I .
$13.. ... . ..
. .2 w ..., 3, . . A... .
. .r 3%}? , . . .. . . . . _ .. ..
. LP? . -

. .._.I
.v 1.9"»). u ....

4:2:
“32%;?

- .'l‘§‘.
335%-

.- f‘
a

.sg'

-
r
-

..::

.V .
k: 1%
.55 .1
. _
"p

7. «mu...
...... .n:
.. w
a H... ...w» ....

y

c.
Q}

. {0.1. .
.lv... .

. Mu ... ., . ., 3, .1»
. .... ...«Mm: _ . a. . x:
L t ... . , . en. ti”
.ﬁWhmawmwx. Rxmﬁmwa . .. .. , a E. ww‘wwwm...
. ....nanwmxi ..gwwugmar- .
.ur La... . «4...: .erfmﬂh‘ .... _
uﬁlﬂuﬁu . 15:...«imw

- .3 . J: . .
. .
I». r v
. EN...
db. or»:

i
;

5
1"};
F. .
y,

-V

'I0
3

; l 3h:

. -' 1N:
. . $2,.
’ ".‘i

, rwf.

: 1 Pat!’

D ,
A. I- IMHJI .‘v

. .
... :1... . .
.hu... 1
. . '4 1 ”V”. D H"! ...I...-.
.1}... ﬁnhumzmr..iu Putx. I. .
ulna! .l 7 .

«4h.
W!!!“ .

 

nvx‘u

1......
{if .

 

 

‘ It! A... . . . ., . . .
. .l..|1 ......rﬂf... . .. . . . r . . . .. , . . , ...:h. J .
. . .. .....u $.94. .. . . r ._ . . .. ... . . . ., .. . u . .. .. ..fv...€..m.m.m .
._ ....UVKMTM . , . . ... . . . ... . . ... .
2 -. ..I...-W....Hx]».u.w....l.u.nr..4....ﬁn.?. . . a . . . A Autumn? ..l!
1 313.13.”..«4. H.- .u...r.ﬂ.ah‘vnl. . . . y . . .. , v P .

 

 

 

 

 

 

 

 

 

 

. .l’.

l 07:11.1

. ...bIX-Qal‘ldo' ‘

... .t I..!.vcr .‘
oVoIIIL

 

..vlv. a.-. .
13.1....
..uWo... 3311.145...“

Inn V‘ALV’.A§‘£iL-V

4:;
’éie ‘

:i.’

t

NH! 1 .
.VUVulruut“v!r

.Iovl

..j L;

.§ h...o.1.
7..-: v .r...
..PII '0 1 «1,91‘
..... ..

. L. . .w u...
. . z .. . .. {$1.121
. 90 o . . v I ~ .. r .0
h... .. .

. I.
"PLMernhnuﬁWJmhnudmw. .

) .- - . .
. : t A. .. ... .. x ...c “4.72: ..v..' -1
335...? ..uail-.l . _ w . - - o

 

"15338

7/

lllill'lllllllmlliilllll'llllllllilll

3 1293 01563 9176

 

 

 

 

 

 

 

 

 

This is to certify that the
dissertation entitled
Rorschach Responses in Older Adults
Differing in Residential Status:
Resource and Trait Theory Perspectives

presented by

Lidia Ann Domitrovic

has been accepted towards fulﬁllment
of the requirements for

PhD degree in Psychol DEL!

Aim 1m

Major professor

Date 8' 7" q 7

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

 

LIBRARY

Michigan State
University

 

 

 

PLACE DI RETURN BOX to tomato this checkout from your record.
TO AVOID FINES return on or baton date duo.

DATE DUE DATE DUE DATE DUE

.- P ' ~ A '
' ,5 . . ~
‘8 I! _.I-l-_. ‘ ___
_ Flat
l .\ ,' i '
l J as -
I

[—qu

       

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l

 

 

 

a ’le

 

 

 

 

 

‘1

 

 

 

 

 

 

: :\\ i: ‘

 

 

MSU I: An Afﬂnnatm Action/EM OppomMy lm
was»:

RORSCHACH RESPONSES IN OLDER ADULTS DIFFERING IN RESIDENTIAL
STATUS: RESOURCE AND TRAIT THEORY PERSPECTIVES

BY

Lidia Ann Domitrovic

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Psychology
1997

ABSTRACT

RORSCHACH RESPONSES IN OLDER ADULTS DIFFERING IN RESIDENTIAL
STATUS: RESOURCE AND TRAIT THEORY PERSPECTIVES

BY

LIDIA ANN DOMITROVIC

Rorschach responses are the product of cognitive processes inﬂuenced
by the underlying personality dynamics of the individual. Age-related changes in
Rorschach responding have been well-documented, as have performance
declines on numerous cognitive tasks. Resource theory posits that age-
associated decrements in information processing may account for these
cognitive differences. This study examined Rorschach responses in older adults
differing in residential status in light of both Resource theory and the five-factor
model of personality (Costa & McCrae, 1992).

Forty community-dwellers and forty care facility residents who had
sufﬁcient sensory and motor abilities to complete all tasks independently, and
who passed mental status criteria, participated. The Digit Symbol Substitution
subtest of the Wechsler Adult Intelligence Scale - Revised served as the
measure of information processing; personality was assessed with the NEO-FFI.
Each participant was administered the Rorschach and selected measures of
general cognition and depression. Differences in personality attributes as a
function of residential status were not expected, but it was predicted that
community-dwellers would process information more rapidly. It was also

hypothesized that faster information processing rates, higher scores on the

personality domains of Openness to Experience, Agreeableness, and
Extraversion, and lower scores on Neuroticism, would be associated with more
productive and higher quality Rorschach protocols.

Residents of care facilities were slower information processors than
community-dwellers, as predicted. Rorschach performance did not differ as a
result of residential status, when age and health status were controlled.
Information processing rate and Openness, both together (r= .58) and
independently (sr = .42 and .26, respectively), accounted for a signiﬁcant
proportion of the variance in a General Response Process factor variable
primarily composed of R, P, and M. Agreeableness was found to be a signiﬁcant
predictor of scores on a Perceptual Accuracy Factor (r = .29) principally deﬁned
by F+%, X+%, and X—%. Openness was signiﬁcantly associated with a factor
variable representing Holistic Integration (r= .28) characterized principally by W
and Zf. The results suggest that information processing rate and certain
personality factors inﬂuence at least some aspects of Rorschach response

formation in older adults.

For my parents, Helen and Vince Domitrovic

iv

ACKNOWLEDGEMENTS

First, and foremost, I extend my deepest appreciation to all the men and
women who gave freely of their time, and of themselves, to serve as study
participants. Much gratitude is reserved for the organizations which allowed me
to recruit participants, including the Seventh Day Adventists church of Lansing,
the Grand Rapids Home for Veterans, Burcham Hills Health Care Center and
Whitehills. Many additional thanks to the following individuals who served as
study liaisons: Laurie Symons, Gary Davis, Brenda Ruddy, and Cathy Carter.

To my committee, I extend my heartfelt appreciation for the guidance and
helpful suggestions that were always readily available, especially during the most
challenging periods of what was often a frustrating project. Special thanks to Dr.
C. L. Winder, who graciously donated his time and expertise to serve on my
committee when unexpected circumstances arose. Many, many thanks to Dr.
John Hurley, who was always available for words of support, cogent comments,
and a Pepsi. To Dr. Abeles: I cannot ﬁnd the words to express my gratitude for
the guidance and support you provided throughout my graduate training; I will
always be grateful to you for all the valuable lessons you taught me.

Last, but far from least, I extend much appreciation to my husband,
Joseph John Cimer, for his never-ending support, encouragement and love.

V

TABLE OF CONTENTS

List of Tables
List of Figures
Introduction

Literature Review
Rorschach Responses In Older Adults
Historical Conclusions
Methodological and Interpretive Issues
Contemporary Considerations and Findings
Resource Theory and the Rorschach
Theoretical Tenets and Findings
Inﬂuence on the Rorschach
Trait Theory and the Rorschach
Personality Findings
A Trait Theory Perspective
Residential Status and Older Adults
The Residential Care Population: General
Characteristics
Mental Health Issues
Psychological Effects of Residential Status
Further Considerations

Methods

Participants

Procedure

Tests and Measures
Digit-Symbol Substitution Subtest, WAlS-R
Exner Comprehensive System for the Rorschach
Geriatric Depression Scaled (GDS)
Mini Mental Status Exam (MMSE)
NEO Five-Factor Inventory (NEO-FFI)
Reading Subtest, WRAT-3

Scoring

vi

viii

vii
Hypotheses and Data Analysis

Results

Sample Selection and Response Rate

Descriptive Statistics
Demographic Data
Cognitive and Information Processing Measures
Personality Variables
Rorschach Responses

Relationships Between Variables: Statistical Analyses
Relationships Among Information Processing Rate,
Mood, and Demographics
Rorschach Variables and lnfonnation Processing
Rate
Rorschach Performance, Information Processing
and Personality

Additional Analyses

Discussion

Sample Selection and Response Rate

Demographic Variables

Depression

Information Processing Rate and Cognition

Personality

Rorschach Performance

Relationships Between Variables
Information Processing Rate, Mood, and
Demographics
Rorschach Variables and Information Processing Rate
Rorschach Variables and Personality
Rorschach Performance, Information Processing
and Personality

Further Considerations

Methodological Issues

Final Considerations

References
Appendix A: Consent Form

Appendix B: Rorschach Determinants: Exner’s (1990)
Normative Values for 700 Nonpatient Adults

35

39
39
41
41
48
50
53
60

60
61

64
7O

73
73
76
80
81
83
85
88

88
91
93
97
100
101
102
106
128

129

Table 1:

Table 2:
Table 3:

Table 4:

Table 5:
Table 6:
Table 7:
Table 8:

Table 9:

Table 10:

Table 11:

Table 12:

Table 13:

Table 14:

LIST OF TABLES

Residential and Gender Differences in Age
and Education

Residential Differences in Medical History
Frequency of Diagnostic Category

Residential Differences in Information Processing
Rate and Cognition

Intercorrelations Between Cognitive Measures
NEO-FFI Descriptive Data

Gender Differences in Personality Domains
Intercorrelations Among Personality Domains
Rorschach Determinants: Descriptive Data

Residential Status Differences in Rorschach
Determinants

Rorschach Determinants: Normative Versus
Sample Differences

Intercorrelations Between Standardized
Rorschach Variables

Factor Pattern Matrix (Promax Solution) for
Rorschach Variables

Intercorrelations Among Rorschach Factor
Variables

viii

42

45

49
50
51
52
53

54

55

56

58

59

59

Table 15:

Table 16:

Table 17:

Table 18:

Table 19:

Table 20:

Table 21:

Table 22:

Table 23:

Table 24:

ix

Regression: Rorschach ACC on Personality
Variables

Regression: Rorschach ORG on Personality
Variables

Regression: Rorschach RSP on Personality
Variables

Regression: ACC on the lnforrnation Processing -
Personality Interaction

Regression: ACC on Information Processing and
Personality - Main Effects

Regression: ORG on the lnforrnation Processing -
Personality Interaction

Regression: ORG on Information Processing and
Personality - Main Effects

Regression: RSP on the Information Processing -
Personality Interaction

Regression: RSP on lnfonnation Processing and
Personality - Main Effects

Rorschach Performance and Selected
Demographics: Correlations

63

63

64

65

66

67

68

69

70

71

Figure 1:

LIST OF FIGURES

Proposed Model of Variable Interaction

37

RORSCHACH RESPONSES IN OLDER ADULTS DIFFERING IN RESIDENTIAL
STATUS: RESOURCE AND TRAIT THEORY PERSPECTIVES
INTRODUCTION

The Rorschach Inkblot method (Rorschach, 1942) is one of the most
popular and common techniques employed in the psychological assessment of
adults across diverse clinical settings (Piotrowski & Keller, 1989; Sweeney,
Clarkin, & Fitzgibbon, 1987). Yet, to date, it has generally been dismissed as an
effective assessment measure with older adults, no doubt because the research
concerning the Rorschach response process in this population has been
equivocal (La Rue, 1992; Scogin, 1994). Our lack of knowledge in this regard is
unfortunate, and needs to be remedied for two reasons: (a) The population of
persons 65 years of age or older is rapidly growing and will represent 17% of the
US. population by the turn of the century (Albert, 1988), and (b) the biological
and social facets of aging, which include physiological changes, increased risk
for chronic diseases (especially dementia), and role changes and/or loss, can
impact psychological factors such as subjective well-being, mood, and coping
skills (La Rue, 1992; Wiggins, 1994). The net effect: an ever-growing segment
of the population requiring accurate assessment and treatment of the sequelae
associated with the biopsychosocial challenges of aging, demanding, in turn, an
accurate understanding of how those challenges may manifest on assessment

tools such as the Rorschach.

2

Rorschach responses are the product of perceptual and cognitive
processes inﬂuenced by the underlying personality dynamics of the individual - in
sum, a result of both perception and representation (Blatt, 1990; Rapaport, Gill,
& Schafer, 1968, chap. 9; Weiner, 1986). Changes in cognitive functioning with
advanced age have been well-documented (Salthouse, 1985; 1991, 1992a).
Recent evidence has suggested that reduced rates of information processing
within the central nervous system can explain the performance decrements of
older adults, relative to younger adults, on a number of cognitive tasks (Cerella,
1985; Salthouse, 1991, 1992a). The present study examined the Rorschach
response process in older adults, living in both the community and in residential
care facilities, within the context of resource theory, which postulates a decline in
the resource of time with increasing age (Salthouse, 1985). Speciﬁcally,
limitations in the rate of processing information as manifested on both a standard
cognitive task and on the Rorschach response process were investigated. As
noted by both Kahana (1978) and Hayslip (1981), there is a need for research on
the factors that uniquely inﬂuence the response process in the elderly.

The mechanism of response formation, however, suggests that potential
age effects could arise not only at the level of perception, but at the level of
representation, or as a result of the interaction of these two processes. While
the accumulated evidence suggests that the ﬁve most commonly replicated
personality domains of Neuroticism, Extraversion, Openness to Experience,

Agreeableness, and Conscientiousness remain stable with age (Costa et al.,

3
1986; Costa & McCrae, 1980, 1992; Costa, Metter, & McCrae, 1994), information
regarding how these constructs may mediate the Rorschach response process is
unknown. Thus, Rorschach responses were also investigated in the context of
trait theory, so that the inﬂuence of personality factors, as well their potential
interaction with reduced processing resources, could be examined in older
aduﬂs.

Including a sample of older adults residing in long-term care facilities was
of interest for several reasons. The most extensive investigations of Rorschach
responses in the elderly (Ames, 1960, 1966, 1974; Ames, Metraux, Rodell, &
Walker, 1973), which were conducted with both community—dwelling and
institutionalized older adults, have never been replicated. This is problematic
because, for years, these atheoretical, and often methodologically ﬂawed
studies, were touted as classics in the ﬁeld (Peterson, 1991). Given that: (a)
The population of adults 65 years and older who reside in care facilities is rapidly
growing (Streim & Katz, 1996), (b) the lifetime risk of at least temporary
placement is anywhere from 25% to 50% (Campion, Ban, & May, 1983;
McConnel, 1984), and (c) substantial mental health needs, demanding accurate
assessment, management, and treatment, exist in this population (Streim,
Rovner, & Katz, 1996; Tariot, Podgorski, Blazina, & Leibovici, 1993), a
contemporary, theoretical re-examination of the Rorschach was long overdue.

The Digit Symbol Substitution subtest of the Wechsler Adult Intelligence

Scale - Revised (WAlS-R) provided a measure of information processing rate,

4
while personality dimensions were assessed with the NEO-FF I. Although
community-dwelling older adults do not necessarily suffer from a greater
frequency of depression relative to their younger counterparts (Myers et al.,
1984), depressive disorders are the second most common psychiatric diagnosis
in residential care facilities (Streim & Katz, 1996). Since depression can
inﬂuence both information processing rate (Kaszniak & Ditraglia Christenson,
1994; La Rue, 1992) and the content of Rorschach protocols (Exner, 1986), level
of depressive symptomatology was also assessed by a self-report measure.
Rorschach protocols were scored with the Exner Comprehensive
System (1986), the most widely and rigorously studied scoring and interpretive
system available.

LITERATURE REVIEW
Rorschach Responses in Older Adults
Historical Conclusions
Early investigations of the Rorschach with older adults tended to support
Rorschach's (1942) hypothesis that normal aging was associated with the
following "pathognomic" signs: (a) a reduced ability to utilize inner resources and
weaker reactions to emotional stimuli as represented by a decrease in human
movement responses relative to chromatic color responses, (b) a decrease in
perceptual acuity and intellectual efﬁciency as represented by vague or poorly
envisioned percepts, and (c) a marked response stereotypy, presumably

demonstrating a narrowing of interests. The majority of the studies were

5

conducted prior to the original publication of Exner's Comprehensive System
(197411986), a strategy for scoring and interpreting the Rorschach which evolved
from the best aspects of the ﬁve extant systems in common use prior to 1957. In
many of the studies, information regarding administration, scoring, and the
nature of any normative data used in comparisons were not offered;
nonetheless, a very crude comparison of the results is possible because many of
the determinants reported had similar meanings across systems. Thus, in
reviewing the ﬂurry of literature largely published during the 1950's and 60's,
Reichlin (1984) found that the Rorschach protocols of older adults were typiﬁed
by fewer total responses (R), increased reaction times, elevations in the number
of responses formed by global stimulus perception (W%), and increases in
animal content (A%). An increase in responses determined solely by the form or
contour of the blots was noted (F%), but was accompanied by a concomitant
decrease in the quality of those forms (i.e., lower F+%). Lastly, fewer of the
responses were those commonly perceived by younger adults (Popular or P%),
and fewer responses were determined by human-like activity or movement (M).
Overall, a decline and constriction of cognitive resources with increasing age
was suggested by the Rorschach protocols of the older adults so examined
(Panek, Wagner, & Kennedy-Zwergel, 1983; Prados & Fried, 1947).
Methodological and Interpretive Issues

The studies were plagued, however, by methodological difﬁculties.

Klopfer (1946) collapsed his sample of 30 community-dwelling and 20 nursing

6

home residents when making conclusions regarding the healthy elderly, even
though there was evidence of severe psychopathology in the latter. The same
institution provided the sample studied by Davidson and Kruglov (1952), who, in
addition to replicating Klopfer's results, observed: (a) an inability to name or
identify the percept; (b) identiﬁcation of the percept by its function, rather than by
name; (0) identiﬁcation of what the blot was not; and (d) immediate rejection of a
given response. Note, however, that the latter four behaviors are frequently
observed on the Rorschach protocols of persons with brain damage (see
Goldfried, Stricker, & Weiner [Chapter 11], 1971, for a review). Indeed, the
changes in the Rorschach variables of reaction time, total R, M, and P%,
traditionally posited to be associated with increasing age, have been long
observed in the protocols of neurologically impaired persons (Hughes, 1948;
Piotrowski, 1936, 1940). Modern research utilizing the Exner system to score
the Rorschach protocols of persons with neurologic disease and/or injury has
found changes in the aforementioned variables, as well as in scores (F+%, X+%,
and X-%) addressing the perceptual accuracy of form dominated responses
(Ellis & Zahn, 1985; Perry, Potterat, Auslander, Kaplan, & Jeste, 1996).

Caldwell's (1954) study of 47 institutionalized women falling in the age
ranges between 61-75 and 76-92 found the typical pattern of results identiﬁed by
Reichlin's (1984) review and observed by Klopfer (1946), and Davidson and
Kruglov (1952); computed correlations between age and Rorschach

determinants, however, reached statistical signiﬁcance only for the variables of

7

M and the number of good original responses (0+), a determinant no longer in
use in the Exner system. Taking a cue from Caldwell's conclusions that
Rorschach differences with age might be a result of unidentiﬁed factors which
may have next to no relationship with age, Ames and her colleagues conducted
extensive studies of the Rorschach response process in older adults (1960,
1966, 1974; Ames et al., 1973). Following participants longitudinally, collecting
information regarding education and employment history, dividing groups by
cognitive capacity into presenile and senile (although on the basis of Rorschach
performance itself), and evaluating the effects of residential status, afforded the
authors the opportunity to observe that poorer quality Rorschach responses (i.e.
increases in F%, F-%, and A%) were more related to demographic factors, such
as low SES, institutionalization, and intellectual capacity, than to age itself.
Similarly, measured intelligence on the Wechsler Adult intelligence Scale was
the most signiﬁcant predictor of Rorschach response quality in a sample of
community-dwelling older adults (Eisdorfer, 1963).
Contemporary Considerations and Findings

Despite increased methodological sophistication, contemporary studies
have tended to generally replicate the equivocal ﬁndings found in the older
research. In a sample of healthy, non-demented, elderly Finnish followed
longitudinally over 10 years, no changes were found in F+%, P%, or animal
movement with advancing age (Mattlar, Knuts, & Virtanen, 1985). Yet,

Shimonaka and Nakazato (1991) found age-related changes in total response

8
number, P%, shading determinants, rejection of blots, poor form quality (F-%),
and content range in their 10 year longitudinal study of 52 healthy Japanese with
an average age 76 at study initiation. These results, which were scored and
interpreted according to the Klopfer system, led the authors to conclude that
advancing age is associated with decreases in productive and creative thinking,
cooperativeness, and interest in society. In contrast, Gross, Newton, and Brooks
(1990), scoring the protocols of a healthy, community—dwelling, American sample
according to the Exner (1986) System, found that older adults with higher IQ
scores tended to produce more active Human Movement (Ma) responses than
the older adults with lower IQ scores. No differences were found with increasing
age in the other determinants in this study, which controlled for the potentially
moderating effects of IQ, sex, depression level, and medication amount on
Rorschach responses.

Gross et al. (1990) also screened study participants for auditory and visual
deﬁcits to control for any possible effects due to the sensory decrements which
frequently accompany old age (see Fozard, 1990, for a review). Klopfer (1974)
argued that the characteristically vague perceptions of older adults were due to
their inability to see them clearly. Eisdorfer (1960a, 1960b), however, provided
evidence that, as long as vision is corrected, Rorschach responses should not be
compromised by normal visual changes, even in those adults with extremely

poor vision.

9

The increased consideration of demographic and physiological factors
which evolved within this research tradition should be noted. It does appear that
many of the so called age effects found in the early studies were largely a result
of unidentiﬁed (or ignored) dementing processes; indeed, studies investigating
Rorschach responses among the healthy elderly and those classiﬁed as
demented on the basis of other measures, found that the two groups could be
distinguished on the basis of determinants such as F+%, A%, and P% (Insua &
Loza, 1986; Kettell, 1976). But little can be conﬁdently stated about the
response process in older adults facing the normal, rather than the pathological,
challenges of aging. The differing Rorschach methodologies employed across
studies, as well as the potential for cultural variability introduced by the
contemporary investigations of Japanese and Finnish populations, have proven
to be less than illuminating, the only seeming commonality across studies
represented by a proclivity to atheoretical research. Indeed, some authors have
insisted that the only conclusion that can be drawn from the accumulated work is
that "extreme caution" should be used in making clinical use of the Rorschach
with older adults (Hayslip & Lowman, 1986; Scogin, 1994).

Resource Theory and the Rorschach

Theoretical Tenets and Findings

A truism in the geriatric literature was nicely summarized by Sharps, who
stated that, "Cognitive declines with age have been documented so thoroughly

that few investigators question their existence or importance" (1990, p. 298).

 

10

Resource theory, which addresses this issue, was the formal appellation applied
by Salthouse (1985, 1991) to "...the belief that age differences in certain
cognitive tasks are not due to impairments in task-speciﬁc components or
strategies, but instead are at least partially attributable to an age-related
reduction in the quantity of some type of general-purpose processing resource
considered necessary for efﬁcient functioning in a broad assortment of cognitive
tasks" (Salthouse, Kausler, & Saults, 1988, p. 158). Salthouse (1985, 1988a,
1988b,1991) proposed three broad classiﬁcations with which processing
resources could be most easily conceptualized: (a) time, as represented by rate
of processing; (b) space, as represented by working memory capacity; and (c)
energy, as represented by attentional capacity.

To date, the majority of related research has investigated limitations in
time, as indexed by reduced rates of processing speed, most likely because a
general slowing of behavior with increased age is one of the most reliable
observations in both everyday behavior and laboratory tasks (Bayles, Kaszniak,
& Tomoeda, 1987; Mazaux et al., 1995; Myerson, Hale, Wagstaff, Poon, &
Smith, 1990). Resource theory, however, asserts that speed is a fundamental
property of an individual's information processing system, representing the
maximum rate at which one is capable of performing, given central, cognitive-
computational parameters, rather than reﬂecting the tempo at which one typically
works (Salthouse, 1985), or the time taken by peripheral sensorimotor processes

(Cerella, 1985). Cross-sectional studies have demonstrated that reduced rates

1 1

of information processing have been associated with performance decrements in
older, as compared to younger adults, on tasks assessing object naming latency
(Thomas, Fozard, & Waugh, 1977); retrieval of semantic information from long-
terrn memory (Madden, 1985); signal detection, visual discrimination, and
temporal-anticipation (Salthouse & Somberg, 1982); perceptual closure
(Salthouse & Prill, 1988); spatial abilities and reasoning (Hertzog, 1989; Schaie,
1989; Salthouse, 1987, 1993); working memory (Salthouse & Babcock, 1991;
Salthouse, 1992a); symbolic arithmetic (Salthouse & Kersten, 1993); serial and
paired associate memory (Salthouse, 1993; Salthouse & Coon, 1993); and study
time, decision time, and decision accuracy (Salthouse, 1994). The age-related
declines in digit symbol performance have been found to be largely independent
of both years of education and self-reported health status and persist despite the
use of similar strategies between younger and older adults in task performance
(Salthouse, 1992b). Furthermore, neither practice (Salthouse & Somberg,
1982), nor unlimited time to perform (Salthouse, 1994), entirely abolish age
differences.
Inﬂuence on the Rorschach

Upon initial appraisal, one might wonder how a theory of cognitive aging
could contribute to knowledge of the Rorschach response process in older
adults, especially given the traditional tendency to view Rorschach proﬁles as
projections of personality (Peterson & Schilling, 1983). Rorschach response

formation, however, demands a "...coordinated series of cognitive activities in a

12
total process of forming, testing, critically evaluating, and communicating
perceptual hypotheses" (Korchin & Larson, 1977, p. 159). Thus, Rorschach
response formation may be considered dependent on ﬂuid intellectual ability,
since the transformation of information is demanded, and it is ﬂuid, rather than
crystallized abilities, that show the greatest decrements in performance with age
(Salthouse, 1985). As a function of the three perceptual elements of sensation,
memory, and association (Rorschach, 1942), there is no reason to suspect that
the Rorschach response process would be any more immune to changes with
age than any other fluid intellectual ability.

Responding to an inkblot bears little resemblance to the types of tasks (for
example, forced choice reaction-time, semantic priming, or memory search
tasks) which have been hypothesized to be compromised in the elderly due to
reductions in processing speed. Other than an increase in response time, it
seems difﬁcult to ascertain where reduced processing rates may manifest on a
Rorschach record. Because of the interdependencies of the human information
processing system, however, a change in speed with age would likely result in
age differences in many other complex, cognitive areas, thus accounting for
qualitative, and not just quantitative differences in performance (Salthouse,
1994). For instance, controlling for perceptual speed has attenuated the age
differences in inductive reasoning tasks (Salthouse & Mitchell, 1990), and in
measures of reasoning and spatial ability (Salthouse, 1993). Bromley (1967)

argued that decreases in processing speed would decrease the number of

1 3
conceptual transformations that can be achieved, thereby also decreasing the
likelihood of unique or high quality transformations. If this is the case, qualitative
changes in the Rorschach with advancing age could be expected.

Given both the theoretical foci of the present investigation and the
literature examining the Rorschach with advancing age, changes would be
expected in those determinants and scores considered indices of information
processing effort, efﬁciency, and cognitive mediation (Exner, 1986). These
include total response number (R), the frequency of Popular responses (P), the
frequency of human movement responses (M), and Whole responses (W).
Additional Rorschach cognitive processing information is provided by (a) Zf, an
index of the complexity of response mediation, or the effort expended to
organize the stimulus ﬁeld; (b) F+%, which represents those percepts that
demonstrate a “superior-overelaborated” use of the contours, resulting in
responses that the majority of individuals would have no difﬁculty seeing; and
(c) X+% and X-%, which represent two components of the Extended Form
Quality distribution (the third being Xu%); X+% represents perceptual accuracy
or conventionality, percepts that do not violate the form demands of the blot,

while X-% reﬂects a distorted or statistically rare use of form.

Trait Theogy and the Rorschach
Personality Findings

In addition to perception, formation of Rorschach responses is also

dependent on representation; while representation itself is based on perception,

14
it goes beyond the latter process (Blatt, 1990). In creating a problem solving,
decision making situation, the Rorschach method provides an opportunity for
persons to respond as they usually do to the problem solving and decision
making arenas of their lives, potentially revealing much about their personality
style, needs, attitudes, and concerns - hence responses are determined by the
meaning or representation assigned to the apprehended percept (Weiner, 1994).
Research examining personality factors and the Rorschach has tended to
compare constructs deﬁned by the clinical and validity scales of the Minnesota
Multiphasic Personality Inventory (MMPI; Hathaway & McKinley, 1940) with
those purported to assess the same dimensions on the Rorschach. Archer and
Krishnamurthy (1993), in their review of the 37 studies published between 1943
and 1992 which examined such interrelationships in adult samples, concluded
that, "...despite their common functions of assessing personality and
psychopathology" (p. 285), only limited or minimal relationships between the two
measures were indicated, a result echoed by others (Acklin, 1993). When
signiﬁcant relationships were found, they tended to be observed in studies that
were theory driven or had explicit research hypotheses (Exner, Annbruster, &
Mittman, 1978). Cross-validated ﬁndings have appeared to be the exception,
however. Research in which the external construct validity of Rorschach
personality variables was examined with other well validated measures of the
same factors found that, with few exceptions, the Rorschach variables thought to

represent ego strength, self-esteem, social desirability, and locus of control failed

15
to correlate signiﬁcantly with self-report measures of those same constructs
(Greenwald, 1990). It has been suggested that the lack of relationship is to be
expected, since responding to the Rorschach depends on processes not tapped
by self-report measures (Acklin, 1993; Weiner, 1993). Meyer (1993), however,
found that Rorschach indices of depression (DEPI), hypervigilant scanning (HVI),
and obsessive cognitive processing style (OBS), and MMPI-2 (Hathaway &
McKinley, 1989) measures of the same respective processes (i.e., scores on
Scales 2, 6, and 7), correlated signiﬁcantly when Rorschach response frequency
(R) was high (R > 29); no correlations were found between the MMPI-2 and
Rorschach variables for the general sample when response frequency was
collapsed. Thus, Meyer (1993) suggested that, at least in his sample of 90 adult
patients with a mean age of 33.5 years, R functioned as a dimension of
openness or ability to acknowledge dysfunction and reveal symptomatology,
especially since, in low R records, the MMPI-2 scales tended to contradict the
Rorschach constellations.

The MMPI in all its versions tends to provide a report of an individual's
current affective and psychological status, rather than an index of underlying,
stable, personality traits. Indeed, from a clinical perspective, the test appears to
be most informative when persons are experiencing state-related needs and
presses. While psychological turmoil and distress obviously inﬂuence Rorschach
responses, Greenwald's (1993) study appears to be the only one that has

assessed how normal variations in personality manifest on the Rorschach.

16

Arguably, the clinical utility of the Rorschach with older adults will only be
realized when it is determined what can be expected in a population of healthy
elderly, and this necessarily includes an investigation of how variations in normal,
non-pathological, personality variables inﬂuence Rorschach responses.
Mt Theory Persggtiv_e

Trait theory (Costa & McCrae, 1980), a comprehensive and widely
researched theory of personality, seems to be a logical point from which to
consider the issue of normal personality variation and the Rorschach for several
reasons: (a) Cross-validation studies examining the Rorschach method have led
to the suggestion that research which is theory or hypothesis driven is more
informative (Archer & Krishnamurthy, 1993), and (b) trait theory has proven to be
remarkably robust. Years of personality research has suggested that ﬁve global
domains are "both necessary and reasonably sufﬁcient" (McCrae & Costa, 1986,
p. 1001) for describing the major features of personality (Digman, 1990; Norman,
1963). Costa and McCrae (1985, 1992) labelled these ﬁve factors Neuroticism,
Extraversion, Openness to Experience, Agreeableness, and Conscientiousness.
Each domain can be further subdivided into more speciﬁc facets, but generally,
Neuroticism refers to emotional stability, Extraversion to sociability and
assertiveness, Openness to intellectual curiosity and independence of judgment,
Agreeableness to the interpersonal tendency to be altruistic and empathic, and

Conscientiousness to the tendency to be obsessive (Costa & McCrae, 1992).

17

Longitudinal studies have demonstrated that these ﬁve general domains
or traits remain relatively stable across the life-span, although speciﬁc facets do
show situational variation (Costa & McCrae, 1980; Costa et al., 1986). Thus,
Rorschach age changes as a result of alterations in personality functioning would
not be expected. Of interest, however, is the manner in which these ﬁve traits
inﬂuence the Rorschach response process. Costa and McCrae (1986) did
investigate longitudinal age changes in the Holtzman Inkblot Technique (HIT;
Holtzman, 1961), a projective test in which response number is a constant,
rather than a variable, since only one response per card is permitted. In this
cross-sectional study of persons aged 25 to 90, little evidence of "maturational"
change in the HIT variables was discovered, nor was there any relationship
between speciﬁc HIT variables and Neuroticism, Extraversion, and Openness as
assessed on the NEO Inventory (Costa & McCrae, 1986).

Myers’ (1993) data on the MMPI-2 and the Rorschach suggests that
individuals high in Openness and low on Neuroticism would provide, at the very
least, a greater number of responses than individuals low on Openness and high
on Neuroticism. Given that Extraversion is thought to represent an individual’s
temperamental level of energy and enthusiasm, and that Agreeableness is a
dimension that contrasts trust and cooperation with cynicism and antagonism
(McCrae, 1991; McCrae & John, 1991), it would seem that persons high in both

factors would demonstrate more eagerness and less apprehension when

18
approaching the task of the Rorschach, yielding more productive protocols both
qualitatively and quantitatively.

Indirect evidence for the latter assumption can be found in the study by
Dollinger and Leong (1993), which found that people high in Agreeableness,
Openness and Extraversion were more likely to volunteer personal information,
as well as consent to be followed longitudinally in a hypothetical psychological
experiment. Other research has demonstrated openness to experience to be
highly negatively correlated with political conservatism (Riemann, Grubich,
Hempel, Mergl, & Richter, 1993); since the features of conservatism include a
lack tolerance for the unfamiliar and a dislike of uncertainty, persons low in
Openness (and therefore high in conservatism) could be expected to produce
less informative and/or productive Rorschach protocols as a result of the
inherent ambiguity of the task. Individuals high in Openness have also been
observed to perform better on tasks assessing divergent thinking skills, tasks
which demand unusual or original responses and/or as many responses as can
be generated (i.e., productivity; McCrae, 1987), arguably all abilities necessary to
Rorschach response formation. Lastly, persons scoring higher on Openness
were found to create personal photo essays (a type of projective task) that were
more self-reﬂective, creative, and interesting than those produced by persons
low in Openness (Dollinger & Clancy, 1993). The paucity of direct data,
however, as well as the lack of any examination of the ﬁve factor model with the

Rorschach, suggests that further research is necessary.

Residential Status and Older Adults
The Residential Care Pogulation: General Characteristics

Since the early 1970’s, the proportion of the population aged 65 years
and over who live in residential care facilities, at any one moment in time, has
remained at about 5% (Brody, 1993a; Butler, Lewis, & Sunderland, 1991). As
already noted, however, this point prevalence estimate of residency is a grave
underestimate of the 25% to 50% of elderly who will eventually experience at
least one episode of institutional long term care (McConnel, 1984). Moreover,
since the risk of institutionalization rises with advancing age (Butler et al., 1991),
increases in life expectancy (Manton & Stallard, 1991), as well as in the cohort of
the “old-old” (i.e., persons age 75 and older) (Brody, 1993b), have contributed to
a residential care facility population that is rapidly growing: Numbering 1.6 million
in 1989, the nursing home population is expected to reach 2 million by the year
2000 and more than double to 4.6 million by the year 2040 (Butler et al., 1991).
Thus, institutionalization is fast becoming one of the commonplace challenges
facing people as they age.

According to the 1996 US. Senate Special Committee on Aging report,
residents of care facilities tend to be predominantly Caucasian (93%), women
(approximately 80%), and very old (45% are 85 years and older), with an
average age of 79 years. Residential facilities provide long-term care for older
adults with chronic illness and/or disability (average length of stay of 2.5 years),
as well as short-term rehabilitative and convalescent services to those elderly

19

20

recovering from acute injury or sickness (Streim & Katz, 1996; Streim et al.,
1996). Placement can also result, however, as a result of non-medical reasons,
such as social isolation, economic considerations, and general apprehension
about the ability to adequately care for oneself (Osterweil, Mulford, Syndulko, &
Martin, 1994; Winocur & Moscovitch, 1990). Cognitive impairment may, or may
not, be a contributing factor to placement (Holland & Rabbitt, 1991; Ostenrveil et
al., 1994). Thus, just like their community dwelling counterparts, the elderly
residing in care facilities are quite heterogeneous in terms of both their individual
characteristics and their service needs (Feifel & Strack, 1987; Manton, Liu, &
Cornelius, 1985; Tobin, 1988).
Me_ntal Heaih Issues

As a group, older adults admitted to residential care facilities since 1989
have been more ill than in the past, with multiple, chronic medical comorbidities
and disabilities the norm (Butler et al., 1991; Manton, Cornelius, & Woodbury,
1995). Not surprisingly, then, such facilities have become a major locus of
geriatric health care in the United States (Liptzin, 1992). In no other branch of
medicine has that focus been more welcome than in the area of mental health.
Estimates of diagnosable psychiatric disorders and their attendant behavioral
disturbances (e.g., confusion, withdrawal, agitation) in the residential care
population run from 80% to 90%; the majority (from 50% to 75%) represent
dementing illnesses, primarily Alzheimer’s disease, followed by depressive

disorders (Liptzin, 1992; Streim & Katz, 1996; Streim et al., 1996; Tariot et al.,

21
1993). Despite the need, however, psychiatric services are often inadequate
and in the past, frequently led not only to misdiagnosis and neglect, but to the
misuse of psychotropic medications and/or physical restraints as behavioral
management tools (Streim & Katz, 1996; Tourigny-Rivard & Drury, 1987).

In 1987, Congress addressed the issue by enacting the Nursing Home
Reform Amendments as part of the Omnibus Budget Reconciliation Act, Public
Law 100-203, a public policy initiative designed to “...improve the lives of nursing
home patients by explicitly recognizing the importance of psychological and
social domains of their lives, restricting the use of psychotropic medications
and physical restraints, and advocating for specialized rehabilitative services to
attain or maintain patients’ highest practicable physical, mental, and
psychosocial well-being” (Streim et al, 1996, p. 921). Accordingly, “the facility
must ensure that a resident who displays mental or psychosocial adjustment
difﬁculties receives appropriate services to correct the assessed problem”
(Health Care Financing Administration, 1991, p. 48,896 as cited by Streim &
Katz, 1996).

Psychological disturbances in residents of care facilities often have
complicated presentations, given the potential mixture of cognitive impairment,
multiple medical diagnoses (and hence medications) and behavioral symptoms
that can arise (Liptzin, 1992). Short-say residents, who are generally younger
and more likely to have been admitted from an acute care hospital with a primary

diagnosis of hip fracture, stroke or cancer, are less likely to suffer from

22
irreversible cognitive impairment or incontinence (Engle & Graney, 1993; Streim
& Katz, 1996). The mental health care needs of the latter population, who are
discharged home or die after relatively brief nursing home residencies, often
differ from those of the longer-term inhabitants, in whom managing behavioral
disturbances are of greatest import. In short-stay residents, screening of
potential cognitive impairments, treatment of depression and/or anxiety, and
helping residents adjust to the sequelae of a new illness or disability, are more
likely to be the foci of mental health services (Streim & Katz, 1996).

Obviously, whatever the particular need, intervention planning for the
management of nursing home residents with psychiatric or behavioral problems
demands a comprehensive assessment integrating information regarding all
aspects of an individual (Liptzin, 1992). Given their training in, and appreciation
of, the biopsychosocial determinants of behavior, clinical psychologists would
seem poised to make unique contributions as mental health providers in
residential care settings. Yet, a search of the Psychinfo database from the
1970’s to the present revealed that, to date, issues in this population have largely
been addressed by social work, nursing, and psychiatry (see for example Brody
& Semel, 1993; IngersoII-Dayton, 1993; Streim et al., 1996). A psychological
perspective is thus sorely needed.

Lezak (1995, p. 8) has stated that psychological assessment provides
the useful functions of diagnosis, patient care (including considerations about

management and planning), treatment (for developing programs and for

23

evaluating their efﬁcacy), and research. It has been asserted that the
information the Rorschach delivers regarding “...an individual’s perceptual,
cognitive, emotional, and social processes is neither matched nor approximated
by any other psychological technique” (Howes, 1981, p.348), making it a useful
adjunct to any psychological assessment. It is an especially attractive tool in
geriatric assessment, both because of its ease of administration, and because it
does not penalize an individual for a lack of testwise skills, or for decrements in
coordination or speed of response (Hayslip & Loman, 1986; Kahana, 1978). The
aforementioned characteristics may also make the Rorschach a useful measure,
either as predictor or criterion, in treatment studies, given the long overdue
interest in evaluating everything from psychotherapy (Salamon, 1986) to
empowerment (Oluwafemi Agbayewa, Ong, & Wilden, 1990), to behavioral skills
training (Hoyer, Lopez, & Goldstein, 1982; Vaccaro, 1990) in the older residential
care facility population. But, before it can be used with any conﬁdence, either
clinically or empirically, information regarding if, and how, the Rorschach
response process may change as a result of age and/or residential status is
desirable.
Psychological Effects of Residential Status

Entering a residential care facility is likely one of life’s least pleasant
experiences. Intuitively, leaving one's home, belongings, and community to
move in with strangers, many of whom one will be dependent upon, must be an

intimidating and often frightening experience. Upon admission, older adults must

24
cope with the cause(s) of their placement (for example, deterioration in health,
new disability), alterations in lifestyle, psychological separation (Dye, 1982), a
social organization with new and distinctive social processes (Moos & Lemke,
1985), and changes in one’s conceptions of self, body, and biographic time
(Golander, 1995). As Dye (1982) stated, “It is an experience prodromal of death
and yet without the ﬁnality or the consciouslessness of death” (p. 533). Given
the stresses of an often sudden change in residence on an already vulnerable
individual, it is not surprising that placement in care facilities is frequently
associated with increased morbidity, disorientation, and mortality (Butler et al.,
1991)

The question remains, however: Does institutionalization exert unique
effects on those older adults that have successfully adapted to (or at least
accepted) their new environments, once thus settled? In terms of the Rorschach
response process, the answer would seem to be yes. As previously noted, the
early studies investigating test performance, which relied heavily on older adults
residing in care facilities, generally found diminished levels of emotional
responsivity, poorer integrative abilities, lower perceptual accuracy, and
decreased intellectual efﬁciency and productivity with advancing age (Ames,
1960, 1966; Ames et al., 1974; Caldwell, 1954; Davidson & Kruglov, 1952;
Klopfer, 1946). Of course, as also already noted, these pre-Exner system
studies were fraught with methodological difﬁculties, not the least of which was

lack of control for cognitive status. Ames’ (1973) better designed study found

25
that institutionalization was related to poorer quality Rorschach protocols, even
when age, sex, and socioeconomic status were controlled. Unfortunately, while
Ames did divide the study participants into presenile and senile groups, she did
so on the basis of their Rorschach protocols, a rather circular method;
furthermore, the relationship between residency and cognitive status was not
reported. It is thus possible that cognitive impairment, rather than place of
residence per se, contributed to her ﬁndings.

Residential status has been found to be an important variable in more
contemporary studies researching neuropsychological functioning in older adults.
In a series of investigations, older adults living in residential care facilities have
performed signiﬁcantly more poorly than their community dwelling counterparts
on tasks of paired associate learning, verbal and design ﬂuency, verbal and
visual memory, and set formation and attentional shift, even when level of
general intellect was controlled (Winocur & Moscovitch, 1983; Winocur &
Moscovitch, 1990; Winocur, Moscovitch, & Stuss,1996). In all of the
investigations, participants from residential facilities were chosen for inclusion
only if they were free of clinically diagnosed cognitive or psychiatric problems,
and judged to be functioning normally by nursing staff. Furthermore, all
participants were equivalent in terms of health, age, education, and
socioeconomic status. Nevertheless, since a proportion of the institutionalized
group performed at a level indistinguishable from the community group, the

authors concluded that it was not likely that institutionalization invariably leads to

26
cognitive impairment (Winocur & Moscovitch, 1990). Indeed, in a separate
investigation, they demonstrated that cognitive function could vary longitudinally
as a result of adjustment to living in a residential care facility, such that cognitive
performance improved with increases in perceived environmental control and
social activity level (Winocur, Moscovitch, & Freedman,1987).

Another group of investigators also found that an elderly residential care
facility population performed more poorly, as compared to older adults living
independently, on neuropsychological tasks assessing visual and verbal learning
and memory, verbal ﬂuency, visuoconstruction, and information processing rate
(as assessed by digit symbol substitution), regardless of cognitive status
(Steenhuis & Ostbye, 1995). Level of education and age were equivalent
between the two groups. In other work, statistically signiﬁcant differences on
task of confrontational naming between community dwelling older adults and
those residing in care facilities were not found for participants with more than 13
years of education; institutionalized subjects with six to nine years of education,
however, did perform more poorly, regardless of age (Neils et al., 1995).

Given that the present study investigated the Rorschach response
process in light of resource theory, the neuropsychological studies are of
particular interest. The ﬁndings suggested that, as a group, persons residing in
care facilities would likely evidence slower rates of information processing. In
terms of the trait theory aspect of the study, however, no difference in personality

functioning was expected between residential groups. Longitudinal studies have

27
demonstrated that, in general, individuals tend to maintain the same rank on a
variety of social and psychological indicators throughout the later years of life
(Maddox & Douglass, 1974). Although some studies have demonstrated group
differences in personality between “young-old” and “old-old” cohorts (Martin et
al., 1992); they have not been in the majority (Costa et al., 1986; Feifel & Strack,
1987). Furthermore, in a review of the literature, Tobin (1988) found that self-
concept was remarkably stable in older adults followed through their transition
from community dwellers to placement in residential care facilities.
Further Considerations

Epidemiological data indicate that, while only about 2% of community
dwelling adults over the age of 65 suffer from major depression, there are
substantial numbers of older people (around 20%) who experience clinically
signiﬁcant depressive symptoms (Scogin, 1994). These rates are much higher,
of course, in the elderly residential care population, where depressive disorders
are the second most common psychiatric disturbance (Streim et al., 1996). The
Rorschach records of depressed persons, as compared to non—depressed
persons, have been found to differ with respect to several determinants (e.g.,
Vista responses (V), color-shading blends (C.Sh)) (Exner, 1986). Additionally,
depressed older adults frequently demonstrate reduced rates of information
processing (Kaszniak & Ditraglia Christenson, 1994). As a result, consideration
of affective status is of import if changes in Rorschach responses with advancing

age are to be understood within a framework of a cognitive theory of aging.

METHODS
2mm

Forty community-dwelling volunteers, 65 years and older, were recruited
from local churches, community groups, or by word-of-mouth in a moderately
sized mid-Western city. Forty long-term care facility participants were selected
from a Home for Veterans and two privately owned nursing homes. Speciﬁcally,
social workers were asked to refer those residents 65 years and older who
appeared functionally intact, and who possessed adequate verbal skills and
motor functions. All subjects had to possess sufﬁcient hearing, vision, and motor
skills to physically perform the Wechsler Adult Intelligence Scale - Revised
(WAlS-R) Digit Symbol Substitution subtest (Wechsler, 1981), and earn a score
of 23 or more out of 30 on the Mini Mental Status Exam (MMSE; Folstein,
Folstein, & McHugh, 1975) to be included in the study. While a score below 24
is generally thought to represent the presence of cognitive impairment on the
MMSE (Folstein, Anthony, Parhad, Duffy, & Gruenberg, 1985), a slightly more
liberal criterion was adopted here, based on research demonstrating that elderly
non-demented institutionalized adults tend to perform at levels considered
impaired by general standards of practice (Ostenrveil et al., 1994). Non-
depressed participants were preferred in both groups, but subjects who
endorsed active depressive symptoms (i.e., scored greater than 10 out of 30 on
the Geriatric Depression Scale) (GDS; Yesavage et al., 1983), were not
excluded. Additionally, no attempt was made to exclude subjects with a history

28

29
of neurologic or active psychiatric disorders , degenerative motor disorders (eg.
Parkinson's Disease), or advanced systemic diseases (eg. severe heart
disease). While it was appreciated that health status could serve to confound
the results, 4 out of every 5 persons over 65 years of age have at least one
chronic medical condition (La Rue & Markee, 1995) and the institutionalized
population tend to suffer multiple morbidities (Manton et al., 1985).
Procedure

Upon initial contact with all potential participants, informed consent for
testing was obtained during an interview (a copy of the consent form can be
found in Appendix A); subjects were told that their participation was voluntary
and part of a university research project on aging. Consenting subjects were
then screened with the MMSE and Digit Symbol subtests. Subjects meeting
inclusion criteria were subsequently administered the Wide Range Achievement
Test 3 (WRAT 3) reading subtest (Wilkinson, 1993), the G08, the NEO-FFI
(Costa & McCrae, 1992), and the Rorschach (Rorschach, 1942). Demographic
information, including age, education, medical and psychiatric history, and
current medications were also collected from each subject or from their medical
chart (in the case of subjects residing in care facilities). All assessments were

administered by the investigator.

Tests and Measures
Digit-Symbol Substitution SubtestI WAIS-R

The Digit-Symbol Substitution Subtest of the WAIS—R (Wechsler, 1981) is
a reliable test (test-retest reliability = .82) believed to be a measure of clerical
speed and accuracy, psychomotor speed, visual short-term memory, and the
ability to follow directions (Kaufman, 1990, p. 411). Of all the WAIS-R subtests.
it shows the greatest difference between old and young adults (Botwinick, 1978
as cited by Gilmore, Royer, & Gruhn, 1983). Evidence suggests, however, that
"...although the digit symbol substitution test clearly involves motor and memory
components, these factors are probably not major contributors to the age
differences in digit symbol score" (Salthouse, 1985, p.310).

With respect to the contribution of memory to digit symbol task
performance, two studies which reduced memory involvement, either by
pretraining younger and older adults to the same accuracy criterion of
associating digit symbol pairs (Erber, Botvvinick, & Storandt, 1981), or by
reducing the number of digit symbol pairs (Salthouse, 1978), still resulted in the
age trends normally observed on this task. Studies in which the motor
components of the task were minimized, either by altering the task to require the
crossing out, rather than the writing out of symbols (Salthouse, 1978), or by
determining coding speed (i.e. handwriting speed) and then subtracting this from
digit symbol performance (Erber et al., 1981), also resulted in the expected age
trends in which healthy younger adults performed better than healthy older

30

31

adults. Indeed, studies in which the difﬁculty of the standard Digit Symbol
Substitution stimulus has been varied have concluded that the performance
decline associated with age was a result of the information-processing
operations of symbol encoding and visual search (Gilmore, Royer, & Gruhn,
1983; Royer, Gilmore, & Gruhn, 1981). Furthermore, Salthouse (1992b) has
recently demonstrated that a composite measure of perceptual processing
speed accounted for the majority of variance observed in Digit Symbol scores; in
contrast, a composite measure of motor speed demonstrated no such
association. In another investigation, Salthouse and his colleagues found that
memory, sustained attention, and manual coordination had little inﬂuence on the
relations between age and speeded substitution performance (Salthouse, Letz, &
Hooisma, 1994). Thus, the available evidence suggests that this subtest is
currently the best measure of information processing speed available (Salthouse,
1985).
E_xner Comprehensive System for t_he Roam

The Exner method (Exner, 1986) is currently the most commonly used
system of scoring and interpreting the Rorschach (Piotrowski, Sherry, & Keller,
1985). Although varying widely, a detailed account of validity measures are
reported by Exner (1986); interrater reliability for scoring is usually reported at or
about .85 (Gross, Newton, & Brooks, 1990). Meta-analytic reviews have
concluded that the reliability and validity of the Rorschach is comparable to that

of the MMPI and WAIS (Atkinson, 1986; Parker, Hanson, & Hunsley, 1988). To

32
date, norms for persons over the age of 60 have not been published; results
from previous studies, however, suggests that differences in Rorschach indices
between the elderly and younger adults are to be expected.
Geriatric Depression Scale (GL5)

The GDS (Yesavage et al., 1983) is a reliable and valid self-rating
depression screening scale for elderly populations both in the community
(Yesavage et al., 1983; Zgourides, Spofford, & Doppelt, 1989) and in institutions
(Koenig, Meador, Cohen, & Blazer, 1988; Parrnelee, Lawton, & Katz, 1989).
Furthermore, the scale has been found to be a valid indicator of depression in
patients with mild to moderate dementia (O'Riordan, Hayes, O'Neill, Shelley,
Walsh, & Coakley, 1990). The GDS consists of 30 items with binary response
categories and takes about ﬁve minutes to complete. A cut-off of 14130 or
greater is representative of clinically signiﬁcant depressive symptomatology.
Mini Mental Status Exam (MMSE)

The 30 questions of the MMSE (Folstein, Folstein, & McHugh, 1975) test
the following ﬁve areas of mental status in a quick and standardized format:
orientation, registration, attention/calculation, recall, and language. Original
standardization revealed no overlap between the scores of senile patients and
aged control subjects and the scale has proven to be a good screening test for
dementia in the institutionalized elderly (Kafonek, Ettinger, Roca,

Kittner, Taylor, & German, 1989). Test-retest reliability (over 24 hours) has been

reported to range between .82 and .98; additionally, MMSE scores have proven

33

valid measures of cognitive impairments correlating with CAT-scans,
neurological examinations, and the WAIS-R (Holzer, Tischler, Leaf, & Myers,
1984).
_N_EO Five-Fgctor Inventory (NEO-FFIJ

The NEO-FF I (Costa & McCrae, 1992) is a 60 item self-report personality
measure yielding scores for the ﬁve domains of Neuroticism (N), Extraversion
(E), Openness (O), Agreeableness (A), and Conscientiousness (C). Internal
reliability is adequate, as represented by coefﬁcient alpha values of .86, .77, .73,
.68., and .90 for N, E, O, A, and C respectively (Costa & McCrae, 1992).
Convergent validity of the respective scales with adjective factors assessing the
same constructs ranged from .56 to .62, with none of the divergent correlations
exceeding .20 (Costa & McCrae, 1992).
Reading Subtest, WRAT 3

The reading subtest of the WRAT 3 (Wilkinson, 1993) consists of a list of

42 words that an examinee reads aloud. There is a time limit of 10 seconds per
word, and the test is discontinued following 10 consecutive failures. This subtest
is based on the assumption that familiar words will be pronounced correctly, thus
reﬂecting vocabulary level. Because vocabulary, in turn, tends to remain stable
with advancing age, it is often used as an index of premorbid functioning (Albert,
Duffy, & Naeser, 1987). Raw scores from the test can be converted into

standard score with a mean of 100 and a standard deviation of 15.

m

Protocols were scored by the examiner according to the conventional
directions accompanying each respective test. Exner’s Comprehensive System
(1986) was used to score the Rorschach protocols. Additionally, a MA. level
psychologist experienced in the administration and scoring of the Rorschach
according to Exner’s method acted as a separate judge, re-scoring 50% of the
protocols; in usual practice, re—scoring 25% to 35% of the total protocols is
considered sufﬁcient to determine reliability (Exner, Kinder, & Curtiss, 1995).
lnterscorer agreement, ascertained by calculating the percentage concurrence
between the following eight segments, as suggested by Exner et al. (1996), was
as follows: (1) Location/Developmental quality - 95%, (2) Determinants - 89%, (3)
Form quality - 82%, (4) Pairs - 98%, (5) Primary content - 91%, (6) Populars -
94%, (7) 2 scores - 90%, and (8) Special scores - 86%. Thus, interscorer

agreement exceeded the recommended 80% (Exner et al., 1995) in all cases.

34

HYPOTHESES AND DATA ANALYSIS

The purpose of this study was to examine how reduced rates of
information processing and variation in normal personality factors manifest in the
Rorschach response process of older adults residing both in the community and
in care facilities. Because of numerous signiﬁcance tests performed on this data
set, precautions were taken to control for alpha inflation. Thus, for all analyses,
an alpha level of .01 was adopted; p 5 .05 were regarded as trends.

Demographic data were examined by descriptive statistics, as were the
variables under investigation. Relationships among the study variables were
also examined by Pearson product-moment correlations. Additionally, in light of
the literature reviewed above, the following hypothesis were tested:
1. It was hypothesized that reduced rates of processing information, as
measured by the WAIS-R Digit Symbol Substitution Subtest, would be
associated with increasing age, increased levels of depressive symptomatology,
and residence in a care facility.

Data Analysis: To test this hypothesis, information processing scores
were collapsed across groups and correlated with age and level of depression; a
correlational analysis was also used to examine the relationship between
residential status and information processing.
2. It was hypothesized that reduced rates of information processing would
affect the productivity and quality of Rorschach responses. Total response
number (R), the frequency of Popular (P) responses, the frequency of human

35

36
movement responses (M), and the value of F+%, all shown to decline with age in
the literature, were expected to be associated with lower Digit Symbol scores, as
were Whole responses (W), which have been shown to increase with advancing
age. The other determinants of interest, considered to represent information
processing effort, efﬁciency, and cognitive mediation, were Zf, X+% and X-%.
The choice of all these determinants also took into consideration those
Rorschach variables suggested to be most appropriate for analysis with
parametric statistics, based on the skew, kurtosis, number of available values,
and the percentage of real zero values of these variables, as reported in Exner’s
(1990) normative and patient data (Viglione, 1995).

Data Analysis: To test the hypothesis that reduced rates of information
processing are associated with lower quality and less efﬁciently processed
percepts, Rorschach performance was regressed on Digit Symbol scores.

Given the meaning assigned to the various constructs by Exner (1986), higher
rates of information processing were expected to be associated with better
quality protocols as represented by lower X-%, and higher R, P, M, W, Zf, F+%
and X+%. In the interest of minimizing the Type I errors inherent with multiple
statistical tests, however, the eight Rorschach determinants were submitted to a
Principal Components Analysis in order to reduce the number of dependent
variables.

3. It was hypothesized that higher quality Rorschach records as deﬁned

above would be associated with persons who are high in Openness (O),

37
Agreeableness (A), and Extraversion (E) and low in Neuroticism (N). No
relationship was expected with Conscientiousness (C).

Data Analysis: This hypothesis was tested by linear regression analyses
in which Rorschach performance was regressed on each of the personality
constructs in turn.

4. Rorschach response formation involves perceptual and cognitive
processes inﬂuenced by the underlying personality dynamics of the individual, or
both perception and representation (Blatt, 1990). Combining the results of the
separate literatures investigating information processing rate, ﬁve-factor
personality theory, and the Rorschach, the aforementioned response process

was conceptualized as:

 

Information Processing Rate

Personality (N. E, O, A, & C) @ch Perf®

Processing Rate x Personality

l)

 

 

ﬂ

 

ﬂ

Figure1. Proposed Model of Variable Interaction

It was expected that processing speed and personality, both together and
individually, would make statistically signiﬁcant contributions to the variance

observed in Rorschach performance when both where included as predictors in

38
the regression equation. In addition, the interaction of the main effects (DSym
by Personality) were considered.

Data Analysis: To test this model, hierarchical multiple regression
analyses were run in which the Rorschach factors were each regressed on
information processing rate, a given personality construct (N, E, O, A, or C), and
the interaction term formed by DSym and said personality variable. Processing
speed and a given personality construct were always added as a set of main
effects, in order to ascertain which predictor accounted for more of the variance

in Rorschach performance; the interaction term was then subsequently added.

RESULTS
§a_rr_1ple Selection and Response Rate

A total of 97 community-dwelling adults (70 women and 27 men) aged 65
years and older were contacted for inclusion in the study. Fifty-ﬁve people
refused to participate, three because of visual difﬁculties, seven due to illness,
and the remainder for reasons ranging from being too busy to a lack of interest.
Overall, 42 people gave consent to participate, but two subjects were excluded,
one because of a failure to comprehend the requirements of the digit symbol
substitution task and a MMSE score below 23, the other because of poor vision.
Forty participants (27 women and 13 men) completed all testing; the resultant
ratio of women to men corresponds to the 15:10 gender ratio observed in the
general population of adults 65 years and older, according to 1990 Census data.
Participants required, on average, approximately two-and-a-half hours to ﬁnish
the assessment. All community dwelling subjects were tested in one sitting.

Permission to collect data at the Home for Veterans was originally sought
through the Director of Social Services, who subsequently garnered the approval
of the head of each of four governing bodies within that facility, with ﬁnal
approval contingent upon the commandant’s decision. Permission to collect data
from the two privately owned and operated nursing homes was obtained by
contacting the chief social workers in each case, who subsequently obtained
permission from their respective managing directors. All of the care facilities
were operating at or near 100% occupancy. Social workers served as study

39

40
liaisons at each facility. In addition to being familiar with the residents’
personalities, attitudes, and overall health status, all social workers regularly
administered mental status exams, similar to the MMSE, to their respective
patients. They were asked to refer residents they believed to be physically and
cognitively capable of completing all assessment measures.

The Home for Veterans provides three levels of care (independent, semi-
skilled, and skilled) to military veterans or their spouses. During the time that
data were collected, the Home had a population of 647 residents, 34 (5%) of
whom were women. Social workers nominated 20 women as being suitable for
the study. Of those, 12 completed the entire assessment battery, while ﬁve
refused, two failed the inclusion criteria, and one was too visually impaired to
complete test items. Of the 24 men who were nominated, eight refused and two
failed inclusion criteria; another protocol was discarded because the participant
refused to take the Rorschach after completing the remainder of the battery.

The other participating facilities, both of which provide basic or skilled care
to their residents, had a census of 113 (86% women) and 133 (87% women),
respectively. Of the 20 women nominated between the two homes, two refused,
one was legally blind, and two failed the MMSE and Digit Symbol inclusion
criteria. Only one of the latter facilities referred men, and both of them refused
participation. As with the community-dwellers, 27 women and 13 men residing in
care facilities completed the testing battery, again approximating the ratio of

women to men over 65 in the general population. The majority required

41
approximately two to three hours and one sitting to ﬁnish the assessment,

though four of the women asked to be tested in shorter spans over two days.

Descriptive Statistics

Demographic Data

Overall, 80 people (54 women and 26 men) completed all measures (40
community-dwellers and 40 residents of care facilities). The total sample ranged
in age from 66 to 95, with a mean age of 80 years (SD = 7.49). Age was
normally distributed across the sample (Kolmogorov-Smirnov statistic = .08, df =
80, p > .20). With the exception of two African American women, the remainder
of the sample was Caucasian. The majority of participants (47.5%) were
widowed; 13.8% of the sample had never married, 27.5% were still married, and
11.3% had been divorced or separated. Fourteen of the sample reported
multiple marriages (six women and eight men). Number of children ranged from
none to 8 with a mean number of 2 births per family.

As a whole, the sample was well-educated: Mean number of years of
schooling was 12.54 (SD = 3.04) and ranged from 6 to 18 years. Approximately
48% of participants reported continuing their education following high school and
almost 28% earned Bachelor’s degrees or higher. Table 1 presents the results
of independent group t-tests evaluating residential status and gender
differences. Residents of care facilities were signiﬁcantly older than community-
dwellers; both groups were comparable in education, however. Women did not

differ from men in either age or level of education.

Table 1

Residential and Gender Differences in Age and Education

42

 

Age
Community"
Care Facility"

Womenb
Men°

Education
Community“
Care Facility"1

Womenb
Men°

 

M SD t df p
77.47 6.06 2.91 72.27 <.01
82.13 8.11
80.54 7.71 1.27 78 ns
78.27 6.90
12.98 2.53 1.29 71.36 ns
12.10 3.46
13.02 2.70 1.90 39.85 ns
11.54 3.50

 

Note. “1n = 40. bn = 54. On = 26. M = mean; SD = standard deviation;
df = degrees of freedom; ns = non-signiﬁcant. All p values are two-tailed.

Length of residency at the time of assessment for the subjects residing in

the long-term care facilities ranged from 1 to 276 months, with a mean of 33.85

months (SD = 47.56); the median length of residence was 22 months. Three of

the female residents, who were convalescing from surgery and hip fractures,

expected to be discharged home in the future; the remainder of the group were

long-stay residents. The main reason reported by subjects for admission into a

given residential care facility was declining health status demanding assistance

with daily living activities (67.5%). Eight subjects reported placing themselves

43
because they felt that their health might deteriorate and thus wanted to be
settled before becoming a burden on family or friends; three were admitted
following surgery; and two subjects entered facilities because of a lack of money
to live independently. Pearson product-moment correlations (all two-tailed)
demonstrated that length of residency was not associated with age (r = .06, p >
.10), number of medical diagnoses (r = .02, p > .10), level of depressive
symptomatology (r = .05, p > .10), or current mental status (r = -.20, p > .10) in
this group of participants.

The number of currently diagnosed medical conditions (excluding
psychiatric diagnoses) for the entire group of subjects ranged from none to 9,
with a mean of 3.46 (SD = 1.95). As a group, participants were taking a mean
number of 4.40 (SD = 3.02) prescribed medications on a daily basis (excluding
vitamins or other nutritional supplements, but including psychotropic agents);
across the sample, daily medication number ranged from a low of 0 to a high of
13. Not surprisingly, a Pearson product-moment correlation analysis
demonstrated that number of medical diagnoses was positively and signiﬁcantly
associated with the number of prescribed medications (r= .64, p < .01, two-
tailed). Table 2 presents the results of independent group t-tests evaluating
differences in medical history between residents of the community and residents
of care facilities: Both number of diagnoses and number of prescribed

medications were signiﬁcantly greater in the latter group. Women did not differ

44
from men in either diagnoses (t(78) = 1.86, p > .05, two-tailed) or in medication

number (t(78) = .03, p > .10, two-tailed).

Table 2

Residential Differences in Medical History

 

 

M SD t df p
Medical Diagnoses
Community 2.42 1.55 -5.58 78 <.01
Care Facility 4.50 1.77
Prescribed Medications
Community 3.13 3.04 -4.14 78 <.01
Care Facility 5.68 2.43

 

Note. N = 80. All p values are two-tailed.

Table 3 presents the percentage of each residential group actively
diagnosed with at least one ailment in the most commonly reported medical
diagnostic categories. Eighteen of the residents of care facilities had either a
history, or current diagnosis, of some type of neurological disorder, as compared
to only two of the community-dwellers. Cancer survivors numbered seven
amongst the community-dwellers and six amongst the residents of care facilities.
Approximately 34% of the entire sample had undergone at least unilateral

cataract removal (10 community-dwellers and 17 care facility dwellers). Only

care facility residents reported experiencing fractures of the hip (six women, one

man) or ribs (one woman).

Table 3

Frequency of Diagnostic Categgry

 

Percent
Communitya Care Facilityb

Arthritis (mild) 57.5 17.5
Cardiac Abnormalities 45 52.5
Cerebrovascular Accident 5 20
Degenerative Joint Disease 5 37.5
Dementia (mild) 2.5 7.5
Diabetes 10 27.5
Gastrointestinal Disorders 5 12.5
Hypertension 22.5 42.5
Osteoporosis 0 7.5
Parkinson’s Disease 0 12.5
Renal/Bladder Disorders 7.5 27.5
Respiratory Disorders 12.5 12.5
Thyroid Abnormalities 15 17.5
Visual Abnormalities 12.5 12.5

 

 

 

Note. an = 40. bn = 40.

All but three participants, who needed reading or driving glasses, were
outﬁtted with corrective bifocal lenses; 67.5% of the sample classiﬁed their visual
acuity when wearing glasses as good or adequate, while 32.5% rated their vision

as fair or poor. Hearing aids were used by only 15% of the entire group, the

46
majority of whom (eight) were men; 95% of the sample rated their hearing as
good or adequate; the remaining 5% reported their hearing to be fair or poor.
Regardless of their sensory self-perceptions, however, all participants had
adequate vision and hearing for testing. Additionally, all subjects had adequate
use of their dominant hand. All but three of the community-dwellers were
ambulatory, either without assistance or with the use of a walker or cane; in
contrast, 12 of the 40 residential facility dwellers were conﬁned to a wheelchair.

In regard to psychiatric history, 12.5% of the community dwelling
participants (four women and one man) described a history of treatment for
depression, while one male subject had a history of bipolar affective disorder.
Twenty percent of community-dwellers (four men and four women) reported
attending counseling for grief or adjustment difﬁculties at least once in their lives,
as compared to only one of the care facility subjects.

In the residential care facility group, 52.5% of the participants had a
history of clinically diagnosed psychiatric illness as follows: depression (six
woman, three men), anxiety (one woman), sporadic psychotic episodes (one
woman), bipolar affective disorder (one man), and alcohol abuse (ﬁve men).
Two of the latter men carried dual diagnoses with the alcohol abuse, one with
depression, the other with anxiety. Three additional male subjects, though not
formally diagnosed, described periods of alcohol abuse early in life. Though four
men experienced at least one hospitalization for alcoholism, the majority

reported that they stopped drinking alcohol “cold turkey”. Only one man reported

47
a recent history of alcohol use which was terminated when he entered the care
facility two months earlier; the remainder had not drunk alcohol for at least one
year, and most reported a 10 year or greater period of abstinence. None of the
men who saw combat while in the service, whether residing in the community or
care facilities, admitted to experiencing symptoms of post-traumatic stress.

Four of the community-dwelling participants (two men, two women) were
taking an anti-depressant at the time of assessment, as compared to eight
women and six men residing in care facilities. Additionally, one community-
dwelling man was taking lithium, while one care facility woman was prescribed a
low-dose anti-psychotic medication.

Level of depressive symptomatology across the entire sample, as
assessed by the Geriatric Depression Scale (GDS), ranged from 0 to 20
endorsements out of 30, with a mean of 4. 69 (SD = 4.16); the mean score fell
within the range of scores considered “Normal“ for the test (Yesavage et al.,
1983). An independent group t-test revealed that mean level of depressive
symptomatology did not differ signiﬁcantly as a function of residential status
(t(78) = -.24, p>.10, two-tailed). An evaluation of gender differences, however
(t(78) = -3.65, p < .01, two-tailed), revealed that mean level of depressive
symptomatology in the men was signiﬁcantly higher (M = 6.96, SD = 4.81) than
in the women (M = 3.59, SD = 3.33), though both average levels fell within
normal limits. Associations between mood and demographic variables, as

assessed by two-tailed Pearson product-moment correlations, revealed that level

48

of depression was not signiﬁcantly associated with age (r = -.15, p > .10) or
number of medical diagnoses (r = .04, p > .10) in this sample; there was a trend
for symptom level to be positively and signiﬁcantly associated with the number of
prescribed medications, however (r = .22, p < .05).
@gnitive and Information Processing MeasLes

Information processing rate was assessed by performance on the
WAIS-R Digit Symbol Substitution subtest (DSym). Scores represent the
number of symbols correctly coded in 90 seconds. DSym scores for the group
ranged from 7 to 59, with a mean of 28.99 (SD = 11.90). Cognitive functioning
was measured with the Wide Range Achievement Test 3 (WRAT 3) Reading
subtest and the Mini Mental Status Exam (MMSE). Raw reading scores were
converted to standard scores with a mean of 100 and a standard deviation of 15;
the scores were used to provide estimates of premorbid verbal functioning.
Standardized reading scores for all the participants ranged from 79 to 117, with a
mean of 101.03 (SD = 9.43), a score that falls in the Average classiﬁcation of
functioning. The MMSE addresses mental status, including orientation and
short-term memory; inclusion criteria necessitated that subjects score 23 or
above out of 30 on the test, a range of scores indicative of intact cognitive status.
Mean MMSE score for the group was 27.55 (SD = 1.98) and ranged from 23 to
30. Table 4 presents the results of independent group t-tests evaluating
differences in information processing and cognitive measures as a function of

residential status; community-dwellers demonstrated statistically signiﬁcant

49
greater scores on all measures as compared to residents of care facilities. There
were no differences between women and men, however, in mean scores on
DSym (t(78) = .79, p > .10, two-tailed), MMSE (t(78) = 1.87, p > .05, two-tailed)

or WRAT 3 reading (t(78) = 1.29, p > .10, two-tailed).

Table 4

Residential Differences in lnforrnation Processing Rate and Cognition

 

 

M SD t df p
DSym
Community 36.33 11.05 6.90 67.99 .01
Care Facility 21.65 7.37
WRAT 3 Reading
Community 104.13 7.61 3.10 72.40 .01
Care Facility 97.93 10.12
MMSE
Community 28.13 1.57 2.70 70.77 .01
Care Facility 26.98 2.19

 

Note. N = 80. All p values are two-tailed.

Pearson product-moment correlations between the cognitive and
information processing measures were determined using standard scores (see
Table 5). Standard scores on DSym were signiﬁcantly and positively correlated
with those on the WRAT 3 Reading subtest and the MMSE; similarly, standard
MMSE scores were signiﬁcantly and positively correlated with standardized

WRAT 3 Reading scores.

50
Table 5

Intercorrelations Between Cognitive Measures

 

 

DSym WRAT 3 Readirlg
DSym
WRAT 3 Reading .45**
MMSE .56" .45**

 

Note. N = 80. **p < .01, two-tailed.

Personality Variables

The NEO-FFI (Costa & McCrae, 1992) was used to assess personality as
represented by the ﬁve global domains of Neuroticism (N), Extraversion (E),
Openness (O), Agreeableness (A), and Conscientiousness (C). The NEO-FFI
proﬁle sheet provides standard Tscores corresponding to a given raw score for
each domain; these Tscores have a mean of 50 and a standard deviation of 10.
Thus, participants raw scores on each domain were converted to Tscores which
were used in these and subsequent analyses. Table 6 provides the descriptive
statistics for each of the personality domains for all subjects. According to the
NEO-FFI manual (Costa & McCrae, 1992), as a group, participants’ scores fell

within the Average range for the all domains, relative to the normative sample.

51
Table 6

NEO-F F I Descriptive Data

 

 

M SD Min Max
Neuroticism 46.03 7.73 26 67
Extraversion 50.56 7.99 29 65
Openness 45.84 8.52 29 65
Agreeableness 48.05 9.14 26 69
Conscientiousness 48.18 6.44 26 64

 

Note. N = 80.

Independent group t-tests (two-tailed) assessing differences in the ﬁve
personality domains as a function of residential status demonstrated that
community-dwellers were not signiﬁcantly different than residents of care
facilities in N (t(78) = -1.07, p > .10), E (t(78) = 1.83, p > .05), O (t(78) = 1.47, p >
.10), A (t(78) = 1.61, p > .10) or C (t(78) = -.59, p > .10). Table 7 presents the
results of independent group t-tests assessing gender differences in the
personality factors: Men were no different than women in E or O. There was a
trend for women to demonstrate slightly higher scores on C. The men’s scores
on N, which fell in the average range, were signiﬁcantly greater than the
women’s scores, which fell in the low average range. Furthermore, the average
level women’s scores on A were signiﬁcantly greater than the men’s scores,

which fell slightly below average.

Table 7

Gender Differences in Personality Domains

52

 

 

M SD 1‘ df p
Neuroticism
Women"I 43.70 6.68 -4.27 78 .01
Menb 50.85 7.54
Extraversion
Womena 50.76 8.15 .32 78 ns
Menb 50.15 7.78
Openness
Womena 46.07 8.24 .36 78 ns
Menb 45.35 9.23
Agreeableness
Women"I 49.96 8.35 2.81 78 .01
Menb 44.08 9.59
Conscientiousness
Womena 49.17 5.71 2.03 78 .05
Menb 45.12 7.44

 

Note. 3n = 54. bn = 26. All p values are two-tailed.

Pearson product-moment correlations between each of the personality

domains are presented in Table 8. As indicated, several of the domains were

statistically associated.

53
Table 8

Intercorrelations Among Personality Domains

 

 

N E O A
N
E -.25*
O -.06 .26*
A -.35** .19 .24*
C -.33** .26* .07 .49**

 

Note. N = 80. *p < .05; **p < .01, two-tailed.

Rorschach Responses

The means and distribution of those Rorschach variables hypothesized to
have an association with information processing rate and personality variables
are presented in Table 9. The determinants examined were: the total number of
responses (R); the ability to organize elements of the environment into a
meaningful impression as represented by the number of Whole responses (W);
Human Movement (M) responses, which are associated with general
intelligence; Zf, an index of the complexity of response mediation, or the effort
expended to organize the stimulus ﬁeld; conventionality as represented by the
number of Popular responses (P); F+%, the percentage of percepts that
demonstrate a superior use of the contours, resulting in responses that the

majority of individuals would have no difﬁculty seeing; and X+% and X-%, two

54
components of the Extended Form Quality Distribution: X+% represents
perceptual accuracy, reﬂecting the percentage of percepts that do not violate the
form demands of the blot, while X-% reﬂects a distorted or statistically rare (as
compared to the Exner (1990) normative sample ) use of form. All community-
dwelling participants produced at least one response to each of the 10 blots; two
residents of care facilities rejected a blot apiece (Card IX and VI, respectively)

while one participant exhibited two rejections (Card II and IX).

Table 9

Rorschach Determinants: DescriptiveQata

 

 

M SD Min Max
R 19.51 8.51 10 59
W 5.01 3.18 1 18
M 2.06 1.63 0 7
Zf 7.09 2.87 2 15
P 5.23 2.02 0 9
F+% .60 .17 .17 1.00
X+% .58 .13 .27 .88
X—% .22 .12 .00 .67

 

80.

Z
9..
.‘D
2
II

The results of independent t—tests evaluating differences in Rorschach
determinants as a function of residential status can be found in Table 10. Mean

scores of the community-dwelling participants differed signiﬁcantly from the care

55
facility residents on R, P, and M, with the latter group tending to produce
signiﬁcantly fewer of all response types. The remainder of the t-tests were not
signiﬁcant. In addition, men did not differ from women on Rorschach
determinants and scores: R (t(78) = 1.60, p > .10), W(f(35.81) = -.89, p > .10), M
(t(78) = .38, p > .10), P (t(78) = -1.94, p > .05), Zf(t(78) = -.47, p > .10), F+%

(t(78) = -.24, p > .10), X+% (t(78) = -.57, p > .10) and X-% (t(78) = .18, p > .10).

Table 10

Residential Statgs Differences in Rorschach Determinants

 

 

 

R W M P Zf F+% X+% X-%

M

Community 22.70 5.15 2.50 5.83 7.45 .62 .60 .21

Care Facility 16.33 4.88 1.63 4.63 6.73 .59 .57 .22
SD

Community 10.15 3.54 1.75 1.69 2.69 .14 .12 .11

Care Facility 4.80 2.81 1.37 2.16 3.03 .19 .14 .14
t 3.59 .38 2.49 2.77 1.13 .78 .87 -.44
df 55.59 78 78 78 78 71.09 78 74.41
p .01 ns .01 .01 ns ns ns ns
Note. N = 80.

Table 11 shows the results of t-tests evaluating differences in Rorschach
determinants between the present sample and Exner’s (1990) normative sample.
Exner’s sample consisted of 700, primarily Caucasian (81%), nonpatient adults

(50% male, 50% female) who were, as a group, very well educated (61% of the

56
sample has 13 or more years of schooling). The age of his normative sample
ranged from 18 years to over 65 years, though only 19 out of the 700 persons
(3%) were over 65 (the upper age limit was not reported) and only 5% of the
sample fell in the 56 to 65 year range. Mean differences in Rorschach variables
between the groups were statistically signiﬁcant in all cases. Compared to
Exner’s normative sample, the present subjects gave fewer R, W, P and M;
values for Zf, X+%, and F+% were also signiﬁcantly lower, while the value of X-%
was higher. Appendix B provides Exner’s means and standard deviations for

these Rorschach determinants.

Table 11

Rorschach Determinants: Normative Versus Sample Differences

 

 

R -3.13
W -9.95
P -7.37
M -12.31
Zf -14.72
F+% -5.78
X+% -14.01
X-% 10.83

 

Note. N = 80. df = 79 all tests. All p < .01 (two-tailed).

57

Clinically, Rorschach determinants and scores are not interpreted in
isolation; thus, it has been recommended that meaningful conﬁgurations of
scores providing multifaceted measures of the variables under study be used for
research purposes (Weiner, 1995). Reducing the number of dependent
variables also provides the practical function of attenuating the risk of alpha
inﬂation attendant when numerous signiﬁcance tests are conducted upon one
data set. A factor analysis, using the eight Rorschach variables under
consideration here, was therefore run.

Prior to computing the analysis, the suitability of the correlation matrix
(see Table 12) for such a statistical method was determined using Bartlett’s Test
of Sphericity, a test recommended when factor analyses with an N of less than
100 will be run (Zillmer & Vuz, 1995). The Bartlett’s sphericity X2 index of 251.71
(p < .001) fell within normal limits, and thus the null hypothesis that the
correlation matrix was an identity matrix unsuitable for factor analysis was
rejected. Initial factor extraction by means of a principal components analysis
revealed 3 factors with eigenvalues 3 1 that accounted for 72.16% of the total
variance. To achieve simple structure, the solutions were rotated by oblique
rotation (promax), a technique useful in psychology where determinants of
behavior, such as Rorschach scores, are rarely uncorrelated (Zillmer & Vuz,

1995)

Table 12

58

Intercorrelations Between Standardized Rorschach Variables

 

 

R W M Zf P F+% X+%
R
W .07
M .17 .01
Zf . 14 .75** .39“
P .17 .04 .43** .25*
F+% -.04 -.12 .16 .01 .34“
X+% -.17 -.17 .05 -.11 .37** .77**
X-% -.06 -.04 -.03 .03 -.31** -.48** -.63**

 

Note. N = 80. *p < .05; **p < .01, all two-tailed.

Table 13 shows the rotated factor pattern matrix for the eight Rorschach

variables under study. The three factors can be described as follows: Factor 1,

Perceptual Accuracy (ACC), primarily composed of X+%, F+%, and

X-%; Factor 2, Holistic Organization (ORG), principally deﬁned by loadings from

Wand Zf, and, Factor 3, General Response Process (RSP), characterized by

loadings for M, R, and P. The intercorrelations among the three factors can be

found in Table 14. Factor scores for the three components were determined by

the regression method, resulting in a new set of factor score variables with a

mean of 0. The three factor scores, named as above, were used as dependent

variables in all subsequent analyses involving Rorschach performance.

59

Table 13

Factor Pattern Matrix (Promax Solution) for Rorschgch Varialleg

 

 

Perceptual Holistic General
Accuracy Organization Response
X+% .95 -.08 -.1 1
F+% .85 -.04 .04
X-% -.80 -.09 .08
W -.006 .99 -.20
Zf -.02 .87 .21
M -.007 .02 .81
R -.26 -.09 .67
P .40 .05 .62

 

Note. N = 80. Bold face indicates factor to which item is allocated.

Table 14

Intercorrelations Among Rorschach Factor Variables

 

 

Perceptual Holistic
Accuracy Organization
Perceptual Accuracy
Holistic Organization -.04
General Response .20 .25*

 

Note. N = 80. *p < .05, two-tailed.

Relationships Between Variables: Statistical Analyses
Relationships Among Information Processim Rate, Mood. and Demographics

Hypothesis 1 predicted that older age, greater level of depression, and
residence in a care facility would be associated with slower rates of information
processing. Pearson product-moment correlations demonstrated that scores on
DSym were not associated with GDS scores (r= -.18, p > .10, two-tailed) in this
sample, although the direction of the relationship was as hypothesized. DSym
scores were signiﬁcantly associated with age (r = -.49, p > .01, two-tailed) and
residential status (r = -.62, p > .01, two-tailed) as predicted, such that both
advancing age and residence in a care facility were associated with less rapid
rates of information processing. Controlling the health status variables did not
attenuate the signiﬁcant negative relationship between DSym and age (r= -.50, p
< .01, two-tailed).

DSym scores also demonstrated statistically signiﬁcant negative
correlations with number of medical diagnoses (r = -.44, p < .01, two-tailed) and
number of medications (r = -.42, p < .01, two-tailed). Partialling out the effects of
these variables as well as age (since participants in care facilities were
signiﬁcantly older and had a statistically greater number of diagnoses and
medications) still resulted in a signiﬁcant semipartial correlation between DSym
and residential status (sr = -.43, p < .01, two-tailed). While residency in a care

facility per se was thus associated with reduced rates of information processing,

60

61

length of stay showed no relationship with DSym scores (r= -.10, p > .10, two-
tailed).
Rorschach Variables and Information Processingﬁai

It was predicted that the productivity and quality of Rorschach responses
would be dependent upon the rate of information processing, as manifested
through performance on DSym (Hypothesis 2). Speciﬁcally, it was hypothesized
that subjects scoring higher on DSym would produce Rorschach protocols with
more total (R), Whole (W), Human Movement (M), and Popular (P) responses,
higher values for Zf, F+%, and X+%, and lower values for X-%. Thus, in terms of
the three factor score variables ACC, ORG and RSP, higher DSym scores were
predicted to be associated with higher values for all three factors. This
hypothesis was tested using simple linear regression, regressing each factor
score in turn onto standardized DSym scores.

Information processing rate did not signiﬁcantly predict scores on either
Acc (beta = .14, R2: .02, p > .10) or ORG (beta = .15, R2 = .02, p > .10). The
regression of RSP on DSym, however, revealed a statistically signiﬁcant positive
association between the two variables (beta = .52, p < .01), with information
processing rate accounting for 27% of the variance observed in Rorschach RSP.
Rorschach Variables and Personaliy

Based on the literature review, hypothesis 3 predicted that better
Rorschach productivity and quality, as manifested in higher scores on factor

variables ACC, ORG and RSP, would be associated with persons high in O, A,

62
E, and low in N. No relationship was expected with C. This hypothesis was
tested using simple regressions, in which each dependent variable was
regressed on each of the personality constructs in turn. Table 15 presents the
results of these analyses examining Rorschach ACC. Only the regression using
A as a predictor demonstrated statistical signiﬁcance; the remainder of the
regressions were not signiﬁcant. Table 16 displays the results of regressing each
personality domain on the Rorschach factor ORG. Only the regression of ORG
on O was signiﬁcant, with persons higher in O demonstrating greater scores on
this factor, as predicted. The regression of RSP on the different personality
constructs is presented in Table 17. As predicted, O was a highly, statistically
signiﬁcant predictor of the Rorschach factor variable RSP, with higher scores on
the 0 scale associated with greater scores on RSP. The remainder of the
regressions were not signiﬁcant, although, there was a trend for persons higher

in A to produce greater scores on RSP.

63
Table 15

ngression: Rorschach ACC on Personality Variables

 

 

[3 R2 t p
N .00 .00 .03 .98
E .05 .00 .43 .67
O .03 .00 -.24 .80
A .29 .08 2.67 <.001
C .13 .02 1.13 .26

 

Note. N = 80. IS = standardized regression coefﬁcient; R2 = coefﬁcient of
determination. df = 78.

Table 16

Regression: Rorschach ORG on Personality Variables

 

 

l3 R2 t p
N .07 .01 .61 .55
E .02 .00 -.13 .90
O .28 .08 2.60 <.01
A .02 .00 -.18 .86
C .08 .00 -.60 .56

 

Note. N = 80. df= 78.

64
Table 17

@ression: Rorschach RSP on Personality Variabl_es

 

 

III R2 t p
N .10 .01 -.89 .38
E .11 .01 .99 .32
O .40 .16 3.86 <.001
A .21 .04 1.89 .06
C .08 .01 .70 .49

 

Note. N = 80. df= 78.

Rorschach Performance, Information Processinland Personality

Hierarchical multiple regressions were used to test the model of
Rorschach response formation outlined by Blatt (1990) and others (Weiner,
1986) using the present variables. Thus, cognitive processes were represented
by DSym scores and personality dynamics by each of the ﬁve personality factors
under investigation. Hypothesis 4 predicted that information processing and
personality would account, both individually and together, for a signiﬁcant
amount of the variance in Rorschach performance when both were used as
predictors. In addition, the interaction of information processing rate and
personality was investigated for each of the ﬁve personality factors.

In each of the hierarchical regressions, DSym and a given personality
construct (i.e., N, E, O, A or C) were entered simultaneously as a set of main

effects, followed by the appropriate two-way interaction (DSym X Personality

65
factor). Thus, for each dependent Rorschach variable (ACC, ORG and RSP),
ﬁve separate regressions were run.

Table 18 presents the results of the respective regressions of ACC onto
the ﬁve interaction terms. The only interaction that approached signiﬁcance
occurred when ACC was regressed on the DSym x 0 term. The latter interaction
accounted for approximately 7% of the variance in ACC above and beyond that
accounted independently by either DSym or 0. Generally, however, as indicated
by the values for Fchange, the addition of the remaining interaction terms into

their respective regressions captured little unique variance.

Table 18

Regression: ACC on the Information Processing - Personality lnteractjpg

 

DSym x N DSym x E DSym x O DSym x A DSym x C

 

[3 .10 -.20 -.26 .-.04 -.02
chhange .01 .04 .07 .00 .00
Fchange .76 2.99 5.77 .11 .03
pchange .39 .09 .02 .74 .87

 

Note. N=80. df=(3, 76).

The results of main effects testing are presented in Table 19. Only the
regression including information processing rate and A approached signiﬁcance,

accounting for approximately 9% of the total variance in ACC. Inspection of the

66
semipartial correlation coefﬁcients indicate that the majority of that variance was
accounted for by A, whose main effect demonstrated a trend towards statistical
signiﬁcance. Information processing rate and N, E, O, or C, respectively, did not
demonstrate signiﬁcant contributions to the variance in ACC, either
independently or when combined together as predictors. The values of the
semipartial correlation coefﬁcients, however, suggest that DSym scores were
responsible for what variance was accounted for when N, E, O, or C were

included in a respective regression equation.

Table 19

Regression: ACC on Information Processing_and Personality - Main Effects

 

 

N E O A C
Multiple R .15 .15 .16 .30 .18
F .88 .84 1.04 3.70 1.36
p .42 .44 .36 .03 .26
DSym sr .15 .14 .16 .06 .13
DSym t 1.32 1.22 1.42 .58 1.19
p .19 .23 .16 .57 .24
Personality sr .04 .03 -.08 .26 .12
Personality t .36 .24 -.67 2.38 1.03
p .72 .81 .51 .02 .30

 

Note. N = 80. sr= semipartial correlation coefﬁcient. df = (2, 77).

67

The same hierarchical multiple regressions were repeated with the
Rorschach ORG variable as the criterion. As before, DSym and a respective
personality construct were entered as a set of main effects ﬁrst, followed by the
appropriate interactions terms. Results of the regression of ORG on each
respective information processing rate by personality construct variable are
presented in Table 20. None of the regressions of ORG on a given interaction
term were statistically signiﬁcant, though a trend was evident for the interaction

terms formed by DSym and N, O, A, and C, respectively.

Table 20

Regression: ORG on the Information Processing - Personality Interaction

 

DSym x N DSym x E DSym x O DSym x A DSym x C

 

B .27 -.12 -.23 -.26 -.26
chhange .07 .01 .05 .07 .06
Fchange 5.65 1.08 4.57 5.60 4.93
pchange .02 .30 .04 .02 .03

 

Note. N=80. df=(3, 76).

The results of main effects testing are presented in Table 21. Only the
regression of ORG on DSym and O approached signiﬁcance, with the latter
combination of predictors accounting for 8% of the variance in ORG. Inspection

of the semipartial correlation coefﬁcients revealed that 0, above and beyond

68
DSym, uniquely accounted for most of that variance (6%). In the remaining
regression equations, neither the main effect of DSym, nor that of N, E, A, and C

respectively, approached signiﬁcance.

Table 21

Regression: ORG on Information Processingand Personality - Main Effects

 

 

N E O A C
Multiple R .19 .16 .29 .17 .17
F 1.44 .99 3.57 1.12 1.20
p .24 .37 .03 .33 .31
DSym sr .18 .16 .07 .17 .16
DSym t 1.59 1.41 .67 1.48 1.43
p .12 .16 .50 .14 .16
Personality sr .11 -.04 .25 -.07 -.08
Personality t .99 -.35 2.27 -.60 -.71
p .32 .73 .03 .55 .48

 

Note. N=80. df=(2, 77)

Table 22 presents the results of regressing Rorschach RSP on each of
the ﬁve information processing rate by personality construct interactions. None
of the interactions were signiﬁcant. Main effects testing can be found in Table
23. In all cases, regressions including information processing rate and a given
personality construct as predictors of RSP were highly signiﬁcant. DSym scores,
in combination with O, accounted for 34% of the variance in Rorschach RSP

process; when combined with either N, E, A, or C as predictors, DSym scores

69
accounted for 27% of the variance in each respective equation. Inspection of the
semipartial correlation coefﬁcients indicates that the majority of that variance was
accounted for by information processing rate, whose main effect reached
statistical signiﬁcance in all 5 regression equations. The only personality
construct which proved to be a signiﬁcant predictor of RSP when combined with

DSym in the equation was 0.

Table 22

Regression: RSP on the Information Processing - Personality Interaction

 

DSym x N DSym x E DSym x O DSym x A DSym x C

 

B -.06 -.00 .06 -.02 .13
chhange .00 .00 .00 .00 .01
Fchange .33 .01 .45 .05 1.52
pchange .57 .93 .51 .82 .22

 

Note. N=80. df=(3, 76).

Table 23

_Fggression: RSP on Information Processing ang Personality - Main Effects

 

N E O A C
Multiple R .52 .52 .58 .52 .52
F 14.02 14.03 19.22 14.24 14.05
p <01 <01 <01 <01 <01
DSym sr .51 .51 .42 .48 .51
DSym t 5.20 5.17 4.46 4.89 5.24
p <01 <01 <01 <01 <01
Personality sr .03 .03 .26 .06 .04
Personality t .32 .34 2.78 .65 .37
p .75 .73 <01 .52 .71

 

 

Note. N= 80. df = (2, 77).

Additional Analyses

Supplementary analyses were conducted to further examine the existence
and nature of any additional relationships among the study variables. Table 24
presents the Pearson product-moment correlations between Rorschach
Perceptual Accuracy (ACC), Holistic Organization (ORG) and General Response
Process (RSP) and selected demographic variables. As indicated, no
statistically signiﬁcant relationships were observed between ACC or ORG and
age, years of education, estimated verbal intellectual ability (as represented by
WRAT-3 reading scores), number of medical diagnoses or number of

medications; there was a trend for current mental status (as represented by

71
MMSE scores) to be positively associated with ORG, but not with ACC. A trend
was also observed between MMSE and the Rorschach RSP factor, such that
higher mental status scores were associated with higher RSP scores.
Additionally, higher RSP scores were statistically signiﬁcantly associated with
younger age, fewer medical diagnoses and medications, higher WRAT-3 scores,
and residence in the community. Holding age, medical history, medication
regimen, and WRAT-3 reading scores constant, however (since all were
signiﬁcantly different in care facility dwellers as compared to community-
dwellers), greatly attenuated the observed relationship between RSP and

residential status (r = -.15, p > .10, two-tailed).

Table 24

Rorschach Performance and Selected Demoggphics: Correlations

 

 

ACC ORG RSP
Age -.00 -.07 -.30**
Education -.03 .14 .07
WRAT-3 .13 .22 .31“
MMSE .13 .23* .24*
Medical Diagnoses -. 16 -.05 -.41**
Prescribed Medications .10 -. 16 -.34**
Residential Status -.10 -.08 -.43**

 

Note. N = 80. *p < .05; **p < .01, two-tailed.

72

The relationship between age and General Response Process (RSP) was
examined further in connection to DSym scores, which had been shown to
decrease with advancing age in this sample. Thus, Rorschach RSP was
regressed on DSym and age. Independently, information processing rate
accounted for 27% of the variance in RSP (r = .52, p < .01, two-tailed). Entering
age into the equation next resulted in a non-signiﬁcant change in R2 of .00 (F(2,
77) = .25, p > .10). Thus, accounting for information processing rate reduced the

magnitude of the inﬂuence of age on Rorschach RSP.

DISCUSSION
Sample Selection and Response Rate

Recruitment of subjects from the community proved to be challenging.
Originally, ﬂiers asking for participation in a study investigating psychology and
aging were posted in several residential complexes geared towards seniors and
in a hospital volunteer organization staffed primarily by persons over 65 years of
age. Response rate was poor, yielding just two participants from these sources.
Several more participants were referred by friends or relatives of colleagues of
the author; several of these subjects in turn served as contacts for additional
subjects by word-of-mouth. In the end, however, the majority of the community
dwelling subjects were recruited from two parishes of the Seventh Day Adventists
faith. Permission to recruit in the church, which had a large membership of older
adults, was obtained through a member who was well known to the investigator
and who announced the study in several meetings and in a newsletter. Even
still, 57% of the persons contacted refused to participate.

It is well known to researchers in the ﬁeld that community-dwelling older
adults are often difﬁcult to locate for research participation (Williams, Vitiello,
Ries, Bokan, & Prinz, 1988). In the present study, persons contacted for
inclusion were asked what factors inﬂuenced their decision to participate. Most
persons refusing participation tended to cite being too busy to take part in the
research, though several admitted to disinterest. There was some indication
among non-participants and subjects alike, however, that the topic under

73

74
investigation made them uncomfortable. Speciﬁcally, several of the
subjects reported a fear that the assessment would show them to be “crazy” or
otherwise mentally unﬁt. Almost 20 years ago, it was suggested that the then
current generation of older adults belonged to a cohort of individuals lacking
psychological mindfulness (Kulka, Veroff, & Douvan, 1979); this would still seem
to be the case, at least for many of the persons contacted for the present study.

It must be noted that the primary source of the community-dwelling
participants may limit the representativeness of the results. According to the
members interviewed, the Seventh Day Adventist religion advocates a healthy
lifestyle in which alcohol use is prohibited and a healthy diet and exercise are
encouraged as part of the faith. Though members were not asked how long they
had belonged to the faith (and hence followed the lifestyle), and though several
members admitted to the occasional glass of wine, it is possible that these
subjects are healthier than their average counterparts in the community.

Gaining permission to collect data from the three residential care facilities
participating in the study was a relatively straightforward, though lengthy process,
reﬂecting each site’s desire to safeguard the rights of their residents. In all
cases, the investigator’s previous experience at each facility helped to garner
approval, as did the fact that all subjects had to be capable of giving their own
informed consent. The greatest hindrance to individual subject recruitment was
the level of impairment of most of the residents. For example, of the 231 women

residing across the three facilities, only 40 were felt by social workers to possess

75
adequate cognitive, motor and visual capabilities to complete the battery. At the
Home for Veterans, only 24 men were nominated, despite a male population
numbering 613. The refusal rate among the care facility group ran about 25%,
less than that observed with the community-dwelling participants. Fewer
opportunities for intellectual stimulation and/or social interaction are not
uncommon in care facilities (Camp, West, & Poon, 1989), a fact of residential life
brought up by many of the participants and perhaps accounting for their greater
likelihood of participation, as compared to community-dwelling individuals.

Many of the community-dwellers reported leading active lives populated by
friends and relatives and ﬁlled with volunteer work and/or recreational pursuits.
Yet, they were as apt to want to “chat” with the investigator, both during and
following completion of the formal battery, as were the residents of the care
facilities. Indeed, the same subjects who complained of fatigue during the
assessment and were relieved when the last measure was administered were
never too tired to talk with the examiner upon its completion. Since all the
participants were donating their time for the study, the examiner was happy to
oblige the subjects’ penchant for showing family pictures or crafts they had
made. This, however, added 30 minutes to one hour to the two to three hours
needed for completion of the formal battery.

It has been observed that older persons are often more loquacious during
a psychological assessment than younger persons (Storandt, 1994), and that this

verbosity is unrelated to loneliness but is related to extraversion (Gold, Andres,

76
Arbuckle, & Schwartzman, 1988), at least in community—dwelling Canadian men.
As a group, the present subjects were average in Extraversion, with individual
scores on this domain of the NEO-FFI ranging from low to high average;
nonetheless, the desire for conversation was evident in almost all subjects.
Furthermore, loneliness and a lack of personal social interaction did appear to
contribute to the quuaciousness of the care facility residents. Thus, the factors
affecting this behavior may be related both to gender (since women
outnumbered men about 3:2) and to residential status, at least in the present
group.
Demographic Variables

Originally, an attempt was made to recruit 10 subjects in each of four age
bands (65-70, 71-75, 76-83, and over 83 years) in each residential group in order
to ensure a continuum of age intervals representing the youngest to the oldest
old (Neugarten, 1975). This eventually proved to be impractical, as it would have
meant turning away willing volunteers and waiting for non-existent participants.
In the end, however, age was normally distributed across the sample and
spanned 29 years from youngest to oldest, with a mean of 80 years. In
accordance with the literature, the residents of care facilities were signiﬁcantly
older, as a group, than were the community.dwelling subjects; their average age
of just over 82 years is slightly higher than the national average for this

population (US. Senate Special Committee on Aging, 1996). Moreover, all

77
the nonagenarians in the sample (11 women) were recruited from the care
facilities.

According to 1990 US. Census data, only 25% of older adults have more
than 12 years of schooling. Relative to the general population then, this sample
may be somewhat atypical; though the average length of schooling for both
residential groups was 12 years, 47% of the subjects reported some post-
secondary education and 28% had university degrees. The study participants
also differed from the general population in terms of ethnic diversity. Save for
two women, the entire sample was Caucasian. Both the education and ethnicity
factors, then, impose limits on the generalizability of the results.

Mean length of residency for the care facility group was similar to the 2.5
years reported in the literature (Streim et al., 1996). Not unexpectedly, nursing
home residents were both more ill and prescribed a greater number of
medications than the community-dwelling participants. It should be noted,
however, that formal medical histories were available only for the care facility
group. In several instances, community-dwellers reported no illness but did
report taking prescribed medication obviously intended for a diagnosed condition.
In addition, community-dwellers may have been unaware of the existence of any
forms of non-debilitating, asymptomatic disorders. For instance, osteoporosis, a
common disorder in older women, was a diagnosis found only among care facility

residents. Concomitantly, however, a tendency for residents of a given home to

78
all be diagnosed with disorders common to aging, perhaps representing the bias
of the attending physician(s), was also observed.

Only four women and one man, all community-dwellers, reported no
medical conditions. Thus, almost 94% of the sample had at least one medical
diagnosis, slightly higher than the 80% expected (La Rue & Markee, 1995).
Cardiac abnormalities were one of the most frequent medical diagnoses amongst
both community-dwellers (45%) and residents of care facilities (52.5%);
hypertension was also common. Signiﬁcant relationships between hypertension
or heart disease and performance on various neuropsychological measures have
been documented (Abrahams & Birren, 1983; Elias, Robbins, Schultz, & Pierce,
1990; Hertzog, Schaie, & Gribbin, 1978; Sands & Meredith, 1992). Additionally,
correlational research has demonstrated that health can account for a greater
proportion of individual differences in older rather than younger adults’ cognitive
performance (Permutter & Nyquist, 1990). Thus, consideration of health status
in general is important, given the information processing focus of the present
invesﬁgaﬁon.

Salthouse and his colleagues (1990; 1992b) found that controlling for
health status, as measured by self-reported ratings, did not attenuate the age-
associated decline in digit symbol performance; similarly, there was no
interaction between age and health status in these studies. While it would be
easy to conclude from this work that health status per se has little inﬂuence on

information processing rate, different results have been obtained in studies

79
examining speciﬁc diagnostic groups. For example, poorer digit symbol
performance was evident in a group of older adults with non-insulin dependent
diabetes mellitus, relative to their age-matched, healthy counterparts (Perlmuter,
Tun, Sizer, McGIinchey, & Nathan, 1987; Tun, Perlmuter, Russo, & Nathan,
1987). Consideration of other diseases have yielded conﬂicting results. Sands
and Meredith (1992) found that blood pressure status was not predictive of digit
symbol performance over time, although it did inﬂuence verbal digit span
performance. In other research, Mitrushina and Satz (1990) found no signiﬁcant
difference in digit symbol performance between older adults with and without
presumable cardiovascular disease; differences were found, however, on more
complex mental operations involving decision making in novel situations. In the
present study, poorer digit symbol performance was associated with increasing
age, as well as with increasing numbers of medical diagnoses and medications
prescribed. As in the Salthouse study, however, controlling for the two health
status variables did not attenuate the negative relationship between information
processing rate and age.

Interestingly, one community-dwelling woman and three women in care
facilities had diagnoses of mild dementia. It is unknown what criteria these
diagnoses were based upon. Inclusion criteria for the study demanded that
subjects score above the cut-off for dementia on the MMSE, which these women
did. Additionally, digit symbol substitution tasks are sensitive to any type of brain

damage (LeFever, 1991 ); again, these women were able to compete the digit

80
symbol test independently. Of course, a complete neuropsychological evaluation
was not part of the present investigation. It is possible that other
neuropsychological testing, particularly of memory, would have supported such
diagnostic conclusions. Regardless, the potential presence of the dementing
illnesses, as well as the ﬁve documented cases of Parkinson’s Disease and the
10 CVA histories, indicates that information processing rate may have been
confounded with disease process, at least in these subjects.
Depression

The distribution of the severity of experienced depressive symptomatology
in the sample, based on published cut-off scores for the GDS (Yesavage et al.,
1983), was as follows: 86.3% within normal limits and 12.5% mild
symptomatology; only one subject, a man, scored in the severely depressed
range. Mean level of depressive symptomatology (approximately 4/30
endorsements) did not differ as a function of residential status, but did differ as a
function of gender. Nevertheless, the mean level of depressive symptomatology
fell within normal limits for both sexes. Fourteen residents of care facilities were
on anti-depressants, as compared to only four community-dwellers Thus, while
clinically signiﬁcant symptoms were absent in the subjects at the time of
assessment, care facility participants were more apt to be formally diagnosed
with depressive disorder. While this may reﬂect the higher incidence of
depression amongst older adults in care facilities (Streim et al., 1996), it may also

reﬂect the fact that the physicians attending such residents are more alert to

81
such symptomatology. It is not uncommon in the current generation of older
adults, given problems with psychological mindfulness, to present psychological
distress as physical symptoms (Schaie & Willis, 1991), making it more
challenging to accurately uncover the nature of presenting symptoms, especially
for practitioners not familiar with this population.
lnformat_ion Processing Rate and Cognition
Results of the WRAT-3 Reading subtest indicated that premorbid verbal
abilities fell in the Average classiﬁcation of functioning for both residential groups,
even though there was a statistically signiﬁcant difference in mean scores; this
suggests that participants were comparable in terms of verbal intellect.
Similarly, while current mental status, as measured by the MMSE, did differ as a
result of residential status, the mean scores of both groups fell in the intact
range. Adopting a more liberal cut-off criterion resulted in the inclusion of three
subjects, all residents of care facilities (and all scoring 23/30), who would
normally be considered to be experiencing mild cognitive impairment (Folstein et
al., 1985). As found in the Osterweil et al. (1994) study, however, 1/3 of an older
non-demented care facility population (mean age 88.9 years), scored 23 or below
and all were functioning well in an independent manner within the living
environment. Even some of the younger residents in that study (average age
80.4 years) scored below the impairment cut-point of 24. In the present case,

the residents earning 23/30 points were 73, 76, and 86 years of age. Mental

82
status for the whole group, then, was comparable across residential status, with
all subjects considered cognitively intact.

The mean score of approximately 29 earned by the total sample on the
WAIS-R Digit Symbol Substitution (DSym) subtest corresponds to a standard
score of 4 (Wechsler, 1981), a performance falling below the ﬁrst percentile of
the young (20-34 years) normative reference sample. When compared to the
oldest provided reference group of 70 - 74 years, however, a standard DSym
score of 10 (50th percentile) is obtained. As a group, then, information
processing rate fell in the Average classiﬁcation, even amongst adults older than
74 years. Processing rate did differ as a function of residential status, however;
care facility residents earned a mean standard score of 8 (as indexed
against the 70 -74 year reference group), relative to the mean standard score of
11 earned by participants residing in the community.

It should be noted that, while the digit symbol task was administered
according to standard procedures as outlined in the WAIS-R manual (Wechsler,
1981), all subjects ﬁlled out a version that was enlarged to legal size (i.e., 8 x 14
inches). As reviewed by Storandt (1994), visual changes common to aging, such
as decreases in acuity and difﬁculty in reading small print, can compromise
performance on psychological measures. All the subjects had to have sufﬁcient
visual ability to complete all assessment measures; in order to accommodate

potential individual differences, however, the digit symbol stimuli were enlarged

83
for everyone, to ensure that what was being assessed was indeed information
processing rate and not difﬁculties with vision.
W
Community-dwellers did not differ from residents of care facilities in the
personality attributes of Neuroticism, Extraversion, Openness, Agreeableness, or
Conscientiousness, as assessed by the NEO-FFI. Relative to the normative data
base for the inventory, which provides norms for both 500 men and 500 women
ranging in age from 21 to 95 years (Costa & McCrae, 1992), the present subjects
scored within the average range for all personality domains. Costa and McCrae
(1992) report that women tend to score higher than men on Neuroticism and
Agreeableness, though the differences tend to be modest in magnitude. In the
present sample, women’s scores were indeed signiﬁcantly greater on
Agreeableness; in contrast, women demonstrated signiﬁcantly lower scores on
Neuroticism as compared to the men. The men in this study endorsed
signiﬁcantly more symptoms of depression than the women; since Neuroticism is
a measure of the tendency to experience negative affects, the greater scores
demonstrated on this domain by the men may have been a function of their
affective status.
Rather than have each subject read and complete the NEO-F F l
themselves, the examiner read the inventory to subjects and marked down their
responses for them, an amendment allowed in the standard procedure of

administration. This was done both in the interest of time, and because, in

84
clinical practice, it has been the investigator’s experience that older adults ﬁnd
ﬁlling and tracking their answers on the small print response sheets visually
taxing.

Early into the investigation, it became apparent that many subjects found it
difﬁcult to respond to negatively keyed statements, such as “I am not a worrier”.
According to the manual, a sixth grade reading level is adequate to complete the
inventory (Costa & McCrae, 1992). Results from the WRAT-3 Reading subtest
indicate that, as a group, subjects were reading at a high school level. It appears
that the difﬁculty arose when choosing one of the ﬁve possible responses, which
ranged from strongly agree to strongly disagree, when confronted with the
negatively keyed statements. This was observed despite the fact that a sheet
listing the ﬁve choices was provided to each subject during the test with explicit
instructions to choose one for each statement. Many subjects tended to state
“Yes, I am...” or “No, I’m not...” as responses, or, alternatively, “True” or “False”.
It was also not uncommon for subjects to ask something akin to “If it says I’m not,
and I am, then shouldn’t I agree?”. Also apparent was a tendency for many of
the subjects to ignore the “not” in a phrase and respond as if the statement was
positively keyed. Fortunately, because these difﬁculties were observed early on,
potential confusion regarding a subject’s response was avoided by asking for

elaboration.

W

In concordance with the literature examining the quality and quantity of
Rorschach responses in older adults, signiﬁcant differences between the present
sample and Exner’s (1990) standardization group were observed. Speciﬁcally,
this sample offered signiﬁcantly fewer total (R), Whole (W), Human Movement
(M) and Popular (P) responses. Similarly, signiﬁcantly lower scores were
demonstrated for organizational activity (2f), good Form Quality (F+%), and good
perceptual accuracy scores (X+%), while poor perceptual accuracy scores (X-%)
were higher. The fewer R, M, and P responses, as well as the lower scores on
F+%, support the results of the literature reviewed by Reichlin (1984).

The scores of X+% or X—% have not been examined in the older adult
Rorschach literature. In this sample, both X+% and X-% were signiﬁcantly
correlated with F+% (r= .77 and r = .-48%, p < .01, respectively), as well as with
each other (r= -.63, p < .01). All three scores describe how well responses
conform to the form demands of the blot; while F+% deals speciﬁcally with
responses which are purely form dominated, the extended form quality scores of
X+% and X-% address the issue for all responses where form is at least one
component of the percept. Not unexpectedly, subjects with high F+% tended to
demonstrate good form on the remainder of the blots; hence the positive
correlation between it and X+%. With respect to the extended form quality
scores, because both scores are based on total response number, when X+% is
high, X-% is usually low, and vice versa; hence the strong negative correlation

85

86
between the two scores. Since F+% was signiﬁcantly lower in this sample,
relative to Exner’s normative base (1990), and given the aforementioned
interrelationships, it is not surprising that X+% was also lower, and X-% was
higher.

Exner’s score of Zf has also not been examined previously in the geriatric
Rorschach literature; indeed, Exner (1996) has stated that, “The scoring for
organizational activity has been one of the most neglected of the Rorschach
components “ (p. 141). Assigning a 2 score depends on the meaningfulness of
the organization of a percept; they are assigned only in percepts that include
form and are either W responses, responses that integrate two detail areas
(adjacent or nonadjacent), or responses that integrate white space meaningfully
into the percept. For most of the cards, responses earning Z scores demand
more cognitive effort and organizing activity on the part of the subject. Not
surprisingly given the criteria of assignment, Zf scores were strongly and
positively correlated with W in this sample (r= .75, p < 01). A signiﬁcant positive
association was also demonstrated between Zf and M in this sample (r= .39, p <
.01), supporting results obtained by Exner, Viglione, and Gillespie (1984) in a
younger adult sample. Projecting human movement into a percept is thought to
demand reasoning, imagination and higher forms of conceptualization (Exner,
1986); hence producing an M response also requires greater and more complex
cognitive processes. Given the intercorrelations amongst these components,

and the result that both Wand M occurred less frequently in the sample than in

87
Exner’s (1990) normative base, the fact that Zf was also signiﬁcantly lower in the
present sample is not unexpected.

Across the studies reviewed, W tended to increase with age; in this
sample, the frequency of W responses was signiﬁcantly less than that reported in
Exner’s (1990) normative sample. Of course, R was signiﬁcantly less as well;
thus, it might be tempting to conclude that the fewer total responses decreased
the probability of observing a Wresponse (or for that matter, an M or P). An
examination of the ratio of Wto D (responses delivered to discrete detail areas of
the blot), however, indicated that many subjects were more apt to deliver D or Dd
replies. Forty-nine of the 80 protocols evidenced W:D ratios of 1:2 or more, with
37 of the that group demonstrating ratios of 1:3 and greater. It is noteworthy,
however, that most nonpatient adults give more D that Wanswers (Exner, 1986);
thus, in this respect, the present older adults Rorschach behavior was similar to
Exner’s normative group qualitatively, if not quantitatively.

Community-dwelling participants differed signiﬁcantly from participants in
care facilities with respect to R, M and P. Speciﬁcally, community women tended
to produce more of all these response types than their counterparts living in care
facilities. The lower P and M may be a function of a lower R. These three
determinants did load onto one factor, which was named the General Response
Process (RSP) factor. In general, the variables composing this factor address
several aspects of responding, including productivity (R), perceptual

conventionality (P), and imagination and cognitive effort (M).

88

The Rorschach was administered following Exner’s guidelines with the
exception of one requirement. According to Exner’s rules, a protocol with less
than 14 responses may be not be interpretively valid . In such cases, the
examiner should normally re-administer the response phase before proceeding
to the inquiry, during which the features inﬂuencing response formation are
explicated. In the present sample, though six subjects produced fewer than 14
responses, the response phase was not re-administered. The reason for this
was two-fold: (1) From the perspective of the research, the number of responses
a subject would spontaneously generate in one administration was of interest,
and (2) In previous research (Domitrovic, 1994) and clinical work with the
Rorschach, it has been my experience that re-administering the test results in
few, if any, additional responses. Instead, re-administration tends to result in a

subject who is defensive and with whom rapport is taxed.

Relationships Between Variables

Information Processing RateI MoodI and Demographics

The results of Hypothesis 1 indicate that, with the current sample, level of
depressive symptomatology was not associated with information processing rate.
This is not surprising, given that, as a group, the present subjects did not express
a clinically signiﬁcant level of depressive symptomatology. Furthermore, eight of
the nine subjects demonstrating mildly elevated levels of symptomatology scored
just outside of normal limits. While studies have demonstrated that elderly

depressed adults can show deﬁcits on neuropsychological testing, including

89
slower rates of information processing, such subjects are usually suffering from
clinically diagnosed major depressive episodes with attendant symptoms greater
in number and severity (see Lezak, 1995 for a review). Even when observed,
cognitive deﬁcits are relatively mild for most depressed older adults (Kaszniak &
DiTraglia Christenson, 1994).

In accordance with Hypothesis 1, scores on the digit symbol substitution
subtest were signiﬁcantly correlated with age and residential status, such that
advancing age and residence in a care facility were both associated with
decreased rates of processing information. This result was found despite the
fact that care facility participants were generally equated with community-dwelling
subjects in years of education, estimated premorbid verbal intelligence, and
current mental status, all variables which could have potentially explained the
observed association. Of course, participants in care facilities were signiﬁcantly
older, as well as more ill and taking more prescribed medications than subjects
in the community. Nevertheless, controlling for age, health and medication
status attenuated the signiﬁcant negative relationship between residential status
and information processing rate only slightly.

The latter result is similar to that observed by Winocur and colleagues
(1983, 1990, and 1996), where residents of care facilities tended to perform more
poorly than their community-dwelling counterparts on various neuropsychological
tests, despite the fact that subjects were equated on health, age, education, and

socioeconomic status, and IQ was controlled. A separate study (Winocur et al.,

90
1987) found that cognitive function could vary longitudinally as a result of
adjustment to the care facility environment. Though adjustment was not formally
assessed here, longer-stay residents presumably would have had more time to
adjust to residential life. In the present sample, however, there was no
association between length of residency and information processing rate.

Another hypothesis suggested by Winocur and Moscovitch (1990) for the
poorer performances of their care facility group, was that, though appearing to be
cognitively intact, individuals may have been in a state of decline upon admission
and that institutionalization either continued or exacerbated this course. The
MMSE has proven to be a reliable and valid instrument in the detection of
cognitive impairment, but it is only a screening instrument. Thus, it is possible
that testing of a broader array of neuropsychological functioning would have
shown care facility residents to be more cognitively impaired than either their
mental status or their level of practical functioning indicated, and that this could
help explain the association between information processing rate and residential
status.

It should be noted that, of the neuropsychological functions examined in
the series of studies by Winocur and colleagues, information processing rate per
se was not included. Data from the Canadian Study of Health and Aging
(Steenhuis & Ostbye, 1995), however, did ﬁnd that residents of care facilities
performed more poorly on the WAIS-R digit symbol substitution task as

compared to community-dwelling older adults. This result was observed even

91
when severity of cognitive deﬁcits (groups studied either demonstrated dementia,
cognitive loss without dementia, or no cerebral pathology) was controlled. The
authors did not comment speciﬁcally on information processing rate, but noted
that since the different residential groups differed on other visuospatial tasks as
well, perhaps unmeasured discrepancies in sensory function were to blame for
decreases in visuospatial functioning.
Rorschach Variables and Information Processing Rate

Researchers of the Rorschach have been encouraged to employ clusters
or conﬁgurations of variables that combine to provide multifaceted measures of
the variables under study; this is because in clinical practice, meaningful
inferences from the test are based upon conjoint examination of the
determinants and their relationship to one another (Weiner, 1995). Thus, in the
remainder of the analyses examining Rorschach performance, the three factors
of Perceptual Accuracy (ACC), Holistic Organization (ORG), and General
Response Process (RSP), which resulted from a principal components analysis
of the eight Rorschach scores and determinants under current study, were used
as dependent variables.

Hypothesis 2 predicted that faster rates of processing information would
be associated with better Rorschach performance as exempliﬁed by higher
scores on the factor variables of ACC, ORG and RSP. This was the case only
for the factor variable RSP, principally deﬁned by R, M, and P, and representing

a complex variable deﬁned by productivity, perceptual conventionality, and

92
creativity. Despite the statistical signiﬁcance, however, Digit Symbol scores
accounted for only 27% of the variance observed in RSP. This is likely because
the cognitive component of forming a Rorschach response is a complex task
demanding not only information processing, but scanning and visuospatial
abilities, concept formation, and adequate verbal and communicative skills
(Perry, Potterat, Auslander, Kaplan, & Jeste, 1996).

The complexity of Rorschach response formation may be responsible for
the non-signiﬁcant correlation observed between information processing rate and
the Holistic Organization factor (ORG). The factor ORG was primarily composed
of the Rorschach variables Wand Zf, variables considered to represent complex
holistic processing, ideation and integration (Exner et al., 1996). While the
processing of information is a prerequisite for all cognitive acts, it may be that the
ORG variable represents higher order cognitive functions not captured solely by
scores on the digit symbol task; hence, the lack of a signiﬁcant association in the
present study.

Information processing rate was not associated with the Perceptual
Accuracy (ACC) factor, determined essentially by F+%, X+% and X-%, either.
Again, while responding to an inkblot demands the processing of information, it
could be that some other visuospatial operations primarily underlie this variable,
thought to represent the perceptual accuracy with which blots were translated. A
recent study examining Rorschach performance in relation to neuropsychological

functioning found that a similar factor variable, composed of P in addition to the

93

three form quality percentages, was not associated with either a verbal-
comprehension or motor/visuospatial factor in a group of 190 psychiatric patients
(Zillmer & Perry, 1996). In other research, 60 subjects suffering from moderately
severe closed head injuries demonstrated scores on X+%, F+%, and X-% that
were no different than those seem in Exner’s (1990) normative sample (Exner et
al., 1996). Thus, it may be that the mediation of the blots, a process represented
by the form quality distribution (Exner, 1990), may only be compromised with
severe and/or focal injuries to areas of the brain responsible for visuospatial
integration; hence, the lack of association between Perceptual Accuracy and
information processing in this group of cognitively intact older adults.
Rorschach Variables and Personality

Hypothesis 3 predicted that better Rorschach quality and productivity, as
represented by higher scores on the factors Perceptual Accuracy, Holistic
Organization, and General Response Process, would be found in persons high in
Openness, Agreeableness, and Extraversion, and low in Neuroticism; no
relationship was expected with Conscientiousness. Results varied somewhat,
depending upon the criterion variable examined. Only persons high on
Agreeableness demonstrated statistically signiﬁcant higher scores on the factor
variable ACC, with this domain predicting 8% of the variance observed in
perceptual accuracy; no signiﬁcant relationships were observed with the other
personality factors. When the Holistic Organization factor was regressed on the

ﬁve individual personality factors, only Openness demonstrated a signiﬁcantly

94
positive association, accounting for 8% of the variance in integrative responding.
Likewise, when General Response Process was regressed on personality, a
signiﬁcant association was again found only with Openness, which predicted
15% of the variance observed in that factor variable.

As the literature review indicated, the ﬁve factor model of personality has
not been previously examined in relation to the Rorschach. The present study
suggests that, at least in this sample of older adults, Rorschach response
formation was inﬂuenced, to some extent, by how agreeable and open to
experience an individual was. This relationship was stronger for the General
Response Process factor than for either Perceptual Accuracy or Holistic
Organization, perhaps reﬂecting the different aspects of response formation that
these variables represent. Though all three factors are composed of
determinants and scores purported to be related to the processing, mediation,
and ideation of the blots (Exner, 1986; Exner et al., 1996), the Perceptual
Accuracy factor probably represents the most “purely” cognitive variable since
F+%, X+% and X-% reﬂect the degree to which percepts conform to the form
demands of the blot and whether or not they violate reality. In contrast, both the
factors of Holistic Organization and General Response Process are composed of
Rorschach variables that can be inﬂuenced to a greater extent by stylistic
tendencies (i.e., R, W, M and Zf). For example, low R records are not
uncommon in adults who are intellectually limited or neurologically impaired

(Exner 1986; Exner et al., 1996; Perry et al., 1996), highlighting the cognitive

95
component of this Rorschach variable. Concomitantly, however, low R records
can indicate resistance or defensiveness (Exner, 1985); while the latter can be
situationally induced, it can represent a personality style of an individual. Thus,
perhaps the domain of Openness was more related to the General Response
Process factor than to Holistic Organization because the former variable is more
inﬂuenced by personality processes.

Openness to experience is believed to be composed of several different
elements, including an active imagination, intellectual curiosity, and attentiveness
to inner feelings, and scores on this domain have been found to be modestly
associated with education and measured intellect (Costa & McCrae, 1992). It is
of note that several of the Rorschach variables under study, particularly R, W,
and M, have been found to be associated with creativity and intellect, though the
results across studies have been equivocal (see Exner, 1996 for a review).
Agreeableness, found to been associated with Perceptual Accuracy but not with
General Response Process or Holistic Organization, is thought to represent an
interpersonal tendency to be sympathetic to others and eager to help them
(Costa & McCrae, 1992); thus, it is possible that the relationship observed here
was due to a social desirability effect; that is, people eager to please the
examiner approached the task with more interest and/or attention. Why this
relationship would then emerge only with the Perceptual Accuracy factor,

however, is unclear.

96

In this older adult sample, the remaining personality domains were not
associated with the aspects of Rorschach performance examined. This does not
mean, of course, that personality is of no consequence in forming a response.
Rather, it supports the results of research examining relationships between the
Rorschach and other “objective” self-report personality tests, such as the MMPI.
As reviewed by Archer and Krishnamurthy (1993), interrelationships between the
MMPI and Rorschach variables have been minimal, a ﬁnding recently replicated
in Zillmer and Perry’s (1996) factor analytic study. Meyer (1992) has pointed out
that self-report inventories in general assess cognitive schemata about the self,
whereas the Rorschach yields tacit, less consciously mediated processes.
Similarly, Dies (1995) has argued that both objective and projective personality
measures furnish a glimpse of the individual, but that self-report measures
reﬂect the capacity to create certain impressions when the “...cues are clear as
to what behaviors are expected by others” (p. 36). In contrast, the Rorschach
makes it difﬁcult to inhibit the expression of underlying liabilities or faulty means
of conceptualizing or processing information, especially when individuals are
psychologically troubled (Dies, 1995). The lack of association between most of
the NEO personality domains and the Rorschach found here, then, emphasizes
the unique, problem-solving nature of this task, with its contributions from both

cognitive and personality realms.

Rorschach Performance, Information Processing and Personality

The intent of this study was to examine Rorschach response formation in
older adults by decomposing and examining the process and its component
aspects of cognition and personality. In actual fact, of course, the perceptual and
representational (Blatt, 1990) processes involved occur simultaneously.
Hypothesis 4 tested this premise, predicting that information processing rate and
a given personality domain would account, both independently and together, for
a signiﬁcant amount of variance in the factors of Perceptual Accuracy, Holistic
Organization, and General Response Process when entered together in a
regression equation. Because there was no a priori reason to expect a greater
inﬂuence of processing speed over personality (or vice versa) on Rorschach
response formation, digit symbol scores and a given personality factor were
entered simultaneously as a set of main effects.

Statistical testing indicated that the combination of information processing
rate and a given personality domain did not account for a signiﬁcant amount of
the variance in the factor variable Perceptual Accuracy, though there was trend
observed with Digit Symbol scores and Agreeableness, with the latter
combination of scores accounting for 9% of the variance in this variable.
Inspection of the semi-partial correlation coefﬁcients indicated that most of the

variance was accounted for by Agreeableness, suggesting that the rate with

97

98
which one processes information has little to do with how accurately perceptual
information is processed, at least in this sample.

Different results were observed with Holistic Organization. When
information processing rate and personality were entered together, only the
equation containing digit symbol scores and Openness approached statistical
signiﬁcance. Furthermore, it was the level of Openness, and not information
processing rate, that demonstrated a trend towards capturing a signiﬁcant
amount of the variance in the Holistic Organization factor when both were in the
equation together. As with the Perceptual Accuracy factor, the results suggest
that personality processes, in particular Openness, have a greater inﬂuence on
holistic processing and integration than rate of processing information.

In contrast to the results observed with the factor variables Perceptual
Accuracy and Holistic Organization, information processing rate and a given
personality construct predicted a signiﬁcant amount of the variation in the
General Response Process factor in all cases. Thus, Digit Symbol scores and
Openness accounted for a highly signiﬁcant 34% of the variance in General
Response Process, while digit symbol scores and Neuroticism, Extraversion,
Agreeableness, and Conscientiousness, respectively, accounted for 27% of the
variance in general responding. In all cases, it was information processing rate
that independently contributed a signiﬁcant amount of the variance to the factor
of General Response Process; only the personality factor Openness also

predicted a signiﬁcant amount of the variance as well.

99

Overall, these results highlight the difference in nature of the Perceptual
Accuracy, Holistic Organization, and General Response Process variables. In
general, the multiple regression analyses indicated that there was a trend for
personality processes, in particular Agreeableness and Openness, to make a
greater contribution to Perceptual Accuracy and Holistic Organization,
respectively, when information processing rate was also considered. In contrast,
information processing rate was a unique predictor of the factor variable General
Response Process, regardless of the personality domain, though Openness was
also a signiﬁcant predictor. These results suggest that, for this sample of older
adults, being open to experience was the key personality factor to better and
more productive protocols, as least as deﬁned by the General Response Process
variable. Since the other domains demonstrated no association with General
Response Process (RSP), it would appear that those personality qualities were
not as important to the aspects of the responding represented by the RSP factor
variable.

The latter analyses were an attempt to statistically test the contributions of
perception and representation to the Rorschach task, especially in regard to the
ﬁve-factor theory of personality. As such, the contribution of each personality
domain was considered. In reality, of course, people are not just open to
experience, or disagreeable, or extraverted. Everyone’s personality is a
composite of all ﬁve domains in varying degrees, and no one is devoid of any

factor. For example, one might be low in agreeableness, but one does not lack

100
the characteristic altogether. In the end, it is this unique combination of an
individual’s personality that interacts with cognitive processes to form a
Rorschach response. This is likely the reason that even when information
processing and personality accounted for a statistically signiﬁcant proportion of
the variance in Rorschach performance, over half of variance was still left
unaccounted for. Teasing out the factors that may inﬂuence Rorschach
response formation is not unimportant, especially with groups such as older
adults, who perform differently than younger adults on many psychological tests.
This research, however, again emphasizes the complex processes involved in
responding to a Rorschach inkblot, and how it is unique among the panoply of
psychometric assessment instruments.
Further Considerations
Post-hoc analyses further examined the nature of the relationship
between the factors of Perceptual Accuracy, Holistic Organization, General
Response Process, and the demographic variables under examination. Notable
ﬁndings were observed only with the Rorschach factor variable General
Response Process, which demonstrated statistically signiﬁcant associations with
age, estimated premorbid verbal abilities, number of medical diagnoses and
medications, and residential status. Holding the ﬁrst four variables constant,
however, attenuated the relationship between residential status and General
Response Process (RSP), suggesting that the association between place of

dwelling and RSP was mediated by age, health status, and premorbid verbal

101
abilities, all of which were signiﬁcantly different in the residential care facility
group. Additionally, General Response Process was regressed on age and
information processing rate, since both of the latter variables were signiﬁcantly
associated with general Rorschach responding. Results of the regression
indicated that age made no signiﬁcant contribution to RSP once information
processing rate was accounted for. The lack of association between Perceptual
Accuracy, Holistic Organization, and the demographic variables examined
suggests that, in these older adults, these factor variables were not inﬂuenced by
years of education, current mental status, estimated premorbid verbal
intelligence, medical diagnoses, or medication regimen.
Methodological Issues

As indicated throughout the discussion of the results, several factors limit
the generalizability of the ﬁndings. Though age was a normally distributed
variable, recruiting limitations resulted in some decades being over-represented:
There were only seven subjects in their sixth decade, 33 in their seventh decade,
29 in their eighth, and 11 in their ninth decade. Thus, an even distribution of age
by decade was not obtained, perhaps attenuating potential age effects among
predictor and criterion variables. Secondly, the sample was not ethnically
diverse; moreover, the majority of community-dwelling adults were recruited from
a religious faith advocating a healthy lifestyle. Both the lifestyle aspect, as well
as membership in a group offering the companionship and support of others,

may make these participants different from other older adults in the general

1 02
community. The entire sample was more educated than is observed with current
cohorts of the same age; thus, other results may have been obtained with a less
educated group. Lastly, difﬁculties with recruitment meant that subjects with
neurologic histories were not excluded, as long as they could independently
complete the MMSE and digit symbol tasks; this created a confound, however,
as digit symbol performance in these individuals could have been inﬂuenced by
age, disease process, or (more likely), some combination thereof.
Final Considerations

In this group of older adults, only the factor variable General Response
Process varied as a function of age; Perceptual Accuracy and Holistic
Organization did not. Thus, at least in terms of perceptual accuracy and holistic
integration, persons in their seventies were producing protocols similar to
persons 10 to 15 years their senior. Nonetheless, this sample did evidence
signiﬁcant differences in the determinants and scores in question as compared to
the relatively young Exner (1990) normative sample. It has been suggested that
it is inappropriate to compare small sample research to the normative data base
because of the differences in reference group patterns (Dies, 1995). While this
may be true, and while clinical interpretation of data does depend to a large
degree on how scores relate to each other, normative reference groups do
provide a set of expectations which guide interpretation.

The results of the present study indicated that, in older adults, differences

in information processing inﬂuenced some aspects of Rorschach performance to

103
some degree. Since older adults are know to be slower information processors
than young adults, perhaps this provides a clue to why this sample produced
signiﬁcantly different frequencies of scores and determinants as compared to
Exner’s normative group. In addition, differences in personality traits, particularly
agreeableness and openness, seemed to contribute to variation in Rorschach
performance in this group of older persons. The fact remains, however, that a
large part of the variance in Rorschach performance was unaccounted for the
variables examined, which included cognitive, personality, and demographic
factors.

Why, then, was this group of older adults, like others in the research,
signiﬁcantly different in all measured Rorschach variables as compared to
Exner’s (1990) normative sample? One possibility is that this particular group of
adults would have always responded in that manner. That is, had they been
followed longitudinally, no signiﬁcant changes would have been observed in their
protocols from time 1 to time 2. Another is that this is cohort effect, unique to this
particular group of older adults and perhaps a function of a lack of psychological
sophistication. Lastly, and perhaps more likely, is that the psychological tasks of
older adults differs from younger adults, and it is this difference that is inﬂuencing
the manner in which older adults respond. Coming to terms with impending
death, possible frailty and deteriorating death, the loss of roles, decreased
autonomy, the loss of loved ones, and how one has lived life is a responsibility

often facing people as they age. For some, considering and resolving these

104
issues can impact their psychosocial and affective adjustment and the quality of
life in their later years. In fact, in Erikson’s developmental theory, the last stage
of life is characterized by a conﬂict between integrity and despair, a conﬂict that
results in wisdom if resolved (Erikson, Erikson, & Kivnick, 1986). Thus, it is
perhaps this psychological process that is inﬂuencing the manner in which older
adults respond to the Rorschach.

The latter observation suggests a direction for future research. Assessing
the developmental tasks of older adults and how they relate to, or inﬂuence,
Rorschach response formation, would be helpful in elucidating normative
expectations for performance. In this way, patterns of responding arising from
developmental issues (if there are any) could be differentiated from responding
indicative of distress or other pathological processes. Of course, large scale,
longitudinal Rorschach studies with older adults, demographically similar to those
in the general population, are sorely needed to elucidate what changes, if any,
are to be expected on the task with age. For example, it could be that an
investigation with a larger sample size would not replicate the differences in
responding observed between Exner’s normative data base and the older adults
in the present study. Furthermore, perhaps the current results are a function of
the cohort studied; it could be that future generations of older adults would
respond differently again, due to presumable changes in psychological
mindfulness and sophistication. These questions demand longitudinal

investigations across the age spectrum, taking into account potentially

105

moderating demographic variables and cohort differences in psychological
development.

In conclusion, the results of this study indicate that the Rorschach can be
a useful assessment tool with older adults residing both within the community
and care facilities. Several caveats must be kept in mind, however. This
investigation suggests that older persons may produce protocols which differ
from younger adults partly as a function of differences in cognitive processes,
and partly due to differences in the personality trait of openness and perhaps
agreeableness. Nonetheless, other as yet unmeasured variables appear to be
contributing to the majority of variance in performance. In sum, these factors
should be considered when clinically interpreting the Rorschach protocol of an
older adult, especially if differences are found between the latter and normative

data.

REFERENCES

Abrahams, J. P., & Birren, J. E. (1983). Reaction time as a function of
age and behavioral predisposition to coronary artery disease. Journal of
Gerontology, 28, 471-478.

Acklin, M. W. (1993). Integrating the Rorschach and the MMPI in
clinical assessment: Conceptual and methodological issues. Journal of
Personalib/ Assessment. 60 (1 ), 125-131 .

Albert, M. S. (1988). General issues in geriatric neuropsychology. In M.
S. Albert & M. B. Moss (Eds), Geriatric neuropsycholggy (pp. 3-10). New York:
Guilford Press.

Albert, M. 8., Duffy, F. H., & Naeser, M. (1987). Nonlinear changes in
cognition with age and their neuropsychologic correlates. Canadian Journal of
Psycholcmy,41. 141-157.

Albert, M. S., & Moss, M. B. (Eds). (1988). Geriatric neuropsycholggy.
New York: Guilford Press.

Ames, L. B. (1960). Age changes in the Rorschach responses of a group
of elderly individuals. The Journal of Gegetic Paycholggv. 97. 257-285.

Ames, L. B. (1966). Changes in Rorschach response throughout the
human life span. Ginetic Psychology Monographs, 74, 89-125.

Ames, L. B. (1974). Calibration of aging. MI of Parsonm
Assessment. 38. 507-519.

Ames, L. B., Metraux, R. W., Rodell, J. L., & Walker, R. N. (1973).

Rorschﬁh rﬁponses in old age. New York: Brunner/Mazel.

106

107

Archer, R. P., & Krishnamurthy, R. (1993). A review of MMPI and
Rorschach interrelationships in adult samples. WM
AssessmenL 61(2), 277-293.

Atkinson, L. (1986). The comparative validities of the Rorschach and
MMPI: A meta-analysis. Canagian Psychology,_5_2_, 238-247.

Bayles, K. A., Kaszniak, A. W., & Tomoeda, C. K. (1987).
Communication 3&1 cognition in normal aging and dementia. Boston: College-
Hill.

Birren, J. E., & Schaie, K. W. (Eds). (1985). Handbook of the
Lsycholgy of agi_ng (2nd ed). New York: Van Nostrand Reinhold.

Birren, J. E., & Schaie, K. W. (Eds). (1990). Handbook of the
psychology of agjgg (3rd ed). San Diego: Academic Press.

Birren, J. E., Sloan, R. B., & Cohen, G. D. (1992). H_an_dp_pok of man_tal
health and aging (2nd ed). San Diego: Academic Press.

Blatt, S. J. (1990). The Rorschach: A test of perception or an evaluation
of representation. Journal of Personality Assessment, 55 (3&4), 394-416.

Brody C. M., & Semel, V. G. (1993). Strategies for therapy with the
m New York: Springer.

Brody, C. M. (1993a). Nursing homes. In C. M. Brody & V. G. Semel
(Eds), gaggies for theraﬂwith the Lkierlv (pp. 3-58). New York: Springer.

Brody, C. M. (1993b). Medicare. In C. M. Brody & V. G. Semel (Eds),
Stat—egies for therapy with the eﬂerly (pp. 141-158). New York: Springer.

Bromley, D. B. (1967). Age and sex differences in the serial production

of creative conceptual responses. Journal of Gerontology, 22, 32-42.

108

Busse, E. W., & Blazer D. G. (Eds). (1996). Taxjpook of Geriatrg
Psychiatg (2nd ed). Washington: The American Psychiatric Press.

Butler, R. M, Lewis, M. l., 8 Sunderland, T. (1991). Aging and Mental
Health. Positive psychosocial and biomedical approaches. New York:
MacMillan.

Caldwell, B. M. (1954). The use of the Rorschach in personality
research with the aged. gpgrnal of Gerontology,_9, 316-323.

Camp, C. J., West, R. L., & Poon, L. W. (1989). Recruitment practices
for psychological research in gerontology. In M. P. Lawton & A. R. Herzog
(Eds), Special research methods for ﬁrontolcmy (pp. 163-189). New York:
Baywood.

Campion, E. W., Ban, A., & May, M. (1983). Why acute-care hospitals
must undertake long-term care. New England Joprnal of Mgljcine. 3084 71-75.

Cerella, J. (1985). Information processing rates in the elderly.
Psychological Bulletin, 98, 67-83.

Costa, P. T., Jr., & McCrae, R. R. (1980). Still stable after all these
years: Personality as a key to some issues in adulthood and old age. Life-span
Development ancLBehavior. 3. 65-102.

Costa, P. T., Jr., & McCrae, R. R. (1985). The NEO Personality Inventory
_nmual. Florida: Psychological Assessment Resources.

Costa, P. T., Jr., & McCrae, R. R. (1986). Age, personality, and the
Holtzman inkblot technique. lnternaLticmal Joprnal of Agjggand Human

Development. E_(2), 1 15-125.

109

Costa, P. T., Jr., & McCrae, R. R. (1992). W
Inventory MFR) agd_l_\IE(_) Five-Factor Inventory (NEO-FFI). Profassional
W. Florida: Psychological Assessment Resources.

Costa, P. T., Jr., McCrae, R. R., Zonderman, A. B., Barbano, H. E.,
Lebowitz, B., & Larson, D. M. (1986). Cross-sectional studies of personality in a
national sample: 2. Stability in neuroticism, extraversion, and openness.
Psychology and Aging, 1, 144-149.

Costa, P. T., Jr., Metter, E. J., & McCrae, R. R. (1994). Personality
stability and its contribution to successful aging. Journal of Geriatric Psychiatgy,
21, 41-59.

Davidson, H. H., 8. Kruglov, L. (1952). Personality characteristics of the
institutionalized aged. Journal of Consrﬂgfaychology. 16, 5-12.

Dies, R. R. (1995). Conceptual issues in Rorschach research. In J. E.
Exner (Ed), lisges ang methcﬁs in Rorschach reseam, (pp. 99-122). New
Jersey: Lawrence Erlbaum Associates.

Digman, J. M. (1990). Personality structure: Emergence of the ﬁve-factor
model. Mal Review of Psychology, 41. 417-440.

Dollinger, S. J., & Clancy, S. M. (1993). Identity, self, and personality: ll.

Glimpses through the autophotographic eye. Journal of Personality and Social
Psycholoml. 64(6), 1064-1071 .
Dollinger, S. J., & Leong, F. T. L. (1993). Volunteer bias and the ﬁve-

 

factor model. The Jognal of Psycholgqy. 12_7(1), 29-36.
Domitrovic, L. A. (1994). lnforrnation processing rate and Rorschach
in_dices of comiitive minctionirmn nufrsing home residents. Unpublished master’s

thesis, Michigan State University.

1 10
Dye, E. J. (1982). The experience of separation at the time of placement
in long-term care facilities. Psychotherapy: Theory. Research and Practice

19(4), 532-537.

 

Eisdorfer, C. (1960a). Rorschach rigidity and sensory decrement in a
senescent population. Journal of Gerontology. 15, 188-190.

Eisdorfer, C. (1960b). Developmental level and sensory impairment in
the aged. Journal of Proiective Techniggesazg, 129-132.

Eisdorfer, C. (1963). Rorschach performance and intellectual functioning
in the aged. Journal of Gerontolggy. 18. 358-363.

Elias, M. F., Robbins, M. A., Schultz, N. R., Jr., & Pierce, T. W. (1990). Is
blood pressure an important variable in research on aging and
neuropsychological test performance? ﬂrnal of Gerontology: Psychological
Sciences, 45LP128-P135.

Ellis, D., & Zahn, B. (1985). Psychological functioning after severe closed
head injury. Journal of Personality Assessment. 49, 125-128.

Engle V. F., & Graney, M. J. (1993). Predicting outcomes of nursing
home residents: Death and discharge home. Journal of Gerontolgqv: Sorﬂ
SciencesAB. $269-$275.

Erber, J. T., Botwinick, J., & Storandt, M. (1981). The impact of memory
on age differences in digit symbol performance. ﬂimal of Gerontology. 36.
586-590.

Erikson, E. H., Erikson, J. M. & Kivnick, H. O. (1986). Vital involvament in
Mg; New York: W. W. Norton & Company.

Exner, J. E. (1986). The Rorschach: A compghensive system, Volume

1: Basic Foundations, (2nd ed). New York: Wiley.

111

Exner, J. E. (1990). A Rorschach wo@ook for the comprehensive
§y§t_erp, (3rd ed). Asheville, NC: Rorschach Workshops.

Exner, J. E. (Ed). (1995). Issues and_ methogs in R0@h_ach research.
New Jersey: Lawrence Erlbaum Associates.

Exner, J. E., Armbruster, G., & Mittman, B. (1978). The Rorschach
response process. Journal of Persoryalitv Assessment. 42, 27-38.

Exner, J. E., Colligan, S. C., Boll, T. J., Stischer, B., & Hillman, L. (1996).
Rorschach ﬁndings concerning closed head injury patients. Assessment 3(3),
317-326.

Exner, J. E., Kinder, B. N., & Curtiss, G. (1995). Reviewing basic design

 

features. In J. E. Exner (Ed), Issues and mettﬂs in Rorschach researm
(pp.145-158). New Jersey: Lawrence Erlbaum Associates.

Exner, J. E., Viglione, E. A., & Gillespie, R. (1985). Relationships
between Rorschach variables as relevant to the interpretation of structural data.
Journal of Personality Assessment, 4; 65-70.

Feifel H., & Strack, S. (1987). Old is old is old? Psychology and Aging,
2_(4), 409-412.

Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). "Mini-mental
state". A practical method for grading the cognitive state of patients for the
clinician. Jamal of Psychiatric Research, 12, 189-198.

Folstein, M., Anthony, J. C., Parhad, l., Duffy, B., & Gruenberg, E. M.
(1985). The meaning of cognitive impairment in the elderly. Journal of the

American Geriatrics Society. 33. 228-235.

1 12

Fozard, J. L. (1990). Vision and hearing in aging. In J. E. Birren & K. W.
Schaie (Eds), Hancﬁiook of the Psycholggy of Aging (3rd ed., pp. 150-165). San
Diego: Academic Press.

Gilmore, G. C., Royer, F. L., & Gruhn, J. J. (1983). Age differences in
symbol-digit substitution task performance. ggu_rnal of Clinical Fstchology,
Q0), 114-124.

Golander, H. (1995). Chronically ill, old, and institutionalized: Being a
nursing home resident. Family and Community Haalth. 17(4), 63-79.

Gold, 0., Andres, D., Arbuckle, T., & Schwartzman, A. (1988).
Measurement and correlates of verbosity in elderly people. Journal of
Gerontology: Psychological Sciences. 43. P27-P33.

Goldfried, M. R., Sticker, G., & Weiner, I. B. (1971). Rorschach
_ha_ngbook of clinical and research applications. Englewood Cliffs, New Jersey:
Prentice-Hall.

Gorsuch, R. L. (1983). Factor Analysis. Hillsdale, NJ: Lawrence
Erlbaum Associates.

Greenley, R. J. (Ed). (1984). Reseaﬁ in community mental healtg,
London: JAI Press.

Greenwald, D. F. (1990). An external construct validity study of
Rorschach personality variables. _Jo_urnal of Personality Assessment. 55(384),
768-780.

Gross, A., Newton, R. R., & Brooks, R. B. (1990). Rorschach responses
in healthy, community dwelling older adults. Mrnal of Personality Assessment.

55, 335-343.

1 13

Hathaway, S. R. & McKinely, J. C. (1989). Minnesota Muﬂhasic
Personality Inventory - L Mapgal for ﬁministration _a_nds_corir_ig. Minneapolis:
University of Minnesota Press.

Hathaway, S. R., & McKinley, J. C. (1940). A multiphasic personality
schedule (Minnesota): I. Construction of the schedule. Journal of Psychology,
_1__Q, 249-254.

Hayslip, B. (1981). Verbosity and projective test performance in the
aged. ﬂirnal of Clinical Psychology. 37(3), 662-666.

Hayslip, B., & Loman, R. L. (1986). The clinical use of projective
techniques with the aged. A critical review and synthesis. Clinical Gerontolggjs_t,
Q. 63-93.

Hertzog, C. (1989). Inﬂuences of cognitive slowing on age differences in
intelligence. Developmental Psychologyﬁ, 636-651.

Hertzog, C., Schaie, K. W., & Gribbin, K. (1978). Cardiovascular disease
and changes in intellectual functioning from middle to old age. Journal of
Gerontology. 33. 872-833.

Holland, C. A., & Rabbitt, P. M. A. (1991). Ageing memory: Use versus
impairment. British Jogrnal of Psycholggy. 82, 29-38.

Holtzman, W. H. (1961). Giggle to administration and scoring Holﬁtzman
inkblot technigua New York: Psychological Corporation.

Holzer, C. E., Tischler, G. L., Leaf, P. J., Myers, J. K. (1984). Community
cognitive impairment. In R. J. Greenley (Ed), Research in commpnity mental
Lew; London: JAI Press.

Howes, R. J. (1981). The Rorschach: Does it have a future? Journal of

Personality Assessment, 45 339-351.

1 14

Hoyer, W. J., Lopez, M. A., & Goldstein, A. P. (1982). Predicting social
skill acquisition and transfer by psychogeriatric inpatients. lntemational Jogrnal
of BehavioraiGerijatrics. 1(1), 43-46.

Hughes, R. M. (1948). Rorschach signs for the diagnosis of organic
pathology. Rorschach Research Exchange and Journal of Personalig
Assessment, 1g. 165-167.

lngersoII-Dayton, B., & Rader, J. (1993). Searching for solutions: Mental
health consultation in nursing homes. Clinical Gerontologist. 13(1), 33-50.

Insua, A. M., & Loza, S. M. (1986). Psychometric patterns on the
Rorschach of healthy elderly persons and patients with suspected dementia.
Perceptual and Motor Skills. 63. 931-936.

Kafonek, S., Ettinger, W. H., Roca, R., Kittner, S., Taylor, N., & German,
P. S. (1989). Instruments for screening for depression and dementia in a long-
term care facility. Jpgrnal of the American Geriatrics Society. 36. 699-706.

Kahana, B. (1978). The use of projective techniques in personality
assessment of the aged. In M. Storandt, I.C. Siegler, & M.F. Elias, (Eds), Illa
clinical ﬁVChOiOQV of agigg (pp. 145-180). New York: Plenum Press.

Kaszniak, A. W., & DiTraglia Christenson, G. (1994). Differential
diagnosis of dementia and depression. In M. Storandt & G. R. VandenBos
(Eds), Neuropsychological assessment of dementiﬂtg depression in olgar:
adults: A clinician's guide (pp. 81-118). Washington: American Psychological
Association.

Kaufman, A. S. (1990). Assessing adolescent ancﬂgult intelligence.

Boston: Allyn and Bacon, Inc.

1 15

Kettell, M. E. (1976). Perceptual and behavioral correlates of "organicity"
in old age. go_urnal of Geriatric Psycmuy. g 85-87.

Klopfer, W. G. (1946). Personality patterns of old age. Rorschach
Research Exchaaqe. 10. 147-151.

Klopfer, W. G. (1974). The Rorschach and old age. Journal of
Personality Assessment, 38. 420-422.

Koenig, H. G., Meador, K. G., Cohen, H. J., & Blazer, D. G. (1988). Self-
rated depression scales and screening for major depression in the older
hospitalized patient with medical illness. Journal of the American Geriatrics

Society. 36. 699-706.

 

Korchin, S. J., & Larson, D. G. (1977). Form perception and ego
functioning. In M. A. Rickers-Ovsiankina (Ed), Rorschach Psychol’ogy.
Huntington, NY: R.E. Krieger.

Kulka, R. A., Veroff, J., & Douvan, E. (1979). Social class and the use of
professional help for personal problems: 1957 and 1976. Journal of Health a_rjg

Social Behavior 20 2-17.

 

La Rue, A. (1992). Aging and Neuropsychological Assessment. New
York: Plenum Press.

La Rue, A., & Markee, T. (1995). Clinical assessment research with older
adults. Psychological Assessment. 7(3), 376-386.

Lawton, M. P. & Herzog, A. R. Special research methﬁ for gerontology.
New York: Baywood.

LeFever, F. F. (1991). Relationships between digit symbol test and a
noncoding motoric equivalent in assessments of brain-damaged patients.

Perceptual and Motor Skills. 72, 65-66.

1 16

Lezak, M. D. (1995). Neuropsychological assessment (3rd ed). New
York: Oxford University Press.

Liptzin, B. (1992). Nursing home care. In J. E. Birren, R. B Sloan, & G.
D. Cohen (Eds), Llﬂgbook of mental health and aging (2nd ed., pp. 833-852).
San Diego: Academic Press.

Madden, D. J. (1985). Age-related slowing in the retrieval of information
from long-term memory. Mrnal of Gerontology. 40. 208-210.

Maddox, G. L., & Douglass, E. B. (1974). Aging and individual
differences: A longitudinal analysis of social, psychological, and physiological
indicators. Mrnal of Gerontolmy,_29(5), 555-563.

Manton, K. G., Cornelius, E. S., & Woodbury, M. A. (1995). Nursing
home residents: A multivariate analysis of their medical, behavioral,
psychosocial, and service use characteristics. ﬂrnal of Gerontolm Me,d_i§_al
Sciences, 50A(5), M242-M251.

Manton, K. G., Liu, K., & Cornelius, E. S. (1985). An analysis of the
heterogeneity of US. nursing home patients. J_ougrnal of (gerontology. 40(1), 34-
46.

Manton, K., G., & Stallard, E. (1991). Cross-sectional estimates of active
life expectancy for the US. elderly and oldest-old populations. Journal of
Gerontolggy: Social Sciences. 46. $170-$182.

Martin, P., Poon, L. W., Clayton, G. M., Lee, H. S., Fulks, J. S., &
Johnson, M. A. (1992). Personality, life events and coping in the oldest-old.

International Journal of Aging and HumanQevelopment. 34(1), 19-30.

117

Mattlar, C. E., Knuts, L. R., & Virtanen, E. (1985). Personality structure
on the Rorschach for a group of healthy 71-year old females and males.
Proiective PsychologL 30(1), 3-8.

Mazaux, J. M., Dartigues, J. F., Letenneur, L., Darriet, D., Wiart, L.,
Gagnon, M., Commenges, D., & Boller, F. (1995). Visuo-spatial attention and
psychomotor performance in elderly community residents: Effects of age,
gender, and education. Journal of Clinical and Experimental Neuropsycholggy,
1_7(1), 71-81.

McConnel, C. E. (1984). A note on the lifetime risk of nursing home
residency. The Gerontologiat,_2_t_l(2), 193-198.

McCrae, R. R. (1987). Creativity, divergent thinking, and openness to
experience. Journal of Personality anc_l__Social PsychologLiZﬁ 258-1265.

 

McCrae, R. R. (1991). The ﬁve-factor model and its assessment in
clinical settings. Journal of Personality Assessment. 57. 399-414.

McCrae, R. R., & Costa, P. T., Jr. (1986). Clinical assessment can
beneﬁt from recent advances in personality psychology. American Psvchologiat,
£19), 1001-1003.

McCrae, R. R., & John, O. P. (1992). An introduction to the ﬁve-factor
model and its applications. Jggrnal of Personality. 60. 175-215.

Menich, S. R., & Baron, A. (1987). Behavioral slowing and
neuropsychological signs in a sample of active older men. _Elgerimental Agi_ng
Research. 13(1), 23-27.

Meyer, G. J. (1992). The Rorschach’s factor structure: A contemporary
investigation and historical review. _gggmal of Personality Assessment. 59. 117-

136.

1 18

Meyer, G. J. (1993). The impact of response frequency on the
Rorschach constellation indices and on their validity with diagnostic MMPI-2
criteria. Journal of Personality Assessment. 60, 153-180.

Mitrushina, M., & Satz, P. (1990). Cardiovascular status and other
factors in speed of mental processes. Brain Dysfunction, 3, 151-155.

Moos, R. H., & Lemke S. (1985). Specialized living environments for
older people. In J. E. Birren & K. W. Schaie (Eds), Handbook of the psychology
W (2nd ed., pp. 864-889). New York: Van Nostrand Reinhold Company.

Myers, J. K., Weissman, M. M., Tischler, G. L., Holzer, C. E., Leaf, P. J.,
Orvaschel, H., Anthony, J. C., Boyd, J. H., Burke, J. D., Kramer, M., & Stoltzman,
R. (1984). Six-month prevalence of psychiatric disorders in three communities:
1980-1982. Archives of General Psychia_try. 41. 959-967.

Myerson, J., Hale, S., Wagstaff, D., Poon, L. W., & Smith, G. A. (1990).
The information-loss model: A mathematical theory of age-related cognitive
slowing. Psychological Review. 97. 475-487.

Neils, J., Baris, J. M., Carter, C., Dell’aria, A. L., Nordloh, S. J., Weiler, E.,
& Weisiger, B. (1995). Effects of age, education, and living environment on
Boston naming test performance. J_o_gmal of Speech and Hearing Research, 38,
1 143-1 149.

Neugarten, B. L. (1975). The future and the young-old. 102
GerontologisL 15. 4-9.

Norman, W. T. (1963). Toward an adequate taxonomy of personality
attributes: Replicated factor structure in peer nomination personality ratings.

_Jpgrnal of Abnormal and Social Psycholgqy. 66. 574-583.

119

Oluwafemi Agvayewa, M., Ong, A., & Wilden, B. (1990). Empowering
long term facility residents using a resident staff group approach. Cli_nical
Gerontologist. 9(3-4), 191 -201 .

O'Riordan, T. G., Hayes, J. P., O'Neill, 0., Shelley, R., Walsh, J. B., &
Coakley, D. (1990). The effect of mild to moderate dementia on the Geriatric
Depression Scale and on the General Health Questionnaire. Age and Ageing,
1a. 57-61.

Osterweil, D., Mulford, P., Syndulko, K., & Martin, M. (1994). Cognitive
function in old and very old residents of a residential facility: Relationship to age,
education, and dementia. Mrnal of the American Geria_t_rics Societyaﬂ, 766-
773.

Panek, P. E., Wagner, E. E., & Kennedy-Zwergel, K. (1983). A review of
projective test ﬁndings with older adults. Mrnal of Personality Assessment 47
562-582.

Parker, K. C. H., Hanson, R. K., & Hunsley, J. (1988) MMPI, Rorschach

 

and WAIS: A meta-analytic comparison of reliability, stability, and validity.
Psycholoﬂal Bulletin. 103. 367-373.

Parmelee, P. A., Lawton, M. P., & Katz, I. R. (1989). Psychometric
properties of the Geriatric Depression Scale among the institutionalized aged.
Psvcholﬂorﬂcal Assessment. 1(4), 331-338.

Perlmuter, L. C., Tun, P., Sizer, N., McGIinchey, R. E., & Nathan, D. M.
(1987). Age and diabetes related changes in verbal ﬂuency. Experimental
AgijResearch. 13(1), 9-14.

120
Perlmutter, M., & Nyquist, L. (1990). Relationships between self-reported
physical and mental health and intelligence performance across adulthood.
Journal of Gerontology: Psychological Sciences, 45, P145-155.
Perry, W., Potterat, E., Auslander, L., Kaplan, E. & Jeste, D. (1996). A
neuropsychological approach to the Rorschach in patients with dementia of the

Alzheimer type. Assessment. 3(3), 351-363.

Peterson, C. A. (1991). Reminiscence, retirement, and Rorschach

Responses in Old Age. Jamal of Personality Assessment. 57(3), 531-536.
Peterson, C.A., & Schilling, K. M. (1983). Card pull in projective testing.

Journal of Personality Assessment. 47. 265-275.

Piotrowski, C., & Keller, J. W. (1989). Psychological testing in outpatient
mental health facilities: A national study. ProfessionﬂsvchologLResearch
and Practice 20 423-425.

Piotrowski, C., Sherry, D. G., & Keller, J. W. (1985). Psychodiagnostic

 

test usage: A survey of the Society for Personality Assessment. Journal of
Personality Assessment. 49. 1 15-1 19.

Piotrowski, Z. (1936). On the Rorschach method and its application in
organic disturbances of the central nervous system. Rorschach Reseaﬂl
Exchaaqe, 1, 23-40.

Piotrowski, Z. (1940). Positive and negative Rorschach organic
reactions. Rorschach Research Exchaaqe. 4. 147-151.

Prados, M., & Fried, E. G. (1947). Personality structure of the older age
groups. _J_9grnal of Clinical Psychology. 3. 113-120.

Rapaport, D., Gill, M. M., & Schafer, R. (1968). Diawstic Psychological
Iaagng (Vol 1). Chicago: Year Book Medical.

121

Reichlin, R. E. (1984). Current perspectives on Rorschach performance
among older adults. Journal of Personally Assessment, 48, 71-81.

Rickers-Ovsiankina, M. A. (1977). _Fg)_r_§ch_ach Psychology (2nd ed).
Huntington, NY: R. E. Krieger.

Riemann, R., Grubich, C., Hempel, S., Mergl, S., & Richter, M. (1993).
Personality and attitudes towards current political topics. Personalig and
Individual Differences. 15(3), 313-321.

Rorschach, H. (1942). Psychodiagnostics. Adiagnostic test based on
perception. Berne, Switzerland: Verlag Hans Huber.

Royer, F. L., Gilmore, G. C., & Gruhn, J. J. (1981). Normative data for
the symbol digit substitution task. Journal of Clinical Psychology. 37(3), 608-614.

Sadavoy, J., Lazarus, L. W., Jarvik, L. F., & Grossberg, G. T. (Eds).
(1996). Comprehensive Review of Geriatric Psychiatry-ll (2nd ed).

Washington: The American Psychiatric Press.

Salamon, M. J., (1986). Measures for psychotherapy outcome in
institutionalized aged. Clinical Gerontologist. 6(1), 62-64.

Salthouse, T. A. (1978). The role of memory in the age decline in digit-
symbol substitution performance. Journal of Gerontology. 33(2), 232-238.

Salthouse, T. A. (1985). A Theory of Cognitive Agi_n_g. New York: North-
Holland.

Salthouse, T. A. (1987). Adult age differences in integrative spatial
ability. Psychology and Aging, 2, 254-260.

122

Salthouse, T. A. (1988a). Effects of aging on verbal abilities:
Examination of the psychometric literature. In L. L. Light & D. M. Burke (Eds),
Language. memory. anc_l_agi_ng (pp. 17-35). New York: Cambridge University
Press.

Salthouse, T. A. (1988b). The role of processing resources in cognitive
aging. In M. L. Howe & C. J. Brainerd (Eds), Cognitive development in
agulthood. Progress in cognitive (_Ieveloament researgp (pp. 185-239). New
York: Springer-Verlag.

Salthouse, T. A. (1991). Theoretical Perspectives on ngnitive Aging.
Hillsdale, New Jersey: Lawrence Erlbaum.

Salthouse, T. A. (1992a). Inﬂuence of processing speed on adult age
differences in working memory. Acta Psycholgaica. 79. 155-170.

Salthouse, T. A. (1992b). What do adult age differences in the digit
symbol substitution test reﬂect? Journal of Gerontology: Psychological
Sciences. 47. P121-P128.

Salthouse, T. A. (1993). Speed mediation of adult age differences in
cognition. Developmental Psychologyﬁ, 722-738.

Salthouse, T. A. (1994). The nature of the inﬂuence of speed on adult
age differences in cognition. Developmental Psychology,_3_(_l. 240-259.

Salthouse, T. A., & Babcock, T. L. (1991). Decomposing adult age
differences in working memory. DevelomneLMI Psychology,_21, 763-776.

Salthouse, T. A., 8 Coon, V. E. (1993). Inﬂuence on task-speciﬁc
processing speed on age differences in memory. _Jo_ug1al of Gerontology

Psychological Sciences, 48. P245-P255.

123

Salthouse, T. A., & Kersten, A. W. (1993). Decomposing adult age
differences in symbol arithmetic. Memom & Cognition, 21(5), 699-710.

Salthouse, T. A., & Mitchell, D. R. D. (1990). Effects of age and naturally
occurring experience on spatial visualization performance. Developmental
Psychology, 26, 845-854.

Salthouse, T. A., & Prill, K. A. (1988). Effects of aging on perceptual
closure. American Joprnal of Psychology. 101. 217-238.

Salthouse, T. A., & Somberg, B. L. (1982). Skilled performance: Effects
of adult age and experience on elementary processes. Journal of Experimental
Psychology: General. 111. 176-207.

Salthouse, T. A., Kausler, D. H., & Saults, J. S. (1988). Utilization of
path-analytic procedures to investigate the role of processing resources in
cognitive aging. Psychology and Aging, 3, 158-166.

Salthouse, T. A., Letz, R., & Hooisma, J. (1994). Causes and
consequences of age-related slowing in speeded substitution performance.
Developmental Neuropsychology. 10. 203-214.

Sands, L. P., & Meredith, W. (1992). Blood pressure and intellectual
functioning in late midlife. Journal of Gerontolog: Psychologica_l§ciencas. 47.

P81 -84.

 

Schaie, K. W. (1989). Perceptual speed in adulthood: Cross-sectional
and longitudinal studies. Psychology and Aging, 4, 443-453.

Schaie, K. W., & Willis, S. L. (1991). Adult development and aging (3rd
ed). New York: HarperCollins.

124

Scogin, F. R. (1994). Assessment of depression in older adults: A guide
for practitioners. In M. Storandt & G. R. VandenBos (Eds), Neuropsychological
assessment of dementia and depression in older adults: A clinician's guide (pp.
61-80). Washington: American Psychological Association.

Sharps, M. J. (1990). A developmental approach to the study of visual
cognition in the elderly. In T.M. Hess (Ed), Aging and cognition. K_nowledga
organization and utilization (pp. 297-341). New York: North-Holland.

Shimonaka, Y., & Nakazato, K. (1991). Aging and terminal changes in
the Rorschach responses among the Japenese elderly. Journal of Personaliy
Assessment, 57. 10-18.

Special Committee on Aging, US. Senate. (1996). _Develcmments in
Aging: 1996 (Vol. 1). Washington, DC: US. Department of Health and Human
Services.

Steenhuis, R. E., & Gstbye, T. (1995). Neuropsychological test
performance of speciﬁc groups in the Canadian Study of Health and Aging
(CSHA). addrnal of Clinical and Experimental Neuropsychology, 17(5), 773-785.

Storandt, M. (1994). General principles of assessment with older adults.
In M. Storandt & G. R. Vandenbos (Eds), Negrggpsvchological assessment of
dementia ancmgression in old_eradr_1lts: A clinician’s guide. Washington, DC:
American Psychological Association.

Storandt, M., & Vandenbos, G. R. (Eds). (1994). Neuropsychological
assessment of dementia ancﬂapression in older adyﬁs: A clinician’s 9%

Washington, DC: American Psychological Association.

125

Streim, J. E., & Katz, l. R. (1996). Clinical Psychiatry in the nursing
home. In E. W. Busse & D. G. Blazer (Eds), Textbook of Geriatric Psychiatry
(2nd ed., pp.413-432). Washington: The American Psychiatric Press.

Streim, J. E., Rovner, B. W., & Katz, I. R. (1996). Psychiatric aspects of
nursing home care. In J. Sadavoy, L. W. Lazarus, L. F. Jarvik, & G. T.
Grossberg (Eds), Comprehensive Review of Geriatric Paychiatgr-II (2nd ed., pp.
907-936). Washington: The American Psychiatric Press.

Sweeney, J. A., Clarkin, J. F., Fitzgibbon, M. L. (1987). Current practice
of psychological assessment. Professional Psychology: Research and Practice
1_8_, 377-380.

Tariot, P. N., Podgorski, C. A., Blazina, L., & Leibovici, A. (1993). Mental

 

disorders in the nursing home: Another perspective. American Journal of
Psychiatry. 150(7), 1063-1069.
Thomas, J. C., Fozard, J. L., & Waugh, N. C. (1977). Age-related

differences in naming latency. Amerigan Journal of Psychology, 90, 499-509.

 

Tobin, S. S. (1988). Preservation of the self in old age. gagial
Casework: The Jodrnal of Contemporary Social Work. 69(9), 550-555.

Tourigny-Rivard, M.F., & Drury, M. (1987). The effects of monthly
psychiatric consultation in a nursing home. The Gerontologistﬂw), 363-366.

Tun, P. A., Perlmuter, L. C., Russo, P., & Nathan, D. M. (1987). Memory
self-assessment and performance in aged diabetics and non-diabetics.
_E_x_perimental Agi_ng Research. 13(3), 151-157.

U. S. Department of Commerce. (1990). Censps of the population:

Social ana economic characteristics - United States. Washington, DC: Author.

126

Vaccaro, F. J. (1990). Application of social skills training in a group of
institutionalized aggressive elderly subjects. Psychology and Aging, 5(3), 369-
378.

Viglione, D. J. (1995). Basic considerations regarding data analysis. In
J. E. Exner (Ed), lsires and methods in Rorschach reseagell (pp. 195-226).
New Jersey: Lawrence Erlbaum Associates.

Wechlser, D. (1981). Wechsler Adult Intelligence Scale - Revised
_Mﬂal. New York: Psychological Corporation.

Weiner, I. B. (1986). Conceptual and empirical perspectives on the
Rorschach assessment of psychopathology. Jo_u_r_dal of Personality Aseessment.
§_(_I, 472-479.

Weiner, I. B. (1993). Clinical considerations in the conjoint use of the
Rorschach and the MMPI. Journal of Personality Assessment 60 148-152.

Weiner, I. B. (1994). The Rorschach inkblot method (RIM) is not a test:

 

Implications for theory and practice. Joumalﬁof Personality Assessment 62

498-504.

 

Weiner, l. B. (1995). Variable selection in Rorschach research. In J. E.
Exner (Ed), Issues and Mettﬂs in Rorschafch Reseam, (pp 73-98). New
Jersey: Lawrence Erlbaum Associates.

Wiggins, J. G. (1994). Introduction. In M. Storandt & G. R. VandenBos
(Eds), Neurmwholgqﬁassessment of dementia and depression in 0kg
ﬁults: A cliniciarﬁgﬂ (pp. 1-6). Washington: American Psychological
Association.

Wilkinson, G. S. (1993). Wide Range Achievement Test 3. Wilmington,

DE: Jastak Assessment Systems.

127

Williams, D. E., Vitiello, M. V., Ries, R. K., Bokan, J., & Prinz, P. N.
(1988). Successful recruitment of elderly community-dwelling subjects for
Alzheimer’s disease research. degrnal of Gerontology. 43. M69-M74.

Winocur, G., & Moscovitch, M. (1983). Paired-associate learning in
institutionalized and noninstitutionalized old people: An analysis of interference
and context effects. gedrnal of Gerontology. 38 455-464.

Winocur, G., & Moscovitch, M. (1990). A comparison of cognitive
function in community-dwelling and institutionalized old people of normal
intelligence. Canadian Journal of Psychology. 44(4), 435-444.

Winocur, G., Moscovitch, M., & Freedman, J. (1987). An investigation of
cognitive function in relation to psychosocial variables in institutionalized old
people. Canadian Journal of Psychology. 41(2), 257-269.

Winocur, G., Moscovitch, M., & Stuss, D. T. (1996). Explicit and implicit
memory in the elderly: Evidence for double dissociation involving medial
temporal- and frontal-lobe functions. Wycholgqy. 10(1), 57-65.

Yesavage, J. A., Brink, T. L., Rose, T. L., Lum, 0., Huang, V., Adey, M., &
Leirer, V. O. (1983). Development and validation of a geriatric depression
screening scale: A preliminary report. _J_ogrnal of Psychiatric Resaarch. 17(1),
37-49.

Zgourides, G., Spofford, M., & Doppelt, L. (1989). The Geriatric
Depression Scale: Discriminant validity and elderly day-treatment clients.
Psychological Reportsy64. 1082.

Zillmer E. A., & Perry, W. (1996). Cognitive-neuropsychological abilities
and related psychological disturbance: A factor model of neuropsychological,

Rorschach, and MMPI indices. Assessment. 3(3), 209-224.

APPENDIX A

Appendix A
Consent Form

MICHIGAN STATE UNIVERSITY
Department of Psychology

RESEARCH CONSENT FORM

Title of Project: Rorschach Responses in Older Adults Differing in Residential
Status: Resource and Trait Theory Perspectives

The following explains your rights as a potential participant in a research project; in
signing this paper, you are acknowledging that the experimenter has explained
these rights to you and that you have understood them.

1. Your participation in this study is voluntary; you may refuse to participate at

any time without penalty, even if you agree to participate now. Furthermore, even
if you agree, you may refuse to answer any of the tests or the experimenters
questions. For residents of nursing homes or the Grand Rapids Home for
Veterans, withdrawing your participation from the study, or refusing to answer tests
or questions, will not affect the care or services you are being provided in any way.
As a volunteer, you will receive no money for participating.

2. The study will take approximately three hours; any time you desire a break,
tell the experimenter and we will stop for as long as you wish. If you want to stop
for the day, tell the experimenter and another appointment will be scheduled at
your convenience.

3. Your participation in the study will be kept conﬁdential; only a code number, not
your name, will appear on the answer sheets. No one else will know how you
responded, or even that you participated, unless you tell them yourself.

4. The results of your individual contribution, the nature of the tests administered,
and/or the results of the study will be explained more at a later date, if you so
desire.

5. This study is being conducted by Lidia Domitrovic, M.A., under the supervision
of Dr. Norman Abeles at the Michigan State University Aging Project. The
measures to be administered are common psychological tests; by participating,
you are helping psychologists understand how people's answers change as they
grow older, what these changes may mean, and what kinds of tests might work
best with older adults.

Signed: Date:

 

128

APPENDIX B

Appendix B

Rorschach Determinants: E_xner’s (1990) Normative VaLres for 700 Nonpatient

 

 

Adults
Determinant/Score M SD
R 22.67 4.23
W 8.55 1.94
M 4.30 1.92
Zf 11.81 2.59
P 6.89 1.38
F+% .71 .17
X+% .79 .08
X-% .07 .05

 

Note. From A Rorschach workbook for the comrehensive system, (3ml ed.,
pp158-159), by J. E. Exner, 1990, Asheville, NC: Rorschach Workshops.

129

"illillllllllIlllllllllli