THE RELAUOR OF SEX, SBEMIHCATKER,

AND SEiF-ACGEFTMCE T8 REGRESS§0fl

3N “5&5 SERVECE *3? "535 {‘30

a

“'S%2:3~ ‘1‘" *"a :-:="~':‘3:: c": ‘5 53‘.
1.33.131: .v. w Avénvv «a axo
nun! :55 f 4'3?” "$7" '13.! "k‘x'fizi
-: #5.:3 363:. 4 ’ ~¢\‘ u
Kuhnuu no. 61-.1'. 9 .st'U‘Aa .J

. clay-1P. “firifr'g?

:3". 343:.- Q :a-ghzagv,‘

£1.1vé Q. “.3. i‘!) [but uni-35“! 8

saw.“

a...“
‘0‘,“

o”‘-H

 

,"..a f.
4, l -
l < IL ' '
4:
I‘WI
‘ O
3 fl '

   
 

 

I. museum. mm ‘11

-~;\/
g

r7 _\._./1-- '

 

 

ABSTRACT
THE RELATION OF SEX, IDENTIFICATION,

AND SELF-ACCEPTANCE TO REGRESSION
IN THE SERVICE OF THE EGO

By

Michael A. Teixeira

Psychoanalytic theory and subsequent empirical research on
what has been conceptualized by Kris as "regression in the service of
the ego," or "adaptive regression," has affirmed the relationship be-
tween regression in the service of the ego and creativity. The present
study undertook to explore the relationship between regression in the
service of the ego and sex, identification, and self-acceptance. Three
hypotheses were proposed: that regression in the service of the ego
would be positively related to cross-sex identification for both sexes;
that regression in the service of the ego would be positively related
to self-acceptance; and that no difference would be found between the
sexes on regression in the service of the ego.

The §§_were 36 undergraduate students equally divided between
males and females and selected from an initial sample pool of 54 males
and 63 females assessed for predominant identification and self-acceptance
using a semantic differential technique. The 18 subjects of each sex
were differentiated by predominant identification into three male and

three female subgroups: maternal, paternal, and mixed identification.

P'.‘
w. ,
-' .‘ fl/ 3'
’ “I? V". r ,' t a

l “W‘”

Michael A. Teixeira

Subjects were administered the Rorschach individually, and protocols
were scored for adaptive regression according to Holt's (1970) scoring
system for primary process.

Comparing males and females, findings suggested that the
capacity for regression in the service of the ego was greater for
females than for males. For both males and females, the capacity for
regression in the service of the ego showed a negative trend with
paternal identification.

For males, the capacity for regression in the service of the
ego appeared to relate significantly more to maternal than to paternal
identification. Maternal-identified males showed the only significant
positive relationship between adaptive regression and maternal identi-
fication, and to a lesser degree, between adaptive regression and
paternal identification. Consistent with this, the maternal-identified
males showed evidence of having the greatest capacity for regression in
the service of the ego of the total male sample.

For females, the capacity for regression in the service of the
ego showed a non-significant negative relationship to paternal identi-
fication, and a significant negative relationship to maternal identifi-
cation. The maternal-identified females showed the only significant
relationships, both negative, between adaptive regression and maternal
and paternal identifications.

Self-acceptance was found to relate positively to the capacity
for regression in the service of the ego for the total sample of males

and females.

Michael A. Teixeira

The mixed-identified males were remarkable in that this parti-
cular subgroup showed negative correlations between self-acceptance,
maternal identification, and adaptive regression, and scored signifi-
cantly lower on adaptive regression than the total male and female
sample, and three of the five other subsamples.

The findings for females with regard to the total adaptive
regression score suggested that this score (ARS) may be more salient
for males due to possible sex-differences related to the components

that constitute it.

THE RELATION OF SEX, IDENTIFICATION,
AND SELF-ACCEPTANCE TO REGRESSION
IN THE SERVICE OF THE EGO

By

Michael A. Teixeira

A THESIS

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

MASTER OF ARTS
Department oprsychology

1976

ACKNOWLEDGMENTS

In completing this thesis, I would like to express my
gratitude and appreciation to the members of the committee--to
Dr. Albert 1. Rabin (Chairman), Dr. Norman Abeles, and Dr. Robert
Zucker--for their suggestions, criticisms, and guidance through
this project.

In particular, I would like to thank Dr. Rabin for his
scholarly and astute observations and comments, perceptive but
firm guidance, and patient support and encouragement.

I would also like to acknowledge the thought-provoking
dialogues and supportive encouragement from my friends and com-

peers, who know how much our continuing friendship means to me.

ii

TABLE OF CONTENTS

Page
LIST OF TABLES ........................ iv
LIST OF APPENDICES ...................... v
INTRODUCTION ......................... l
THEORETICAL BACKGROUND .................... 2
Regression in the Service of the Ego ........... 2
Identification ...................... 4
The Relation between Regression in the Service
of the Ego and Identification .............. 6
Empirical Studies of Adaptive Regression and
Identification Utilizing the Rorschach Test ....... 8
HYPOTHESES .......................... 13
METHOD . . . . . . . . . . . . . . . . . . . . . ....... 16
The Sample ........................ 16
The Instruments and Scoring ............... 16
Procedure ........................ 19
Statistical Analysis ................... 21
RESULTS ........................... 22
Adaptive Regression and Identification .......... 22
Adaptive Regression and Self-Acceptance ......... 23
Adaptive Regression and Biological Sex . ......... 24
Additional Results .................... 26
DISCUSSION . ......................... 32
Adaptive Regression and Identification .......... 32
Adaptive Regression and Self-Acceptance ......... 35
Adaptive Regression and Biological Sex .......... 35
-Identification and Self-Acceptance ............ 36
Summary and Integration of Findings ........... 37
APPENDICES .......................... 41
REFERENCES .......................... 54

iii

Table

1C.

2C.

LIST OF TABLES

Means and standard deviations of identification

scores for the original sample pool ..........

Correlations between identification scores and

adaptive regression score (ARS) ............

Correlations between the measure of self-acceptance

(S-A) and adaptive regression score (ARS) .......

Analysis of variance of adaptive regression scores

(ARS) for all subjects ................

Correlations between adaptive regression score (ARS)

and identification scores for males ..........

Correlations between adaptive regression score (ARS)

and identification scores for females .........

Correlations between the measure of self-acceptance

(S-A) and identification scores ............

Correlations for male groups between the measure of

self-acceptance (S-A) and identification scores . . . .

Correlations for female groups between the measure of

self-acceptance (S-A) and identification scores . . . .

Comparison of the male groups on adaptive regression

score (ARS) ......................

Comparison of the female groups on adaptive

regression score (ARS) ................

iv

Page

20

22

24

25

27

28

29

3O

31

52

52

LIST OF APPENDICES

Appendix Page
A. Listing of All Scores Contained in the Holt

Manual for Primary Process Scores .......... .41

B. The Semantic Differential .............. 48

C. Comparison of Groups on Adaptive Regression Score . . 53

INTRODUCTION

Much of the theoretical and empirical literature on the
adaptive mental process first described by Kris (1952) as "regression
in the service of the ego," or by Hartmann (1958) as "adaptive regres-
sion," has dealt with the relation of regression in the service of the
ego to creativity, notably to artistic creativity.

In general, the findings of experimental studies designed to
test the psychoanalytic hypothesis of a relationship between regression
in the service of the ego and creativity, using the Rorschach test,
have been positive (Pine and Holt, 1960; Cohen, 1960; Dudek, 1968;
Holt, 1970; and Borofsky, 1971).

The present study will investigate a related hypothesis
suggested by psychoanalytic studies of creative personalities; namely,
that associated with the capacity for regression in the service of the
ego is salient crossesex identification.

The relationships between regression in the service of the ego,
biological sex, self-acceptance, and predominant identification will

be examined as well.

THEORETICAL BACKGROUND

Regression in the
Service of the Ego

In The Interpretation of Dreams (1914 revision), Freud distin-

guishes between three kinds of regression: "(a) topographical

regression, in the sense of the schematic picture of the W-systems. . . °

(b) temporal regression, insofar as what is in question is a harking
back to older psychical structures; and (c) formal regression, where
primitive methods of expression and representation take the place of
the usual ones." It is formal regression--in which there is a regres-
sive shift in psychic functioning from the secondary to the primary
process--which has been emphasized in theoretical discussions of
"regression in the service of the ego" (Holt, 1970; Borofsky, 1971).

Kris (1952) acknowledges Freud with advancing the first general
hypothesis of what would later be described by Kris as "regression in
the service of the ego." In his 1915 paper on "The Unconscious,"

Freud states that:

Cooperation between a preconscious and an unconscious impulse,
even when the latter is subject to very strong repression,

may be established if the situation permits of the unconscious
impulse operating in harmony with one of the controlling ten-
dencies. The repression is removed for the occasion, the
repressed activity being admitted as a reinforcement of the

one intended by the ego. In respect of this single constella-
tion the unconscious becomes ego-syntonic, falls into line with
the ego, without any change taking place in the repression
otherwise. The effect of the unconscious in the cooperation is

unmistakable; the reinforced tendencies reveal themselves
as, in spite of all, different from norma1--they make pos-
sible achievements of a special perfection, and they mani-
fest a resistance in the face of opposition similar to that
of obsessional symptoms.

Later, in Introductory Lectures on Psychoanalysis (1917),
Freud remarked on the constitutional disposition of the artist as com-
bining a powerful capacity for sublimation with a certain flexibility

of repression. However, Kris maintains that "this flexibility, or

 

whatever other and satisfactory characteristics we might establish,
is clearly not limited to the artist: These characteristics are
related to those conditions in which id impulses intrude upon the
ego. . . ." [underlining mine, MT.].

Schafer (1958) mentions what can be taken as examples of the
foregoing "conditions in which id impulses intrude upon the ego."
_They are artistic creativity and the audience's response to it, pro-
ductive fantasy and imaginative processes, wit and humor, sleeping and
dreaming, problem solving, ego-building identifications, empathy,
motherliness, intimacy and love, capacity for orgastic experience, the
therapeutic process including both patient and therapist, the hypnotic
process and hypnotic state, and the response to and interpretation of
projective test material. He gives the following definition of regres-

sion in the service of the ego:

Regression in the service of the ego is a partial, temporary,
controlled lowering of the level of psychic functioning to
promote adaptation by maintaining, restoring, or improving
inner balance and organization, interpersonal relations, and
work. It is a process which increases the individual's access
to preconscious and unconscious contents, without a thorough-
going sexualization or aggressivization of major ego functions,
and therefore without disruptive anxiety and guilt. In other

words, the primary and secondary (relative) autonomy of
higher ego functions is not impaired; the encroachment of id
tendencies is circumscribed. The process implies central
controlling functions in the ego which may suspend some

other functions, such as defensive functions and logical func-
tions, and may emphasize genetically primitive mechanisms,
such as projection and introjection (p. 122).

. . it refers, then, to the ego's permitting relatively

free play to the primary process in order to accomplish its
adaptive tasks. The ego detours through regression toward
adaptation. It is warranted to speak here of regression in-
sofar as primary process or its close derivatives, normally
warded off, are allowed a place in conscious experience; and it
is warranted to speak of the process being in the service of
the ego insofar as the regression serves ego interests (such
as being creative or empathic), is relatively easily revers-
ible, and is amenable to productive working over by the ego

in terms of its adaptive pursuits (p. 125).

Kris and Schafer seem to be affirming that the capacity for re-
gression in the service of the ego (characterized by fluidity of
repressive mechanisms, shifts in levels of psychic functioning, etc.)
is a differential human potential or disposition existing in other than
only artistically creative personalities. We may wonder, then, if
there is not a dynamic core related to this capacity for adaptive regres-
sion. Considering the psychoanalytic hypothesis of bisexuality in the
artist and research findings of cross-sex identification in creative
personalities, a relationship is suggested between the capacity for

regression in the service of the ego and cross-sex identification.

Identification

 

According to psychoanalytic theory, there is a biological basis
for sexual disposition; however, "it is the strength of identification
with one or the other of the parents which is the key determinant of

masculinity and femininity" (Levy, 1961). In the course of normal

development it is expected that the child's predominant identification
will be with the parent of the same sex.

Freud (1923) outlines the process of identification in The Ego
and the Id. He observes that in the male child, there develops at a
very early age an object-cathexis of the mother based on the anaclitic
model. The little boy relates to his father by identifying himself
with him. The oedipal conflict develops as the boy's sexual wishes
towards his mother become more intense and he begins to perceive his
father as an obstacle and rival for the mother. The boy's identifica-
tion with his father changes to a more ambivalent attitude towards him,
with the wish to get rid of and replace the father in relation to the
mother. Freud suggests that because of the constitutional bisexuality
originally present in the child, the oedipal complex usually takes a
twofold form, both positive and negative. The male child develops not
only an affectionate object-relation towards his mother and an ambiva-
lent relation to his father, but an affectionate feminine attitude
towards his father and a jealous attitude towards his mother as well.

Freud (1923) states that:

Along with the dissolution of the Oedipus complex the
object-cathexis of the mother must be given up. Its place
may be filled by one of two things: either an identifica-
‘tion with the mother or an intensified identification with
the father. We are accustomed to regard the latter outcome
as the more normal. . . . In this way the passing of the
Oedipus complex would consolidate the masculinity in the boy's
character. In a precisely analogous way, the outcome of the
Oedipus attitude in the little girl may be an intensification
of the identification with the mother (or such an identifica-
tion may thus be set up for the first time)--a result which
will stamp the child's character in the feminine mould (p. 41).

And,

The broad general outcome of the sexual phase
governed by the Oedipus complex may, therefore, be taken
to be the forming of a precipitate in the ego, consisting
of these two identifications with the mother and with the
father in some way combined together. This modification
of the ego retains its special position; it stands in con-
trast to the other constituents of the ego in the form of
the ego-ideal or super-ego (p. 44).

It can be seen that the ego-ideal arises from the internal-
ization (or introjection [Knight, 1940]) of attributes of both the
mother and the father in the form of a mother-identification and a
father-identification. Moreover, "the relative intensities of the
two identifications in any individual will reflect the preponderance
in him of one or the other of the two sexual dispositions" (Freud,
1923). Thus, an individual's identification is related to the inten-
sity of the identification with one or the other, or both, of the

parents.

The Relation between Regression in the
Service of the Ego and Identification

Freud, in his studies of Leonardo da Vinci (1910) and Fyodor
Dostoevsky (1928), established and gave emphasis to a particularly
strong feminine (bisexual) component in the personalities of both
these creative men. Lowenfeld (1957) suggests that "this formula
[heightened bisexuality] may well hold true for the artist in general."
Kris (1952) comments that bisexuality in the artist is related to
another of Freud's hypotheses about artistic creativity, that of the

artist's "flexibility of repression" (1917).

The hypothesis of bisexuality in the artist receives parallel

recognition in the Analytic Psychology of Carl Jung and his followers:

We find a fundamental, not a pathological, phenomenon in
the dominance of the mother archetype, i.e., of a suprapersonal
mother image, in the creative man. In normal development,
the man's 'feminine component' is largely repressed and con-
tributes to the constellation of the anima in the unconscious.
But in the creative man this process is incomplete. By his
very nature he remains in high degree bisexual, and the
retained feminine component is manifested by his increased
'receptivity,‘ by his sensibility and a greater emphasis in
his life on the 'matriarchal consciousness' expressed in in-
ward processes of parturition and the formation that essentially
condition his creativeness (Neumann, 1959, p. 18).

In studies of creative personalities discussed in his paper "On
the Nature and Nurture of Creative Talent" MacKinnon (1962) presents
research findings which would seem to be consistent with the hypothe-
sized relationship betWeen the capacity for regression in the service
of the ego and salient cross-sex identification. MacKinnon studied dis-
tinguished architects selected for their acknowledged creativity and
found them to have "ambiguities in identification with the parents."

The architects tended to have identified either with both parents or
with neither, rather than making the more usual clear identification

with the parent of the same sex. Moreover, MacKinnon states that:

The most striking aspect of the MMPI profiles of all
our male creative groups is an extremely high peak on the
Mf (femininity) scale of the California Psychological In-
ventory (CPI) (Gough, 1957) and on the masculinity-
femininity scale of the Strong Vocational Interest Blank
(Strong, 1959). Scores on the latter scale (where high
scores indicate more masculinity) correlate -.49 with
rated creativity (p. 488).

Highly creative individuals were also found to score high on a scale

measuring basic acceptance of the self.

Helson (1966), a colleague of MacKinnon, studied undergraduate
women with imaginative and artistic interests since childhood. The
most creative women in the sample were not found to be more assertive,
dominant, logical, or generally more masculine than the others.

Helson comments on a recurrent finding in creative women:

This pattern-~paternal influence in background findings
but inconspicuous masculine traits in the S--suggests a
solution to the paradox that women are said to be uncreative
(rigid) if they have masculine traits but also to be un-
creative (unproductive) if they lack them. The solution
adopted by creative women of this study was not to take an
overall masculine identification or to develop masculine
traits which could not coexist with other parts of a fem-
inine personality (p. 21).

Thus, Helson's creative women were not found to be more masculine than
the less creative women, but to have more identification with their

fathers than the comparison subjects.

Empirical Studies of Adaptive
Regression and Identification
Utilizing the Rorschach Test

'The hypothesis of a relationship between cross-sex identifi-
cation and the capacity for adaptive regression receives indirect
support from the pioneering work of Roe (1946, 1956, 1960). Roe
studied twenty eminent male painters by means of interview, the
Rorschach, and the TAT. Her Rorschach findings for the group included
overproduction of responses with sexual content, and a predominance of
M over sum C which Rorschach considered an indication of a tendency
toward regression, and Piotrowski relates to creativity and original-

ity.

Summarizing her findings, Roe (1956) states that "Both tests
indicated a nonaggressive and rather immature type of social and
sexual adaptation. There were many indications of insufficient freeing
of emotional ties with parents, particularly mothers, and considerable
confusion over their own sexual and personal roles." Roe points to the
frequent finding of unresolved oedipal conflicts and a strong tendency
toward femininity in these painters. She states that she does not
doubt that these characteristics are intimately related to the artists'
choice of vocation. Reviewing Roe's study, Burchard (1952) comments
that "there were many indications of the bisexually-toned passivity
which psychoanalytic theory would lead us to expect."

Sex differences in amount of drive expression were found by
Pine and Holt (1960), who investigated the relationship of individual
modes of expression and control of primary process on the Rorschach
Test to the quality of productions in tests of imagination (TAT,
Science Test, Humor Test, Rorschach, Animal Drawing, and Guilford's
Brick Uses and Consequences Test). The tests were administered to 13
male and 14 female undergraduate students who were participating in an
on-going research program. The subjects had been selected on the
basis of approximate indications on the MMPI of emotional stability,
and for intelligence (upper 60% of entering freshmen) on the Ohio
State Psychological Examination. Subjects had no particular excellence
in creative arts or science.

For each subject, Pine and Holt correlated both individual
ratings and an overall creativity score with scores for the amount of

primary process material, the effectiveness of control over primary

TO

process, and adaptive versus maladaptive regression-~a11 derived from
the Rorschach scoring. Rank correlations between expression and con-
trol of primary process indicated that these two variables were sta-
tistically independent. The adaptive regression score was found to
correlate far more strongly with its control score component than with
its expression score component. The overall creativity score correlated
significantly with the control scores for both sexes, and with the
adaptive regression score for males. Pine and Holt point out that

other data (Holt and Havel, 1960) on the female sample suggested that
the adaptive regression score was not particularly meaningful for this
female sample. Correlations of the overall creativity scores with
scores for amount of expression of primary process, using Holt's scoring
of the Rorschach, were positive but negligible.

Results on the individual tests paralleled those with the overall
creativity scores for males. For each test, control score correlated
more highly with score for the rated quality of production than with the
score for amount of primary process expression. The correlations with
control tended to be consistently positive for both groups, although
correlations for the females were smaller. The females had larger
correlations with the amount of primary process expression.

Myden (1960) attempted to differentiate personality character-
istics between creative and noncreative individuals using a battery of
projective tests including the Rorschach. He reports quantitative
results for the Rorschach only. Although Myden's results must be
evaluated in the light of the methodological criticisms pointed out

by Holt (1970), one of his findings for the creative group was a

11

significantly higher number of sexually ambivalent responses to the
Rorschach. Myden uses the concept of sexual ambivalence ". . . to
indicate failure to achieve a positive identification with the appro-
priate sex role." He reports that he did not find passivity which,
he says, Freud associated with sexual ambivalence, to be significantly
higher in the creative group.

The study to be presented is similar in purpose to a study by
Levy (1961). Levy suggested a reformulation of the psychoanalytic con-
cept of regression in the service of the ego as a complex ability
existing as a disposition in individuals, and consisting of three in-
dependent component abilities: regression, oscillation, and progres-
sion. He hypothesized a relationship between the regressive and pro-
gressive components of regression in the service of the ego and sexual
identification, expecting that ". . . regression would be associated
with predominantly feminine identification, while progression would be
associated with predominantly masculine identifiCation" for both-sexes.

The subjects were 34 female and 47 male undergraduates. Each
subject was assessed for the variables of ability to regress, using a
multiple choice group Rorschach; to oscillate, using the Stroop color-
word interference task; and to progress, using a sentence writing task;
predominant sexual identification, using the adjective check list; and
cognitive control style, using the color-word score.

Levy's expectations regarding the tasks used were generally con-
firmed. He found that each component measure seemed to be highly reli-
able, measured a fairly unitary function, and with the exception of one

subgroup of the sample, measured an independent variable. However,

12

there were important but unresolved questions about the validity of the
measures tapping only the function defined. The hypotheses concerning
the relationship of predominant sexual identification to both the
regression and progression components were not confirmed. Furthermore,
the variables of biological sex and cognitive control did not appear to
have any relationship to the regression component. Some tentative
findings did emerge suggesting relationships between progression, sexual

identification, sex, and cognitive control.

HYPOTHESES

The evidence of empirical studies just reviewed supports the
psychoanalytic hypothesis of a relationship between creativity and
regression in the service of the ego, and offers some support for the
related hypothesis of salient Cross—sex identification in creative
personalities. Taken together, these findings in turn suggest a pos-
sible relationship between the capacity for regression in the service
of the ego and cross-sex identification. Therefore, it is proposed

that:

Hypothesis IA. For males, the capacity_for regression in the
service of the ego is positively related to the degree of maternal

identification.

Hypothesis 18. For females, the capacity for regression in
the service of the ego is_positively related to the degree of paternal

identification.

 

A second hypothesis is suggested from the finding that highly
creative individuals have been found to score high on self-acceptance
(MacKinnon, 1962). It is hypothesized, therefore, that there is a
positive relationship between the capacity for regression in the ser-

vice of the ego and self-acceptance.

13

14

There is no inherent contradiction between the proposals that
the capacity for regression in the service of the ego is related to
cross-sex identification, and to self-acceptance. Cross—sex identifi-
cation is generally regarded as being more predisposing to anxiety,
defensiveness, maladjustment, perversion, and low self-acceptance than
identification with one's own sex. However, Chang and Block (1960)
found no difference in self-acceptance between same-sex and cross-sex
identified males, and Lazowick (1955) and Dyal (Osgood, gt_gl,, 1957)
did not find that neurotics were more closely identified with the
cross-sex parent than with the same-sex parent. Dyal did find greater
identification in normal males with the same-sex parent than in highly
anxious males, and Lazowick found that normals as a group tended to
form stronger identification with the same-sex parent than did neurotics.

Low self-acceptance would be expected to detract from the
capacity for regression in the service of the ego. Individuals with
cross-sex identification and low self-acceptance could be conflicted
at some level about the ambiguity of their identification, and the
greater their conflict and defensiveness, the less might be the capac-
ity for regression in the service of the ego--since this capacity
assumes the availability of autonomous or more conflict-free ego func-
tioning. Therefore, it is proposed that:

Hypothesis II. Self-acceptance isgpositivelyirelated to the

capacity for regression in the service of the ego.

The last hypothesis is based on developmental and clinical obser-
vations that biological sex alone is no guarantee of sex-appropriate

identification. Moreover, it is expected from Hypotheses I and II

15

that biological sex will show no relation to the capacity for regression
in the service of the ego, but rather, that differences between subjects
will be found to relate to the variables of identification and self-

acceptance. Therefore, the following hypothesis is proposed:

Hypothesis III. There is no relationship between the capacity

for regression in the service of thepego and biological sex.

METHOD

The Sample

 

The sample was drawn from male and female undergraduates
enrolled in introductory courses in general psychology, for which
students receive course credit for participation in departmental
research. The total sample consisted of 36 subjects equally divided

between males and females.

The Instruments and Scoring

The measure of the capacity for regression in the service of
the ego was the total Adaptive Regression Score (ARS) from Holt's
{(1970) Manual for the Scoring_of Primary Process Manifestations in
Rorschach Responses, Tenth Draft. The scoring for adaptive regression
recommended in this latest edition of the scoring manual was used by
Borofsky (1970), and replaced an earlier scoring procedure utilized by
Cohen (1960) and Zukowsky (1961).

To score for adaptive regression using the Holt scoring system,
Rorschach responses are first scored for form-level, and then scored
for primary process according to content and fgrmgl_categories. A
listing with some description of these scores may be found in Appendix A.

The formula for the total Adaptive Regression Score (ARS)
recommended by Holt (1970) is:

16

17

ARS = Sum (00 X DE)
total # offprimary
process responses

where, for responses containing scorable primary process, 00 (Defense
Demand) is the score for the shock value of the response as an inter-
personal communication. Defense Effectiveness (DE) is the score for
successful or unsuccessful control and defense in the response. Defense
Demand (00) is scored on a scale of 1-6. Defense Effectiveness (DE) is
scored on a 6-point scale, ranging from +2 to —3, with positive ratings
indicating good control or defense and negative ratings indicating less
to poor control and defense in the response. According to Holt, the
theoretical limits of the sum Adaptive Regression Score (ARS) are +12
to -18, with positive net sums indicating adaptive regression-

and negative values indicating maladaptive regression.

The split-half reliability for ARS reported by Zukowsky was
.648, which was significant beyond the .01 level. Inter-judge reliabil-
ities for the scoring of ARS were reported by Holt to generally range
between .85 and .95.

The measure of identification was derived-from Osgood's
semantic differential technique (Osgood, Suci, and Tannenbaum, 1957).
The semantic differential technique has been used to assess identifica-
tion by Lazowick (1955); Block (1958); Bieri, Lobeck, and Galinsky..
(1959); Endler (1961); Shell, O'Mally, and Johnsgard (1964).

According to Lazowick (1955):

Identification has been defined as the relation between
the meaning systems (mediating processes) of an S and his
model, and inferred identification as the relation between
meaning systems of an S and his model as perceived by the S.

18

If the semantic differential is a valid technique for
tapping these meaning systems, then identification should
be measurable in this way. The data seem to indicate this
is so.

The semantic differential employed consisted of 20 scales, each

scale defined by bipolar adjectives on a seven-interval rating. The
twenty scales were represetative of the three major connotative factors
of meaning isolated in factor analytic studies by Osgood, gt_gl,:
I Evaluative; II Potency; and III Activity. Factor I (Evaluative) was
represented by Items 1, 2, 9, ll, 16, 19, and 20. Factor 11 (Potency),
by Items 3, 4, 7, 13, 17, and 18. Factor III (Activity) by Items 5, 6,
8, 10, 12, 14, and 15.

The four concepts to be rated, "Ideal Self," "Your Mother,"
"Your Father," and "Your Self as You Actually Are," were presented
separately using the same twenty rating scales. A copy of the semantic
differential may be found in Appendix B.

Maternal, Paternal, or Mixed identification was differentiated
by comparing the maternal and paternal identification scores derived
from the "distance" score measure between the rating profiles of
perceived-self and respective parent ratings.> Assigning numerical
values of from +1 to +7 for the seven intervals of the twenty item
rating scales yielded a theoretical range of 0 - 120 for the semantic
differential d-score measure of identification, with 0 indicating a per-

fect identicalness in identification, and progressively larger numbers

 

indicating lesser degrees of identification.
The measure of self-acceptance was also derived from the seman-

tic differential, using the degree of correspondence between the

19

perceived-self and ideal self to define self-acceptance (Chang and
Block, 1960). Again, the theoretical limits of the d-score measure of

self-acceptance would be 0 - 120, with low scores indicating_high self-

 

acceptance andpprogressively larger scores indicating lesser degrees of

self-acceptance.

Procedure

The semantic differential was administered to groups of approx-
imately 20 subjects in a classroom setting. The general instructions
(from Chang and Block, 1960) were read aloud by the experimenter at the
beginning of the test session, and subjects were encouraged to ask
questions then if they needed to for clarification. The subjects were
asked to identify their test booklets with their student number, age,
and sex.

Correctly completed semantic differentials were obtained in
this manner from 63 females and 54 males before the six experimental
groups were filled. Shown in Table 1 are the means and standard devi-
ations of the maternal and paternal identification scores for the
original subject pool of males and females.

The maternal and paternal identification scores for each subject
were compared and predominant identification was determined. Predomi-
nant identification, either maternal or paternal, was defined in terms
of a cutting score of a difference of six points or more between the
maternal and paternal identification scores. A difference of less than
six points, or equivalent identification scores, was considered a mixed

identification.

20

Table 1. Means and standard deviations of identification scores
for the original sample pool.

 

 

Paternal Maternal
Identification Identification
Score Score
Male (N = 54)
X 26 30
SD 9.97 13.93

Female (N = 63)

X 29 26
SD 10.20 9.85

 

From the original subject pool of 63 females and 54 males,
18 females and 18 males were selected to complete the three male and
three female experimental groups: I Maternal Identification; II Pa-
ternal Identification; and III Mixed Identification. Each of the six
groups contained six subjects, with a total N = 36. The criteria for
selection, once predominant identification had been differentiated,
was that the samller identification score (indicating the predominant
identification), or both identification scores be within one standard
deviation of the sample mean identification scores in an attempt to
make the subjects relatively comparable, and to control for extreme
cases of identification or disidentification in the sample. Selection
between two or more likely protocols was then made randomly.

Student numbers from the test booklet were then matched with
subjects' phone numbers by personnel in the Registrar's Office, and

subjects were re-contacted and asked to return for a second part of

the experiment.

21

In the second half of the testing, the experimenter met with
each of the subjects individually and began by informing them that
the second half of the experiment involved "working with inkblots,‘l
and would they mind if a verbatim tape recording was made. No sub-
jects refused to comply with this request, although a few wary indi-
viduals were only reassured after the experimenter told them he could
tell them about the experiment at the end of the session, and that
the recording tape would be reused or otherwise erased.

The student number was again used to identify the subject's
Rorschach protocol, and the Rorschach was then administered to the
subject following Holt's suggestions for administration, the most
important of which is a verbatim protocol of everything said by the
subject.

When all the subjects had been tested, the Rorschach protocols
were then blind-scored by the experimenter for ARS, and the adaptive
regression score, identification scores, and self-acceptance score

for each subject were then compared.

Statistical Analysis

The hypotheses proposed in this study were tested using
Pearson product-moment correlations, and analysis of variance with
t-tests, on the variables of adaptive regression scores, identifica-
tion scores, and self-acceptance scores. It is noted that the direc—
tion of the obtained correlations between adaptive regression scores

and semantic differential scores had to be reversed in order to

correct for the inverse magnitude of the semantic differential scoring.

RESULTS

Adaptive Regression
and Identification

 

 

Hypothesis IA proposed that for males, the capacity for
adaptive regression, or regression in the service of the ego, would
be positively related to the degree of maternal identification.

Hypothesis IB proposed that for females, the capacity for
regression in the service of the ego would be positively related to
the degree of paternal identification.

Table 2 presents the Pearson product-moment correlation co-
efficients between paternal and maternal identification scores and
adaptive regression score (ARS) for males, females, and the total

sample (N = 36).

Table 2. Correlations between identification scores and adaptive
regression score (ARS).

 

 

Adaptive Regression Score (ARS)

 

Male Female Total

(n = 18) (n = 18) (N = 36)
‘ Paternal Score -.23a -.28 -.3o
Maternal Score .51* a,b -.53*b -.O3

 

*p<.05
adifference between these r's p<.05.
bdifference between these r's p<.01.

22

23

' As shown in Table 2, for males the correlation between ARS and
maternal identification score was found to be positive and significant.
This finding was supportive of Hypothesis IA for males. The correla-
tion between ARS and paternal identification score was found to be
negative and non-significant, and was significantly different from the
positive correlation between ARS and maternal identification score for
males.

For females, results showed a significant negative correlation
between ARS and maternal identification score, and a negative but non-
significant correlation between ARS and paternal identification--this
finding was not supportive of Hypothesis 18 for females.

Comparing the correlations of male and female subjects showed
a significant difference between the positive correlation for males,
and negative correlation for females, of ARS and maternal identification
score. Regarding the total sample of males and females (N = 36), the
negative correlation between ARS and paternal identification score

approached significance (p<.10) for the combined sample.

Adaptive Regression
and Self-Acceptance

 

 

Hypothesis 11 proposed that self-acceptance would be positively
related to the capacity for regression in the service of the ego.
Table 3 presents correlations between the measure of self-acceptance
and ARS for the combined sample of males and females (N = 36), and for

males and females separately.

24

Table 3. Correlations between the measure of self-acceptance (S-A)
and adaptive regression score (ARS).

 

 

Adaptive Regression Score (ARS)

 

Male Female Total
(n = 18) (n = 18) (N = 36)
Self-Acceptance (S-A) .41 .30 .35*

 

*p<.05

As the results in Table 3 show, the correlation for all sub-
jects between the measure of self-acceptance and ARS was found to be
statistically significant. The correlation between self-acceptance
and ARS for the male sample approached significance (p<.lO). Findings,

therefore, appeared to confirm Hypothesis II for the total sample.

Adaptive Regression
andTBiBTOgical sax

Hypothesis III proposed that no significant relationship would
be found between the capacity for regression in the service of the ego
and biological sex. Rather, it was expected from Hypothesis I and II
that the capacity for regression in the service of the ego would be
related to the variables of identification and self-acceptance. The
findings relevant to Hypothesis III are presented in Table 4.

As shown in the second row of Table 4, a significant differ-
ence was found between sexes on ARS, with females showing the greater
capacity for adaptive regression (see Appendix C). This finding was
therefore contradictory of Hypothesis III. However, additional com-

parisons between the male and female groups strongly suggested that

25

the significant difference obtained between the total male and female
samples was a function of comparisons involving one particular sub-
sample of the total male sample, the mixed-identified males. More-
over, comparison between the mixed-identified males and the total male
and female samples showed the mixed-identified males to score signi-
ficantly lower on ARS than the total sample, which would have affected

the overall comparison between males and females.

Table 4. Analysis of variance of adaptive regression scores (ARS)
for all subjects.

 

 

 

 

Source Sum of Sq. df Mean Sq. F
Between identif 3.53 2 1.77 6.79**
Between sex 2.30 1 2.30 4.43*
Identif x sex 1.03 2 .52 1.00
Error 15.60 30 .52
Total 22.46 35

Comparison _j;_
Maternal-Female vs Maternal-Male .28
Maternal-Female vs Paternal-Male 1.47
Maternal-Female vs Mixed-Male 3.4l**

Paternal-Female vs Maternal-Male .12

Paternal-Female vs Paternal-Male 1.42
Paternal-Female vs Mixed-Male 3.46**
Mixed-Female vs Maternal-Male -.5
Mixed-Female vs Paternal-Male .40
Mixed-Female vs Mixed-Male 1.66
Total vs Mixed-Male 2.98**

 

*p<.05 **p<.01

Therefore, while the overall difference found between males

and females on ARS was contradictory of Hypothesis III, additional

26

analyses call into question the validity of this finding and point to
the need for replication of this finding. The interaction between

identification and sex was nonsignificant.

Additional Results

Adaptive Regression and Identification. Hypothesis IA proposed

 

that for males, the capacity for regression in the service of the ego
would be positively related to the degree of maternal identification.

Using the identification scores derived from the semantic
differential to differentiate male subjects by prominent identification
into maternal, paternal, or mixed identified groups, it was expected
from Hypothesis IA that the maternal idenified males would show the
strongest positive relationship between maternal identification and
adaptive regression, and that in turn, the mixed-identified males
would show a greater relationship between maternal identification and
adaptive regression than the paternal-identified males.

Table 5 presents the correlations between identification
scores and adaptive regression score (ARS) for the three male groups

and the total male sample.

As shown in Table 5, the maternal-identified males showed the
only significant positive correlation between maternal identification
and adaptive regression, and a lesser significant positive correlation
between adaptive regression and paternal identification.

Comparisons between the three male groups showed the positive
correlations of the maternal-identified males to be significantly
different from the negative correlations shown by the mixed-identified

males.

27

Table 5. Correlations between adaptive regression score (ARS)
and identification scores for males.

 

 

Adaptive Regression Score (ARS)

 

 

Paternal Maternal Mixed Total
(n = 6) (n = 6) (n = 6) (N = 18)
Paternal Score .42 '81*a,b -.69a -.23b
Maternal Score .47 .93**C -.46c .51*
*p<.05 **p<.01
adifference between these r's p<.05.

bdifference between these r's p<.05.

cdifference between these r's p<.Ol.

Thus, the findings for maternal-identified males appeared to be
supportive of Hypothesis IA for males. The negative correlation between
ARS and maternal identification shown by the mixed-identified males was
contrary to expectations from Hypothesis IA. Results for paternal-
identified males were not significant. The means, SD, F, and t-tests
for the three male groups are presented in Appendix C. The maternal-
identified showed the highest mean ARS, with the paternal-identified
‘males following, and the mixed-identified males showing the lowest
mean ARS. Analysis showed the significant F-ratio obtained between
the groups to be between the maternal and mixed-identified males;
differences between these two groups and the group of paternal-
identified males were not significant.

Hypothesis 18 proposed that for females, the capacity for

regression in the service of the ego would be positively related to

28

the degree of paternal identification. It was expected from Hypothe-
sis 18 that paternal-identified females would show the strongest
positive relationship between paternal identification and adaptive
regression, and that in turn, the correlation between paternal iden-
tification and adaptive regression would be greater for mixed-
identified females than for maternal identified females.

Table 6 reports correlations between adaptive regression
score and identification scores for the three female groups and the

total female sample.

Table 6. Correlations between adaptive regression score (ARS)
and identification scores for females.

 

 

Adaptive Regression Score (ARS)

 

Maternal Paternal Mixed Total
(n = 6) (n = 6) (n = 6) (N = 18)

Maternal Score —.84* -.57' .10 A -.28
Paternal Score -.81* .23 -.23 -.53*

 

*p<.05

As shown in Table 6, significant correlations were found for
only the maternal-identified females, who showed negative correlations
between ARS and both paternal and maternal identification scores.

The negative correlations between ARS and paternal identification for
the maternal and mixed-identified females were contrary to expecta-
tions from Hypothesis 18. The difference between the negative correla-

tion for the maternal-identified females and the positive correlation

29

between ARS and paternal identification shown by the paternal-identified
females approached significance (p<.lO). The means, SD, and F for the
three female groups are presented in Appendix C. Differences between
the three female groups were not found in the direction predicted from
Hypothesis 18, nor were the differences significant.

Identification and Self-Aegeptance. The relationship between
identification and self-acceptance was not considered directly in the
hypotheses, but is relevant to the hypotheses put forth regarding the
relationships between adaptive regression and identification, and
adaptive regression and self-acceptance. Moreover, the relationship
between identification and self-acceptance is of considerable theoreti-
cal and clinical interest on its own merit.

Shown in Table 7 are the correlations between the measure of
self-acceptance and identification scores for male and female subjects

and total correlations for all subjects (N = 36).

Table 7. Correlations between the measure of self—acceptance (S-A)
and identification scores. .

 

 

Self-Acceptance (S-A)

 

Male Female Total

(n = 18) (n = 18) (N = 36)
Paternal Score .64**a -.07a .16
Maternal Score .13 .35 .26

 

**p<.01
adifference between these r's p<.05

30

As Table 7 shows, the only significant correlation found was
for males, between self-acceptance and paternal identification. For
females, self-acceptance showed a non-significant positive correla-
tion with maternal identification and a non-significant negative cor-
relation with paternal identification. The difference in the correla-
tions for males and females between self-acceptance and paternal
identification was significant.

Table 8 shows the correlations between the measure of self-

acceptance and identification scores for the three male groups.

Table 8. Correlations for male groups between the measure of self-
acceptance (S-A) and identification scores.

 

 

Self-Acceptance (S-A)

Paternal Maternal Mixed Total
(n=6) (n=6) (n=6) (N=l8)

 

 

Paternal Score .16 .91* .38 .64**
Maternal Score .62 .73a -.09a .13
*p<.05 **p<.01

adifference between these r's p<.05.

As shown in Table 8, only the correlation for maternal—
identified males between self-acceptance and paternal identification
was significant. The correlation for maternal-identified males between
self-acceptance and maternal identification approached significance
(p<.lO). Comparison of the correlations of self-acceptance with

maternal identification showed an almost significant difference

31

between maternal and paternal identified males (p<.lO). The correla-
tions between the measure of self-acceptance and maternal identifica-
tion showed a significant difference between maternal-identified
males, who showed a positive correlation, and mixed-identified males,
who showed a negative correlation.

Table 9 presents correlations between the measure of self-

acceptance and identification scores for the female groups.

Table 9. Correlations for female groups between the measure of
self-acceptance (S-A) and identification scores.

 

 

Self-Acceptance (S-A)

Maternal Paternal Mixed Total
(n=6) (n=6) (n=6)(N=18)

 

Maternal Score -.27 .21 .62 .35
Paternal Score -.38 .49 .51 -.O7

 

As the results in Table 9 show, none of the correlations for
the female groups between self-acceptance and respective identifica-
tions were significant. The maternal-identified females showed non-
significant negative correlations between self-acceptance and both

maternal and paternal identifications.

DISCUSSION

Adaptive Regression
and Identification

 

Hypothesis IA predicted that for males, the capacity for
regression in the service of the ego would be positively related
to the degree of maternal identification.

The overall findings for males were supportive of Hypothesis
IA. Adaptive regression showed a significant positive correlation
with maternal identification, and a negative correlation with pater-
nal identification. The difference between the direction and magni-
tude of these correlations was significant.

Some interesting secondary findings emerged between male
subjects when differentiated by predominant identification into
maternal, paternal, or mixed identification groups. Findings were
supportive of Hypothesis IA for the maternal-identified males, who
showed the highest level of adaptive regression, although the only
significant difference found between the male groups was between the
maternal and mixed-identified males. Contrary to expectations from
Hypothesis IA, the mixed-identified males showed the lowest level of
adaptive regression of the three male groups.

The maternal-identified males showed the only significant
correlations between adaptive regression and identification. The

significantly larger correlation between adaptive regression and

32

 

 

33

maternal identification for the maternal-identified males was support-
ive of Hypothesis IA. For the mixed-identified males, the correlations
were remarkable not because they were significant, but because they
were in a negative direction, which was contrary to expectations based
on Hypothesis IA.

Hypothesis IB predicted that for females, the capacity for
regression in-the service of the ego would be positively related to
the degree of paternal identification.

The overall findings for females were not supportive of Hypo-
thesis IB. Adaptive regression showed a non-significant negative cor-
relation between adaptive regression and paternal identification, and
a significant negative correlation between adaptive regression and
maternal identification.

Comparison of the three female groups showed no significant
differences between groups on the level of adaptive regression. 0f
the three female groups, only the maternal-identified females showed
significant correlations, both negative, between adaptive regression
and paternal and maternal identifications.

The total correlation for all subjects between adaptive re-
gression and paternal identification was negative and approached sig-
nificance (p<.lO), suggesting that for both males and females the
capacity for regression in the service of the ego is negatively re-
lated to paternal identification. This finding is consistent with
expectations from Hypothesis IA for males, and contrary to Hypothesis
18 for females.

With regard to the lack of significant findings for females,

34

Pine and Holt (1960) report similar results and conclude that the
adaptive regression score was not meaningful for their female sample.
In support of the hypothesis put forth in the present study, it is
suggested that perhaps the adaptive regression score j§_more salient
for male than for female subjects, or is not meaningful for female
subjects at all.
Pine and Holt report in their study that:
The combined (adaptive regression) Rorschach score cor-
relates far more strongly with the control score component
(rho is .90 for males and .79 for females) than with the

expression score component (rho is .11 for males and .13
for females).

And yet these experimenters observe that:

The major difference in results between males and fe-
males is in the correlation with the amount score. For
both groups, the correlations with control tend to be con-
sistently positive although the correlations for the females
are smaller. The correlations nevertheless seem to fall
into a pattern for the females. The tests which at least
in appearance permit more open-ended and expressive verbal
responses, and in particular permit more drive expression
in the response, are just those where the correlation for
the female S5 is higher with the Rorschach score for amount
of primary expression than with the control score.

Thus, in terms of the limited findings for females reported
in the present study, it may be that differences between the female
subjects were not fully reflected in their adaptive regression scores
as a function of the low correlation reported by Pine and Holt for
females between the adaptive regression score and its expression
score component. Pine and Holt found that their female subjects
(female undergraduates) showed the greatest differences in expression,

rather than control, of primary process.

35

Adaptive Regression
and Self-Acceptance

 

 

Hypothesis II predicted that self-acceptance would be pos-
itively related to the capacity for regression in the service of the
ego.

The significant positive correlation obtained for the total
sample of males and females appeared to confirm HypotheSis II. More-
over, the positive correlation for males approached significance

(despite a negative correlation fOr the mixed-identified males).

Adaptive Regression
and Biological Sex

 

 

Hypothesis III predicted that no significant relationship
would be found between the capacity for regression in the service
of the ego and biological sex.

Comparison between male and female subjects on adaptive re- _
gression showed a significant difference between the sexes, with
female subjects showing the greater capacity for adaptive regression.
Additional comparisons between the male and female groups showed
significant differences in comparisons involving only one sub-group of
the male sample, the mixed-identified males. A further comparison
showed the mixed-identified males to score significantly lower than
the total sample on adaptive regression, which would have affected
the overall comparison between the male and female subjects on adaptive
regression. Therefore, although Hypothesis III was contradiced by the
overall findings, additional findings make the relationship between

the capacity for adaptive regression and biological sex appear more

36

equivocal, and suggest further study.

Identification and Self-Acceptance

The relationship between identification and self-acceptance
was not directly predicted in the hypotheses, but was explored
because the data were available and findings were thought to be
relevant to the hypotheses being tested.

For males, a significant positive relationship was found be-
tween self—acceptance and paternal identification. Comparing the male
groups, no significant differences were found between groups on the
measure of self-acceptance. Maternal-identified males showed the only
significant correlation between self-acceptance and paternal identi-
fication, and a correlation which approached significance between self-
acceptance and maternal identification. The former finding was of
interest because the correlation between self-acceptance and paternal
identification for maternal-identified males was almost significantly
greater than the same correlation for paternal-identified males.
Another finding of interest was the negative correlation between
self-acceptance and maternal identification shown by the.mixed-
identified males. This inverse relationship was significantly differ-
ent from the positive correlation between self-acceptance and mater-
nal identification shown by the maternal-identified males.

For females, no significant correlation was found between
self-acceptance and either maternal or paternal identification.

There were no significant differences between female groups on the

measure of self-acceptance. Comparing the female groups, the

37

paternal and mixed-identified females showed non-significant positive
correlations between self-acceptance and both maternal and paternal
identifications, while the maternal-identified females showed non-
significant negative correlations.

The study being reported was intended to be exploratory in
purpose and design; one limitation of the findings, or lack of find-
ings, therefore, is the small sample size. The significance levels
established would minimize the probability of Type I error (rejecting
the null hypothesis when it is true); however, the small sample size
and the small significance levels would increase the probability of
Type II error (accepting the null hypothesis when it is false). A
more extensive study is suggested and supported by the tentative find-

ings reported.

Summary and Integration
of’Finding§_

 

 

Comparing-males and females, findings suggested that the
capacity for regression in the service of the ego was greater for
females than for males.

For males, the capacity for regression in the service of the
ego appeared to relate significantly more to maternal than to pater-
nal identification. Maternal-identified males showed the only signi-
ficant positive relationship between adaptive regression and maternal
identification, and to a lesser degree, between adaptive regression
and paternal identification. Consistent with this, the maternal-

identified males showed evidence of having the greatest capacity

 

 

38

for regression in the service of the ego of the total male sample
(which may have been demonstrated more cogently with a larger male
sample).

For females, the capacity for regression in the service of
the ego showed a non-significant negative relationship to paternal
identification, and a significant negative relationship to maternal
identification. There were no significant differences between female
groups on the capacity for regression in the service of the ego. The
maternal-identified females showed the only significant relationships
between adaptive regression and respective identifications; both neg-
ative.

For both males and females, the capacity for regression in
the service of the ego showed a negative trend with paternal identi-
fication.

Self-acceptance was found to relate positively to the capacity
for regression in the service of the ego for the total sample of males
and females. However, results for the group of mixed-identified males
showed a negative relationship between self-acceptance and adaptive
regression. The positive relationship between self-acceptance and
adaptive regression appeared to be more salient for males than for
females, perhaps as a function of the possible construct limitations
of the adaptive regression score (ARS) for female subjects.

Regarding the relationship between identification and self-
acceptance: For males, self-acceptance showed a significant positive
relationship to paternal identification, which was almost significantly

greater than the relationship between self-acceptance and maternal

39

identification. Maternal-identified males showed a significant posi-
tive relationship between self-acceptance and paternal identification,
and an almost significant secondary relationship between self-
acceptance and maternal identification. The paternal-identified males
showed less relationship between self-acceptance and paternal identi-
fication than did the maternal-identified males (p<.lO). The mixed-
identified males showed a negative relationship between self-acceptance
and maternal identification, which was significantly different from the
positive relationship between self-acceptance and maternal identifica-
tion shown by the maternal-identified males.

For females, the direction of the non-significant correlations
suggested a positive relationship between self-acceptance and maternal
identification. However, for the maternal-identified females, results
showed non-significant but negative relationships between self-
acceptance and both paternal and maternal identifications.

Regarding the mixed-identified males, results obtained for
this group were clearly contrary to expectations from Hypotheses IA
and II. Findings can be interpreted as suggesting that the mixed-
identified males were more conflicted about their maternal identifi-
cation to the extent that their degree of self-acceptance and capacity
for regression in the service of the ego were impaired. The mixed-
identified males showed negative correlations between self-acceptance,
maternal identification, and adaptive regression, and scored signifi-
cantly lower on adaptive regression than the total male and female
sample, and significantly lower than three of the five other male and

female groups. The mixed-identified males also showed inverse

4O

relationships between adaptive regression and both maternal and
paternal identifications, which were significantly different from
the positive findings for paternal and maternal-identified males.

The results suggest that the maternal—identified males may
have in part shown the greatest capacity for regression in the ser-
vice of the ego because, like the mixed-identified males, they
showed salient maternal identification. However, unlike the mixed-
identified males, the maternal-identified males appeared to be more
accepting of both their maternal and paternal components, which did
not seem to be true for the mixed-identified males, who seemed defen-

sive and rejecting of their maternal identification.

APPENDICES

APPENDIX A

Listing of All Scores Contained in the Holt Manual
for Primary Process Scores

Primary Process Categories

CONTENT CATEGORIES

 

L. Libidinal 00
L10 2,3 Oral, Level 1
L10-Ag 2-4 Oral-Agressive, Level 1
LlA 3,4 Anal, Level 1
L18 4 Sexual, Level 1
LlE-V 3 Exhibitionistic-voyeuristic, Level 1
L1H 4,5 Homosexual, Level 1
LlM 3,4 Miscellaneous libidinal, Level 1
L20 1-3 Oral, Level 2
L20-Ag 2,3 Oral-Agressive, Level 2
L2A 1-3 Anal, Level 2
L2S 1-3 Sexual, Level 2
L2 E-V 1-3 Exhibitionistic-voyeuristic, Level 2
L2H 2,3 Homosexual, Level 2
L2M 1-3 Miscellaneous libidinal, Level 2

Ag. Aggressive

Ang 3-5 Sadistic Aggression, Level 1

AglOb 3-5 Masochistic ("Object") Aggression,
Level 1

AglR 3-5 Results of Aggression, Level 1

Ag2S 2,3 "Subject" Aggression, Level 2

A920b 1-3 "Object" Aggression, Level 2

Ag2R 1-3 Results of Aggression, Level 2

FORMAL CATEGORIES

C. Condensation

 

C-ctm 1 4,5 Fusion of Two Separate Percepts
(Contamination)

C-ctgn l 4 Contagion, Level 1

C-int 1 3 Interpenetration (Partial Fusion),

Level 1

41

APPENDIX A (continued)

C. Condensation (continued)

 

C-co 1 3,4
C-co 2 1,2
C-a-c 2 2,3
C-arb 1 1-3
C-arb 2 1
0. Displacement
D-chain l 4
D-dist 2 3
D-Clang l 4
D'fig 2 2,3
D-time 2 2

Sym. Explicit Symbolism

Sym-Cl 3
Sym-C2 l
Sym-Sl 3
Sym-Il 3
Sym-12 l

Ctr. Contradiction

Ctr Al 4
Ctr L1 4
Ctr R1 3,4
Ctr R2 2,3

42

Composition, Level 1

Composition, Level 2

Arbitrary Combination of Separate
Percepts

Arbitrary Combination of Color and
Form, Level 1

Rationalized Arbitrary Color

Chain Association, Level 1

Distant Association, Level 2

Clang Association, Level 1

Figures of Speech, Level 2

Displacement in Time (Anachronism),
Level 2

Color or Shading Symbolism, Idio-
syncratic, Level 1

Color and Shading Symbolism, Con-
ventional, Level 2

Spatial Symbolism

Image Symbolism, Idiosyncratic,
Level 1

Image Symbolism, Conventional,
Level 2

Affective Contradiction

Logical Contradiction ,.

Contradiction of Reality (Delib-
erate Molding)

Contradiction of Reality
(Inappropriateness)

Miscellaneous Distortions of Thought and Perception

Au Lg 1 4,5
M L 1 4
Intr l 3
Impr 2 1,2
Do 2 l

Autistic Logic, Level 1

Memory Loosening, Level 1

Intrusion of Irrelevancy, Level 1

Impressionistic Response, Level 2

Fragmentation (00 Response),
Level 2

43

APPENDIX A (continued)

Miscellaneous Distortions of Thopght and Perception (continued)

Un Rel l 3,4 Unrealistic Relationships, Level 1

Trans 1 3 Fluid Transformation of Percept,
Level 1

S-R l 3 Self-Reference (Magical), Level 1

Au El 1 4 Autistic Elaboration, Level 1

Au El 2 3 Autistic Elaboration, Level 2

F-msc l 3,4 Miscellaneous Formal Deviations,
Level 1

F-msc 2 2,3 Miscellaneous Formal Deviations,
Level 2

V. Verbalization Scores

V11 5 Verbal Incoherence, Level 1

VCl 4 Verbal Condensation, Level 1
V01 4 Queer Verbalization, Level 1
VP2 3 Peculiar Verbalization, Level 2
V82 2 Verbal Slips, Level 2

CONTROL AND DEFENSE CATEGORIES

R. Remoteness

 

R-min Minimal Remoteness

R-eth Remoteness in Person-Ethnic

R-an Remoteness in Person-Animal (ego-Syntonic)
R-(an) Remoteness in Person-Animal (ego-alien)
R-pl Remoteness in Person-Plant

R-ia Remoteness in Person-Inanimate

R-dep+ Remoteness-Depiction

R—geo+ Remoteness-Geographic

R-time+, time Remoteness in Time

R-fic 5+, 5, s- Reference to Specific Fictional Character, etc.
R-fic n+, n, n- Remoteness, Fictional, Nonspecific Character, etc.
R-rel+, rel, rel-Use of Religious Character or Context
R-fan+, fan, fan-Characters or Context from Dream or Explicit

Fantasy
F-fig+, fig, fig—Figurative Remoteness (Figures of Speech)
R-cond Remoteness-Conditional

Cx. Context

 

Cx C+, C, C- Cultural Context

Cx E+, E, E- Esthetic Context

Cx I+, I, I- Intellectual Context

Cx H+, H, H- Humorous Context (and Pathos)

44

APPENDIX A (continued)
Refl. Reflection

 

Ref1+, Refl Reflection on Response (Introspective or
Self-Critical

Postponing Strategies

Del Delay
Blkg- Blocking

Miscellaneous (Mostly Pathological) Defenses

 

Eu Euphemism

Mod+ Adaptive Modification of Response toward
Secpro

Ratn+, Ratn Rationalization

Neg+, Neg Negation

Minz+, Minz Minimization

Chpb- Counterphobic Defense

Self-D- Self-Depreciation

Rep- Repudiation or Disavowal of a Response

Va— Vagueness of Percept or Communication

Pri- Projection (of Responsibility; Paranoid Rage)

Obs- Obsessional Defenses

Iso- Isolation

Eva- Evasiveness and Avoidance

Imp- Impotence

S. Seguence

 

S C 1-0 Sequence--Change from Level 1 to Unscorable
S C 2-0 Sequence-~Change from Level 2 to Unscorable
S C 1-2 Sequence--Change from Level 1 to Level 2

O. Overtness

 

O-beh Overtness--Behavioral
.0-vbl 0vertness--Verbal

O-exp Overtness--Experiential
0-pot Overtness--Potential

X. No control (record to distinguish lack of control from failure
to score)

OVERALL RATINGS OF TOTAL RESPONSE

 

FL Form Level Scoring
c, i Combinations and Integrations
Cr Creativeness and Originality
DD Demand for Defense

DE Effectiveness of Defense

45

APPENDIX A (continued)

Form Level Scoring Categories

The first column on the scoring sheet after the identification of
response by card and number is for the form level symbol, which must
be given to each response. We follow the system developed by Mayman
out of proposals by Rapaport (1968). It is fully presented in an
appended manual by Mayman and Holt, but the following set of brief
definitions may be helpful.

The column at the left presents numerical equivalents, which may be
used to derive an overall quantitative summary score for Form Level.
All scores with numbers less than 5 may be regarded as R-secpro.

Rating EL

7 F+ Sharp, convincing forms, easily seen by E.

6 F0 Popular and near-popular forms; fixed list in manual.

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

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

1 F- Arbitrary forms, very little or no resemblance.

5 Fv+ Vague forms that fit the blot quite well (see Mayman,
p. 24n.), and non-definitive form combined with appro-
priate use of color or shading; good CF, ChF, C'F, or
(C)F responses ('fire'; 'flowers'; 'dark clouds';
'splashing water'; 'ink stain, running down a water
color.')

4 Fv Vague forms with no other determinant or forced use
thereof (as in C/F or CFarb): 'clouds'; 'islands';
'cave mouth'; 'piece of dough.I

3 Fa Amorphous responses, in which form plays ng_role (and

could not, by the nature of the concept). Usually
pure C, C' or Ch. "Sky'; 'water'; 'night'; 'spring'
(and other abstract concepts); 'urine' (but 'wine
stain' seen as having some sort of shape, Fv+; like-
wise, 'blood' mgy_be Fa but is usually Fv+).

46

APPENDIX A (continued)

Rating EL_
2 F5 Spoiled form responses, to be used when the subject

gives what is basically a familiar and good response
(which would 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.

DE. Effectiveness of Defense

Finally, each response that contains any scorable content or
formal element is to be rated on the effectiveness of controlling and
defensive measures in reducing or preventing anxiety and making a
successful, adaptive response to the examiner's demand to interpret the
blots. Logically, DD and DE are independent, though in practice it is
sometimes difficult to consider them as separately as would be desirable.

DE is rated on a 6-point scale, ranging from +2 to -3, by half-
point steps. Postive values indicate good control and adaptive regression,
negative ratings indicate more pathological defensive efforts and mal-
adaptive regression. Distinguish carefully the successfulness of control
or defense, and the response's creativeness, which is separately rated.

A completely successful response mey_be at the bottom of the creative-
ness scale, or at the top.

 

Note that two types of ratings are given, distinguished by the
letter e_following the undefended type. An undefended response is de-
fined as one that lacks any control and defense category except Remin,
any of the Overtness and/or Sequence categories. It is assumed that if
someone is quite mature and unthreatened by the primary process, he may
feel no need to defend a response containing a good deal of it; therefore,
the lack of any scorable defenses is not considered a detriment if the
response is in other respects suCcessful.

 

 

Rating

+2 Completely successful control and defense, in a successful
response. -

+2a Highly successful response, undefended.
+1.5 Successful control and defense.

+1.5a Successful response, undefended.

47

Rating (continued)

+1

+1a

+.5a

Successful control and defense.

Successful response, undefended.

Fairly successful control and defense.

Fairly successful response, uncontrolled.

Only moderately successful control and defense.
Only moderately successful response, undefended.
Slightly unsuccessful control and defense.
Slightly unsuccessful response, undefended.
Relatively unsuccessful control and defense.
Relatively unsuccessful control and defense, undefended.
Mostly unsuccessful control and defense.
Unsuccessful attempts at control and defense.
Unsuccessful response, undefended.

Very unsuccessful control and defense.

Very unsuccessful response, undefended.

Disorganized responses with only pathological attempts at
defense.

Pathological disorganized responses, undefended.

APPENDIX B

The Semantic Differential

 

Department of Psychology
Michigan State University

Number: Age: Sex:
INSTRUCTIONS

First, identify the pages of this booklet with your student
number. No names are to be recorded on these sheets since numbers
serve our purpose of identification and besides, many individuals find
it easier to respond when more anonymity is provided.

The procedure with which you are being requested to cooperate
represents a preliminary attempt to gather some information on the ques—
tion of just how people perceive or experience themselves, their personal
ideals, and their parents. Necessarily, we have to resort to roundabout,
verbal means of collecting this information and it is hoped that you
will go along with the instructions as sincerely and conscientiously as
you can. Please do not be flippant or random in your performance since
you would be defeating a serious purpose.

A The instructions are simple. At the top of the next page, in
capital letters, is the phrase "Your Ideal Self." Below it are a number
of seven-interval scales or dimensions. The dimensions are defined at
each end. For example, "high"-"low," "active"-”passive," "angular"-
"rounded," and so on. You are to describe your ideal self, the person
that ideally you would like to be, in terms of these scales. For each
scale listed place an X in the one of the seven intervals which you feel
most closely approximates your association with regard to your ideal self.
Thus, you may feel your ideal self is relatively "low" rather than "high,"
more "passive" than "active," and approximately in the middle on the
"angular"-"rounded" dimension. Place your X's in the interval appro-
priate to express the intensity of your feeling or association.

Please make your descriptions rapidly, using your first, intuitive
reaction as a basis for marking each scale. Certain aspects of the task
are necessarily ambiguous but proceed nevertheless to the best of your
ability. There are no right or wrong responses.

Following the description of your ideal self, turn the page and
in the same manner, describe "Your Mother," "Your Father," and "Your
Self as You Actually Are."

In closing, we would like to emphasize again that the worth of
this research is dependent upon how well and conscientiously the people
participating in it carry through their tasks. Thank you for your
cooperation.

48

high
green
weak
rough
active
empty
small
cold
clear
young
good
peaceful
sick
angular
tense
sad

soft

wet
beautiful

fresh

49

YOUR IDEAL SELF

low

red
strong
smooth
passive
full
large
hot
hazy
old

bad
ferocious
healthy
rounded
relaxed
happy
loud
dry
ugly

stale

high
green
weak
rough
active
empty
small
cold
clear
young
good
peaceful

sick

angular’

tense

sad

soft

wet
beautiful

fresh

50

YOUR MOTHER

low
red
strong

smooth

epassive

full
large
hot
hazy
old

bad
ferocious
healthy
rounded
relaxed
happy
loud
dry
ugly

stale

high
green
weak
rough
active
empty
small
cold
clear
young
good
peaceful
sick
angular
tense
sad

soft

wet
beautiful

fresh

51

YOUR FATHER

low
red
strong

smooth

epassive

full
large
hot
hazy
old

bad
ferocious
healthy
rounded
relaxed
happy
loud
dry
ugly

stale

high
green
weak
rough
active

empty

52

YOUR SELF AS YOU ACTUALLY ARE

low
red
strong

smooth

passive

full

 

small

large

 

cold

hot

 

clear
young
good
peaceful
sick
angular

tense

hazy

old

bad
ferocious
healthy
rounded

relaxed

 

sad

soft

wet
beautiful

fresh

happy
loud

dry
ugly

stale

 

APPENDIX C

Comparison of Groups on Adaptive Regression Score

Table 1C. Comparison of the male groups on adaptive regression

score (ARS).

 

 

 

 

Groups
Maternal Paternal Mixed Total
Mean 2.61 2.13 1.43 2.06
SD .48 .79 .76 .82
F = 4.35*
Comparison t
Maternal Vs. Total 1.53
Maternal vs. Paternal 1.30
Maternal vs. Mixed 8.43**
Paternal vs. Mixed 1.56
Mixed vs. Total 1.66

 

*Significant at the .05 level.
**Significant at the .001 level.

Table 2C. Comparison of the female groups on adaptive regression

score (ARS).

 

 

 

Groups
Maternal Paternal Mixed Total
Mean 2.69 2.64 2.36 2.56
SD .51 .37 1.13 .72

F = .34 N.S.

 

53

REFERENCES

REFERENCES

Bieri, J.; Lobeck, R.; and Galinsky, M. A comparison of direct,
indirect, and fantasy measures of identification. Journal
of Abnormal and Social Psychology, 1959, 58, 253-258.

Block, J. An unprofitable application of the semantic differential.
Journal of Consulting Psychology, 1958, 22, 235.

Borofsky, G. L. Regression and ego functioning in creative normals
and psychotics. Unpublished doctoral dissertation, Michigan
State University, 1971.

Burchard, E. M. L. Projective techniques in the analysis of creativ-
ity. Journal of Projective Techniques, 1952, 16, 412-427.

Chang, J., and Block, J. A study of identification in male homo-
sexuals. Journal of Consulting Psychology, 1960, 24, 307-310.

Cohen, I. H. An investigation of the relationship between adaptive
regression, dogmatism, and creativity, using the Rorschach
and dogmatism scale. Unpublished doctoral dissertation,
Michigan State University, 1960.

Dudek, S. 2. Regression and creativity: A comparison of the Rorschach
records of successful vs. unsuccessful painters and writers.
Journal of Projective Techniqpes and Personality‘Assessment,
1968, 32, 453-461.

Endler, N. S. Changes in meaning during psychotherapy as measured by
the semantic differential. Journal of Counseling Psychology,
196], 8, 105-11].

 

Freud, S. (1900) The Interpretation of Dreams (trans. James Strachey).
New YorkngVOn Books, 1965.

Freud, S. (1911) Formulations regarding the two principles of mental
functioning. In Collected Papers, Vol IV, London: Hogarth
Press, 1946.

Freud, S. (1915) The unconscious. In Collected Papers, Vol IV, London:
Hogarth Press, 1946.

54

55

MacKinnon, D. W. The nature and nurture of creative talent. American
Psycholpgist, 1965, 20, 273-281.

Maes, J. Identification of male college students with their fathers
and some related indices of affect expression and psychosexual
adjustment. Unpublished doctoral dissertation, Michigan State
University, 1962.

Myden, W. An interpretation and evaluation of certain personality
characteristics involved in creative production. In M. H.
Sherman (ed.), A Rorschach Reader. New York: International
Universities Press, 1960.

Neumann, E. Art and the Creative Unconscious. New York: Pantheon
Booksj—1959.

Osgood, C. E.; Suci, G. J.; and Tannenbaum, P. H. The Measurement of
Meaning. Urbana: University of Illinois Press, 1957.

 

Pine, F., and Holt, R. R. Creativity and primary process: a study of
adaptive regression. Journal of Abnormal and Social Psyphology,
1960, 61, 370-379.

 

Rabin, A. I. A contribution to the ''meaning" of Rorschach's inkblots
via the semantic differential. Journal of Consulting ngch-
ology, 1959, 23, 368-372.

 

Roe, A. Painting and personality. Rorschach Research Exchange, 1946,
10, 86-100. _

Roe, A. The Psychology of Occupations. New York and London: John
Wiley and Sons, Inc., 1956.

Schafer, R. Regression in the service of the ego: the relevance of a
psychoanalytic concept for personality assessment. In G.
Lindzey (ed.), Assessment of Human Motives. New York: Rine-
hart, 1958.

Shell, S. A., Jr.; O'Mally, J. M.; and Johnsgard, K. W. The semantic
differential and inferred identification. Psychological
Reports, 1964, 14, 547-558.

 

Snider, J. G., and Osgood, C. E. (eds.). Semantic Differential
Technigue. Chicago: Aldine Publishing Co., 1969.

 

Wild, C. Creativity and adaptive regression. Journal of Personality
and Social Psyphology, 1968, 9, 266-271.

 

56
Freud, S. (1917) Introductory lectures on psychoanalysis. Standard
Edition, Vol. XVI, London: Hogarth Press, 1966.

Freud, S. (1923) The ego and the id. Standard Edition, Vol. XIX,
London: Hogarth Press, 1966.

 

Freud, S. Leonardo da Vinci: a study in_psychosexuality_(trans. A. A.
Brill). New York: Random House, Inc., 1947.

 

Freud, S. Doestoevsky and Parricide (trans. D. F. Tait). In W. Phil-
lips (ed.) Art and Peychoanalysis. Cleveland and New York:
The World Publishing Co., 1963.

 

Gamble, K. R., and Kellner, H. Creative functioning and cognitive
regression. Journal of Personality and Social Psychology,
1968, 9, 266-271.

 

Helson, R. Personality of women with imaginative and artistic inter-
ests: the role of masculinity, originality, and other charac-
teristics in their creativity. Journal of Personality, 1966,
34, 1'25.

 

Holt, R. R. Artistic creativity and Rorschach measures of adaptive
regression. In M. Meyer and B. Klopfer (eds.) Develppments in
the Rorschach Technique. Vol. III. New York: Harcourt, Brace
and World, 1970.

 

 

Holt, R. R. Manual for the scoring of primary process manifestations
in Rorschach responses. Draft 10, New York: Research Center
for Mental Health, Unpublished manuscript, 1970 (mimeographed).

Holt, R. R. and Havel, J. A method for assessing primary and secondary
process in the Rorschach. In M. A. Rickers-Ovsiankina (ed.)
Rorschach Psychology. New York: Wiley, 1960. '

 

Knight, R. P. Introjection, projection, and identification. Psycho-
analytic Quarterly, 1940, 9, 334-341.

 

Kris, E. Psychoanalytic Explorations in Art. New York: International
Universities Press, 1952.

 

Lazowick, L. M. On the nature of identification. Journal of Abnormal
and Social Psychology, 1955, 51, 175-183.

 

 

Levy, J. Regression in the service of the ego, cognitive control, and
sexual identification. Unpublished doctoral dissertation,
Michigan State University, 1961.

Lowenfeld, H. Psychic trauma and productive experience in the artist.
In W. Phillips (ed.), Art and Psychoanalysis. Cleveland and
New York: The World Publishing Co., 1963.

57

Zucker, R., and Fillmore, K. Masculinity-femininity in fantasy pre-
ferences: an indirect approach. Journal of Projective Tech-
niqpes and Personality Assessment, 1969, 33, 424-432.

 

 

Zukowsky, E. Measuring primary process and secondary process thinking
in schizophrenics and normals by means of the Rorschach.
Unpublished doctoral dissertation, Michigan State University,
1961.

 
  
 

Typed and Printed in the U.S.A.
Professional Thesis Preparation
Cliff and Paula Haughey

144 Maplewood Drive

East Lansing. Michigan 48823
Telephone (517) 337-1 527

. ~‘7

  

MICHIGAN STATE UNIV. LIBRQR ES

‘ 111115Mlillllllll111

:11
JL293106663978