University of Southern California Dissertations and Theses

A two-dimensional model of cognitive empathy: An empirical study

(USC Thesis Other)

A two-dimensional model of cognitive empathy: An empirical study

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content A TWO-DIMENSIONAL MODEL OF COGNITIVE EMPATHY:
AN EMPIRICAL STUDY
by
Kristin Elisabeth Zethren
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PSYCHOLOGY)
August 2002
Copyright 2002 Kristin Elisabeth Zethren
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 3094405
UMI
UMI Microform 3094405
Copyright 2003 by ProQuest Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY OF SOUTHERN CALIFORNIA
The Graduate School
University Park
LOS ANGELES, CALIFORNIA 900894695
Thi s dissertation, w ritte n b y
K R IS T / k] j?enJ_________
Under th e d irectio n o f h£- . C. D isserta tio n
Com m i ttee, and approved b y a ll its members,
has been presented to and accepted b y The
G raduate School , in p a rtia l fu lfillm e n t o f
requirem ents fo r th e degree o f
D O CTO R OF PH ILO SO PH Y
D a te I ' ( 0 • &
DI SSER TA T IO N C O M M ITT E E
I
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEDICATION
for Lois Corte
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
IN MEMORIAM
September 11, 2001
Empathy was for Berlin the essential liberal aptitude--the ability to be
open, receptive, unafraid in the face of opinions, temperaments,
passions alien to one’s own. Empathy committed you to leaving the
stockades of liberal reason in order to understand reason’s limited
possession of a world propelled by passionate conviction. Empathy of
this sort did not imply relativist indulgence: it meant coolly establishing
what could be negotiated across the frontiers between convictions and
what could not.
Michael Ignatieff on Isaiah Berlin in The New Yorker,
September 28, 1998, p. 73.
The truth, perhaps, is that Billy the Kid didn’t really care about religions
or nations. He was not in it for faith or fatherland but for guns and
money. This strange feud gave him a chance to perform, an excuse to
kill, a living. Perhaps he was just a brilliant opportunist, a scrawny kid
with few prospects who suddenly found himself in an exciting,
dangerous place and decided to make the most of it. He seized the
moment, improvised a destiny for himself, and kept moving until his luck
ran out.
Fintan O’Toole on Billy the Kid in
The New Yorker,
December 28,1999, p. 97.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ACKNOWLEDGEMENTS
I would like to thank the National Charity League and the NCL
Assessment Committee under the direction of Dr. Gretchen Guiton, for
awarding me a grant in financial support of this project.
I am especially grateful for the guidance of my Chair, Dr. Michael
Newcomb, for sharing his formidable expertise, as well as the resources of a
grant from the National Institute of Drug Abuse. Dr. Newcomb helped me gain
access to two formal university subjects pools, without which there would not
have been sufficient subjects for this study; and he further allowed me access
to his data entry firm, Advanced Data Entry Consultants.
Sincere gratitude is extended to members of my committee: Dr. Rodney
Goodyear, for his insights and suggestions; and to Dr. Dallas Willard, for his
attention to the structure of argument. Thank you, gentlemen, for your
forbearance, and empathy.
Thanks also to Dr. Michael Laurent for helping me enlist subjects in the
direct mail campaign; and to my mother, Lois Corte, who helped me with the
practical demands of direct mail solicitation.
Special thanks are extended to Dr. Thomas Locke for helping me with
data management and restructuring the questionnaire for the college subject
pools; to Dr. Kevin Kim, for guidance on data cleanup; and to Linda Long, for
empathic help with statistical analysis.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
I would also like to thank all the individuals who participated in this
study. Their willingness to spend time filling out a lengthy questionnaire,
sharing highly personal information, is recognized and appreciated.
Finally, I thank my family and friends, particularly Christina Adam (and
her timely white tulips), Curt Batiste, and Chic Wolk, for their love and support.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
DEDICATION..............................................................................................................ii
IN MEMORIAM...........................................................................................................iii
ACKNOWLEDGEMENTS........................................................................................ iv
LIST OF TABLES.....................................................................................................vii
LIST OF FIGURES................................................................................................... ix
ABSTRACT................................................................................................................x
CHAPTER
1 . INTRODUCTION................................................................................................... 1
A Model of Empathy..................................................................................... 3
2. REVIEW OF THE LITERATURE......................................................................10
Cognitive Complexity................................................................................. 10
Altruism......................................................................................................... 16
Empathy...................................................................................................... 22
Sidebar on Measurement.......................................................................... 33
3. METHOD............................................................................................................. 35
Participants..................................................................................................35
Protocol ................................................................................................... 35
Description of the Sample......................................................................... 36
List of Measures.........................................................................................41
Instrumentation and Sample Reliabilities.................... 43
Research Design and Analysis................................................................ 54
Data Management ......... 56
Reliability of the Sample Measures......................................................... 57
Statistics and Data Analysis.......................... 58
4. RESULTS............................................................................................................ 60
The Independent Variables.......................................................................60
Correlations Among Dependent Variables ............................................68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Hierarchical Regressions Controlling for Age, Gender
And Social Desirability............................................... 77
Multivariate Analysis of Covariance........................................................90
5. DISCUSSION.....................................................................................94
Limitations of the Study........................................................................... 104
SELECTED REFERENCES................................................................................105
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
1: Demographic Characteristics by Sample Location...................................... 38
2: Continuous Demographics by Sample Location......................................... 40
3: Reliability for Study Measures.......................................................................44
4: Summary Statistics for Study Measures................................................... 45
5: Pearson Correlations Among Dependent Measures............................... 62
6: Pearson Correlations Among Independent Measures.............................63
7: Pearson Correlations Among Dependent and Independent
Measures.................................................. ..................................................... 64
8: Pearson Correlations Among Dependent Measures and
Controls............................................................................................................. 65
9: Hierarchical Regression on Splitting............................................................78
10: Hierarchical Regression on Psychopathy...................................................79
11: Hierarchical Regression on Receptive Giving............................................80
12: Hierarchical Regression on Narcissism...................................................... 81
13: Hierarchical Regression on Responsibility................................................. 82
14: Hierarchical Regression on Control.............................................................83
15: Hierarchical regression on Enmeshment .........................................84
16. Hierarchical regression on Intimacy ...................................................85
17: Hierarchical regression on Empathic Concern.........................................86
18: Hierarchical regression on Fantasy...........................................................87
19: Hierarchical regression on Personal Distress..........................................88
20: Hierarchical regression on Perspective Taking........................................89
viii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
21: Two (High versus Low Cognitive Complexity) by Two (High
versus Low Altruism) Multivariate Analysis of Covariance,
Controlling for Age and Social Desirability.................. 91
22. Correlations Among Dependent and Independent Variables
Among the Three Sample Subgroups.......................................................... 95
ix
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
1: Circumplex Model: Theory.................................................................................4
2: Circumplex Model: Theory Operationalized With Measures.......................9
3: Circumplex Model: Results on Cognitive Complexity.................................61
4: Circumplex Model: Results on Altruism........................................................69
5: Circumplex Model: Results on Altruism and Cognitive Complexity 101
6. Circumplex Model: Results Substituting Authoritarianism for the
4 x 6 Repertory Grid........................................................................................103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ABSTRACT
This study was designed to investigate the validity of a proposed
theoretical model of cognitive empathy. The model consists of a circumplex
with two separate constructs free to vary independently of one another, creating
a contingency table: Altruism (A) on the X-Axis, measuring the ability to see
similarities between self and other; and Cognitive Complexity (CC) on the
Y-Axis, the capacity to see differences.
Four interpersonal relationship styles were hypothesized to result from
the combination of A and CC:
Cognitive Empathy-high scores on both A and CC. This is the quadrant
of resonant understanding and describes the “l-Thou” relationship.
Psychopathy or a less malevolent Expedience-high scores on CC but
low on A, describing a contractual, contingent style characterized by the “l-You”
relationship. Extremely low scores on A would indicate a more antisocial
stance.
Co-Dependence~\ovj scores on CC and high on A: a state of emotional
confiscation characterized by the “l-We relationship, without much of an “I.”
Narcissism-low scores on both A and CC, suggesting an incapacity to
decenter and reflect. It is characterized by an “l-lf relationship style.
A convenience sample of 344 subjects was solicited by direct mail, and
obtained from two formal university subjects pools-a large state university and
a large private Catholic university.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The study relied on a correlational design, augmented by hierarchical
regressions and a 2x2 MANCOVA. To determine if the model was correct, the
independent variables of Altruism (a subscale of the Helping Orientation
Questionnaire) and Cognitive Complexity (measured by the 4 x 6 Repertory
Grid) were examined for their relationship to the key dependent variables:
Cognitive Empathy, four subscales of Co-Dependence, Primary Psychopathy,
and Narcissism. Several additional measures were added to assess
discriminant and convergent validity: Authoritarianism (CC); Splitting
(Narcissism), Receptive Giving (Expedience); Empathic Concern (A); and
seven subscales of the Love Attitudes Questionnaire. The Crowne Marlowe
was added to control for social desirability; age and gender were also used as
control variables.
The results of the study did not support the model as measured.
Altruism appeared to measure what it purported to measure, yielding positive
relationships with Empathic Concern and Perspective Taking, and negative
relationships with Psychopathy, Personal Distress, and Splitting. Cognitive
Complexity, on the other hand, contrary to previous studies, was positively
related to Splitting and unrelated to Authoritarianism, suggesting that the 4x6
Repertory Grid did not measure Cognitive Complexity, a potential failure of
construct validity. However when the four-item F-Scale measuring
Authoritarianism was substituted for Cognitive Simplicity, the main hypothesis
was validated: That to adopt an empathic stance, one has to see both
similarities in self and other, as measured by Altruism, and differences between
xii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
self and other, as measured by Cognitive Complexity (or negative scores on
Authoritarianism). This last finding suggests that a more robust measure of
Cognitive Complexity might yet, in future studies, validate the hypothesized
model.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 1
INTRODUCTION
Empathy has long been considered an essential element in the
counseling process (Gladstein, 1983; Rogers, 1975; Williams, 1990), helping to
build the therapeutic alliance; allowing the counselor to sense and evaluate the
client's experience. Indeed, the capacity for empathy is important in fields other
than counseling, such as management and medicine, and is especially valuable
in education. Hence, numerous measures have been developed for screening
or training purposes based on numerous theoretical models.
However empathy has been notoriously difficult to measure (Gladstein
1983, Williams 1990). Either the theoretical model has been unidimensional
and too simplistic; or, by way of correction, the model is viewed as
polydimensional and becomes unfocused; or the measure has required judges
observing and rating the behavior of others in situ (e.g., Kagan & Schneider,
1987), and therefore becomes too time and labor-intensive for practical
purposes.
A persistent problem concerns the observation that there appear to be
two necessary but frequently unrelated kinds of empathy: affective and
cognitive. Affective is the ability to feel what another is experiencing and
cognitive is the ability to understand what another is experiencing. Both
cognitive and emotional empathy appear essential to the counseling process
yet are often negatively correlated when measured (Chlopan, McCain,
Carbonell, & Hagen, 1985; Williams, 1990). This is because, I shall argue,
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
each developed at different times in the course of evolution-affective empathy
during the emergence of the Mammalia in the service of rearing the young in a
social setting, and cognitive empathy emerging only with the advent of Homo
Sapiens in the service of greater behavioral flexibility.
Established measures include the Hogan Empathy Scale (Hogan,
1969), a measure of cognitive empathy or perspective-taking; and the
Mehrabian-Epstein Emotional Empathy Scale (Mehrabian & Epstein, 1972).
The latter purports to measure affective empathy, but actually measures a sort
of emotional contagion; for instance, items like “I tend to get emotionally
involved with a friend’s problems” could load onto codependent pathologies
(Roehling & Gaumond, 1996). A more recent, polydimensional measure used
in this study, the Interpersonal Reactivity Index (Davis, 1983), reconceptualizes
the construct by positing a four-factor model: perspective-taking (cognitive
empathy), empathic concern (emotional empathy), fantasy (the capacity to
imagine or identify with fictitious scenarios), and personal distress (sympathetic
arousal at another’s misfortune). As will be seen, emotional empathy is really
the positive form of personal distress-an often unconscious arousal
engendered by the arousal of another, much like the resonance of harmonically
related strings when one string on a harpsichord is plucked.
In thinking about other ways of conceptualizing empathy that could
capture its core attributes, we drew inspiration from an ancient model that
evolved out of spiritual traditions, particularly from the Tibetan Buddhist model
of spiritual attainment, symbolized by an embracing couple as the marriage of
2
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
wisdom and compassion. It is like the Jungian notion of the conjunctio
oppositorum, the conjunction of opposites, or, perhaps more accurately, the
marriage of complements.
And so this study proposes a similar model that, theoretically, describes
the more basal constructs out of which empathy arises, from the perspective of
evolutionary psychology (Zethren, 2000). Accordingly, empathy is conceived
as bidimensional, consisting in two empirically separate constructs free to vary
independently of one another. Compassion is operationalized as Altruism (A)
on the x-axis, and wisdom operationalized as Cognitive Complexity (CC) on the
y-axis (Figure 1), creating a circumplex or contingency table with four
quadrants. One quadrant is hypothesized to capture cognitive empathy, a
relationship style and an expression of the self in social contexts. As a
byproduct, three other relationship styles, perhaps diminished expressions of
self, hypothetically emerge to fill the circumplex.
A Model of Empathy
Altruism (A), on the x-axis, is defined as "behavior carried out to benefit
another without anticipation of rewards from external sources (Macaulay &
Berkowitz, 1979, p. 3), a willingness to put aside one’s own agenda for the
agenda of another. Altruism includes behaviors such as active listening and
helping interventions. It is the ability to see similarities between self and other.
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FIGURE 1. Circumplex Model: Theory
Cognitive Complexity
- Altruism + Altruism
Narcissism
Expedience
Splitting
Psychopathy
Enmeshment
Codependence
Social Desirability
Perspective-taking
Cognitive Empathy
Cognitive Simplicity
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Now, this assignment is problematic, because the current assumption
(Batson, Batson, Slingsby, Harrell, Peekna, & Todd, 1991) among researchers
is that, from the point of view of causality, empathy precedes altruism, perhaps
an anthropomorphic prejudice. In addition, altruism is difficult to measure in
paper and pencil tests. Undoubtedly in the real world there is a recursive, back
and forth process in the altruism-empathy interaction, but Gladstein (1987) has
shown that an empathic understanding does not necessarily lead to an altruistic
intervention.
So I shall argue that, from an evolutionary point of view, altruism is basic
because it is hardwired much early in phylogeny, conserved across the socially
organized arthropods, for instance, and conserved again across the Mammalia
as a way of improving the survivability of the increasingly complex individual
within a social species. With the development of the subcordical limbic system
in the higher Mammalia, one sees the development of rudimentary forms of
emotional empathy--a sort of sympathy, in the sense of sympathetic arousal
and feeling what others feel, in the postnatal service of rearing the young. It is
like a sympathetic arousal that then becomes conserved not only for its
contribution to the survival of the young, but also for its contribution to the
development of complexity. But as this arousal can have negative as well as
positive consequences, we return to the capacity for altruism as our
independent variable on the x-axis, in that to empathize in the ordinary-
language sense, one has to first take the more basal altruistic stance-a form of
generosity-and be prepared to orient toward the other with a benign attitude.
5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
But something else is then required, something akin to wisdom.
Cognitive Complexity (CC), on the y-axis, is that ability to perceive nuance, to
see several aspects of a situation, to see the gray areas in a black-and-white
picture, to tolerate ambiguity, to differentiate, as opposed to fuse, elements of a
construct. In counseling, one may argue, it is the capacity for differential
diagnosis and case conceptualization. CC captures the ability to see
differences between self and other.
Cognitive simplicity, on the other hand, consists in a reduction of the
capacity for multidimensional conceptualization-an incapacity to perceive
nuance, an intolerance of the anxiety of ambiguity, the reduction of experience
to black or white polarities or simple constructs, and the tendency to assume
that one’s own experience is also the experience of the other (Bieri, 1955).
From the point of view of evolution, cognition is increasingly elaborated
as we move up the phyla. Cognitive complexity is a highly evolved capacity,
part of what we call consciousness. What is its purpose? It is part of the non
hardwired decision-making ability of the cortex to help the organism arrive at
the best in-the-moment choice, to aid in prediction of others’ behaviors. It is an
addition to the adaptive hardwired, such as the reflexive rise of the hand to
deflect a blow, for instance, or the flight-or-fight arousal system, or the
sometimes unwelcome sympathetic resonance of emotional empathy, as when
we feel manipulated by a maudlin tear-jerker movie.
So, with Altruism on the x-axis, which one can summarize as the
capacity to see similarities; and with Cognitive Complexity on the y-axis, which
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
we can characterize as the ability to detect differences, what sort of interaction
can we expect? When assessed together, the following results are
hypothesized:
Empathy. High scores on both cognitive complexity (CC) and altruism
(A). This primarily captures the capacity for cognitive empathy, the ability to
see things from another person’s point of view, and to partake of another’s
experience. It is the capacity to understand a foreign consciousness.
Borrowing from Buber and Smith (1937), we call this the "l-Thou" relationship,
where one subjectivity meets another, where one Self encounters another Self.
Expedience, or Psychopathy. High score on CC and low score on A.
This is the quadrant of everyday transactions, the quid pro quo of the business
world, a contractual, contingent style characterized by the "l-You" relationship.
It is normal to evaluate situations in a cost-beneflt framework; but if there is a
very low score on A, then the tendency slides more into psychopathy, a selfish,
uncaring state where a person might have sufficient cognitive complexity to
detect vulnerabilities to exploit.
Co-Dependence: Low score on CC and high score on A. This quadrant
describes a state of emotional contagion, or sympathetic arousal, where one
may be overwhelmed by another's emotion but does not have sufficient
distance to discriminate others’ affect from one’s own. It is a kind of emotional
confiscation characterized by the "we" relationship of enmeshment in another's
life, such as with spouses and families of alcoholics, without much of an “I.”
(Note: In measuring this construct, we used subscales of a codependence
7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
measure rather than the full scale, in order to achieve greater refinement in our
analysis (Figure 2).
Narcissism: Low score on CC and on A. This describes those whose
capacity for both altruism and cognitive understanding is limited. In this state,
one can neither decenter nor distinguish nuances of others’ experience. This
quadrant describes the "i-lt" relationship, where the other is perceived as either
good or bad in the context of one’s own pleasure or pain. One becomes an
object, to be manipulated or perfected, rather than the subject of a life.
Most of us fall somewhere in the middle of the circumplex, depending on
mood or state. Don’t we often think, as Gore Vidal quipped, that a narcissist is
someone better looking than we are? Or define a bore, as Budd Schulberg
does, as someone who, when you ask him how he is, tells you? And we
Americans still can’t decide, according to humorist Larry Brown, whether capital
punishment is an affront to humanity or a potential parking place. But
persistent trait extremes indicate that somewhere along the developmental line
our manner of relating to each other got off track. The exception, perhaps, is
on the empathic quadrant, where high spiritual attainment hypothetically occurs.
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FIGURE 2. Circumplex Model: Theory
Operationalized With Measures
Cognitive Complexity
■Altruism
Personal Distress
Splitting Index:
Narcissism
Responsibility
Intimacy
Enmeshment
Control
Cognitive Simplicity
' Receptive Giving _ _ Perspective-taking \
\ Social Desirability
Psychopathy _ _ Empathic Concern \ Age
+ Altruism
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 2
REVIEW OF THE LITERATURE
In the review of the literature on empathy, altruism, and cognitive
complexity, I have focused on those aspects which reveal three central
arguments: that there is an evolutionary aspect to the capacity for both altruism
and complexity; that altruism consists in the capacity to see similarities between
oneself and another, while cognitive complexity is the ability to see differences;
and that empathy of others helps one develop a self concept.
Cognitive Complexity
The model was created after inspecting the diverse literature on
empathy. Each study of which constituted of a single tessera but when fit
together in aggregate, revealed the bidimensional mosaic. For instance,
Gladstein (1983) noted that part of counseling's empathic response includes not
only emotional resonance (affective empathy), but also an intellectual capacity
for role-taking (cognitive empathy); imagining what it is like to be the client;
which, in turn requires cognitive suspension, a provisional letting-go of personal
beliefs and bias in the service of greater comprehension. This suspension of
belief and disbelief requires an intellectual flexibility that theoretically is
captured by cognitive-complexity.
Indeed, measures of CC negatively correlate with measures of rigidity
and dogmatism (Hinze, Doster, & Joe, 1997; Scott, 1962). It is the ability to be
receptive to novel information at hand, rather than anxiously wedded to
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
precepts chiseled in stone. In the development of constructs, or schemata, it is
the capacity to bracket preconceptions and perceive multiple dimensions when
one conceptualizes another person, a situation, or a task (Scott, 1962).
Part of the value of bracketing is that one can look at another person’s
pattern of behavior more objectively and thus better predict what she or he will
do. In a discussion about the uses of CC, Bieri (1955) states:
It is assumed that a basic characteristic of human behavior is its
movement in the direction of greater predictability of an individual’s
interpersonal environment. It is further assumed that each individual
possess a system of constructs for perceiving his social world. These
constructs are invoked and form the basis for making predictions....
Thus, under the supposition that person X is perceived as “hostile”
(construct), an individual may make one kind of prediction about his
behavior, while if he were perceived as “ friendly,” another kind of
prediction might be made. (p. 263)
If an assumption were validated, that would tend to preserve the
construct, while if that assumption were false, i.e., if the prediction failed, then
the construct on which the prediction was based would best be changed or
modified. One’s predictive efficiency, then, is a function of the versatility of
one’s construct system-one’s ability to dance to the tune of empirical trial-and-
error. “Inasmuch as constructs represent differential perceptions or
discriminations of the environment,” writes Bieri (1955), “it would be expected
that the greater the degree of differentiation among the constructs, the greater
will be the predictive power of the individual” (p. 263.
Cognitive complexity is related to measures of g, or general intelligence
(Gottfredson, 1997), which, in turn, has a genetic basis and therefore, from an
evolutionary point of view and functions in the service of fitness. Higher levels
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of CC predict a greater ability to prevail over the petty hazards of life; higher
levels of CC systematically improve the odds of successfully negotiating
complex situations-an evolutionary advantage. Those with lower CC, and thus
a diminished capacity to deal with the complexity of life, appear to rely more on
algorithms and if-then contingency tables to tell them what to do; sticking with
the tried-and-true compensates somewhat for an inability to process complex
information in-the-moment.
Indeed, CC as a trait is inversely related to conservatism (Hinze, Doster
& Joe, 1987). Subjects scoring high on measures of conservatism demonstrate
a restricted range of “construct assignment flexibility," that is, they tended to
view others in a preset categorical manner, losing differentiating information.
This lack of flexibility seems to constitute a defense against the anxiety of
ambiguity, or may be a general strategy to reduce anxiety and stress. Subjects
with social anxiety, for instance, demonstrate less construct differentiation, i.e.,
cognitive complexity, than controls (Sanz, Avia & Sanchez-Bernardos, 1996).
CC, as a trait, is subject to environmental states and thus can vary over
time in all subjects. Stress, as measured by high levels of cortisol (Rotton,
Dubitzky, Milov & White, 1997), is negatively related to current levels of
cognitive complexity. All of us resort to canned recipes in times of confusion
and distress.
Cognitive Complexity, however, does not appear to inoculate against
intellectual laziness, such as racial stereotyping or a reductive bias. In
language attitude assessment, for instance, racial belief exerted a greater
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
influence than CC (Robinson, 1996) on attitudes towards nonstandard English.
In acquiring competence and expertise in a field, subjects have a tendency to
fall into a reductive bias-that is, there is a tendency to become more rigid in
one's thinking as one masters a field of endeavor (Feltovich, Spiro, & Coulson,
1997). A medical expert, for instance, may dismiss the possibility of statistical
outliers and thus miss subtle differentiating information that could be of help to
the patient. In both instances, flexibility is lost and adaptability is compromised.
Cognitive Complexity, then, appears to work as an adaptive strategy by
focusing on differences (Millington, Strohmer, Reid, & Spengler, 1996), rather
than similarities, in the service of making advantageous choices. Those who
focus only on similarities tend to project their own emotions and so on to the
other, often with insufficient evidence as to the accuracy of their assumptions.
This has been defined as “ assimilative projection.” Cameron (1947) states “ The
less practiced a person is in the social techniques of sharing the perspectives of
others, the less opportunity he will have of finding out how different from himself
other ordinary people can be” (p. 167). Cognitive simplicity “reflects an
incomplete differentiation of the boundaries between self and the external
world, leading to unwarranted assumptions of the similarity between self and
others” (Bieri, 1955, p. 267). Cognitive complexity refers more to an accurate
prediction of the differences between self and others, rather than the
similarities.
The late philosopher Robert Nozick (1998), in an invited address at
APA, suggested why this might be advantageous. Consciousness, he
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
hazarded, evolved in the service of fitness in a complex environment. It is
awareness that data is registering, held in mind as a mental model while
possible consequences are calculated.1 Behavior is hardwired, or unconscious,
he suggested. When there is one best choice in a given situation: the hand
goes up reflexively, for instance, to shield the head from a blow. Nozick
proposes that operant conditioning, a higher order decision mechanism, is cued
by discriminant stimuli. In the shifting sands of more complex situations,
however, adaptation requires conscious review of that which is both novel and
changing. Mental representations, or awareness that sensory data is
registering, help achieve increasing precision in reaching a non-hardwired best-
choice point.
The ability to fashion mental representations with high cognitive
complexity, however, does not necessarily motivate engagement with another
or inspire interest in the other's welfare. In two studies designed to assess CC
and attraction, subjects with high CC were attracted to other targets (situations,
1 The capacity to be aware that sensory data are registering is a
property of consciousness. This can be illustrated by the phenomenon of blind
sight, where a person whose retina is intact but whose visual cortex is not
working data into awareness, can nonetheless react to objects in the visual field
yet not know that they're there. These people are blind. When asked what is
there, they might say they can't see, but when asked to guess, generally guess
correctly beyond chance. It could be argued that simple existence, with a
repertoire of reflexes and conditionings, would not require consciousness, or
mental representations, since behavior would be hardwired and you would not
have to think about alternatives. Complex existence seems to require
consciousness in order to calculate the best in-the-moment choices to novel
situations: a metaphorical space where situations can be seen, behaviors tried
out and outcomes imagined. Julian Jaynes (1978) presents a lucid discussion
of the functions of consciousness.
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
tasks) or people with similar high CC, but were not drawn to those with low CC
(Burleson, Kunkel & Szolwinski, 1997)-a boredom factor? Low CC subjects
were more attracted to other low CC partners, but unexpectedly preferred high
CC targets, perhaps sensing the novel possibilities embedded in complexity.
The core evolutionary advantage conferred by the capacity for cognitive
complexity appears to be an ability to predict dangerous. In a study designed
to see whether one differentiates more among people who evoke negative than
positive affect (Irwin, Tripodi & Bieri, 1967), subjects displayed a heightened
capacity for differentiation when exposed to negative stimulus persons, i.e.,
mean or dangerous people, as opposed to positive persons. Neutral persons
elicited more differentiation than positive persons, but less than negative
persons.
An important behavior arises from the capacity to differentiate among
constructs and mental representations of self and other. This is the capacity to
shape communication in conscious ways, as opposed to the tendency to
articulate often unconscious emotional reactions in a haphazard manner.
Communications analyst Ray Preiss (1998, personal communication) explains:
Because complexity allows communicators to act strategically,
cognitively complex communicators seem to have fundamentally
different notions of what communication can be used to ‘do’ (when
compared to their low complexity counterparts). In other words, low
complex communicators seem to view communication as an
“expressive” act. They say what they feel. Complex communicators,
however, see communication as a rhetorical event. They
simultaneously pursue multiple goals and talk in strategic, tactical ways.
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Altruism
While CC may be a necessary element in the empathic response, it is
not a sufficient one. Crooks and con-artists can be aware of others' emotional
states but may use that information for self-serving purposes. Cognitive
complexity focuses on those differences which can serve to alienate. Another
quality is required which motivates one to make an effort and to care about
someone else's welfare in the absence of immediate external reward--a stance
of beneficent cooperation, a sense that one is part of a human family. It is the
ability to perceive similarities.
Altruism, derived from the Latin alter, “other,” was chosen to capture that
quality of connectedness, in no small part because there is evidence that it, too,
is supported by a biological substrate, conserved across the phyla, as far back
as the social arthropods, in the genetic legacy that includes bees, birds, bats,
baboons, and humans (Trivers, 1985). Twin studies tend to confirm the
heritability of altruism (Loehlin, 1992), in that, even when raised apart,
monozygotic twins are twice as likely to share personality traits such as
cooperation, empathy, and altruism.
However, altruism as a human behavior is not wholly accounted for by
gene studies; it appears to not be completely hardwired. Rather, it is a
propensity, barring organic dysfunction, which is then developed through
socialization. The genotype requires environmental input to express its
phenotype. It is partly an acquired attitude and a mechanism of evaluation
based on the pleasure principle, with a valence and an intensity.
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Altruism as an attitude, like a trait, would have variable expression,
depending on the circumstance or state. Mood influences behavior, as does
setting and state. Game theory (Ekeland, 1999) suggests that under certain
conditions, altruism would not necessarily accrue to the benefit of the individual,
e.g., in competitive situations such as exam taking; nor would it be
disadvantageous for society if an individual did not exercise it in that situation.
The individual looks after his genes; society looks after a gene pool by
encouraging competition designed to give advantages to greater fitness. In
other arenas, however, altruistic behaviors are expected, such as during a
natural disaster, when individuals seem more ready to forgo personal
advantage to help others and repair the social fabric.
Sociobiology offers evidence that related ness inspires altruism: children
elicit greater sacrifice from their parents, who share 50% of their genes, than
from aunts and uncles, who share just 25% (Dawkins, 1976). We obey the
evolutionary command to pass on our genes by helping those who do
(Haldane, 1955; Hamilton, 1972). Helping behaviors and genetic heritability do
not make a perfect correlation, however; it is the individual, not her genes, who
makes altruistic choices, and in doing so, may serve a social rather than
individual agenda, such as adopting a child who has no immediate genetic
relationship.
Actual helping behaviors, as opposed to altruistic attitudes, appear to be
motivated by sameness. Relatives inspire more help than friends but strangers
who at least look the same elicit help as well. People are more motivated to
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
prevent a drunk person from driving, and possibly harming himself, for instance,
if they know and like the inebriate, or find him similar in appearance and dress
(Rabow, Newcomb , Monto & Hernandez, 1990).
It is useful to remember that altruism ups the ante from cooperation,
adding an element of selflessness which geneticists find inimical to canonical
evolutionary theory. Cooperation, a quid-pro-quo strategy, arises among
unrelated individuals, even species, if there is mutual benefit (Ridley, 1988). It
is calculated in a cost-benefit analysis, elicited as behavior when its Darwinian
benefits outweigh the effort expended (Trivers, 1975; one should note, that
calculating cost vs. benefit occurs mostly outside awareness, for humans no
less than bees and bats).
Altruism serves the genetic interests of relatives and the gene pools of
groups. From an evolutionary perspective, altruism manifests in group
practice-even when there is a high cost to the individual, such as adopting a
child, or going to war-when selection at the level of the group is stronger than
at the level of the individual (Sober & Wilson, 1998). Thus, one could argue,
altruism at the tribal level would help conserve a gene pool; global altruism
would help survival of the species. Conservation of a culture, such as western
European civilization, would conserve a set of survival strategies for a gene
pool; maintaining a diversity of gene pools would accrue to the genetic health of
a species.
Since members of Homo sapiens share most genes in common, it could
be further argued that moral tenets extolling altruism, as expressed universally,
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
for instance, in the Golden Rule, arise among religions as a general societal
strategy to insure survival of the species. This famously sets up potential
conflict between the desires of the individual and the needs of the species as a
whole (Freud, 1930).
Though altruism may be held as socially desirable, there is ample
motivation to cheat, especially if one feels alienated from society. Thus
measures of altruism must be weighed against mere lip service to it. If the
tenets of that attitude are actually violated, as in antisocial behavior or
excessive self-service, then guilt should normally arise as a counter-motivator.
The presence of antisocial behavior may be tolerated and dealt with in the
context of civil law, but absence of guilt is seen as deviant. (Remorseless
antisocial behavior, it could be argued, arises when, in the absence of organic
dysfunction, there is no environmental support for the genotype. Examples
may include lack of maternal bonding due to postpartum depression, or
excessive neglect or egregious random punishment.)
Many measures of altruism rely on observers and raters, on the theory
that, qua behavior, altruism can be observed only in situ. From an
anthropological point of view, when behavior is our target, observation in the
field is thought to be more accurate. This is because, at least on paper, we
often present our persona painted according to the tenets of social desirability;
but in real life we may act in unkind or self-serving ways. Now it could be
argued that, even though one may give lip service to altruism without
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
actualizing it, at least attitude shows one's aims, if not achievements, and
demonstrates that one has internalized the moral tenets of society.
Altruism is part nature, part nurture. Behind the nurture aspect, there is
an attitude, a propensity that, though not directly observable, nevertheless can
be assessed. This kind of assessment, a stimulus, then person scaling
technique (Himmelfarb, 1993), was adapted for the measurement of
hypothetical constructs (MacCorquodal & Meehl, 1948) from already
established models of psychophysical measurement, such as measurement of
response to an auditory tone stimulus. Both measures atomize psychic
phenomena by calibrating them into discrete, quantifiable valences; both aspire
to some sort of representational, or correspondence, statement, though
measurement of psychological constructs, in the statistical language of the
social sciences, tolerates more than 50% error.
Philosopher Robert Nozick (1998) offers an hypothesis as to why
measurement is so difficult in psychology, which bears on our discussion of
altruism. He noted wryly that our measurements are poor and predictions off--
from the point of view, say, of a chemist, for whom the statement H2 0
corresponds exactly with molecular fact--not because we're stupid (!) but
because there may be an evolutionary advantage to being unpredictable.
Nozick hazards a guess that in arenas when cooperation confers an advantage,
individuals will be more predictable than in competitive situations where one
may stand to gain from an element of surprise. By extrapolation, one could
guess that altruistic behaviors operate like clockwork in situations that favor
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
selection, such as rearing one's own children. Indeed, that appears to be the
case, and society thinks so, too. Parents who seem unable to sacrifice a
measure of self-interest for the sake of their children soon come to the attention
of government social services.
For the purposes of measurement, attitude is defined as "a
psychological tendency that is expressed by evaluating a particular entity with
some degree of favor or disfavor" (Eagly & Chaiken, 1993, p. 1). It is a bias
that predisposes, based on the pleasure principle, like a channel cut through
rock through which a river is invited to flow. When measured, the attitude of
altruism has been associated with self-actualization and internal locus of control
(Sharma & Rosha, 1992), which reflects the fact that altruistic behavior is
defined as undertaken without expectation of external reward. It is not that
altruistic actions are without reward; human behavior is shaped by the pleasure
principle. Rather, that reward is internal: it feels good to help others. Proverb
therefore confirms sociobiology in stating that virtue is its own reward; French
wit Mme. Du Barry underestimated the power of internalized pleasure when she
retorted that virtue is its only reward. Inner rewards appear, in fact, to be more
rewarding than external ones (Batson et al, 1991).
Subjects who test as having a high degree of altruism, when asked to
look at the same situations presented to subjects testing with low altruism,
tended to see those situations as more rewarding and less costly (Kerber,
1984), suggesting more optimism in situation evaluation. Altruistic people seem
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to perceive helping situations favorably in terms of a cost-benefit analysis.
Helping is seen as less effortful, less threatening, and more pleasurable.
Altruism, then, could be seen as a tendency to focus on the benefits of a
Darwinian cost-benefit analysis, and to find internal rewards for helping
behaviors. Altruism appears to operate from a position of strength, finding
pleasure, rather than threat, from differences and diversity. It is the strong who
can afford to be kind.
Empathy
When measured, subjects testing high on A and high on CC are
hypothesized to possess that capacity called empathy. It should be noted at
the outset that the kind of empathy we are after here is a mature empathy,
more like the perspective-taking cognitive empathy, that normal adults seem to
acquire in stages over the life span.
An even deeper form of empathy, evoked in counseling or pastoral
settings, can be developed through study and experience. It requires an extra
measure of effort in that understanding must go deeper than face value into
what psychologists call unconscious motives, or behaviors mediated by
learning which occur outside awareness. Philosopher Thomas Nagel (1994)
offers a sense of what we mean by mature adult empathy:
To understand someone else’s thoughts, feelings or behavior requires
that we make sense-even if only irrational sense-of his point of view,
by using our own point of view as an imaginative resource. Imagination
enables us to make internal sense of beliefs, emotions, and aims that
we do not share~to see how they hang together so as to render the
other’s conduct intelligible.... When we interpret other people in this
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
way by making sense of their point of view, we are not merely imagining
things as when we see animals in the clouds or ascribe malice to a
defective toaster. Rather, we are trying to understand, within the limits
of a nonscientific psychology, what really makes people tick, and we
often hope to be confirmed by the person’s own self-understanding.”
(Nagel, 1994, p. 34)
Counseling empathy goes a step farther, and attempts to put social-
scientific research behind attempts to understand the client’s unconscious
mental processes. For instance, an analyst may couch empathic
understanding in the language of transference such as the patient’s idealizing
or acting out. Behaviorists may see a pattern of maladaptive operant
conditioning behind a client’s unfortunate behavior and so on. But it is not
merely research that is brought to bear on counseling empathy, for human
behavior is far too complex with multiple causes and outcomes for any given
data set, to chart a certain etiology or stance. “ We may not be able to run
controlled experiments,” writes Nagel, “but we can still try to make internal
sense of what people do, in light of their circumstances, relying on a general
form of understanding that is supported by its usefulness [italics mine] in
countless other cases, none of them exactly the same” (Nagel, 1994, p. 35).
In this regard it is useful to remember that the word empathy comes to
us from the field of aesthetics, and refers metaphorically to a sense of
belonging to or being carried away by an experience~Einfuhlung, to feel with,
according to German philosopher Theodore Lipp (1891). The sense conveyed
here is that "Aesthetic pleasure is an enjoyment of our own activity in an object"
(Gregory, 1987, p. 221). This captures the notion of internal pleasure at the
knowledge of an object of contemplation, and more: a need to identify with it,
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and a sense of discomfort if that object feels off. In art and architecture, writes
Gregory, “It has been suggested that when we look at the columns of buildings,
such as Greek temples, we identify ourselves with the columns: if they are very
thin, we feel uncomfortable as though they, like us, must be inadequate to take
the weight they support; and if they are very fat, we feel uncomfortable as some
people do when large and clumsy” (Gregory, 1987, p. 220). In the counseling
setting it is the pull of altruism and the harmonic resonance of empathy that
makes us want to feel congruent with the client. When the client feels “off”--
when her columns, as it were, won’t hold up the weight of life-we feel
uncomfortable and want to restore balance and harmony to the chaos and
discord we sense. Counselors can identify these sensations as instances of
counter transference, or projective identification, used qua information in the
counseling setting with regard to the mental state of the client.
Indeed, this very capacity can be used on oneself, as informed self-
reflective consciousness, and constitutes a sort of utilitarian validation of the
psychological explanation for an empathic hunch. Using Freudian theory as his
example, Nagel writes:
For most of those who believe in the reality of repression and the
unconscious, whether or not they have gone through psychoanalysis,
the belief is based not on blind trust in the authority of analyst and their
clinical observations but on the evident usefulness of a rudimentary
Freudian outlook in understanding ourselves and other people,
particularly erotic life, family dramas, and what Freud called the
psychopathology of everyday life. Things that would otherwise surprise
us do not. Behavior or feelings that would otherwise seem simply
irrational become nevertheless comprehensible. You feel miserable all
day and then discover that it is the forgotten anniversary of the death of
someone who was important to you. You find yourself repeatedly
becoming absurdly angry with certain women in your professional life,
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and come to realize that your anger is a throwback to a childhood
struggle with your mother. In the end, if we are to believe that Freud
was getting at the truth, we must be able in some degree to make use of
his approach ourselves. Since controlled and reproducible experiments
are impracticable here, the kind of internal understanding characteristic
of psychoanalysis must rely on the dispersed but cumulative
confirmation in life that supports more familiar psychological
judgments.... So I regard as fatally benighted the epistemological
position taken by Frederic Crews.... He treats psychoanalysis as if it
were a free-standing explanatory system [like chemistry], rather than an
extension of familiar forms of psychological insight to new domains and
new phenomena. (Nagel, 1994, p. 35)
Counseling empathy and mature empathy derive from ordinary
experience. Part of that experience, in turn, is derived from a biological
predisposition for emotional resonance. Babies, for instance, will cry upon
hearing another baby cry (Zahn-Wexler & Radke-Yarrow, 1979)--somewhat like
strings of an instrument humming in sympathetic vibration when a harmonically
related chord is played. Researchers have called this “ emotional empathy”
(Barrett-Lennard, 1981), in that it is a precursor to cognitive empathy and often
inspires it, much as affect in general precedes cognition (Zajonc, 1980).
However ‘emotional empathy’ is somewhat of a misnomer, as this biological
capacity constitutes a form of unconscious nonverbal communication, a way to
gain information about another’s emotions, and is not necessarily a benign form
of intelligence, as the word empathy has come to mean in the ordinary
language sense.
Emotional resonance is a capacity that is conserved across the
Mammalia for the nurturance of the young in the service of survival in a social
species. It is hardwired, that is, occurs outside awareness. Psychologist
Martin Hoffman (1998) has described three mechanisms that are known to elicit
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
emotional resonance: the sympathetic cry of a newborn in resonance to
another baby’s cry; facial mimicry, where one automatically assumes the
posture and facial expression of someone else’s sadness, fear, or delight; and
classical conditioning, where another’s experience is associated with a similar
experience of one’s own, eliciting the emotion that was present at the time.
Something closer to mature empathy is evoked by mediated association, when
one hears about someone else’s fortune or misfortune, and has an emotional
reaction based on having had a similar experience. Role taking, or cognitive
empathy, can be of two types: (1) self-focused role taking, where one uses
one’s own experience to imagine what another person is experiencing, using
mediated association, and (2) counseling empathy, or other-focused role taking,
where mediated association plus knowledge of another’s history, as well as the
mechanisms of interior life, come together to form a picture of another’s
experience.
But consider that emotional resonance, as a form of knowledge
acquisition, could be used in the service of competition as well as when an
attorney or athlete senses discouragement in an opponent and uses that
information to gain an advantage; or when a sociopath senses vulnerability in a
victim. It is altruism, a sense of connectedness, which guides one towards
beneficent intentions, if not helpful interventions. Therefore we have chosen
altruism, rather than emotional resonance, for the x-axis, on the theory that
emotional ‘empathy’ is really a form of arousal (Chlopan et al, 1985) and
underlies all forms of interpersonal interaction, much like the binary computer
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
language of DOS underlies the more analogic language of Windows. For this
binary language of emotional resonance is based on the binary pleasure
principle, and can elicit understanding, aversion, predation, fear, or concern.
We shall see later that this has consequences for empathic development.
From a developmental point of view (Hoffman, 1975), emotional
resonance starts as a part of the maternal matrix, where the infant cannot
distinguish self from other and cries when another baby cries. After the first
year, babies begin to perceive their sensorium as separate from that of others
but still get upset and seek maternal comfort when another child is distressed.
In normal development, as the child gets older, altruism inspires her to offer
help to the distressed child, using self-soothing strategies that may have proved
useful for herself, such as offering the use of a comforting teddy bear. By the
end of the second year, a child begins to grasp that everyone has his own
internal reality. The emergence of a separate identity coincides with more
cognitively complex empathic strategies in response to the emotions of another.
These strategies appear to be learned through experience. For instance,
conduct-disordered children, who often mature into antisocial adults, do not
suffer from physiological underarousal, as was hypothesized (Zahn-Waxler,
Cole, Welsh, & Fox, 1995). They show just as much endogenous arousal and
emotional resonance as normal children in response to another’s distress.
However they do not remain engaged with the other person, and when exposed
to forced-choice situations involving another’s distress, their responses lack
altruism and are more antisocial.
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Goleman (1995) describes this process in his discourse on empathy in
the context of a child’s attempt to sooth his distressed sibling:
Of course, as every parent of young children knows, Jay’s display of
empathy and soothing is by no means universal. It is perhaps as likely
that a child his age will see a sibling’s upset as a chance for vengeance,
and so do whatever it takes to make the upset even worse. The same
skills can be used to tease or torment a sibling. But even that mean
spiritedness bespeaks the emergence of a crucial emotional aptitude:
the ability to know another’s feelings and to act in a way that further
shapes those feelings. Being able to manage emotions in someone
else is the core of the art of handling relationships. (Goleman, 1995,
p. 112)
It may be that to possess a concept of mental states--as an attribute of
others or oneself-is a capacity unique to humans that emerges
developmentally. While chimpanzees, our genetic close kin, may recognize
themselves in mirrors (Gallup, 1995), displaying a rudimentary kinesthetic body
awareness, they do not entertain a self-concept, according to biological
anthropologist Daniel Previnelli (1998):
Consider the simple act of seeing. When we witness other people
turning their eyes toward a particular object, we automatically interpret
this behavior in terms of their underlying psychological states--what they
are attending to, what they are thinking about, what they intend to do
next. These inferences are often solely based on fairly subtle
movements of their eyes and heads. Do chimpanzees understand
seeing in this manner?... A chimpanzee who follows your gaze leads
you to assume that the animal is trying to figure out what you are
looking at. But what excludes the possibility that evolution has simply
produced “mind-blind” mechanisms that lead social primates to look
where other animals look, without entertaining any ideas about their
visual perspective?...
Despite the fact that many of our chimpanzees had displayed evidence
of self-recognition for more than four years, we had no evidence that
they genuinely understand one of the most basic empathic aspects of
human intelligence: the understanding that others see....
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Other studies in our laboratory ... have suggested that chimpanzees
may not understand any behavior in a psychological manner. For
example, careful tests revealed that our apes do not comprehend
pointing gestures as referential actions, nor do they understand the
difference between accidental and intentional behavior....
Our research suggests that only one primate lineage-the human
one-evolved the unique cognitive specialization that enables us to
represent explicitly our own psychological states and those of others.
But in evolving this specialization, we did not discard our array of basic
primate behaviors. Our new awareness of the mental dimension of
behavior was woven into our existing neural circuitry, forever altering our
understanding of our own behavior and the behavior of those around us.
(Previnelli, 1999, p. 67, 72-75.)
A child’s normal empathic development can be arrested or distorted,
usually by maladaptive interpersonal interaction. If antisocial attitudes and
behaviors are mostly learned, one might hypothesize that parental empathic
understanding was denied as when a mother suffers from major depression; or
that punishment was arbitrary or severe; or that outright abuse occurred; or that
the social and peer environment was especially hostile. In narcissistic states,
empathic development is thought be arrested due to the absence of parental
mirroring, where the caretaker acknowledges and responds to emotions in a
benignly empathic manner, neither overwhelmed nor unmoved by the child's
affect (Kohut, 1978; Miller, 1981). In co-dependent situations, however, too
much emotional resonance can blur the distinction between self and other
(Hoffman, 1989; Zahn-Wexler, 1979), resulting in a person’s being
overwhelmed by another’s emotions, or mistaking her own resonant distress for
guilt and therefore blaming herself unjustly for another’s misfortune or
misdeeds.
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In any instance where the empathic response arises, in daily life as in
the counseling setting, it has been suggested that there is also a multistage
process that recapitulates within a matter of seconds or minutes the
developmental stages that occur across the lifespan over months and years,
much as ontogeny is said to roughly recapitulate phylogeny. Carl Rogers
(1975) proposed two: first, sensing and imagining what it must be like to be in
the client’s shoes, then communicating that experience to the client. It is this
form of communication, he posited, that is the core mechanism of therapy, in
that the client’s feeling understood is the beginning of his healing. Indeed, it
appears that it is the client’s perception of counselor empathy, the feeling of
being known, that is most related to positive counseling outcomes (Barrett-
Leonard; Gladstein, 1987). It is worth describing in Rogers’ own words, which
convey with Proustian sensitivity, the intricate process of counseling empathy:
It means entering the private perceptual world of the other and
becoming thoroughly at home in it. It involves being sensitive, moment
to moment, to the changing felt meanings which flow in this other
person, to the fear or rage or tenderness or confusion or whatever, that
he/she is experiencing. It means temporarily living in his/hr life, moving
about in it delicately without making judgments, sensing meanings of
which he/she is scarcely aware, but not trying to uncover feelings of
which the person is totally unaware, since this would be too threatening.
It includes communicating your sensings of his/her world as you look
with fresh and unfrightened eyes at elements of which the individual is
fearful. (Rogers, 1975, p. 4)
Kohut (1978) advanced these sentiments when he stated, dramatically,
that empathy “ constitutes a powerful psychological bond between individuals
that-more perhaps than love, the depression and sublimation of the sexual
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
drive--counteracts man’s destructiveness against his fellows” (p. 705). The
therapist
uses his sensory impressions, of course, as he hears the analyst’s
words and observes his gestures and movements, but these sensory
data would remain meaningless were it not for his ability to recognize
complex psychological configurations that only empathy, the human
echo to human experience, can provide. (Kohut, 1978, p. 700)
Psychoanalyst Stewart (1956) described a more detailed 4-stage
process which the therapist may experience: raw identification (emotional
resonance), deliberate identification (cognitive empathy), resistance (taking
perspective), and de-identification (clinical judgment). A 5-stage process, which
describes the counselor/client empathic dialogue, was proposed by Barrett-
Lennard (1981): empathic set (situation), empathic resonation, expressed
empathy, received empathy, and feedback. Both of these acknowledge that
some sort of unconscious process precedes the more conscious empathic
stance.
One could hypothesize that the stage process of empathy, like its
developmental cousin, could be sideswiped or distorted by various states. For
instance, stress, busyness, fear, and so on, could short circuit an emerging
empathic set. And, too, there are limits to attempts at imagining what it is like to
be someone, without making inferential errors. Philosopher Douglas Hofstadter
(1981) warns that
There is something very beguiling about this concept of mapping,
projection, identification, empathy-whatever you want to call it. It is a
very human trait, practically irresistible. Yet it can lead us down very
strange conceptual pathways, (p. 404)
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
As a delicate process, empathy may not be robust to the petty hazards
of living, nor to the dangers of “over facile self-projection,” as Hofstadter puts it
(p. 404). However it serves, in daily life as well as in psychotherapy, to give us
some measure of ongoing insight into self and other within the social context.
If we return to proposed circumplex model of empathy, we see the
addition of three common deviations from the empathic stance-narcissism,
psychopathy, and codependence--each with emotional tones and behaviors
that imply a lack, one might hazard, of having been empathized with, along the
way. This is perhaps because, developmentally, we do not first gain a sense of
self, and then generalize to the existence of other selves; but rather, as the
philosopher Edith Stein (1917) has suggested, we learn that we are a
subjectivity, a self, by seeing it first in others. It is then that we learn that we,
too, are the subject, rather than an object, of qualitative experience.
The psychologist Julian Jaynes (1976) sees it this way, too, when he
says that “ the tradition in philosophy that phrases the problem [of
consciousness] as the logic of inferring other minds from one’s own has it the
wrong way around. We have first unconsciously (sic) suppose other
consciousnesses, then infer our own by generalization” (p. 217).
Otherwise, why would mirroring be so important for the child (Miller,
1981), and cognitive empathy so healing in therapy (Kohut, 1978; Rogers,
1975)? And why would its absence, by some lack of altruism or complexity,
manifest in the pathologies we hypothesize in the other three quadrants of the
circumplex? Empathy of the parent helps the child organize his sense of self
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
around the chaos of outer perception and inner sensation. Empathy of others
during our lifespan helps us continue to grow.
When we hear the stories others have to tell about their experience,
when we note the facial expression or gesture that may contradict this story, we
learn something about what might be possible or impossible for our own self
development. As Edith Stein notes:
It is possible for another to ‘ judge me more accurately’ than I judge
myself and give me clarity about myself. For example, he notices that I
look around me for approval as I show kindness, while I myself think I
am acting out of pure generosity. This is how empathy and inner
perception work hand in hand to give me myself to myself. (Stein, 1917,
p. 89)
Sidebar on Measurement
It appears that, for measurement purposes, both altruism and cognitive
complexity are half-attribute, genetically based like intelligence; and half
attitude, learned though socialization. In addition, altruism is so prized that it is
subject to social desirability response sets-we tend to answer in ways we think
will make us look altruistic.
Measurement is in the stimulus, then person tradition (Himmelfarb,
1993), where a stimulus (statements of affect, behavior, or belief) elicits a
scaled response on the altruism dimension, the total of which responses then
places the respondent somewhere on an attitude continuum--e.g., a high or low
A. Rather than fixed as in psychophysical measurement that assumes that
every person is alike anatomically, each stimulus statement about a
hypothetical construct produces a normal distribution of endorsements to
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
account for individual differences and variability over time. "The distribution
arises," says Himmlefarb (1993), "from the fact that the same statement may
elicit somewhat different degrees of judged favorableness in different
individuals or in the same person, from one occasion to another" (p. 33).
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 3
METHOD
Participants
A convenience sample was acquired by (a) direct mail solicitation, (b)
access to a formal undergraduate subjects pool sponsored by the Department
of Psychology at a major state university, and (c) access to a formal
undergraduate subjects pool sponsored by the Department of Psychology at a
major private Catholic university.
Protocol
Direct Mail: One thousand questionnaires, along with self-addressed
stamped envelopes, were sent to acquaintances, colleagues, and students of
colleagues. Most of these subjects were from the Southern California area,
with a smattering from the Northwest, Midwest, and the Northeast.
There were 170 responses, of which 161 were sufficiently intact to
include in data processing. That is, nine respondents omitted more items than
reasonably could be substituted for, or else they failed to complete the entire
Cognitive Complexity (CC) measure, and were deleted from the sample.
The purpose of the direct-mail campaign was to increase overall sample
size, as well as increase the variability of the sample with regard to age,
ethnicity, and years of education.
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
State University. One hundred thirty-five undergraduate psychology
students at a major state university volunteered to participate in this study to
fulfill the research requirement imposed by the Department of Psychology.
Each cohort of volunteers was asked to read and sign a disclaimer stating that
the study was confidential, that they could stop taking the test at any time, that
the questionnaire was potentially challenging in that subjects are asked
sensitive questions which might evoke negative affect, and that should anyone
have subsequent emotional difficulties, several counseling centers were
available to help them. One hundred twenty-three students completed or
sufficiently completed the questionnaire.
Private Catholic University Ninety-two undergraduate psychology
students at a major private Catholic university volunteered to participate in this
study to fulfill the research requirement imposed by the Department of
Psychology. Students were similarly advised of their right to confidentiality,
right of refusal, and notified of the availability of counseling centers. Eighty-
seven subjects sufficiently completed the questionnaire and the responses
were retained in the final study sample.
Description of the Sample
The results of this study are based on a final sample of 344 subjects
after data were entered and subjects with excessive missing data were deleted.
Initially the two college locations were collapsed into a single demographic
group on the theory that they would be similar to one another with regard to
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
age, years of education, living arrangements, and so on. However it transpired
that students from a large state university are different in interesting ways from
students at a private Catholic university, and so the two college groups were
separated for location comparisons (Tables 1 and 2).
Subjects in the full sample included 134 males and 210 females. Sixty-
eight percent were single and the rest partnered. Significantly more of the
direct mail respondents lived with a partner [ x2 (2) = 59.74, p <.001 ]; 95% were
heterosexual; 210, or about two thirds, were college students; 180 were
unemployed, but 162 had full-time or part-time employment. With regard to
ethnicity, 66 were Asian, 171 were of European descent, 24 were African
Americans, 39 were of Hispanic origin, 10 were Middle Eastern, 6 were Native
Americans, and 28 considered themselves of mixed origin. Politically, 16%
were conservatives, 48% were moderates, and 39% were liberals. With regard
to spiritual orientation, 37% considered themselves religious and observant,
28% occasionally observant, 3% preferred New Age approaches, 7% endorsed
meditation, 4% preferred science to religion, and 15% said they endorsed no
religious orientation at all.
After an examination of the frequency distributions across demographic
variables, two of the categories were dichotomized: “Relationship” was
redefined as “living with someone” (married or partnered) or “living alone”
(single or widowed). “Sexual Orientation” was dichotomized into “heterosexual”
and “homosexual” (gay, lesbian or bisexual).
37
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 1
Demographic Characteristics by Sample Location
Characteristics
Direct Mail
(n = 134)
f %
Location
State Universitv
(n = 123)
f %
Private Universitv
(n = 87)
f %
x !
Gender
Male 54 43.3 49 39.8 3 1 35.6
0.5
Female 80 59.7 74 60.2 56 64.4
Age Group
Younger 24 18.8 114 96.6 83 96.5
215.7**
Middle 38 29.7 4 3.4 2 2.3
Older 66 51.6 0 0 1 1.2
Relationship
Single 50 37.3 104 84.6 79 90.8
144.9***
Married 62 46.3 3 2.4 1 1.1
Partnered 6 4.5 16 13.0 7 8.0
Widowed 16 11.9 0 0.0 0 0.0
Sexual Orientation
Heterosexual 123 92.5 121 98.4 84 96.6
10.2*
Gay 7 5.3 1 0.8 0 0.0
Bi-sexual 3 2.3 1 0.8 3 3.4
Ethnicity
Asian 12 9.0 43 35.0 1 1 12.6
87.4***
European 83 61.9 37 30.1 5 1 58.6
Black 19 14.2 2 1.6 3 3.4
Hispanic 5 3.7 24 19.5 10 11.5
Middle Eastern 1 0.7 8 6.5 1 1.1
Native 5 3.7 1 0.8 0 0.0
Mixed 9 6.7 8 6.5 1 1 12.6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 1 (continued)
Demographic Characteristics by Sample Location
Characteristics
Direct Mail
(n = 134)
F %
Location
State Universitv
(n = 123)
£ %
Private Universitv
(n = 87)
f %
x!
Spirituality
Religious 33 24.8 54 43.9 4 1 47.1
53.6***
Occasionally 35 26.3 36 29.3 28 32.2
New age 9 6.8 1 0.8 1 1 .1
Meditation 2 1 15.8 2 1.6 1 1 .1
Preferred science 6 4.5 6 4.9 2 2.3
None 18 13.5 22 17.9 13 14.9
Multiple 1 1 8.3 2 1.6 1 1 .1
Political Orientation
Conservative 20 15.6 13 10.7 20 23.0
12.5*
Moderate 52 40.6 7 1 58.2 37 42.5
Liberal 56 43.8 38 31.1 30 34.5
Employment Status
Unemployed 40 29.9 88 72.1 52 60.5
48.6***
Employed 94 70.1 34 27.9 34 39.5
* £ < .05 ** £ < .01 *** £ < .001
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2
Continuous Demographics bv Sample Location
Location
Direct Mail State Universitv Private Universitv
(n = 134) (n = 123) (n = 87)
Characteristics
M
SD
M
SD
M
SD
Age 45.7 19.0 19.8 2.9 19.5 3.5
Years of Education 16.4 3.4 13.7 1.2 13.5 0.96
Frequency of Drug Use3 7.6 6.3 7.3 6.4 9.7 7.2 4.0*
* E < .05 *** E < -001
3 7-point Likert scale (1 = never; 7 = more than once a day) summed for cigarettes,
alcohol, and illicit drugs
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
List of Measures
The study entailed the use of self-report questionnaires. For the
purposes of this study, each subject was asked to complete a demographic
section followed by eleven previously published instruments selected to
address different conceptual domains:
1. Helping Orientation Questionnaire (HOQ), with four subscales:
Altruistic, or selfless helping: Receptive Giving, or a quid pro quo
style of helping; Selfish, a preference for receiving help rather than
giving it; and Inner Sustaining, an absence of either seeking or
offering help.
2. Four-Item F Scale (Authoritarianism).
3. Codependent Questionnaire (CdQ), a full scale with four subscales:
Control, the need to influence/control self or other; Responsibility,
meeting others’ needs rather than one’s own; Intimacy, boundary
ambiguity between self and other; and Enmeshment, or over
involvement with others’ problems.
4. A 4 x 6 Repertory Grid (Cognitive Complexity).
5. Interpersonal Reactivity Index (IRP), with four subscales:
Perspective Taking, or Cognitive Empathy; Empathic Concern, or
Emotional Empathy; Fantasy, or the capacity to imagine fictitious
scenarios; and Personal Distress, sympathetic arousal at another’s
misfortune.
6. Love Attitudes Scale (LAS), with seven subscales.
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
7. Marlowe-Crowne (control for socially desirable response set).
8. Narcissistic Personality Inventory (NPO.
9. Primary Psychopathy Scale (PPS). a measure of subclinical
antisocial tendencies.
10. Splitting Index (SI), measuring the defense of splitting one’s
evaluation of self and other into polar opposites.
11. Drug Use Frequency
In addition to the instruments named above, the two college groups were
administered the following additional instruments; these, however, were not
included in the study, but are named for reference purposes.
12. Questionnaire for Eating Disorder Diagnoses (Q-EDD).
13. Interpersonal Guilt Questionnaire.
14. Collins & Reed Attachment Questionnaire.
15. Locke Marital Adjustment Test— Revised.
16. Strauss Conflict Tactics Scale.
17. Dyadic Adjustment Scale.
18. Psychological Maltreatment of Women Inventory.
19. Alexithvmia Scale.
20. Male Role Norms Scale-
21 . Coercive Sex Scale.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Instrumentation and Sample Reliabilities
The Helping-Orientation Questionnaire (Romer, Gruder & Lizzadro,
1986) consists of 23 categorical items designed to pull four response
alternatives to situations that could elicit helping behaviors: Altruistic
(preference for the inner rewards of ethical and emotional satisfaction, without
expectation of reciprocity or external rewards such as gratitude or material
gain); Receptive Giving (quid pro quo stance, with the expectation of
reciprocity), Selfish (helping only if easy or if external rewards are lavish;
expecting others to help instead), and Inner Sustaining (detached; not
interested in helping or being helped).
Response alternatives are varied so as to avoid a response set.
Subjects are assigned to the orientation for which their frequency score is
highest; most subjects generate a score on each scale, though most have their
highest score on Altruistic. Internal consistency for the HOQ was achieved by
coding each item for the socially desirable Altruistic response (1 = yes, 0 = no),
yielding a coefficient alpha of .56, suggesting an acceptable level of reliability
for the measure. Table 3 presents the internal consistency reliability for study
measures and Table 4 presents the summary statistics for study measures.
Reexamination of the internal consistency of the sample measure
showed that only the Altruistic measure was sufficiently stable for use (alpha =
.63). Therefore we ran a factor analysis on the remaining three subscales and
selected items with the factor loadings above .4 to create new subscales.
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 3
Internal Consistency Reliability for Study Measures fn = 344)
Cronbach’s Alpha
Dependent Measures
Splitting Index 0.79
Psychoticism 0.85
Narcissism 0.60
Codependent Scales:
Responsibility 0.66
Control 0.59
Enmeshment 0.52
Intimacy 0.66
Interpersonal Reactivity Indices:
Empathic Concern 0.70
Fantasy 0.75
Personal Distress 0.74
Perspective-taking 0.74
Independent Measures
Altruism 0.63
Authoritarianism 0.46
Control Measure
Crowne Marlowe Social Desirability 0.80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4
Summary Statistics for Study Measures
M S D
M edian M ode
SK/SE* K U /S E *
Dependent Measures
Splitting Index 28.19 7.40 28 22 4.08 0.36
Psychoticism 28.57 7.73 28 22 6.10 3.22
Narcissism 5.15 2.44 5 5 1.15 -1.45
Helping Orientation Factors:
Receptive 2.23 1.36 2 2 1.97 -1.35
Selfish 0.13 0.48 0 0 37.53 115.45
Inner Sustaining 0.38 0.76 0 0 19.10 26.23
Codependent Scales:
Responsibility 24.40 4.85 24 25 3.59 3.08
Control 29.84 4.82 30 30 0.88 1.02
Enmeshment 20.13 3.61 20 19 3.87 2.64
Intimacy 27.54 5.11 27 26 1.39 0.32
Interpersonal Reactivity
Indices:
Empathic Concern 20.08 4.06 20 19 -5.04 4.68
Fantasy 16.59 5.34 17 12 -0.42 -1.52
Personal Distress 10.95 4.81 1 1 10 1.75 1.47
Perspective-taking 16.87 4.58 17 18 -0.61 -0.33
Independent Measures
Cognitive Complexity 21.83 8.79 20 16 7.03 3.08
Altruism 10.65 3.35 1 1 10 -1.35 0.57
Authoritarianism 1.47 1.19 1 1 3.34 -2.72
Control Measure
M-C Social Desirability 13.88 5.63 14 14 1.61 -1.21
* Skewness and kurtosis divided by their standard errors
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Of these, only the new 6-item Receptive Giving subscale achieved
adequate reliability (alpha = .46), and we dropped the other two measures from
the study.
The Four-Item F Scale (Lane, 1955) is a brief test of authoritarianism
based on a large cross sectional national sample. Authoritarians, Lane found,
are less tolerant of ambiguity--a marker of cognitive simplicity. Democrats,
independent of education, tended to be slightly less authoritarian than
Republicans. The original instrument had an “agree” or “disagree” forced
choice format which we recoded as “1” or “0;” scores ranged from 4 to 0.
Higher scores showed higher levels of authoritarianism. In the original study the
coefficient for the 4-item Guttman scale was 90.4, suggesting a decent level of
reliability for a unidimensional measure. A reassessment of the sample
measure yielded an alpha coefficient of .46.
The Codependent Questionnaire (Roehling & Gaumond, 1996) is a 36-
item measure rated on a Likert-type scale from 1 (“I never feel this way”) to 5 (“I
always feel this way”), yielding a full scale score and scores on four subscales:
Control (need to influence/control feelings of self and other, even if
maladaptive); Responsibility (meeting others’ needs to the exclusion of one’s
own); Intimacy (difficulty with boundaries); and Enmeshment (overly involved,
especially with troubled individuals). To control for bias, four items were
reversed scored.
Discriminant validity was shown by Roehling and Gaumond, comparing
the CdQ and scores on the Millon Clinical Multiaxial Inventory (MCMI).
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Codependent characteristics are negatively correlated with acting-out type
personality disorders such as narcissistic and antisocial, or psychopathic.
Strong convergent validity included correlations with borderline avoidant and
passive-aggressive personalities. Alpha coefficients of .86 (total) reveal good
reliability (.77 Responsibility, .65 Control, .66 Enmeshment, and .67 Intimacy).
Test-retest reliability for the total score was r = .80).
For this sample, the full CdQ achieved a stable alpha coefficient of .79.
Subscale alphas proved to be sturdy as well: Responsibility = .66; Intimacy =
.66; Enmeshment = .52; and Control = .59. As the sample CdQ subscales
achieved adequate stability, we used the subscales, rather than the full scale,
to achieve greater refinement in the analysis.
A 4 x 6 Repertory Grid (Spengler & Strohmer, 1994) is designed to
measure Cognitive Complexity. This measure was derived from the 10 x 10
repertory grid originally developed by Bieri, Atkins, Briar, Leaman, and Tripodi
(1966), shortened to provide researchers with a smaller, less time-consuming
instrument. “Cognitive complexity,” note Bieri et al., “refers to the relative
number of dimensions an individual uses in construing events, objects, or
persons” (Bieri, et al., 1966, p. 263). In other words, complexity entails
multidimensional judgments, the use of the entire spectrum of grays on a black-
white continuum; an object is seen, for instance, to have both good and bad
attributes. Simplicity, on the other hand, consists in polarizing judgments or
focusing a single, unidimensional aspect; a clinical example would be the failure
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to see the possibility of co-morbidity in a client profile (Reiss, Levitan &
Szyszko, 1982; called diagnostic overshadowing).
The test consists of four roles (mother, friend of the opposite sex,
person with whom you feel most uncomfortable, and boss or supervisor). Each
role is assigned six column constructs on a continuum of six row valences:
outgoing/shy, adjusted/maladjusted, decisive/indecisive; calm/excitable,
interested in others/self absorbed, and cheerful/ill humored. Scoring is
accomplished by summing the valences column-wise and calculating
redundancy by means of a binomial equation. The results of that equation are
summed row-wise for each role, then all four roles scores are summed column
wise for the total CC score. Because of this scoring difficulty, subjects who did
not complete the entire measure were dropped from the study. Scores range
from 0 to 60, with low scores representing less redundancy and a higher level
of cognitive complexity; higher scores reflected cognitive simplicity.
Complexity scores from Bieri’s (1955) original 10 x 10 matrix were cross
validated with scores from the 4 x 6 grid, yielding a correlation of r= .89,
indicating that the two grids provide comparable measures. A one-week test-
retest yielded a reliability correlation of .82, suggesting adequate stability. The
authors note, however, that this grid may provide only one aspect of cognitive
complexity (multidimensionality). Other aspects of complexity, such as the
capacity for hierarchical organization, were not assessed.
For the sample measure, because of the complex scoring, we had to
rely on the original test-retest reliability correlation of .82.
48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Interpersonal Reactivity Index (IRI) (Davis, 1983). This is a 28-item
instrument scored on a 5-point Likert-type scale from 1 = does not describe me
well, to 5 = describes me very well. Nine items were reversed in order to
moderate a possible response bias.
The IRI seeks to measure four aspects of the construct empathy:
perspective taking, or cognitive empathy, the ability to see situations from
another’s point of view; empathic concern, or other-oriented emotional
empathy; fantasy, or imagining hypothetical events; and personal distress, or
self-oriented sympathetic arousal due to another’s distress. Respondents are
assigned to the subscale on which they score highest; most receive a score on
each of the four subscales. There is no full-scale score for the following
reason: Perspective-taking correlates with Empathic Concern but is negatively
related to Personal Distress; high perspective-taking/low distress subjects
would score equally with high distress/low perspective-taking subjects, even
though the two tendencies are opposites.
Convergent validity was assessed by comparing subscales with the
Hogan Empathy Scale (Hogan, 1969), a measure of cognitive empathy; and
with the Mehrabian Epstein Emotional Empathy Scale (Mehrabian & Epstein,
1972). As predicted, perspective taking correlated with the Hogan (r = .42 for
males, r = .37 for females); and empathic concern related to the Mehrabian (r =
.63 for males, r = .56 for males). In the subscale intercorrelations, personal
distress, as expected, was negatively correlated with perspective taking.
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Stability was assessed with standardized alpha coefficients for each
scale: fantasy (.78 for males, .75 for females); perspective taking (.75 for
males, .78 for females); empathic concern (.72 for males, .70 for females); and
personal distress (.78 for males and females) showing that each scale provided
a reliable measure. Test-retest (60 - 75 days) correlations ranged from .61 to
.79 for males, and .62 to .81 for females, offering evidence for temporal
reliability.
The sample subscales achieved adequate stability in a reliability
analysis: Perspective Taking, Alpha = .74; Empathic Concern, Alpha = .70;
Fantasy, Alpha = .75; and Personal Distress, Alpha = .74.
Love Attitude Scales (LAS) (Hendrick & Hendrick, 1986) is a 42 item
instrument on a five-point Likert-type scale from 1 = strongly agree with the
statement, to 5 = strongly disagree with the statement. Low scores signify
greater endorsement; only subscales are used.
The LAS measures attitudes on six styles of love, which form the
subscales: Eros (passionate love); Ludus (game-playing, manipulative love);
Storge (friendship); Pragma (logical “shopping-lisf love); Mania (possessive,
dependent love); and Agape (spiritual or selfless love). Significant differences
were found between men and women on every scale except the Eros and
Agape scales. Internal consistency with alpha coefficients from .74 to .84
suggested good levels of reliability.
For the sake of brevity, we selected 2 items with the highest factor
loadings on each subscale and created a shorter measure of 14 items, in the
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
hopes of reducing subject fatigue. Subsequent analysis revealed that all
subscales were unreliable for this sample, except for Eros, which had no
theoretical relationship to the model. Thus the entire measure was deleted for
the analysis.
Marlowe-Crowne (Crowne & Marlowe, 1964) is a 33 item forced-choice
(T = 0, F = 1) measure of the tendency to respond in socially-desirable ways,
that is, to look good or appear virtuous, independent of psychopathology. It
reveals a self-deceptive component as well as avoidance of disapproval.
Scores range from 0 to 33, with high scorers demonstrating a need for
approval. Fifteen items are reversed to control for a response set
The measure contains items of minor misconduct, neglect or hostility
that all people experience from time to time and are part of normal functioning.
Examples include “Before voting I thoroughly investigate the qualifications of all
the candidates,” or “I have never intensely disliked anyone.” This measure
was included in the battery to control for an expected socially desirable
response set elicited by measures of empathy and altruism. In addition, a low
score was thought to relate positively to cognitive complexity. In the original
study, the mean for college students in anonymous situations was 13.3; college
students in a public disclosure situation = 15.5; depressed students = 12.3;
Peace Corps volunteers male = 16.1 and female = 16.4.
For this sample, the mean response was 13.88; a reliability analysis
revealed a robust Alpha of .80.
51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Narcissistic Personality Inventory (Raskin & Hall, 1979; Raskin & Terry,
1988) is a 40-item forced-choice measure of the characteristics of narcissism:
an inability to decenter and attend to another’s feelings or agenda, but rather,
the tendency to exploit others in order to shore up a false, idealized sense of
self. Of two possible answers to each item (A or B), respondents are asked to
choose one, which are then recoded as A = 1, B = 2. Seven items were
reversed to avoid response bias. The measure yielded a full-scale score and
subscale scores; we used the full scale score, which had a Guttman lambda
three estimate of internal consistency of .83, demonstrating good reliability.
For the sake of brevity, we chose 2 items on each of 7 subscales with
the highest factor loadings, creating a smaller measure of 14 items, thus
altering the structure of the original measure. Subsequent reliability analysis of
the sample measure, after deleting one item to improve the reliability
coefficient, yielded a new 13-item measure of adequate stability, with an Alpha
of .60.
Primary Psychopathy Scale (PPS) (Levenson, Kiehl & Fitzpatrick, 1995)
is a 16-item measure of antisocial tendencies in non-prison populations. It
assesses core elements of coldness, lying, exploitiveness, and lack of foresight.
The PPS focuses on populations that exhibit these tendencies, called “primary
psychopathy” (Karpman, 1948; Levinson, Kiehl, & Fitzpatrick, 1995), without the
extreme impulsivity that marks antisocial behaviors of criminals in the justice
system (“secondary psychopathy”).
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The total scale is scored by summing endorsements to items on a 4-
point Likert-type scale, ranging from 1 = “ disagree strongly” to 4 = “ agree
strongly. Five items are reversed to avoid a response bias. Convergent validity
was established by comparing it with two subscales of the Zuckerman
Sensation Seeking Scale (Zuckerman, 1979), with a correlation of r= .39 on
Boredom Susceptibility and r = .34 on Disinhibition. Discriminant validity was
established by comparing the PPS with more acting-out subscales from the
Zuckerman Scale, with near zero correlations with subscale Thrill and
Adventure Seeking (r = .01), “ thus discriminating [primary] psychopathy from a
global sensation seeking construct.” (Levinson et al, 1995, p. 155) The alpha
coefficient in the original study was .82. A reliability analysis of the current
sample measure replicated the original with a robust Alpha of .85.
The Splitting Index (Gould, Prentice, & Ainslie, 1996), is a 24-item self-
report scale on a Likert-type scale ranging from 1 = “strongly disagree” to 5 =
“ strongly agree.” It is designed to elicit tendencies to adopt the defense
mechanism of splitting: regarding self and others via alternating bipolar
judgments of good or bad, based on the moment’s pleasure or displeasure,
creating identity diffusion (Kernberg, 1975). Convergent validity was revealed
by significant correlations with narcissism, borderline, and self image instability.
Discriminant validity was demonstrated by near zero correlations with two
measures of cognitive complexity: The Intolerance of Ambiguity Scale (Budner,
1962), r = -.7; and the California F Scale (Adorno, Frenkel-Brunswik, Levinson,
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
& Sanford, 1950), measuring authoritarianism, r = -.04. An alpha coefficient of
.90 showed robust stability for the total index.
For the present study, in the interests of brevity we selected the four
highest factor loadings on three subscales (splitting of Self, Family and Others)
to create a 12-item measure. Subsequent reliability analyses revealed
adequate stability of the sample measure with an Alpha of .60.
Drug Use Frequency is a column-wise list of substances ingested during
the last six months, including alcohol, cigarettes, marijuana, cocaine, opiates,
and pharmaceuticals, with row-wise frequencies, from “never” to “more than
once a day.” High scores reflect greater substance use.
Research Design and Analysis
The study relied on a correlational design to determine if the
hypothesized model was accurate.
Hypothesis One: The two independent variables, Cognitive Complexity
(CC) and Altruism (A), were assumed to vary independently of one another;
but in concert, were hypothesized to demonstrate that relatively high levels of
each would correlate with relatively high levels of the key dependent variable
cognitive empathy, measured by the Perspective Taking subscale of the HOQ.
The hypothesis rests on the assumption, as stated above, that CC helps one
assess differences between self and other, that A encourages one to look at
similarities between self and other; and that one needs both capacities in order
to achieve a state of cognitive empathy
54
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Hypothesis Two: In addition, it was theorized that low CC and high A
would correlate with Codependence; high A and low CC would relate to
Psychopathy; and low scores on both A and CC would relate to Narcissism.
The Marlowe-Crowne was added to see if a socially desirable response
set would emerge, hypothetically, on measures of altruism and empathy; that is,
there would be a positive relationship between social desirability and
endorsements of altruism and empathy. Demographic variables such as age
and gender were other hypothesized control variables. The Marlowe-Crowne
was further thought to positively relate to complexity, in that lower scores,
demonstrating an acknowledgment of one’s normal frailties, was a feature of
cognitive complexity.
For the purposes of discriminant or convergent validation, a measure of
the defense mechanism of splitting, as well as a measure of authoritarianism,
were added, on the theory that they would relate to cognitive simplicity. The
LAS was included in that some of its subscales were thought to relate to other
dependent variables: Pragma, or pragmatic love, was theorized to relate to
moderate measures of A and CC, a sort of expedience; Mania, or dependent
love, was thought to relate to the dependent variable codependence; Storge
was expected to relate to cognitive empathy; Ludus, or manipulative love, was
thought to moderately relate to psychopathy.
A subscale of the HOQ--Receptive Giving-was hypothesized to relate to
Pragma, or moderate levels of both A and CC. A subscale of the IRP, Personal
Distress, was thought to relate to codependence. Two subscales of the HOQ--
55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Selfish and Inner Sustaining--were hypothesized to relate to psychopathy. A
measure of drug use frequency was expected to correlate with psychopathy.
Hypothesis 3: Finally, we sought to examine the theory that narcissists
tend to be cognitively simple. According to Emmons, in a reevaluation of the
NPI,
Narcissistic individuals may have simple cognitive representations of
themselves, resulting in their showing more extreme swings in mood
following success or failure....Whether narcissistic individuals actually
do have a simple cognitive representation of themselves will have to be
empirically documented in future research.” (Emmons, 1987, p. 16)
This study attempted to shed some light on that hypothesis.
Data Management
Data were entered by Advanced Data Entry Consultants and logged
onto the Statistical Package for the Social Sciences (SPSS version 6.1 for
Windows). Of the original 376 subjects, 32 were subsequently eliminated from
the sample after it was discovered that these had excessive missing data for
which substitution could not be made. In addition, because of the peculiarity of
its scoring, all items on the Cognitive Complexity measure had to be filled or the
subject was dropped. The final data set included 344 subjects.
Continuous missing data were substituted with the sample item mean selected
by the three demographics judged most variable: age; ethnicity (Asian = 1,
European = 2, African-American = 3, Hispanic = 4, Middle Eastern = 5, Native
American = 6; and Mixed = 7); and gender (male = 1, female = 2). Categorical
missing data were substituted with the sample item mode selected by age,
ethnicity and gender. To reduce the number of algorithms to a manageable
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
number (42 in all) for the purpose of data substitution, the continuous
demographic age was collapsed into three categories: college = 1 (ages
1 - 25), adult - 2 (ages 26 - 45), and mature = 3 (ages 45 - 90). All statistics
were evaluated at the <.05 level of significance.
Reliability of the Sample Measures
Reliability studies were done on all the instruments used in the current
study. This was particularly important in measures that were abbreviated for
this study by selecting three or four items with the highest factor loadings, as
determined by the original researcher, and deleting the rest. This maneuver
then altered the reliability of already established measures. Measures thus
abbreviated, to reduce fatigue effects, were the LAS, the SI, and the NPI.
We could not run reliability studies on the 4 x 6 Cognitive Complexity
measure because of its unusual scoring; we had to rely instead on the original
test-retest numbers.
For the LAS, six subscales with two items each had such low reliabilities
that we deleted them from the study. Only Eros attained sufficient reliability;
however, it was the lone subscale with no pre-study theoretical relationship to
any other measure. Indeed, a preliminary correlation matrix showed that it had
no significant relation to any other measure of interest, it too, was eliminated.
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Statistics and Data Analysis
To test the applicability of the linear model for the data set, preliminary
analysis involved assessments of the distributions for normality for the
dependent and independent variables, using skewness and kurtosis divided by
their respective errors (SK/SE, KU/SE; values within +/- 2.0 are indicative of
normality). On the Helping Orientation Questionnaire, the two refactored
subscales, Selfish (SK/SE = 38) and Inner Sustaining (SK/SE = 26), were so
skewed that we eliminated these subscales from the analysis.
Pearson correlations were run on dependent by dependent measures,
dependent by independent measures, and independent by independent
measures. Independent T-tests were run to determine if the dependent
variables differed by the dichotomous variable gender. Pearson correlations
were used to assess the relationship between the dependent variables and the
continuous control variables age and social desirability.
To further test the main hypothesis that relatively higher levels of
altruism and cognitive complexity would result in higher levels of cognitive
empathy (Perspective Taking), a split-half multivariate comparison (MANCOVA)
was run to assess interactions between Altruism and Cognitive Complexity,
using median splits on Cognitive Complexity (<20, >20), and Altruism (< = 10,
>10). One would expect to get a median split classification of subjects along the
four theoretical subgroups generated by the interaction of Altruism and
Cognitive Complexity: Cognitive Empathy, Psychopathy, Narcissism, and the
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
various subscales of Codependence. Finally, hierarchical regressions were
run, controlling for the effects of age, gender, and social desirability on each of
the dependent variables, using Cognitive Complexity and Altruism as
predictors.
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 4
RESULTS
The results of the present study did not support Hypothesis One: that a
positive synergy of the independent constructs Altruism and Cognitive
Complexity would correlate with Cognitive Empathy. The two independent
variables, Cognitive Complexity and Altruism, were orthogonal, or unrelated, as
expected. However, while Altruism was related to Cognitive Empathy
(Perspective Taking, r = .27, p<.001), Cognitive Complexity was not (Figure 3).
In addition, there were no significant multivariate interactions; the
independent variables taken together could not predict the theoretical
subgroups, as stated in Hypothesis Two. Indeed, in the hierarchical regression,
the control variable, Social Desirability, proved to be a better predictor (Beta =
.24, p <.001) of Perspective Taking than either of the independent variables or
both combined. Finally, the Emmons hypothesis, that narcissists are
cognitively simple, was not supported by this study. Tables 5 through 8 present
the Pearson correlations among measures and variables.
The Independent Variables
The independent variable Cognitive Complexity did not behave, or
rather, relate as predicted. In noting the correlations, it must be remembered
that high scores on CC refer to simplicity, and low scores to complexity. CC as
measured was related to Splitting (r = -.15, p = .01) which, by definition as well
as theory, should not be the case. To split one’s worldview into good or bad,
60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FIGURE 3. Clrcumplex Model: Results
on Cognitive Complexity
Cognitive Complexity +
Control
Splitting Index
.Receptive Giving
Cognitive Complexity -
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 5
Pearson Correlations Among Dependent Measures (n = 344)*
PS Y N AR R E C R E S C O N E N M IN T E C FA P D P T
Splitting Index (SI) .16 - . 1 0 .08 A0
A l
.19
M
-.04 .15 -.13
Psychoticism (PSY) M
.28
. 0 0 . 1 0 - .0 1 M -.49 -.07 .1 1 -.33
Narcissism (NAR) .07 -.15 .0 1 - .0 1 - .1 1 -.09 .08 -.24 -.15
Receptive (REC) - .0 1 .14 .04 .1 1 - . 1 0 .53 M -.18
Responsibility (RES)
A8
M
M
J 5
M
.09
Control (CON) A0
M
.17
M
.23 . 0 0
Enmeshment (ENM)
H
.26 .18 .19 . 1 0
Intimacy (INT) .04 M .44 - . 1 0
Empathic Concern (EC)
M
-.04 A5
Fantasy (FA) .13 .16
Personal Distress (PD) -.22
Perspective-taking (PT)
* coefficients in bold type are significant at the .05 level
coefficients in bold italic type are significant at the .0 1 level
coefficients in bold italic underlined type are significant at the .0 0 1 level
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 6
Pearson Correlations Among Independent Measures
Altruism Authoritarianism
Cognitive Complexity - . 0 2 . 1 0
Altruism -.18**
** B < .01
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 7
Pearson Correlations Between Dependent and Independent Measures
Cognitive Complexity Altruism Authoritarianism
Splitting Index -.15** -.17** . 0 0
Psychoticism .04 -.40*** .32***
Narcissism -.07 -.07 .16**
Receptive -.1 1 *
_ 52***
.07
Responsibility -.09 .08 . 0 0
Control -.16** -.08 .07
Enmeshment -.08 .13* -.04
Intimacy - . 1 0 -.15** .13*
Empathic Concern -.04 .33*** -.24***
Fantasy - . 1 0 .07 . 0 2
Personal Distress . 0 0 -.19** .1 2 *
Perspective-taking -.05 .27*** -.16**
*£<.05 **£<.01 *** £ < .001
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 8
Pearson Correlations Between Dependent Measures and Control Variables
Contro! Variables
Social Desirability Age
Drug Use
Deoendent Measures
Splitting Index - 24 ***
- 2 o ***
.1 1
Psychoticism -.23 *** -.30 ***
2 2 ***
Narcissism -.14** -.09
* j g * * *
Receptive - 2 0 *** -.38 *** .05
Responsibility -.13 * -.19 *** -.06
Control -.28 *** - 24 *** -.03
Enmeshment -.07
_ 2 5 ***
. 0 2
Intimacy - . 2 0 ***
- 2 9 ***
- .0 1
Empathic Concern .15** . 0 0 -.1 2 *
Fantasy -.15 ** ~ 2 1 *** .0 1
Personal Distress
. 24 ***
-.23 *** -.08
Perspective-taking .28 *** .1 1 * - .1 1 *
Independent
Measures
Cognitive
Complexity
. 1 0 .17 ** - .1 1 *
Altruism .28 *** .1 2 * -.03
Authoritarianism .06 -.08 -.04
* E < .05 **£< .01 *** E < -001
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
black or white, is a hallmark of cognitive simplicity. Moreover, CC was also
related to a codependent subscale, the neurotic need for Control (r = -.16, p =
.001). This, too, was a bit puzzling, as it implies a need to make things a
certain way, a form of construct assignment inflexibility. The third correlation
among dependent variables was with Receptive Giving (r = -.11, p = .05.
Receptive Giving constitutes a quid pro quo stance and is one of the HOQ
subscales, the realm of everyday give-and-take transactions; this, however, is
in the direction we surmised, as the world of business should relate to
complexity rather than simplicity. Finally, cognitive simplicity was not
significantly correlated, as expected, with Authoritarianism; the latter often is
used, negatively, as a marker of Cognitive Complexity, and used in this study
for the purpose of convergent validity.
Complexity did not relate to social desirability. It was expected to
negatively relate, as knowledge and acknowledgement of one’s faults was
theorized to be a feature of complexity. Simplicity related to age (r = .17, p =
.01); apparently, one’s capacity for construct differentiation diminishes with
time. This could be a function of diminished cognition; it could be a cultural
phenomenon characteristic of a generational cohort; or it could be that one’s
view of the world, through trial and error experience, solidifies with time. In
addition, complexity was related to drug use (r = -.11, p = .05), the interpretation
of which is difficult to make. Is the capacity for nuance so stressful that it
necessitates self-medication (Table 8).
66
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
While it is not clear just what construct Cognitive Complexity, or the 4x6
Repertory Grid, is measuring, the other independent variable, Altruism,
appeared to measure what it purported to measure. Over all locations Altruism
related, as predicted, with Empathic Concern (r = .33, p = .001) and
Perspective Taking (r = .27, p = .001). As predicted, Altruism was negatively
correlated with Psychopathy (r = -40, p = .001),and the defense mechanism of
Splitting (r = .17, p = .01). Subjects who experienced Personal Distress at
another’s misfortune also, as predicted, had negative scores on Altruism. This
is because distress would engender anger or anxiety reactions rather than
helping ones. Indeed, subjects who scored high on Personal Distress also had
higher scores on Authoritarianism, in that firm rules may function to allay
anxiety. Altruists, in fact, were different from Authoritarians (r = -.18, p = .01).
Those who scored higher on Receptive Giving, a quid-pro-quo attitude, also
had negative Altruism scores. Altruists tend not, apparently, to indulge in self-
serving cost-benefit analyses. Though Receptive Giving is negatively related to
Altruism (r = -.52, p = .001), as predicted, the huge correlation is due to the fact
that it is, after all, a subscale, along with Altruism, of the HOQ. Altruists are
liable to get enmeshed in others' troubles (r = .13, p = .05); however, they are
able to maintain self/other boundaries and are not drawn to compulsive
Intimacy ( r = -.15, p = .01).
Altruists like to look good, as measured by the Crowne Marlowe (r = .28,
p = .001). They have internalized the moral tenets of society, though they may
67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
seek external validation of their ethical stance. Age was related to Altruism
(r= .12, p = .05), suggesting that one is more accepting as one grows older
(Figure 4).
Correlations Among Dependent Variables
Perspective Taking, measuring the dependent variable cognitive
empathy, related as expected to other dependent variables. It was negatively
related to Personal Distress, Splitting, and Psychopathy. To understand a
foreign consciousness, one has to be calm, detect nuance, and maintain a
benign attitude. It was also strongly related, of course, to its subscale mate,
Empathic Concern (r = .45, p = .001), a sympathetic arousal in concert with
another’s feelings. It is interesting to note that cognitive empathy may make
use of sympathetic arousal as a form of information gathering, without the
confiscation of distress.
Perspective taking is moderately related to social desirability. While
there may be a sort of denial of menial faults at play, there is also evidence that
perspective takers have internalized society’s moral tenets. Indeed, both
Perspective Taking and Altruism achieved identical correlations with the
Crowne Marlowe (r = .28, p = .001). The capacity for taking perspective also
improves with age (r= .11, p = .05), validating a folk assumption that wisdom is
acquired. Perspective taking apparently inoculates against excessive drug-
taking as well (r = -.11, p = .05).
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FIGURE 4. Circumplex Mode!: Results
On Altruism
- Altruism
Receptive Giving Perspective-taking
Social Desirability
Psychopathy Empathic Concern
Intimacy
Personal Distress
Splitting Index Enmeshment
Narcissism
+ Altruism
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Empathic Concern, or emotional empathy, was strongly negatively
related to Psychopathy (r = -.49, p = .001). This tends to validate an
assumption that people with uncaring tendencies lack the capacity to be
sympathetically aroused by another’s emotions. But because this measures a
form of arousal, it is also related with two subscales of codependence:
Responsibility, or meeting others’ needs to the exclusion of one’s own (r = .23,
p = .001); and Intimacy, or difficulty with boundaries (r = .26. p = .001). It was
also strongly related to another form of arousal and its subscale mate, Personal
Distress (r = .44, p = .001). It should be noted that the discrimination of self-
other boundaries, i.e., Am I sad or are you, requires, theoretically, the capacity
for differentiation that was supposed to have been captured by Cognitive
Complexity.
Those who demonstrate the capacity for emotional empathy also score
high on Altruism (r = .33, p = .001). Authoritarianism, as expected, is negatively
related to Empathic concern (r = -.24, p = .001). People with emotional
empathy tend to respond with a socially desirable response set (r= .15, p =
.01). However they also tend not to use drugs (-.11, p = .05).
Narcissism, contrary to theory, did not relate to Splitting; indeed, though
non significant, it related in the opposite direction one would expect, given the
fact that splitting is considered a key defense mechanism of the narcissistic
stance. However, Narcissism did negatively relate as expected to Receptive
Giving (r = -.15, p = .01). Those who cannot decenter cannot fairly deal in the
marketplace. In addition, Narcissism was negatively related to compulsive
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Intimacy, indicating an aversion to any sort of bonding, maladaptive or not; and
negatively related to Personal Distress, where there seems to be no arousal at
the distress of another.
Narcissists, it transpires, have not internalized the moral tenets of
society, in that Narcissism was negatively correlated with the Crowne Marlowe
(r = -.14, p = .01). It may be that they can acknowledge minor faults, but, more
likely, are so self-involved that they cannot even grasp a good deal of what
society expects of them. Narcissists appear to deal with the pain of their
alienation by self-soothing with drugs (r = .19, p = .001); they also take refuge in
the rigid tenets of Authoritarianism (r = .16, p = .01).
Receptive Giving, a quid pro quo stance, was predicted in the
hierarchical regression (controlling for age, gender and social desirability) by
Cognitive Complexity and selfishness (as the putative opposite of Altruism),
suggesting that it is the enlightened aspect of self-interest that plays a part in a
cost-benefit analysis.
But Receptive Giving, far from being a benign form of expedient give-
and-take, turns out to be correlated with compulsive Control, compulsive
Intimacy, and, amazingly, to Personal Distress. Indeed, Receptive Giving was,
also as predicted, moderately related to Psychopathy (r= .16. p = .01).
Receptive Giving was negatively related to social desirability (r = -.20,
p = .001). One has the capacity to acknowledge one’s minor transgressions
when taking the quid pro quo stance of the marketplace, or else, the moral
71
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
tenets of society are disregarded. The desire to seek tit-for-tat appears to be
moderated by age (r = -.38, p = .001).
The tendency to meet others’ needs at the expense of one’s own, as
measured by Responsibility, was related, as expected, to its three sister
subscales of Codependence. It was also related to Personal Distress (r = .28,
p = .001), suggesting that maladaptive helping behaviors are motivated by
distress. One helps not to be of use to the other, but to rid oneself of anxiety. It
related to Fantasy, a subscale of Interpersonal Reactivity; perhaps one
imagines too vividly what one feels oneself, and projects that onto the other
indiscriminately. Responsibility was also strongly related to Splitting (r = .40,
p = .001); if a need to take care of others at one’s own expense is the result of
some sort of anxious distress, splitting may be summoned as a way of
achieving some sort of anxiolytic clarity amid the ambiguity of interpersonal and
intrapersonal feelings.
Subjects who endorsed Responsibility also displayed a capacity for
admitting minor faults, as measured by the Crowne Marlowe, (r = -.13, p = .05).
As one grows older, one apparently feels less compelled to meet others’ needs
at one’s own expense (r = - .19, p = .001); rather, one becomes more altruistic
(r= .12, p = .01). While this last sentence may seem to state a contradiction. It
must be remembered that people who feel compelled to help others do so with
a self-serving motive: to allay the arousal of anxiety, while people in an
altruistic mood have an other-oriented motive for helping.
72
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The need to Control the behavior and affect of self and other is
unrelated to Perspective Taking. With perspective, the ability to see a larger
picture, one feels less compelled to resort to control. Control is, of course,
related to its sister subscales Enmeshment, Intimacy, and Responsibility. It is
so strongly related to Splitting (r = .47, p = .001) that one can only guess that
these two variables are related to a more basic third variable, such as anxiety.
Indeed, that appears to be the case: Personal Distress is related to Control (r =
.28, p = .01). That Responsibility and Control are both related to Cognitive
Complexity, at least in this sample, was puzzling, in that complexity was
thought to give one sufficient cognitive distance to obviate the need for
compulsive control. However, it does make sense that Control was related to
Receptive Giving (r = .14, p = .05), in that one needs to manipulate others to
achieve the best outcome in a marketplace transaction.
Those with a need to control others also display a capacity to admit
minor faults of their own as measured by the Crowne Marlowe (r = -.28, p =
.001), or else they have not sufficiently internalized aspects of the moral tenets
of society. The need for control apparently diminishes with age (r = -.24, p =
.001).
Enmeshment, or over-involvement with the troubles of others, is often
seen in those who facilitate substance abuse among family members, as a way
of protecting what love there is to be had. It is related to its three other
codependent subscales, as well as to the arousal states Personal Distress (r =
.19, p = .01) and Empathic Concern (r = .26, p = .001). It is unrelated, by
73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
definition, to Perspective Taking. It constitutes a benign, if maladaptive, effort
to be kind: Enmeshment is related to Altruism (r = .13, p = .05).
As we get older, it appears that we are less likely to get maladaptively
involved with the troubles of others (r = -.25, p = .001), and instead, tend to gain
perspective (r= .11, p = .05).
Difficulty with self-other boundaries is captured by the codependent
subscale Intimacy, where a subject cannot quite grasp the affective differences
between self and other. Theoretically, Intimacy should negatively relate to
Complexity but does not. However it is, as predicted, negatively related to
Altruism (r = -.15, p = .01). Intimacy is so strongly related to Splitting (r = .51,
p = .001) that one suspects a third factor to be at play, which is, again, anxiety-
perhaps the anxiety of abandonment. This appears to be the case, in that it is
also strongly related to Personal Distress (r = .44, p = .001). It is, surprisingly,
related to Psychopathy (r = .20, p = .01). Could this be a tendency to project
one’s paranoid ideation onto another? Not surprisingly, compulsive Intimacy is
negatively related to Narcissism (r = -. 11, p = .05), where there is an incapacity
for intimacy of any sort, however maladaptive.
Those with boundary ambiguities and the distress that engenders take
refuge, not surprisingly, in the stern rules of Authoritarianism (r = .13, p = .05).
They admit their own frailties, or else, have not grasped aspects of social
desirability as measured by the Crowne Marlowe (r = -.20, p = .001). One
seems to gain boundary differentiation with age (r = -.29, p = .001).
74
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Primary Psychopathy, characterized by a selfish, uncaring and
manipulative stance towards others, was negatively related, as predicted, with
Empathic Concern (r = -.49, p = .001) and Perspective Taking (r = -.33, p =
.001). Not surprisingly, it was negatively related to the generous spirit of
Altruism (r = -.40. p = .001). Of all the variables measured, it had the strongest
relationship with Authoritarianism (r = .32, p = .001). Perhaps this is because
as one’s means become harsh, so does one’s moral code. As Psychopathy
was also correlated with Personal Distress (r= .11, p = .05), as well as the
boundary ambiguity of Intimacy (r = .20, p = .001), it may be that
authoritarianism then becomes a refuge of stem certainty adopted to give
definition to one’s own vague paranoid projections. Indeed, primary
psychopaths tend to resort to the defense mechanism of Splitting (r = .16, p =
.01) in their relationships with others.
Psychopathy was also related to Receptive Giving (r = .28, p = .001), in
that both lack the spirit of altruism: the former, due to a pathological incapacity
for generosity; the latter, due to an acknowledged stance of self-interest.
However it was thought that the quid pro quo stance of the marketplace was a
mild, even socially sanctioned form of selfishness, a frank cost-benefit analysis,
while psychopathy was a more virulent, exploitative form. This assumption of
relative pathology appears to be in question, as measured against Altruism;
Psychopathy was less negatively related (r = -.40, p = .001) than Receptive
Giving (r -.52, p = .001). Of course the relationship of Altruism and Receptive
Giving are highly correlated because they are both HOQ subscales. Perhaps
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
another more pertinent measure of relative malevolence is the harsh law of
Authoritarianism, unrelated to Receptive Giving but strongly related to
Psychopathy.
The subscale of the Interpersonal Reactivity Index least understood in
the context of this study is, perhaps, Fantasy, or the capacity to imagine or
identify with fictitious scenarios. This is because it was originally conceived as
the ability to “become deeply involved in the fictitious world of books, movies
and plays” (Davis, 1983, p. 115). As such, Fantasy does not distinguish, in the
original measure, among the ability to indulge in armchair adventure,
projections of one’s own fantasies, or the capacity to imagine possible
outcomes in a mental model of real-world situations. In the original study,
fantasizers displayed a tendency toward emotional reactivity. And, indeed, in
this study, Fantasy was correlated with all four subscales of the codependence
measure, given the reactive tone of the scale, as well as to all three of its IRI
sister subscales. As a measure of arousal it also related to Splitting (r = .15,
p = .01), a defense of emotional triage. When one thinks of our preference for
the thorough virtuousness of protagonist good guys, and our collective
resistance to well-rounded protagonists with faults as well as virtues (as
measured, say, by movie gross receipts), we come to understand the tropism
towards splitting. Considered in this context, it is difficult to interpret the fact
that Fantasy negatively related to Social Desirability (r = -.24, p = .001). Can
fantasizers admit to frailties, or do they suspend some of the usual societal
tenets in the process of identifying with characters in fiction?
76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The aging process seems to diminish the capacity for fantasizing
(r = -.21, p = .001): perhaps we become more interested in nonfiction truths as
we age, or we lose our passion for armchair fantasies.
Hierarchical Regressions Controlling for Age.
Gender, and Social Desirability
In the effort to control for the effects of age, gender and social
desirability, we ran regressions on the dependent variables using Cognitive
Complexity and Altruism as predictors (Tables 9 through 20). The results
should be read with caution, as CC may not be measuring what we supposed.
The dependent variable of interest, cognitive empathy, as measured by
Perspective Taking, was not predicted, as stated in Hypothesis One. However
the two independent variables, controlling for age, gender; and desirability,
together predicted two of the other dependent variables.
Splitting is somewhat predicted by the interaction of CC (read Simplicity)
and Selfishness (as the putative opposite of Altruism) (F = 4.50,
p = .05). Though they do not account for much of the variance in Splitting, the
results are consistent with the model as stated in Hypothesis Two.
Psychopathy is predicted (F = 28.16, p = .001) by the interaction of
Complexity, as envisioned in the original model, and by Selfishness (the
opposite of Altruism), validating one aspect of Hypothesis Two.
77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 9
Hierarchical Regression on the Splitting Index. Controlling for Age and Social
Desirability and Usina Coanitive Complexity and Altruism as Predictors (n = 344)
Block Predictors R R 2
R 2
Change
F for
Change
Beta
1 Control Measures .29 .08 .08 9.70***
Gender -.05
Age -.14*
Social Desirability -.16**
2 Predictors .33 .1 1 .03 4.50*
Cognitive Complexity -.1 2 *
Altruism -.1 2 *
* D < .05 ** fi < .01
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 10
Hierarchical Regression on Psvchoticism. Controlling for Age and Social Desirability
and Using Cognitive Complexity and Altruism as Predictors (n = 3441
Block Predictors
R
R 2
R 2
Change
F for
Change
Beta
1 Control Measures .40 .16 .16 20.23***
Gender - 2 0 ***
Age
_ 2 7 ***
Social Desirability -.09
2 Predictors .53 .28 . 1 2 28.16***
Cognitive Complexity .1 0 *
Altruism -.35***
*£ < .05 ***£<.001
79
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 11
Hierarchical Regression on Receptive. Controlling for Age and Social Desirability and
Using Cognitive Complexity. Altruism and Their Cross-Products as Predictors (n = 344)
Block Predictors R R
R F for
Change Change
Beta
Control Measures
Gender
Age
Social
Desirability
Predictors
Cognitive
Complexity
Altruism
Cross-product of
Cognitive
Complexity and
Altruism
.42
.62
.63
.18
.39
.39
.18
.21
.01
23.58*
55.05*
3.30
.05
-.31*
.02
-.33*
- . 68 * * *
.33
p < .05 p < .001
80
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 12
Hierarchical Regression on Narcissism. Controlling for Age and Social Desirability and
Using Cognitive Complexity. Altruism and Their Cross-Products as Predictors (n = 344)
Block Predictors
R
R 2
R 2
Change
Ffor
Change
Beta
1 Control Measures .19 .04 .04 4.04**
Gender - . 1 0
Age -.06
Social Desirability - .1 1
2 Predictors . 2 0 .04 . 0 0 .49
Cognitive Complexity .09
Altruism .06
Cross-product of
3 Cognitive Complexity . 2 0 .04 . 0 0 .49 -.16
and Altruism
** a < .01
81
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 13
Hierarchical Regression on Responsibility. Controlling for Aae and Social Desirability
and Using Cognitive Complexity. Altruism and Their Cross-Products as Predictors
(n = 344)
Block Predictors R R 2 Beta
Change Change
Control Measures .22 .05 .05 5.47**
Gender -.04
Age -.18
Social Desirability -.12
Predictors .27 .07 .02 4.3*
and Altruism
' g < .05 **g< .0 1
* *
Cognitive Complexity -.40*
Altruism -.12
Cross-Product of
Cognitive Complexity .29 .08 .01 3.64 .43
82
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 14
Hierarchical Regression on Control. Controlling for Age and Social Desirability and
Using Cognitive Complexity. Altruism and Their Cross-Products as Predictors (n = 344)
Block Predictors R R 2 Beta
Change Change
1 Control Measures .33 .1 1 .1 1 13.76***
Gender -.25
Age -.18*
Social Desirability 23’ »
2 Predictors .36 .13 .02 2.92
Cognitive
Complexity
a <.oi ***£< .001
-.32
Altruism -.15
Cross-Product of
Cognitive 3 6 .1 3 .00 1.31 .25
Complexity and
Altruism
83
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 15
Hierarchical Regression on Enmeshment. Controlling for Age and Social Desirability
and Using Cognitive Complexity. Altruism and Their Cross-Products as Predictors
(n = 344)
Block Predictors B B; chf nge c^ " e Bela
1 Control Measures .26 .07 .07 7.77***
Gender .05
Age -.25
Social Desirability -.10
2 Predictors .32 .11 .04 7.00**
Cognitive
Complexity
Altruism
P < .01 *** q < .001
* * *
.23
Altruism .4 1
Cross-Product of
Cognitive . 3 3 ^ .00 2.37 -.34
Complexity and
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 16
Hierarchical Regression on Intimacy. Controlling for Age and Social Desirability and
Using Cognitive Complexity. Altruism and Their Cross-Products as Predictors (n = 344)
Block Predictors R
R 2
R 2
Change
F for
Change
Beta
1 Control Measures
Gender
Age
Social Desirability
.34 .1 1 .1 1 14.08***
-.06
-.25***
-.13*
2 Predictors
Cognitive Complexity
Altruism
.35 .1 2 .0 1 1.82
-.40*
-.34*
3
Cross-Product of
Cognitive Complexity
and Altruism
.37 .13 .0 1 3.7 .42
*fi< .05 *** e < .0 0 1
85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 17
Hierarchical Regression on Empathic Concern. Controlling for Age and Social
Predictors (n = 344)
Block Predictors R R 2 &
— Change Change Beta
Control Measures .23 .05 .05 5.94*
Gender
Age
Social Desirability
Predictors
Cognitive
Complexity
Altruism
Cross-Product of
Cognitive
Complexity and
Altruism
.40 .16
.4 1 .17
.11
.01
21 .20*
. 19* * *
-.40
.06
-.33
.11
3.3 .39
.01 B < .001
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 18
Hierarchical Regression on Fantasy, Controlling for Age and Social Desirability and
Usina Coanitive ComDlexitv. Altruism and Their Cross-Products as Predictors fn = 344)
Block Predictors R
R 2 R 2
— Change
F for 0 .
Change e a
1 Control Measures .29 .08 .08 9.70***
Gender .15**
Age -.2 0 ***
Social Desirability -.19**
2 Predictors .33 .1 1 .03 4.61*
Cognitive Complexity -.07
Altruism .14
Cross-Product of
3 Cognitive Complexity .33 .1 1 .0 0 .03 .04
and Altruism
* £ < .05 ** g < .0 1 ***£<.001
87
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 19
Hierarchical Regression on Personal Distress. Controlling for Age and Social
Desirability and Using Cognitive Complexity. Altruism and Their Cross-Products as
Predictors (n = 344)
Block Predictors R R
R
Change
F for
Change
Beta
1 Control Measures .33 .1 1 .1 1 13.70***
Gender .13*
Age
■ j Q ***
Social Desirability -.2 0 ***
2 Predictors .35 .12 .01 2.08
Cognitive Complexity -.07
Altruism -.19
Cross-Product of
3 Cognitive Complexity .35 .12 . 0 0 .45 .15
and Altruism
* B < .05 e < .001
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 20
Hierarchical Regression on Perspective-taking. Controlling for Age and Social
Desirability and Using Cognitive Complexity. Altruism and Their Cross-Products as
Predictors (n = 344)
Block Predictors R
R F for
Change Change
Beta
Control Measures
Gender
Age
Social Desirability
Predictors
Cognitive
Complexity
Altruism
Cross-Product of
Cognitive
Complexity and
Altruism
.30
.36
.36
.9 1
.13
.13
.09
.04
.00
10.97*
6.95*
1.03
-.01
.07
.24*
-.24
.05
.22
E< .01 £ < .001
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Multivariate Analysis of Covariance
There were no significant multivariate interactions between Complexity
and Altruism (Table 21).
90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 21
Two (High versus Low Cognitive Complexity) bv Two (High versus Low Altruism) Multivariate
Analysis of Covariance. Controlling for Age and Social Desirability
MANCOVA Effects3
CC A CCxA
E E E
Covariates Groups M
SD n
Age Cognitively Complex 1.2 9.1*** 1.1
Low Altruism 24.39 11.40 78
High Altruism 29.88 18.16 89
Cognitively Simple
Low Altruism 29.33 17.33 80
High Altruism 34.64 20.09 85
Social Desirability Cognitively Complex
Low Altruism 12.23 5.12 78
High Altruism 14.89 5.10 89
Cognitively Simple
Low Altruism 12.68 6.18 80
High Altruism 15.40 5.60 85
ANCOVA Effects3
CC A CCxA
Dependent
Measures Groups M
SD n F
U-l
u _ l
Splitting Index Cognitively Complex 3.6 3.5 0.0
Low Altruism 30.27 7.59 78
High Altruism 27.89 7.08 89
Cognitively Simple
Low Altruism 28.48 7.50 80
High Altruism 25.87 7.09 85
Psychoticism Cognitively Complex 6.07* 14.0*** 0.0
Low Altruism 30.12 7.75 78
High Altruism 26.00 6.81 89
Cognitively Simple
Low Altruism 31.54 8.69 80
High Altruism 27.12 6.76 85
* p < .05 *** p < .001
91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 21 (continued)
Two (High versus Low Cognitive Complexity) bv Two (High versus Low Altruism) Multivariate
Analysis of Covariance. Controlling for Age and Social Desirability
Dependent
Measures Groups M
SD n
ANOVA
CC
F
Effects3
A CCxA
E E
Receptive Cognitively Complex 0.1 65.1*** 0.0
Low Altruism 2.96 1.15 78
High Altruism 1.74 1.10 89
Cognitively Simple
Low Altruism 2.80 1.42 80
High Altruism 1.55 1.23 85
Narcissism Cognitively Complex 0.1 0.0 2.9
Low Altruism 5.00 2.52 78
High Altruism 5.27 2.28 89
Cognitively Simple
Low Altruism 5.46 2.44 80
High Altruism 4.82 2.63 85
Responsibility Cognitively Complex 3.1 5.7* 5.6*
Low Altruism 25.21 4.23 78
High Altruism 24.71 4.82 89
Cognitively Simple
Low Altruism 22.79 4.91 80
High Altruism 24.72 5.02 85
Control Cognitively Complex 2.2 0.1 0.0
Low Altruism 30.87 4.19 78
High Altruism 29.91 5.14 89
Cognitively Simple
Low Altruism 29.75 4.95 80
High Altruism 28.82 4.82 85
Enmeshment Cognitively Complex 0.6 7.1** 0.2
Low Altruism 20.13 3.53 78
High Altruism 20.60 3.91 89
Cognitively Simple
Low Altruism 19.38 3.43 80
High Altruism 20.20 3.60 85
* j3 < .05 ** J 2 < .01 ***£<.001
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 21 (continued)
Two (High versus Low Cognitive Complexity) bv Two (High versus Low Altruism) Multivariate
Analysis of Covariance. Controlling for Age and Social Desirability
Dependent
Measures Groups M
SD
Q
ANOVA Effects3
CC A CCxA
F F F
Intimacy Cognitively Complex 1.6 1.9 1.1
Low Altruism 29.13 4.54 78
High Altruism 27.05 5.77 89
Cognitively Simple
Low Altruism 27.46 50.8 80
High Altruism 26.52 4.72 85
Empathic Concern Cognitively Complex 0.3 24.2*** 2.0
Low Altruism 19.35 3.75 78
High Altruism 21.07 3.77 89
Cognitively Simple
Low Altruism 18.48 4.68 80
High Altruism 21.41 3.30 85
Fantasy Cognitively Complex 0.2 3.9* 1.1
Low Altruism 17.04 5.36 78
High Altruism 16.87 5.31 89
Cognitively Simple
Low Altruism 15.79 5.54 80
High Altruism 16.82 5.22 85
Personal Distress Cognitively Complex 0.1 4.3* 0.5
Low Altruism 12.21 4.21 78
High Altruism 10.10 4.64 89
Cognitively Simple
Low Altruism 11.40 5.26 80
High Altruism 10.00 4.70 85
Perspective-taking Cognitively Complex 1.3 5.8* 2.4
Low Altruism 16.54 4.45
High Altruism 17.63 4.78
Cognitively Simple
Low Altruism 15.4 4.7
High Altruism 18.0 4.2
* E < .05 *** E < .001
93
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 5
DISCUSSION
The main source of puzzlement concerns the fact that it is not clear just
what the key independent variable, Cognitive Complexity, is in fact measuring.
Cognitive simplicity barely related, contrary to expectations, to Authoritarianism
(r = .10, p = .07). Moreover, complexity appeared to relate significantly to the
defense mechanism of Splitting (r = -.15, p = .01. High scores on CC, it must
be remembered, refer to cognitive simplicity). The test of splitting was included
in the battery as a form of discriminant validity. According to theory as well as
to outcomes of previous studies, splitting should be negatively correlated with
complexity. While none of these correlations is especially large, they
nonetheless suggest a direction; and it is the opposite direction we surmised in
selecting these variables as convergent correlates. One would, for instance,
expect the key dependent variable, Perspective Taking, to relate to complexity,
but it does not. However, an aspect of Codependence does relate: compulsive
Control over self and other (r = -.16, p = .01). As this trend suggests a potential
failure of construct validity, it may be that what we are measuring is a form of
scanning for information, in the service of control. That is, the ability to form
complex characterizations is developed to moderate anxiety.
This last observation leads one to an arresting notion when one looks at
correlations for the sample location consisting of students at a large private
Catholic university (Table 22). There one finds a significant correlation between
complexity and splitting (r = - .212, p = .049). The sample location of the
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 22
Correlations Among Dependent and Independent Variables Among the Three Sample Subgroups
Corrections - Direct Mail
ALTRUISM A U tH tO T TOTALCC
SPUTTOT Pearson Correlation -.220* .012 -.138
Sig. (2-tailed) .011 .889 .111
N 134 134 134 |
PSYCHO Pearson Correlation -.424" .266" .088 !
Sig. (2-tailed) .000 .002 .309
N 134 134 134
NARCISSM Pearson Correlation -.029 -.049 -.132
Sig. (2-tailed) .742 .573 .128
N ' 134 134 134
RECEPFAC Pearson Correlation -.526** -.055 -.170'
Sig. (24ailed) .000 .530 .049
N 134 134 134
RESPON Pearson Correlation .116 -.037 -.240" j
Sig. (2-tailed) .180 .669 .005
N 134 134 134
CONTROL Pearson Correlation -.039 -.070 -.320"
Sig. (2-tailed) .653 .424 .000
N 134 134 134
ENMESH Pearson Correlation .165 -.043 -.185*
$ig. (2-tailed) .056 .619 .033
N 134 134 134
INTIM Pearson Conrelation -.136 .092 -.110
Sig. (2-tailed) .118 .292 .208
N 134 134 134
CODEP Pearson Correlation .018 -.016 -.269"
Sig. (2-tailed) .840 .855 .002
N 134 134 134
ECONCERN Pearson Correlation .3 4 2 " -.308" -.105
Sig. (2-tailed) .cm .000 .229
N 134 134 134
FANTASY Pearson Correlation .066 .001 -.155
Sig. (2-tailed) .446 .991 .074
N 134 134 134
PDISTRES Pearson Correlation -.154 .162 -.045
Sig. (2-tailed) .076 .062 .605
N 134 134 134
PERSPECT Pearson Correlation .295** -.276" -.096
Sig. (2-tailed) .001 .001 .26 8
N 134 134 134
CROWNE Pearson Correlation .2 2 2" .210' .125
Sig. (2tailed) .010 .015 .149
N 134 134 134
AGE Pearson Correlation .309" -.002 .140
Sig. (2tailed) .000 .983 .115
N 128 128 128
DRUGUSE Pearson Correlation -.209* -.016 -.081
Sig. (2tailed) .017 .853 .359
N 131 131 131
* Correlation is significant at the 0.05 level (2-tailed).
" Correlation is significant at the 0.01 level (2-tailed).
95
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 22 (continued)
Correlations Among Dependent and Independent Variables Among the Three Sample Subgroups
Corrections - State University
I ! ALTRUISM AUTHTOT TOTALCC j
SPUTTOT Pearson Correlation -.19 8* -.041 -.132 i
Sig. (2-tailed) .028 .652 .146 j
N 123 123 123 j
PSYCHO Pearson Correlation -.418** .309" .066
Sig. (2-tailed) .000 .001 .470 j
N 123 123 123
NARCISSM Pearson Correlation -.126 .264" -.054
Sig. (2-tailed) .165 .003 .551
N 123 123 123
RECEPFAC Pearson Correlation -.626" .105 -.062
Sig. (2-tailed) .000 .248 .495 I
N 123 123 123
RESPON Pearson Correlation -.029 .044 .146 I
Sig. (2-tailed) .754 .632 .107
N 123 123 123
CONTROL Pearson Correlation -.105 .158 .072
Sig. (2-tailed) .246 .080 .431
N 123 123 123
ENMESH Pearson Correlation .059 -.059 .068
Sig. (2-tailed) .515 .520 .458 |
N 123 123 123
INTIM Pearson Correlation -.113 .127 -.004
Sig. (2-tailed) .215 .162 .966 |
N 123 123 123
CODEP Pearson Correlation -.069 .096 .090
Sig. (2-tailed) .446 .289 .325
I N 123 123 123
ECONCERN Pearson Correlation .201' -.080 .026
Sig. (2-tailed) .026 .378 .772
N 123 123 123
FANTASY Pearson Conelation -.034 .147 .065 |
Sig. (2-tailed) .708 .105 ■475
N 123 123 123
PDISTRES Pearson Correlation -.203' .060 .099
Sig. (2-tailed) .025 .507 .275
| N 123 123 123
PERSPECT Pearson Correlation .205' -.046 -.054
Sig. (2-tailed) .023 .614 .553
N 123 123 123
CROWNE Pearson Conelation .3 9 6 " -.019 .088
Sig. (2-tailed) .000 .834 .333
N 123 123 123
AGE Pearson Correlation -.201* .108 .068
Sig. (2-tailed) .029 .244 .466
N 118 118 118
DRUGUSE Pearson Correlation .032 -.076 -.163
Sig. (2-tailed) .726 .402 .073
N 122 122 122
* Correlation is significant at the 0.05 level (2-tailed).
". Correlation is significant at the 0.01 level (2-tailed).
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 22 (continued)
Correlations Among Dependent and Independent Variables Among the Three Sample Subgroups
Corrections - Private University
SPUTTOT Pearson Correlation -.060 .028 -.212*
| Sig. (2-tailed) .584 .796 .049 I
I N 87 87 87
PSYCHO Pearson Correlation -.384** .320** .035
J Sig. (2-tailed) .000 .003 .745
I N 87 87 87
NARCISSM Pearson Correlation -.040 .289** .032
I Sig. (2-tailed) .712 .007 .770
I
N 87 87 87
RECEPFAC Pearson Correlation -.494** .066 .004 I
Sig. (2-tailed) .000 .546 .967 !
N 87 87 87
JRESPON Pearson Conelation .195 -.062 -.158 !
Sig. (2-tailed) .071 .566 .143
| N 87 87 87
j CONTROL Pearson Correlation -.123 .125 -.129
Sig. (24ailed) .257 .250 .234
N 87 87 87
j ENMESH Pearson Correlation .184 -.102 -.059
Sig. (2-tailed) .089 .348 .584
| N 87 87 87
llNTIM Pearson Correlation -.242* .095 -.13 8
Sig. (2-tailed) .024 .380 .204
I N 87 87 87
\\C0DEP Pearson Correlation -.016 .030 -.164
Sig. (2-tailed) .881 .782 .128
| | N 87 87 87
I e c o n c e r n Pearson Conelation .454** -.316** .004
I Sig. (2-tailed) .000 .003 .973
II N 87 87 87
I f a n t a s y Pearson Conelation .216* -.184 -.205
I 1 Sig. (2-tailed) .045 .087 .057
N 87 87 87
jPDISTRES Pearson Conelation -.267* -.004 .064
Sig. (2-tailed) .013 .973 .556
N 87 87 87
1PERSPECT Pearson Conelation .331** -.090 .000
Sig. (2-tailed) .002 .406 .998
N 87 87 87
tCROWNE Pearson Correlation .264* -.088 .012
Sig. (24ailed) .013 .420 .913
| N 87 87 87
AGE Pearson Conelation .013 .178 .389**
1 Sig. (24ailed) .908 .101 .000
I N 86 86 86
WRUGUSE Pearson Correlation .135 -.076 -.092
J Sig. (24ailed) .213 .485 .395
II N 87 87 87
* . Correlation is significant at the 0.05 level (2-tailed).
* * . Correlation is significant at the 0.01 level (2-tailed).
97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CC/splitting relationship freighted the entire sample with this correlation. It
could be argued that this finding suggests that, although one can be relatively
complex, one’s lifelong education in a moral theology of good and evil
socializes one to split the world into polarizations: God versus the Devil,
Madonna versus whore, saint versus sinner, heaven versus hell, and so on.
Complexity serves, then, to detect nuances of relative good or evil, to scan the
environment for moral dangerousness, as one’s mortal soul may be on the line.
Perhaps a more instructive explanation lies in the process of test taking
itself. In this study, the researcher was able to gain access to two university
subject pools, participating in the privileges of her dissertation Chair’s grant. In
doing so, numerous additional measures were added to the front of an already
long questionnaire, which then required from one to two hours to complete.
These additional measures were of a highly sensitive nature, asking for
histories of family substance abuse, sexual attitudes, behaviors, and
experiences of sexual abuse, parental neglect, and family emotional and
physical abuse.
One can then speculate that the exigencies of taking a one to two hour
self-report test, in addition to the emotional turmoil caused by confronting
questions one never may have asked oneself, may have aroused feelings of
intense anxiety. A person in a state of negative sympathetic arousal is likely to
dispense with the niceties of nuance and cut to the chase: the splitting of
experience into good and bad, safe or dangerous. This defense no doubt has
survival value, for it focuses the organism on what is important to attend to in
98
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the environment. A person who in calmer moods may enjoy the variations of
gray in his landscape may, under stress, focus on the black and white. A
leisurely stroll in the countryside takes on a different cast when a strange dog
barks aggressively. There, we are likely to focus on one polar aspect-
dangerousness-and spend our efforts predicting the likelihood of attack.
Thus, while splitting may be chronic in certain personality disorders.
Where inner turmoil is constant, it is a defense anyone can adopt when
confronted with something immediately threatening: a dog, a memory, a
sudden insight, a sickening fear. One can speculate that sufficient numbers of
the college sample experienced childhood trauma of one sort or another, such
that in their re-aroused state they responded in a splitting manner.
It is likely, therefore, that we failed to fully appreciate the central
evolutionary purpose of complexity, according to Irwin, Tripodi, and Bieri
(1967), and that is, to predict dangerousness, whether concrete, as in the case
of the encounter with a hostile dog; or diffuse, as in the case of a memory of
abuse. As Millington et at (1996) found, cognitive complexity serves to focus on
differences between self and other. And yet Rotton, Dubitsky, Milov, and White
(1997) demonstrated that people who experience stress, as measured by
elevated serum levels of cortisol, also experience a diminished capacity for
complexity. If normally complex people find themselves in a stressful situation,
could a measure of splitting be a state-dependent measure of their distress?
Indeed, the Personal Distress measure correlated with Splitting (r = .33, p. =
.001). So, it could be that, from the point of view of measurement, Cognitive
99
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Complexity reflects a general capacity or trait; while Splitting, at least in the
case of this study, measures a state-dependent defense against anxiety. It
may be that there is a nonlinear relationship between complexity and the
perception of dangerousness.
In spite of these conjectures, Cognitive Complexity, in the end, could be
measuring nothing more than a sort of conscientiousness. Or, lastly, it could be
a poor measure for what is admittedly a factorially complex construct.
An intriguing finding, which could be explored in further studies,
concerns the fact that if Authoritarianism is substituted for Cognitive Simplicity
as one of the key independent variables, along with Altruism, a part of the
model is validated (Figure 5, Table 7). Authoritarians, it must be remembered,
are generally intolerant of ambiguity and prefer a restricted range of construct
differentiation (Table 22).
Now this assignment is problematic, as the current measure (Lane’s F-
Scale) consists in a mere four items and achieved a rather meager sample
reliability (Alpha = .46). Nonetheless, that it has numerous significant
relationships with dependent variables demonstrates that there is some power
behind the construct. And, previous studies have used other measures of
Authoritarianism as a marker of simplicity (Gould et al, 1996).
In addition, given the fact that Authoritarianism is negatively correlated
with its dependent mate, Altruism (r = -.18, p = .01), one has to deal with the
possible effects of multicollinearity: shaky model specification or measurement
problems, or very likely both.
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
FIGURE 5. Circumplex Model: Results
on Altruism and Cognitive Complexity
Cognitive Complexity
” Altruism
Splitting Index
Receptive Giving
+ Altruism
Cognitive Simplicity
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Given these caveats, however, Table 7 is instructive: Altruism and
Authoritarianism, together, predict part of the model. Let us examine these
relationships. Perspective Taking (cognitive empathy) and Empathic Concern
(a sympathetic resonance) are both negatively related to Authoritarianism and
positively related to Altruism; this is as predicted by the model. Primary
Psychopathy is negatively related to Altruism, as predicted, but positively
related to Authoritarianism, contrary to prediction. It appears that people with
predatory dispositions have an intolerance of ambiguity and, by extrapolation,
are cognitively simple. Likewise, Personal Distress (a negative sympathetic
arousal) and compulsive Intimacy are negatively correlated with Altruism--as
they are markers of anxiety-while positively correlated with Authoritarianism-a
potential moderator of anxiety.
Other correlations, though not significant, nonetheless suggest the
predicted direction: Narcissism is significantly related to authoritarianism but
unrelated to Altruism; Enmeshment (overinvolvement in others’ problems) is
significantly related to Altruism, as expected, but unrelated to Authoritarianism;
and Receptive Giving (quid pro quo) is negatively related to Altruism and
unrelated to Authoritarianism.
Taken together, these suggest that a more conventional measure of
complexity might yield richer results (Figure 6).
102
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
- Altruism
FIGURE 6. Circumplex Model: Results
Substituting Authoritarianism for the 4 x 6 Repertory Grid
Cognitive Complexity
+ Altruism
Intimacy
Psychopathy
Perspective-taking
Empathic Concern
Personal Distress
Authoritarianism
103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Limitations of the Study
The present study was, perhaps, over ambitious; moreover, the two
independent variables, Altruism and Cognitive Complexity are no doubt both
factorially complex and difficult to measure in paper and pencil tests. While
Altruism appeared to measure what it purported to measure, Cognitive
Complexity did not; as complexity falls, so falls the model.
In addition, several of the measures were truncated by selecting items
with the highest factor loadings, thus altering their published reliability. Sample
reliabilities proved adequate, but nonetheless lower than the published ones.
More damaging, we could not run sample reliability tests on the 4x6 Repertory
Grid measuring Cognitive Complexity and so have no real understanding of its
internal consistency for this sample.
There were differences among subjects in the three sample locations on
variables such as age, years of education, and gender. Now the original
purpose for seeking demographic heterogeneity with regard to sample subjects
was to improve our chances for achieving external validity, or generalizability.
However the problems generated by non equivalent groups suggest that, for
the purposes of predoctoral research, modest generalizability allows for greater
ease and clarity of discourse.
Lastly, it is no doubt advisable to limit the number of anxiety-provoking
measures one gives to subjects within the context of a single study, thereby
avoiding an exaggeration of random error in measurement due to test taking
effects, as appeared to be the case in this study.
104
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
SELECTED REFERENCES
Adomo, T. W., Frenkel-Brunswik, E., Levinson, D. J., & Sanford, R. N. (1950).
The authoritarian personality. New York: Harper & Row.
Barrett-Lennard, G. T. (1981). The empathy cycle: Refinement of a nuclear
concept. Journal of Counseling Psychology, 28, 91-100.
Batson, C. D., Batson, J. G., Slingsby, J. K., Harrell, K. L. Peekna, H. M., & Todd,
R. M. (1991). Empathicjoy and the empathy-altruism hypothesis. Journal
of Personality and Social Psychology, 61,413-462.
Berkowitz, L. (1972). Social norms, feelings, and other factors affecting helping
and altruism. In L. Berkowitz (Ed.), Advances in experimental social
psychology (Vol. 6, pp. 63-108). New York: Academic Press.
Bieri, J. (1955). Cognitive complexity-simplicity and predictive behavior. The
Journal of Abnormal and Social Psychology, 50 (1), 263-268.
Bieri, J., Atkins, A. L., Briar, S., Leaman, R. L., & Tripodi, T. (1966). Clinical and
social judgment: The discrimination of behavioral information. New York:
Wiley.
Bowlby, J. (1973). Attachment and loss. New York: Basic Books.
Buber, M. & Smith, R. G. (1937). I and thou. Edinburgh: T&T Clark.
Budner, S. (1962). Intolerance of ambiguity as a personality variable. Journal of
Personality, 30, 29-50.
Burleson, B. R., Kunkel A. W., & Szolwinski, J .B. (1997). Similarity in cognitive
complexity and attraction to friends and lovers: Experimental and
correlational studies. Journal of Constructivist Psychology, 10(3),
221-248.
Cameron, N. (1947). The psychology of behavior disorders. Boston: Houghton
Mifflin.
Chlopan, B. E., McCain, M. L., Carbonell, J. L., & Hagen, R. L. (1985). Empathy:
Review of available measures. Journal of Personality and Social
Psychology, 48 (3), 635-653.
Crowne, D. & Marlowe, D. (1964). The approval motive. New York: Wiley.
105
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Davis, M. H. (1983). Measuring individual differences in empathy: Evidence for a
multidimensional approach. Journal of Personality and Social Psychology,
44, 133-126.
Dawkins, R. (1976). The selfish gene. New York: Oxford University Press.
Eagly, A. H. & Chaiken, S. (1993). The Psychology of attitudes. New York:
Harcourt Brace & Jovanovich.
Ekeland, I. (1999). Agreeing on strategies. Nature, 400 (6745), 623-624.
Emmons, R. A. (1987). Narcissism: Theory and measurement. Journal of
Personality and Social Psychology, 52,11-17.
Feltovich, P. J., Spiro, R. J., & Coulson, R. L. (1997). Issues of expert flexibility in
contexts characterized by complexity and change. In K. M. Ford & P. J.
Feltovich (Eds.), Expertise in context: Human and machine (pp. 125-146).
Menlo Park, CA: American Association for Artificial Intelligence.
Freud, S. (1930). Civilization and its discontents. London: Hearth.
Gallup, G. G. (1995). Self-awareness and the evolution of social intelligence.
Behavioral Processes, 42 (2-3), 239-247.
Gladstein, G. A. & Associates (1987). Empathy and counseling: Explorations in
theory and research. New York: Springer-Veriag.
Gladstein, G. A. & Fieldstone, J.C. (1983). Using film to increase counselor
empathic experiences. Counselor Education and Supervision, 23 (2),
125-131.
Gladstein, G. A. (1983). Understanding empathy: Integrating counseling,
developmental and social psychology perspectives. Journal of Counseling
Psychology, 30, (4), 467- 482.
Gladstein, G. A. (1984). The historical roots of contemporary empathy research.
Journal of the History of the Behavioral Sciences, 20(1), 38-59.
Goleman, D. (1995). Emotional Intelligence. New York: Bantam Books.
Gottfredson, L. S. (1997). Why g matters: The complexity of everyday life.
Intelligence, 24 (1), 79-132.
Gould, J. R., Prentice, N.M., & Ainslie R. C. (1996). The splitting index:
Construction of a scale measuring the defense mechanism of splitting.
Journal of Personality Assessment, 66 (2), 414-430.
106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Gregory, R. L. (1987). Empathy. The Oxford Companion to the Mind. Oxford
University Press.
Haldane, J. B. S. (1955). Population genetics. New Biology, 18, 34-51.
Hamilton, W. D. (1964). The genetical evolution of social behavior. Journal of
Theoretical Biology, 7,1-52.
Hamilton, W. D. (1972). Altruism and related phenomena, mainly in the social
insects. Annual Review of Ecological Systems, 3, 193-232.
Hendrick, C. & Hendrick, S. (1986). A theory and method of love. Journal of
Personality and Social Psychology, 50 (2), 392-402.
Himmelfarb, S. (1993). The measurement of attitudes. In A. H Eagly & S.
Chaiken (Eds.), The psychology of attitudes. New York: Harcourt Brace
Jovanovich.
Hinze, T., Doster, J., & Joe V. C. (1997). The relationship of conservatism and
cognitive-complexity. Personality and Individual Differences, 22 (2),
297-298.
Hoffman, M. L. (1989). Empathy, social cognition and moral action. In W.
Kurten’s, & J. Pewits (Eds.), Moral behavior and development: Advances
in theory, research, and application, Vol. 1. Hillsdale, NJ: Erlbaum.
Hofstadter, D. R. (1981). Reflections. In D. R. Hofstadter & D.C. Dennett (Eds.),
The mind’ s I: Reflections on self and soul. New York: Bantam.
Hogan, R. (1969). Development of an empathy scale. Journal of Consulting and
Clinical Psychology, 33, 307-316.
Irwin, M., Tripodi, T., & Bieri, J. (1967). Affective stimulus value and cognitive
complexity. Journal of Personality & Social Psychology, 5, (4), 444-448.
Jaynes, J. (1976). The origin of consciousness in the breakdown of the bicameral
mind. Boston: Houghton Mifflin.
Johnson, J. G., Cohen, P., Brown, J., Smailes, E. M., & Bernstein, D. P. (1999).
Childhood maltreatment increases risk for personality disorders during
early adulthood. Archives of General Psychiatry, 56,600-606.
Kagan, J., & Schneider (1987). Towards the measurement of affective sensitivity.
Journal of Counseling and Development, 65 (9), 459-464.
Karpman, B. (1948). The myth of the psychopathic personality. American Journal
of Psychiatry, 104, 523-534.
107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Kerber, K. W. (1984). The perception of nonemergency helping situations: Costs,
rewards, and the altruistic personality. Journal of Personality, 52 (2),
177-187.
Kernberg, O. P. (1975). Borderline conditions and pathological narcissism. New
York: Aronson.
Kohut, H. (1978). The Search for the se/f (Vol. 1). New York: International
Universities Press.
Lane, R. E. (1955). Political personality and electoral choice. American Political
Science Review, 4 9 ,173-190.
Levenson, M. R., Kiehl, K. A., & Fitzpatrick, C. M. (1995). Assessing psychopathic
attributes in a noninstitutionalized population. Journal of Personality and
Social Psychology, 6 8 ,151-158.
Lipp, T. (1891). Aesthetik. Berlin.
Loehlin, J. C. (1992). Genes and environment in personality development.
Newbury Park, CA: Sage.
Macaulay & Berkowitz (Eds.). (1979). Altruism and helping behavior. New York:
Academic Press.
MacCorquodale, K., & Meehl, P. E. (1948). On a distinction between hypothetical
constructs and intervening variables. Psychological Review, 55, 95-107.
Mehrabian, A., & Epstein, N. (1972). A measure of emotional empathy. Journal
of Personality, 40, 525-543.
Miller, A. (1981). The drama of the gifted child. New York: Basic Books.
Millington, M. J., Strohmer, D.C., Reid, C.A., & Spengler, P.M. (1996). A
preliminary investigation of the role of differential complexity and response
style in measuring attitudes towards people with disabilities. Rehabilitation
Psychology, 41 (3), 243-254.
Millon, T. (1983). Millon Clinical Multiaxial Inventory Manual, Third Edition.
Minnesota: Interpretive Scoring Systems.
Nagel, T. (1994). Freud’s permanent revolution. The New York Review of Books,
X L /(2), 34-38).
Nathanson, D. L. (1993). About emotion. Psychiatric Annals, 23(10), 541-555.
108
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Nozick, R. (1998, August). The Place of Consciousness. Paper presented at the
meeting of the American Psychological Association, San Francisco.
Preiss, R. (1998). Personal communication.
Previnelli, D. A. (1998). Can animals empathize? Scientific American, 9 (4),
66-76.
Rabow, J, Newcomb, M. D., Monto, M. A. & Hernandez. (1990). Altruism in drunk
driving situations: Personal and situational factors in intervention. Social
Psychology Quarterly, 53 (3), 199-213.
Raskin, R. & Hall, C. S. (1979). A narcissistic personality inventory.
Psychological Reports, 45, 590.
Raskin, R., & Terry, H. (1988). A principal-components analysis of the Narcissistic
Personality Inventory and further evidence of its construct validity. Journal
of Personality and Social Psychology, 54, (5), 890-902.
Reiss, S., Levitan, G. W. & Szyszko, J. (1982). Emotional disturbance and mental
retardation: Diagnostic overshadowing. American Journal of Mental
Deficiency, 86 (6), 567-574.
Ridley, M. (1998). The origins of virtue: Human instincts and the evolution of
cooperation. New York: Viking.
Robinson, J. R. (1996). The relationship between personal characteristics and
attitudes towards Black and White speakers of informal non-standard
English. Western Journal of Black Studies, 20 (4), 211-220.
Roehling, P. V., & Gaumond, E. (1996). Reliability and validity of the
Codependent Questionnaire. Alcoholism Treatment Quarterly, 14 (1),
85-95.
Rogers, C. R. (1975). Empathic: An unappreciated way of being. Counseling
Psychologist, 5, 2-10.
Romer, D., Gruder, C. L., & Lizzadro, T. (1986). A person-situation approach to
altruistic behavior. Journal of Personality and Social Psychology, 51(5),
1001- 1012.
Rotton, J., Dubitsky, S.S., Milov, A., & White, S. M.; et al., (1997). Distress,
elevated cortisol, cognitive deficits, and illness following a natural disaster.
Journal of Environmental Education, 17 (2), 85-98.
109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Sanz, J., Avia, M. D., & Sanchez-Bernardos, M. L. (1996). The structure of the
construct system in social anxiety: Qualifications due to affective
confounding. Journal of Constructivist Psychology, 9(3), 210-212.
Scott, W. A. (1962). Cognitive complexity and cognitive flexibility. Sociometry, 25
(4), 405-415.
Sharma, V., & Rosha, J. (1992). Altruism as a function of self-actualization and
locus of control of benefactor. Psychological Studies, 37 (1), 26-30.
Sober, E. & Wilson, D. S. (1998). Unto others: The evolution and psychology of
unselfish behavior. Cambridge: Harvard University Press.
Spengler, P. M., & Strohmer, D. C. (1994). Stability of a 4 x 6 Repertory Grid for
Measuring Cognitive Complexity. Journal of Constructivist Psychology 7,
137-145.
Stein, E. (1917/1989). On the problem of empathy. In W. Stein, (Trans.), The
Collected Works of Edith Stein, Vol. 3, pp. 1-131. Washington, D.C.: ICS
Press.
Stewart, D. (1956). Preface to empathy. New York: Philosophical Library.
Tellegen, A. P. (1982). Manual for the Multidimensional Personality
Questionnaire. University of Minnesota: Department of Psychiatry.
Tripodi, T., & Bieri, J. (1964). Information transmission in clinical judgments as a
function of the stimulus dimensionality and cognitive complexity. Journal
of Personality, 3 2 ,119-137.
Trivers, R. (1985). Social evolution. Menlo Park, CA: Benjamin Cummings.
Williams, C. A. (1990). Biopsychosocial elements of empathy: A
multidimensional model. Issues in Mental Health Nursing, 11, (2),
155-174.
Zahn-Waxler, C. & Radke-Yarrow, M. (1979, August). A developmental analysis
of children’ s responses to emotions in others. Paper presented at the
biennial meeting of the Society for Research in Child Development, San
Francisco, CA.
Zahn-Waxler, C., Cole, P. M., Welsh, J. D. & Fox, N. A. (1995).
Psychophysiological correlates of empathy and prosocial behaviors in
preschool children with behavior problems. Development and
Psychopathology, 7 (1), 27-48.
110
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences.
American Psychologist, 35 (2), 151-175.
Zethren, K. (2000, July). A bidimensional model of empathy. Paper presented
at the XXVII International Congress of Psychology, Stockholm.
Zuckerman, M. (1979). Sensation seeking: Beyond the optimal level of
arousal. Hillsdale, NJ: Erlbaum.
111
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Gender differences in motivation for sexual intercourse: Implications for risky sexual behavior and substance use in a university and community sample

PDF

Effects of test interpretation style and need for cognition on elaboration, favorability, and recall

PDF

Integration of science and practice: A collective case study of scientist -practitioner programs in counseling psychology

PDF

Implementation of tobacco prevention programming and smoking related outcomes

PDF

Cognitive and non-cognitive factors as predictors of retention among academically at-risk college students: A structural equation modelling approach

PDF

An integrative analysis of racism and quality of life: A comparison of multidimensional moderators in an ethnically diverse sample

PDF

Assessment of racial identity and self -esteem in an Armenian American population

PDF

An evaluation of a brief HIV /AIDS prevention intervention using normative feedback to promote risk reduction among sexually active college students

PDF

Gender differences in symptom presentation of depression in primary care settings

PDF

Behavioral and demographic predictors of breast cancer stage at diagnosis

PDF

Heterosexuals' attitudes toward lesbians and gay men and the attitude functions they serve: Correlates, stability and gender differences

PDF

Anger expression in men with elevated blood pressure: Relationships to assertiveness

PDF

Attachment style, interpersonal guilt, parental alcoholism, parental divorce and eating disordered symptomatology in college women

PDF

A phenomenological inquiry into the essential meanings of the most intense experience of religiosity and spirituality

PDF

Impact of language and culture on a neuropsychological screening battery for Hispanics

PDF

A Positive Psychology goal intervention using strengths and values to enhance goals and increase well -being

PDF

Gay men with eating disorders and food, body image, and exercise concerns: A group treatment approach

PDF

An investigation of a new diagnostic sub-type: Post traumatic stress disorder with psychotic features

PDF

Effects of parenting style and ethnic identity on European American and Asian Indian adolescents' academic competence and self esteem

PDF

Intergenerational conflict, family functioning, and acculturation experienced by Asian American community college students

Asset Metadata

Creator Zethren, Kristin Elisabeth (author)

Core Title A two-dimensional model of cognitive empathy: An empirical study

School Graduate School

Degree Doctor of Philosophy

Degree Program Psychology

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag OAI-PMH Harvest,psychology, cognitive,psychology, personality

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Newcomb, Michael (committee chair), Goodyear, Rodney (committee member), Willard, Dallas (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-550556

Unique identifier UC11334956

Identifier 3094405.pdf (filename),usctheses-c16-550556 (legacy record id)

Legacy Identifier 3094405.pdf

Dmrecord 550556

Document Type Dissertation

Rights Zethren, Kristin Elisabeth

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA