreement:

A PPENDIX C

Primary Process Categories Used for Scoring the Rorschach
Protocols, and the Defense Demand (DD) Associated with Each

Content (D155)
L. Libidinal

1 O. Oral l ........................ (2-4)

L

L 1 0. Anal 1 ............ . ........... (3,4)
L 1 S. Sexual 1 ....................... (4)
L l E-V. Exhibitionistic-voyeuristic 1 ............ (3)
L l H. Homosexual l ............ . . . . . . . . (4)
L 1 M. Miscellaneous libidinal 1 .............. (3,4)
L 2 0. Oral 2 ................. . ...... (1-3)
L 2 A. Anal 2 ............... . ........ (1-3)
L 2 S. Sexual 2 ....................... (1)

L 2 E-V. Exhibitionistic-voyeuristic 2 ............ (1-3)
L 2 H. Homosexual 2 .................... (1-3)
L 2 M. Miscellaneous libidinal 2 .............. (2, 3)

Ag. Aggressive

 

Ag 1 P-S. Potential--subject 1 . . . . . . . . ........ (4, 5)
Ag 1 P-O. Potentia1--Object 1 ................. (3,4)
Ag 1 A-S. Active-~subject l .................. (4, 5)
Ag 1 A-O. Active--object 1 .......... . ........ (4, 5)
Ag 1 R. Results 1 . . . . . ........ . . . ...... (4,5)
Ag 2 P-S. Potentia1--subject 2 ................ (2, 3)
Ag 2 P-O. Potentia1--object 2 ................. (1, 2)
Ag 2 A-S. Active--subject 2 ......... . ....... . (2)

Ag 2 A-O. Active--object 2 .................. (2)

Ag 2 R. Results 2 . . . . . . . ............... (1-3)
Aff 2. Affective drive-derivatives 2 . . . . . . ........ (3)

 

 

*DD 2: Defense Demand
80

81

 

Formal (DD)
C. Condensation
C f-p 1. Fusion of percepts ................. (4)
C i-e 1. Internal-external view ............... (4)
C p-f 1. ' Partial fusion ..... . . ............ ‘ (3)
C u-p 1. Unrelinquished percepts .............. (2)
C-co 1. Compositionl . . . . ............... (3,4)
C-co 2. Composition 2 ............... . (2)
C a-l 2. Arbitrary linkage of percepts ........... (2, 3)
C a-c 12. Arbitrary impossible combinations . (3)
C a-c u2. Arbitrary unlikely combinations ...... . (2)
Arbitrary combinations of color and form ......
FCarbl.FCarb....... ................ .(3)
FCZ 2. F5 or F/C . ................... . (2)
DO 2. Fragmentation . . . . ........... (1)
Imp 2. Impressionistic response . . . .......... (2, 3)
Trans 1. Fluid transformation of percept ...... . (3)
C-sym 1. Color symbolism ..... I ............. (3)
C-sym 2. Color symbolism .............. (1)
S-sym 1. Spatial symbolism .............. . (3)
I-sym 1. Concrete image symbolism . . ..... . (3)
I-sym 2. Concrete image symbolism . . ........ (1)
Au Lg 1. Autistic Logic .............. . (5)
DW 1. DW ............ . . . ..... . (4)
Ctr A 1. Affective contradiction . . . . . . . . . . . (4)
Ctr L 1. Logical contradiction ........ (3, 4)
Ctr In. Inappropriate activity . . . . . ...... (3, 4)
VS 2. Verbal slips ...... . . . ........... (2)
VP 2. Peculiar verbalizations . . . . .......... (3)
V0 1. Queer verbalizations . . . . . .......... (4)
VC 1. Verbal condensation . . . . . . ...... . (4)

Formal

VI 1.

Au El 1.

Au E12.

S-R 1.

F-Msc l or 2.
ML 1.

Ii.

82

Verbal incoherence, confusion . .

Autistic elaboration, Level 1 . . .

Autistic elaboration, Level 2 . . .

Self-reference ........

. Miscellaneous formal aspects

Loosening Of memory .....

Intrusion of irrelevancy . .

00000

(DD)

. (5)
. (5)
. (3)
. (2,3)
. (3,4)
. (4)

APPENDIX D

Control and Defense Categories Used to Score the
Rorschach Protocols, and the Defense Contribution

(DC) Associated with Each

 

Sequence (DC)
Recovery, modifying percept . . ..............
. S M 1-O. Level 1 - unscorable . . ....... (+1.5)
S M 2-O. Level 2 - unscorable ....... (+1)
S M 1-2. Level 1 - Level 2 . . . . ............ (+.5)
S M R+. . Rationalization of percept ....... (+2)
S M R-. Rationalization of percept ....... (-2)
Recovery, changing percept ....... . .........
S C l-O. Level 1 - unscorable . . . . ....... (+1)
S C 2-0. Level 2 - unscorable . . . . ....... . (+1)
5 C 1-2. Level 1 - Level 2 . ............. (+.5)
Regression, modifying percept ......
S M 0-1. Unscorable - Level 1 . . ....... . (-l)
S M 0-2. Unscorable - Level 2 .............. (-l)
SM2-l. LevelZ-Levell. . . . .......... (-.5)
Regression, changing percept . ...............
S C 2-1. Level 2 - Level 1 ......... . (-.5)
122.131
Inhibition .................... (+1;+l. 5)
Reflection on Response
Introspection . . . ............... (+Z;-1)
Criticism of response . .......... . (+2;-2)

 

3’6
DC = Defensive Contribution; X = No control

83

84

 

 

 

Remoteness (DC)>§<
R-min. Minimal remoteness . ........... . (0)
R-eth. Ethically remote from S ....... . (+. 5)
R-an. Animals .................. . (+1)
R-pl. Plants . . ..... . .......... . (+1.5)
R-ia. Inanimate ............... . (+2)
R-dep. Depictions ................. . (+1)
R-geo. Geographical remoteness ........ . (+2)
R-tm. Remoteness in time ............ . (+2)
Remoteness in level of reality
R-fic 5+, -. Specific fictional ................ (+2;-1)
R-fic n+, -. Non- specific fictional .......... (+2;-l)
R-rel +, -. Religious character or context ......... (+2;-l)
R-fan +, -1. Explicit fantasy or dream ....... (+2;- 1)
R-fig+. Figures of Speech ............... (+2)

Context of reSponse (DC)
Cx C+, -. Cultural context .............. . . (+2;-l)
Cx E+, -. Esthetic context . .............. . (+2;-1)
Cx 1+, -. Intellectual context ....... - ....... . (+2;- 1)
Cx H+, -. Humorous context ............. (+2;-1)

Pathological defenses
Va- Vagueness of percept . . ...... (-3)
Prj-. Projection of responsibility ........ . (-3)
Neg+, -. Negation and undoing . . . . ......... (+1; - 2)
Eu- . Euphemism ................... (-1)
Minz-. Minimization . . . . . . . . ..... . . . (-l)
Obs-. Obsessional defense . ............. (-1)
Den-. Attempted denial ........ . ..... . (-2)
Ev- Evasion, ..... . . . . . . . . . . . . . .(-2)
Imp- . Impotenc e .................... (- 2)

A PPENDIX E

Form Level Scores (FL) and Examples for Scoring
From Holt (1960)

Form Level Score Category Explanation
+3 F+ Sharp, convincing forms, easily
seen by E.
+2 Fo Popular and near popular forms,

Fixed list in manual (Mayman, 1959)°

+1 Fw+ Reasonably plausible, but not terriby
convincing forms; takes a little
stretching to see.

-1 Fw- Forms that bear only a slight re-
semblance to the blot area; not very
plausible, or based on one point of
resemblance.

-2 Fv Vague, non-definitive forms--things
that intrinsically do not have specific
shapes. "Clouds"; "blood stain”;
"piece of dough".

-3 Fa Amorphous responses, in which form
plays no role (and could not, by the
nature—Of the concept).

Usually pure C.
"Sky”; "water"; “night” ”spring";
(and other abstract concepts).

-3 F5 Spoiled form responses, to be given
' where the subject gives what is basically
a familiar and good response (which
could have been scored F0 or F+) but
introduces some specification that has
the effect of markedly lowering the
acceptability of the response as a whole.

-4 F- Arbitrary forms, bearing very little or
no resemblance to the blot.

85

 

 

.l.

v“

 

. ....y., n .\.. , E‘tawfiklfi...

. L

. , 1.3;...
.‘t.... Var? . p w... , v
. . . zifisnfilrgcwxufiubu

    

alums...

 

 

4“

0 01.." llulllll'

9
7
9
2
5
7
1
3
0
3
9
2
4|
3

w.
Rl
A“
R!!!
m."
Ll”
V"
T“
S“
RI!”
E”
v"
N“
U"
u
M
I“
H
"