.3; uflmflfiw‘mq. V

:V

‘e‘

:7t
e. 1?.

i .
‘31:; £5;3
ll ; a v. 1“:

3 9:39: a: "V.
. .
tutti

«

F .V 2f A.

V i $qu.304...
or? -.+.L V . , a) he
an»? 31 A5... 3.3.

men‘.

wt”; .2. :

. .., . .M . .

$2,155.,

.,z.n;.;.., at

a; V. .a 3V

.1. . . a 3...,
592,30!)
; f...

\«
. .3: .
2 23?;

... .
. s, :5."f
5.3: .. x . .
V: {7463 , . . . . V .Hmm'
. .1: . ‘ . ‘
V .rtitw. |.. . . 1! V
plfiiii . . V u~ MW .
;.l.:vum.i.. . . a V . V , . VMYMILU 1‘ V 88.. .
n.3hhf13fa. , , . . 17‘!!! V . A :1 V
7 0 .Yr . . V , V ‘n Am. erI
. . . )2. \ .4.
, ... 5&9th
. V .me kn... .
.5...

t.
V!
. .

:. .:.|:.n..,
unnatuh V . AMP}:
.. V V . h?- is:
. «8%,...9‘5
. 6%.. J . {~53
. 1;. sun“;
. . Es.

> .
3555
.3 am

. V .8 at:
43.31.»).
4.9.. .V.
.Piriy

ax‘aukuvn.’ .

I . 5
.3? y
in. :5 a JV:

 

9 .
ts. £53}:

 

.\ .V
:V. T 5. V. {TV
.11.. in 3‘ .l.

. ‘

:iafioaa ,a , . V V 1.... 2.2: . k3
. H:
:0 ..-. ~ . 2 .

S.:v...:.c.1.w..§ fig 4*

 

THES:S

ZCC l

This is to certify that the

thesis entitled

RISK FACTORS OF COGNITIVE DECLINE IN A COMMUNITY
BASED SAMPLE OF OLDER ADULTS

presented by

Terry R. Barclay

has been accepted towards fulfillment
of the requirements for

 

 

 

 

M.A . degree in C11n1ca1 Psychology
Major professor
Date / LT/é/ZQOo
1
0-7 639 MS U i: an Affirmative Action/Equal Opportunity Institution

 

 

LEBRARY
Michigan State
University

 

 

PLACE IN REI'URN BOX to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE ' DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

11/00 cJCIRCIDatoDuepes-p.“

RISK FACTORS OF COGNITIVE DECLINE IN A COMMUNITY-BASED SAMPLE
OF OLDER ADULTS

By

Terry Randy] Barclay

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

MASTER OF THE ARTS

Department of Psychology

2000

ABSTRACT

RISK FACTORS OF COGNITIVE DECLINE IN A COMMUNITY-BASED SAMPLE
OF OLDER ADULTS

By

Terry Randyl Barclay

Most cross-sectional studies of human aging find that the majority of cognitive
abilities continuously decline after the third decade of life. Research has been
particularly consistent in showing decrements in the areas of speed of processing and
higher-order reasoning. This thesis investigated the hypothesis that individual
differences among older adults, including premorbid intelligence, level of education, type
of occupation, health status, and gender, moderate age-related cognitive decline in the
areas of speed of processing and executive functioning. The sample included 203
community-dwelling older adults between the ages of 54 and 87 (M=69.11 years). All
variables were measured using standard neuropsychological tests of performance and a
short demographic questionnaire. Multiple regression analyses were used to determine
the moderating effect of each demographic variable on both the relationship between age
and executive functioning and the relationship between age and speed of processing.
Results indicated that age was significantly correlated to a decline in executive
functioning (r=.38; p<.Ol) and speed of processing (5-.437, p<.01). However, none of
the demographic variables were identified as significant moderators of either relationship.
These results are discussed in light of their practical, theoretical, and methodological

implications and in the context of suggestions for future research.

ACKNOWLEDGMENTS

I would like to thank my advisor, Norman Abeles, Ph.D., for his valuable contributions
and commitment throughout this endeavor. It was his guidance and determination that
made this possible.

I would also like to thank the other members of my committee, Ray Frankmann, Ph.D.
and Michael Lambert, Ph.D., for their generous support, constructive feedback, and
conceptual assistance.

I would like to thank Josh Sacco for his time and statistical assistance.

Special thanks to Tara for her friendship and support and to Allison for her kindness and
generosity.

To my mother and sister, thank you for the love and encouragement you have given me
from the very beginning.

A very special thanks to Michelle for her patience, love, and understanding. Your
support means the world to me.

Finally, this work was made possible by generous funding provided by the Pearl J.

Aldrich Endowment in Gerontology as well as a Student Award Program grant from the
Blue Cross and Blue Shield of Michigan Foundation.

iii

TABLE OF CONTENTS

List of Tables .................................................................................... vi
Introduction ....................................................................................... 1
Individual Differences ................................................................. 4
Risk Factors of Cognitive Decline .................................................... 6
Education ....................................................................... 7
Premorbid Intelligence ......................................................... 9

Gender ........................................................................... 11

Health Status ................................................................... 12

Occupation ..................................................................... 13

Cognitive Decline Among the Elderly .............................................. 15

Age-Related Decline in Executive Functioning ........................... 16

Anatomical and Biological Evidence ...................... ‘ ........ 18

Structural Changes of the Aging Brain ............................ 18

Functional Changes in the Aging Brain ........................... 20

Neuropsychological Evidence ...................................... 21

Stroop Color-Naming Task ................................. 22

Wisconsin Card Sorting Test (WCST) ................... 23

Age-Related Decline in Speed of Processing .............................. 25

Summary .......................................................................................... 27

Hypotheses ....................................................................................... 27

Methods ........................................................................................... 29

iv

Participants .............................................................................. 29

Procedures ............................................................................... 29
Instruments .............................................................................. 30
Screening Measures ........................................................... 30

Beck Depression Inventory (BDI) ................................. 30

Geriatric Depression Scale (GDS) ................................. 31

Mini-Mental State Exam (MMSE) ................................. 32

Demographic Questionnaire ................................................. 33
Premorbid Intelligence ........................................................ 33
American New Adult Reading Test (AMN ART) ................ 33

Speed of Processing ........................................................... 34
Symbol Digit Modalities Test (SDMT) ........................... 34

Executive Functioning ........................................................ 35
Wisconsin Card Sorting Test (WCST) ..................... A ...... 35

Trail Making Test (TMT), Part B .................................. 36

Stroop Test (SD ........................ _ .............................. 36

Results ............................................................................................. 38
Descriptive Statistics ................................................................... 38
Power Analysis .......................................................................... 38
Hypothesis 1: Age, Executive Functioning and Speed of Processing ........... 39
Composite Scores .............................................................. 40

Hypothesis 2: Age, Demographic Variables, and Executive Functioning 40

Tests for Linearity ............................................................. 40

Regression Analyses .......................................................... 41

Hypothesis 3: Age, Demographic Variables, and Speed of Processing ........ 42

Tests for Linearity ............................................................. 42

Regression Analyses .......................................................... 43

Discussion ........................................................................................ 45
Methodological Considerations ...................................................... 48

Future Directions ....................................................................... 50
Summary .......................................................................................... 5 1
Appendix ......................................................................................... 53
List of References ............................................................................... 60

'V' l

LIST OF TABLES

Table 1 Means and standard deviations for subject’s age, education,
occupational level, premorbid IQ, health status and scores on the
BDI, GDS and MMSE ....................................................... 54

Table 2 Means and standard deviations for subjects’ performance on all
dependent measures, including WCST, ST, TMT and SDMT ......... 55

Table 3 Correlation matrix for all measured variables ............................ 56

Table 4 Correlation matrix between age and measures of executive
functioning and speed of processing ........................................ 57

Table 5 Multiple regression analyses predicting the impact of age, education,
occupation, gender, health status, and premorbid IQ on the
relationship between age and executive functioning ..................... 58

Table 6 Multiple regression analyses predicting the impact of age, education,

occupation, gender, health status, and premorbid IQ on the
relationship between age and speed of processing ....................... 59

V11

INTRODUCTION

Cognitive changes associated with older adulthood have long been the focus of
substantial research in the fields of psychology, human development, neuropsychology
and medicine. As early as the 1920’s, psychological research on aging and cognition was
producing a wealth of interesting, if not surprising, results on the abilities of adults
relative to their younger cohorts. In 1928, Hollingworth published data from one of the
first studies showing that the performance of adults aged 35 to 40 on a symbol
substitution test was 16% lower than scores of individuals in their late teens
(Hollingworth, 1928). This controversial finding created an emergence of empirical
work in the psychological literature and over the course of the next 20 years, researchers
from varied disciplines studied age-related changes in the scores of different types of
intelligence tests and replicated the findings of Hollingworth (Lezac, 1987). However,
despite years of study and accumulated evidence of the effects of aging, it was:not until
1949. when the popular Wechsler-Bellevue Intelligence Scale incorporated age
differentials into its IQ score norms, that age was accepted as a significant variable in the
study of cognition (Lezac, 1987).

Since 1949, the effects of aging on human potential have taken an even greater
importance in the United States and other parts of the world as the population of
individuals over the age of 65 has increased at a dramatic and unprecedented rate.
According to some estimates, by the year 2000, one American in seven will be over the
age of 65 and as early as 2025, the proportion of these individuals will become 1 in 5
(Tranel, Benton, & Olson, 1997). This age group is the fastest growing segment of the

population as advances in the field of medicine, as well as an improved standard of

living, continue to prolong the life expectancy of men and women into the 8th, 9th and 10th
decades. This trend is bound to have serious economic, political, and social implications
(Powell & Whitla, 1994), making research in this area of even greater importance.

It is not surprising that the product of over 70 years of empirical research on adult
cognition has shown that age has a profound effect on virtually all biological,
psychological, and social functioning, including cognition. Many studies, like that of
Hollingworth, have shown age-related cognitive decline in numerous abilities using a
variety of measures. In fact, most cross-sectional studies of human aging find that the
majority of cognitive abilities continuously decline after the third decade of life
(Wechsler, 1956). Research has been particularly consistent in showing decrements in
the areas of speed of processing (Cerella, 1990; Hartley, 1986, 1993; Hultsch, Hertzog, &
Dixon, 1990; Kail & Salthouse, 1994; Salthouse 1985a, 1992; Schaie, 1989, 1994; Van
Gorp, Satz, & Mitrushina, 1990) and higher-order reasoning (Haaland et al, 1987;
Daigneault, Braun, & Whitaker, 1992; Ardila & Rosselli, 1989; Whelihan & Lesher,
1985; Shimamura & Jurica, 1994; Salthouse & Mitchell, 1990; Burke, 1972; Barr &
Giambra, 1990; Pierce et al, 1989). However, despite the overwhelming evidence of late-
life cognitive decline, it is now known that age is associated with greater variability in
performance (Christensen et al., 1994; Morse, 1993; Rabbitt, 1993). In other words,
differences in performance on a variety of cognitive measures between the lowest and
highest scoring individuals has been found to increase as the age group being studied
increases. One important question raised by these findings is whether increasing age
necessarily leads to a decrease in cognitive functions. More specifically, are there

individuals who do not demonstrate cognitive decline as they age? Or, are there

10

individuals whose decline is less than normal and who show a flatter trajectory of
deterioration than their peers? Finally, since individual differences do exist, in what
ways are individuals who show little or no cognitive decline different from those who
decline more rapidly? Stated another way, what are the risk and protective factors of
late-life cognitive decline?

The purpose of the following research protocol is to determine whether individual
differences among older adults, including premorbid intelligence, level of education, type
of occupation, gender, or health status, moderate age-related cognitive decline in the
areas of speed of processing and executive functioning. The present work will attempt to
elucidate the most important risk and protective factors associated with late-life cognitive
deficits and determine the relative strength of each of these variables in predicting the
rate of intellectual deterioration in a sample of community dwelling older adults.

Attempting to uncover the relationship between age and cognitive decline at the
individual level has a number of important benefits and practical implications. First,
instead of making broad generalizations about the way age negatively impacts cognition,
we can begin to tackle the complexities of the aging experience and identify those factors
which have the most direct impact on intellectual decline. Secondly, once the risk and
protective factors have been identified, we can use that information to target at-risk
populations and begin to explore more appropriate prevention and treatment options.
Finally, identifying at-risk populations will benefit the field of gerontology as we can
begin to more clearly focus our research on those factors most relevant to the aging

process and each individual’s quality of life.

Before describing the methodology and specific hypotheses of the present study,
it is necessary to provide a brief review of the literature on the risk factors associated with
cognitive decline in older individuals. In addition, since speed of processing and
executive functioning will be used as measures of cognitive decline in this study, a brief
introduction to the research in these areas is also needed.

Individual Differences

In a typical laboratory study of aging and cognitive decline, the performance of
elderly adults on a variety of cognitive tasks is compared with that of a group of young
adults. The usual findings are that the old remember less than the young and that age
differences are greatest for tasks that measure speed of processing (Hertzog, 1991;
Salthouse, 1985b) or executive functions (Mittenberg et al., 1989). Although these
demonstrations provide valuable information about how task variables affect age
differences in performance, they tell relatively little about the aging process as it relates
to cognitive decline and they fail to account for the marked individual differences among
older adults in the degree to which their cognition is impaired relative to young adult
standards.

Aging is a developmental process and, as such, it reflects not only the biological
changes that take place over the life span but also the environmental contexts in which
those changes are occurring (Baltes, 1987). Researchers studying human aging from a
contextualist perspective (e.g. Baltes, 1987; Schaie, 1983) have viewed developmental
changes in cognition as resulting from a lifelong series of mutually causal interactions
between the characteristics of the individual and the demands and supports provided by

the environment. The basic proposition of contextualism is that “any particular course of

individual development can be understood as the outcome of the interactions among three
systems of developmental influences: age-graded, history-graded, and nonnormative”
(Baltes, 1987). Age-graded factors include biological aging and variables closely
associated with it. History-graded influences include changes among different cohorts
(e.g. level of education) as well as period-specific events (e. g. World War II).
Nonnormative influences, which are the focus of this study, include all those contextual
variables that contribute to the individuality or idiosyncrasy of development. In more
recent literature, these variables have usually been described as “individual difference
variables” or “psychosocial variables”. The usefulness of the contextual perspective in
the study of aging is supported by longitudinal data on intellectual change. For example,
Schaie (1983) reported that, although mean performance on intellectual tasks does
decline with age, individual differences in maintenance of intellectual functioning over
time are predicted by a complex interaction among favorable life experiences, greater
education and socioeconomic status, active engagement in social and intellectual
activities, and a flexible personality style.

Adoption of a contextualist position implies that, to gain a fuller understanding of
the aging process as it relates to cognitive decline, it is necessary to take into
consideration the individual characteristics and life experiences which may greatly
impact the level and rate of intellectual decline for any given individual. Although
contextual variables have generally not been the object of study within the traditional
literature, the situation appears to be changing. Although a relatively few number of
studies have been conducted, previous research and theory within the contextual

framework (e. g. Schaie, 1983) suggest that at least two broad categories of contextual

(J:

factors are relevant to age changes in cognitive abilities: demographic and health factors,
including gender, level of education, occupation, socioeconomic status, and general
physical health, and psychological factors, including such things as the presence of
anxiety or depression.

Risk Factors of Cognitive Decline

Although it is suspected that a number of risk factors are directly related to the
development of late-life cognitive decline, there is a paucity of information with respect
to the influence of demographic, health, and psychological factors on neuropsychological
test performance (Boone et al., 1993; Golden, 1981; Reynolds, 1997). This seems to be
the norm rather than the exception in neuropsychology. That is, research has, in large
part, neglected the influence of demographic variables on neuropsychological test results.
Demographic variables, such as gender, premorbid intelligence, level of education, type
of occupation, socioeconomic level, and other factors, such as health status, can have an
impact on the integrity of cognitive functions. Knowledge about the relative influence of
these variables is critical for our understanding of the individual differences that are
associated with intellectual decline in old age (Boone et al., 1993).

While there has been little research on risk and protective factors associated with
the processes of normal aging, in recent years there have been a number of community-
based epidemiological studies of dementia and associated risk factors. These studies
have identified a number of psychosocial characteristics, such as low levels of education
and premorbid intelligence, that are predictive of brain pathology in later life. This
research, while also limited, does provide clues as to the specific demographic factors

which may increase the risk for, or provide protection against, cognitive decline. The

C3\

empirical findings for studies which have investigated the impact of gender, premorbid
intelligence, level of education, type of occupation, and health status on cognitive
changes for both normal aging and dementia are reviewed below.

mm

Community longitudinal studies using large numbers of subjects have reported an
association between low educational attainment and decline in cognitive function in later
life, noting that less well educated people deteriorate to a greater extent than better
educated people over various follow-up periods (Colsher and Wallace, 1991; Evans et al.,
1993; White et al., 1994; Farmer et al., 1995; Bomstein and Suga, 1988; Wiederholt et
al., 1993). There have also been numerous investigations of educational attainment as a
risk factor for dementia in Alzheimer’s disease. Several population-based investigations
of the prevalence of AD have reported higher rates of AD among individuals with fewer
years of education (Sulkava et al., 1985; Zhang et al., 1990; Korczyn et al., 1991).
Similarly, a study by Evans et al. (1993) found that the risk of disease decreased by
approximately 17% for each year of education.

The results of some incidence studies, however, have been conflicting. In a
community-based cohort in New York, subjects with less than eight years of education
were at significantly greater risk for the development of AD during the four years of
follow-up than subjects with eight or more years of education (Stern et al., 1994). In
contrast, the incidence of AD in the F ramingham Study (Cobb et al., 1995) and the Mayo
Clinic Rochester Epidemiology Project (Beard et al., 1992) was not affected by

educational attainment. Differences between these studies in the range of education in

‘J

the populations studied and the methods used to assess cognition and diagnose dementia
may have contributed to these opposing results.

The existence of a strong association between educational attainment and
cognitive functioning is gaining more support but remains controversial. In order to
explain this finding, some have suggested that individuals with lower education levels
may perform more poorly because of impaired test-taking ability (Escobar et al., 1986;
Kittner et al., 1986; Anthony et al., 1982). However, the interpretation that test bias is
primarily responsible for the observed association between educational attainment and
cognitive functioning is weakened by the accumulating evidence supporting an inverse
association between educational level and the prevalence and incidence of dementia, as
well as between educational attainment and the rate of cognitive decline (Colsher and
Wallace, 1991).

If education does indeed reduce the risk of cognitive decline and the prevalence of
dementing illnesses such as AD, a number of mechanisms may be offered to explain the
link. Education might in some way protect against neurodegeneration (Schmand, 1997)
or the onset of dementia might be delayed because education had improved neuronal
networking so that when neurons died others could carry out similar functional tasks and
thereby minimize the signs of cognitive impairment (Swaab, 1991). This “brain reserve”
theory has been reviewed by Katzman (1993), who noted that education-could increase
brain reserve by increasing the density of neocortical synapses in the brain and delay the
onset of symptoms in Alzheimer’s disease by up to five years. He pointed to several
studies showing that the brains of elderly rats reared in an enriched environment had

increased cortical thickness and weight and increased dendritic branching and that these

rats’ general performance was better than that of control rats from a non-stimulating
environment. Dendritic growth continues in humans well into old age (Buell et al.,
1979). Therefore, previous education as well as continued mental activity may be
important for older adults. Educational attainment and stimulating mental activity may
help improve coping skills and strategies for solving problems, and in turn these may
help offset the cognitive effects of normal aging and delay the clinical symptoms
associated with Alzheimer’s disease.
Premorbid Intelligence

Another factor that may influence the rate of cognitive decline and the expression
of dementia in old age is premorbid intellectual ability. It has been hypothesized that,
similar to level of educational attainment, premorbid intellectual ability reflects a
cognitive reserve that can alter the effects of age on cognitive functioning as well as the
clinical expression of dementia in Alzheimer’s disease (Schmand, 1997). While the
research in this area is quite sparse, there is some evidence to support the theory that a
high level of premorbid ability may indicate more cognitive resources, which allow for
greater compensatory skills in response to advancing decline. It is also possible that
premorbid experience directly influences brain function at the neuronal level. Increasing
levels of mental stimulation during life can lead to alterations in brain physiology, such
as production of greater synaptic interconnectivity among neurons (Greenough et al.,
1981; Diamond, 1990). A high level of premorbid intellectual function may reflect
greater availability or efficiency of functional brain systems that are recruited when the

performance of a task requires greater mental effort.

9

One of the first studies to examine the effect of premorbid intelligence on
cognitive decline was a five-year longitudinal study of initially normal 75-85 year-old
World War II veterans (Plassman et al., 1995). They found that cognitive status at the
beginning of the study was the best predictor of subsequent development of AD
symptoms. Other investigations have found similar results, including the well-publicized
longitudinal investigation of 678 elderly Catholic nuns. Using linguistic measures
derived from essays written by sisters at the time they took their final vows (around age
22), Snowdon et al. (1996) found that sisters with poor linguistic ability were up to 31
times more likely to be cognitively and functionally impaired in old age. Verbal ability,
especially ‘idea density’, was more strongly related to cognitive functioning than other
demographics, such as educational attainment, some sixty years later. Idea density was
even related to the number of neurofibrillar tangles in various regions of the brain.

More recent studies have also supported these earlier investigations, though more
research is needed to confirm the findings. Schmand et al. (1997) conducted aifour year
longitudinal study of over 2,000 elderly subjects and found that premorbid IQ was the
second strongest predictor, after age, of incident dementia. Another study by Alexander
et al. (1994) investigated the relation between estimates of premorbid intellectual
function and cerebral glucose metabolism in patients with Alzheimer’s disease to test the
effect of differing levels of premorbid ability on neurophysiological dysfunction. Using
positron emission tomography (PET), the authors found that premorbid intellectual
abilities were predictive of cerebral metabolism and the neurophysiological heterogeneity

observed in AD. Overall, the few studies to investigate the relation between intelligence

10

and later cognitive functioning lend some support for the theory that premorbid cognitive
ability plays a role in determining the level of cognitive reserve in old age.
M91

The impact of gender on neuropsychological measures has been studied for many
years and the typical finding is that women outperform men on measures of verbal ability
while men perform better than women on tasks that involve visuospatial reasoning and
speed of processing (Wiederholt, 1993). However, such differences do not provide
information relating to the nature of cognitive deterioration associated with gender. In
fact, only recently has the study of gender differences been incorporated into the
investigation of risk factors of cognitive decline.

The influence of gender on intellectual stability remains unclear as the research is
not only limited, but has also produced mixed results. While some studies have shown a
relationship between female gender and greater cognitive deterioration and occurrence of
AD (Launer et al., 1999; Sulkava et al., 1985; Rocca et al., 1990; Zhang et al., {1990),
other investigations have challenged these findings (F ratiglioni etal., 1991; Bowler et al.,
1998; Letenneur et al., 1994; Paykel et al., 1994). In fact, there is evidence that men with
AD may be at disproportionate risk for death compared to female cases and that the
apparently higher prevalence of AD among women may simply reflect this differential
mortality (Perls et al., 1993; Corder et al., 1995).

Most studies that have investigated the association between cognitive decline and
gender have done so in an attempt to predict the development of some form of dementia,
and thus little research has been devoted to the impact of gender as a risk or protective

factor of normal aging. One study by Wiederholt et al. (1993) examined the effects of

11

gender on selected neuropsychological tests in a community cohort of normal elderly and
found that cognitive deterioration was significantly greater for men than for women on a
number of intellectual measures. However, another study conducted by Lowe and
Reynolds (1999) found that gender differences had no influence on level of cognitive
ability in their sample of healthy elderly individuals. Thus, it is not currently known
whether gender has any influence on the rate of cognitive decline associated with
advancing age.
Health Status

While there is great speculation that health status is related to the deterioration of
cognitive abilities in the elderly, there has been little investigation into the strength of this
variable in predicting the level of intellectual decline with advancing age, though this has
recently begun to change (Abate et al., 1998). Among the few studies examining the
relationship between health status and intellectual performance during adulthood is the
work of Perlmutter and Nyquist (1990). Using forward and backward digit span as a
measures of memory, they found that education together with gender explained less than
1% of the variance in memory performance of subjects aged 20 to 50 but explained a
statistically significant 12% of the variance in performance of subjects aged 60 and over.
More interesting, however, was the finding that when measures of physical health were
added to the regression equation, the explained variance in memory performance showed
a non-significant increase to 7% for the group aged 20 to 50 but a significant increase to
27% for the group aged 60 and over. The authors of this study hypothesize that health

factors have a moderating effect on other variables that may impact cognition and should

therefore be included in any investigation of risk factors associated with cognitive decline
among older adults.

This hypothesis was recently confirmed by at least one study by Arbuckle et al.
(1992). The researchers investigated the impact of demographic factors on cognition,
including age, education, social support, etc., and found that many of these variables,
including health status, contributed to intellectual performance only indirectly through
their relationship with current intelligence. While there was no a priori theoretical reason
for predicting that health status would be related to cognition through general
intelligence, the authors mention that in hindsight it seems obvious that poor health could
have a generalized debilitating effects on all cognitive abilities. Therefore, it is important
that health status be measured when possible because, although the research is in its
infancy, it is clear that such a variable has the potential to greatly impact the stability of
cognitive functions throughout adulthood.

Occupation

The psychological literature reflects few studies of the association between
occupational attainment and cognitive decline and those studies that have been conducted
show substantial variation (Evans et al., 1997). Some investigations have shown that
lifetime occupation is a significant predictor of Alzheimer’s disease, independent of its
association with education (Evans et al., 1997; Bonaiuto et al., 1995; Schmand et al.,
1997). In addition, individuals with both low educational and low occupational
attainment have been shown to be at increased risk of AD (Stern et al., 1994). In one
study, Stern and colleagues followed a cohort of non-demented elderly individuals for l-

4 years. At the initial visit, each subject’s primary occupation was classified according to

the US census categories. Based upon these categories, subjects were grouped into low
(unskilled/semiskilled, skilled trade or craft and clerical/office worker) and high
(manager business/ government and professional/technical) occupational levels. Subjects
with lower occupational attainment were at more than twice the risk of developing AD
and the increased risk for subjects with both low educational and occupational attainment
approached threefold. In a later study of occupation and regional cerebral blood flow
(rCBF) of the parietal lobe, Stern et al. (1995) demonstrated that, among patients matched
for clinical disease severity, those who had occupations involving more substantive
complexity, more interpersonal skills, and higher physical demands, showed greater
cognitive reserve against the clinical expression of AD. These and other investigators
have hypothesized that lifetime occupation could supply a reserve that allows individuals
to cope longer before intellectual deterioration is experienced. This reserve could be an
acquired set of skills or repertoires or the result of an increased synaptic density
developed on the basis of long-term stimulation. It has even been speculated that lifetime
occupation is more important than years of education in protecting against cognitive
decline given the relatively brief period of life spent in school compared to time spent on
the job (Stern, 1994).

Despite the findings from these studies, however, other researchers have not been
as successful in demonstrating a relationship between occupation and cognitive reserve.
In fact, a number of investigations have reported no significant association between the
two variables (Brayne and Calloway, 1990; O’Connor et al., 1991; Bowler, 1998). The
study by Bowler and colleagues (1998) investigated the impact of occupation on the

development of Alzheimer’s disease and found that the age of onset and the rate of

14

progression of the disease were not significantly influenced by lifetime occupation.
Similarly, an analysis of data from the Kungsholmen study (Fratiglioni et al., 1991) and a
prevalence study among African Americans (Callahan et al., 1996) also suggested no
association. More research is needed to determine the relative influence, if any, of
occupational attainment on cognitive functioning in older adults.
Cognitive Decline Among the Elderly

As mentioned previously, most cross-sectional studies of human aging find that
the majority of cognitive abilities continuously decline after the third decade of life
(Wechsler, 1956). Research has been particularly consistent in showing decrements in
the areas of speed of processing (Cerella, 1990; Hartley, 1986, 1993; Hultsch, Hertzog, &
Dixon, 1990; Kail & Salthouse, 1994; Salthouse 1985, 1992; Schaie, 1989, 1994; Van
Gorp, Satz, & Mitrushina, 1990) and higher-order reasoning/executive functioning
(Haaland et a1, 1987; Daigneault, Braun, & Whitaker, 1992; Ardila & Rosselli, 1989;
Whelihan & Lesher, 1985; Shimamura & Jurica, 1994; Salthouse & Mitchell, I990;
Burke, 1972; Barr & Giambra, 1990; Pierce et a1, 1989). Since these processes are
known to decline with age, both speed of processing and higher-order reasoning will be
measured in the present study in order to provide an index of intellectual functioning and
to test the hypothesis that demographic variables, such as gender, premorbid intelligence,
level of education, type of occupation, and health status, impact their integrity as
individuals grow older. It is thus necessary to provide a brief summary of the literature
on the effects of age on executive functioning and speed of processing before describing

the methodology in more detail.

H
(J:

Age-Related Decline in Executive Functioning

Research over the last two decades has shown that normal aging is accompanied
by neurological changes in the brain (Moscovitch & Winocur, 1992). Often, these
changes are both structural and functional and are likely to influence cognitive
fimctioning during the later part of life. One finding in particular, which has had great
impact on the field, is that normal aging is often connected with frontal lobe and
prefiontal lobe changes (Albert & Kaplan, 1980). Since this was first discovered in 1980,
there has been a great deal of evidence suggesting that there is a selective, age-related
decline in performance on neuropsychological tests which measure executive functioning
of the frontal lobe. Interestingly, a number of studies have shown that older adults
perform poorly on measures of frontal lobe function when compared to younger adults,
while performing similarly to younger adults on most tasks not related to fi'ontal lobe
function (Ardila & Rosselli, 1989). These findings have led researchers and theorists to
suggest that the cognitive processes associated with the frontal lobes are among the first
to decline with age. Although a decrease in frontal lobe function has been shown to
occur with old age, it most likely does not affect all individuals equally. However, the
extent to which the decrease in executive function can be predicted by individual
premorbid variables is not yet known.

The idea that executive functions are associated with the frontal lobe of the brain
is perhaps an old one, stemming from case studies involving patients with frontal
damage. Undoubtedly, the most popular such account is that of Phineas Gage, a now
famous subject who suffered damage to the frontal region of his brain when a steel pipe

was shot through his head. Phineas Gage survived the accident, however, major changes

16

in his personality and behavior left him a different person. Perhaps the most significant
and most noteworthy change in Gage’s life following the injury was his inability to take
control and organize his life in a meaningfiil way.

Clinical reports of the behavior of individuals sustaining frontal lobe damage, like
that of Phineas Gage, have produced a wealth of descriptive information regarding the
nature of functioning of that area of the brain. Deficits of attention, memory, and
planning have been well documented characteristics of fiontal lobe damage. These
“higher order” capacities associated with the frontal lobe were named the “central
executive” by Baddeley and Hitch in 1974 (Baddeley, 1998) and have since been referred
to as “executive functions”.

The executive functions of the brain have been described in many ways. Perhaps
one of the most comprehensive and understandable models is that given by Muriel Lezak.
Lezak (1993) describes the functions of the central executive as a four-part classification
schema which includes volition, planning, purposive action, and performance 3
effectiveness. Volition includes the ability to be aware of one’s self, the surrounding
world, and one’s level of motivation. The act of planning involves being able to
“conceptualize change (look ahead), be objective, conceive of alternatives and make
choices, develop a plan conceptually, and sustain attention” (Lezak, 1993). Purposive
action is described as self-regulation and productivity. Finally, executive functions are
thought to involve performance effectiveness which is, in other words, maintaining
quality control over one’s actions. Impairment or loss of these functions can cause a
person to loose the ability to maintain an independent, constructive, and socially

productive life, even when other cognitive abilities are intact.

p—c‘
\J

These higher order abilities have been the focus of an enormous amount of
empirical work over the years and much has been learned about these processes. Support
for an age-related decline in executive functioning comes from extensive evidence from

both biological and anatomical research as well as recent neuropsychological findings.

Anatomical and Biological Evidence

StructuralfiChanges of the Aging Brain.

Anatomical and biological studies have investigated the aging brain from multiple
perspectives, which range from the measurement of the gross volume of various regions
to the neurochemistry and synaptic density. Numerous researchers have found evidence
suggesting a significant reduction of brain mass in persons over the age of 70 (Haug &
Eggers, 1991). This reduction in volume of the entire brain has been estimated by
researchers to be approximately 6% when compared to brain mass of young adults. More
interesting, however, is that this reduction is not uniform but rather varies by region.
While the reduction in volume of the temporal, parietal, and occipital cortices of
individuals over 70 is estimated at 1%, the reduction of the frontal cortex in the same
individuals, is estimated to be between 10-15% (Haug & Eggers, 1991). Haug and
Eggers have shown that this reduction of brain volume occurs because of the reduction of
neuronal mass, and not actual loss of cells. This “shrinkage” of neurons in the frontal
cortex appears to more severe than in other areas of the brain. These researchers also
reported that below the age of 45, there is not much evidence of any cell deterioration.
However, noticeable levels of neuronal shrinkage were seen in subjects who were 50
years of age, and the reduction of cell mass is greatly pronounced in those 65 years of age

and older.

Consistent with this body of research is the finding that there may be a reduction
in the number of neuronal synapses in the frontal region of aging adults. A small study of
four individuals, aged 74-90 years, found that there was a significant reduction of
synaptic density in certain areas of the frontal lobe (Huttenlocher, 1997). This reduction
of the number of synapses could account for some of the neuronal shrinkage found in
earlier studies, however, due to the small sample size, more research in this area is
necessary before any specific conclusions can be drawn.

The reduction of cortical volume that has been observed with advancing age is
also accompanied by an increase in the number of pathological structures present in the
brain. For example, the existence of senile plaques in the brains of older adults has been
well documented. Evidence from non-human (primate) research has shown that the
distribution of plaques in the brain is not uniform, but again, varies by region. Greater
numbers of these plaques have been revealed in the prefrontal cortex than in other areas
of the brain (Struble, et a1, 1985). Plaques in other areas of the brain have been
associated with Alzheimer’s Disease. Although the effect of plaques on brain tissue is not
fully understood, it is interesting to note that they seem to have a greater affinity for
frontal tissue.

All of the above findings support the hypothesis that the frontal cortex is more
sensitive to the effects of aging than other areas of the brain. Research has demonstrated
accelerated loss of neuronal volume, possible reduction of synaptic density, greater
number of plaques, and changes in the neurochemistry of the frontal regions of the aging

brain. Although the implications of these findings are not completely understood, it is

19

clear that as we age, there are a number of biological changes that take place in the brain
which may have an impact on cognitive functioning.

WM

Investigations of the functional changes of the aging brain have incorporated two
kinds of neuroimaging techniques, namely the Xenon method and positron emission
tomography (PET). Both are used to measure cerebral blood flow and oxygen use in the
brain, but PET has the added capability of measuring the cerebral metabolism of glucose.
Unfortunately, according to Gur and colleagues (1987), these two methods have not
always provided concordant results with regard to the effects of increasing age. Studies
incorporating the Xenon and PET methods have in general found decreases in oxygen use
associated with age, while PET studies do not typically find significant age-related
differences in glucose metabolism (Gur et al, 1987).

Studies of brain function using PET and Xenon have provided evidence that
increases in age are associated with a decline in cerebral blood flow in certain areas of the
brain. This age-related reduction in blood flow has been estimated to exceed 27% in
some regions (Melamed et al, 1980). Interestingly, whereas. reduction in cerebral blood
flow has been observed in the entire cortex, there is evidence to suggest that levels of
decline are specific to different regions of the brain. In young and middle-aged
individuals, a pattern of “hyper-frontality” is often observed (Gur et al, 1987). Hyper-
frontality is characterized by a greater amount of blood flow in the frontal portions of the
brain. However, in older individuals, the opposite is true — a state of “hypo-frontality” is
often seen. This reduction of cerebral blood flow in the frontal region may have a

negative effect on executive functioning in old age.

In addition to examining the rate of cerebral blood flow while an individual is at
rest as in the above mentioned studies, researchers have also investigated cerebral blood
flow while individuals perform various cognitive tasks. These tasks include things like
solving mathematical problems, verbal analogies, and perception activities (Gur et al,
1987). Interestingly, these studies have not demonstrated consistent declines in blood
flow in older adults who are actively performing cognitive tasks, but have found declines
in the frontal regions while individuals are at rest (Gur et al, 1987). Unfortunately, these
findings are difficult to interpret in light of the limited literature available on these
relatively new investigative techniques.

The evidence presented in this section suggests that there are both structural and
functional changes of the cerebral cortex associated with increasing age. Research has
shown that these changes involving the aging brain do have a fairly high degree of
regional specificity. In support of the frontal lobe hypothesis, it appears that the frontal
cortex is one of those regions most affected by the aging process. I

Neuropsychological Evidence.

The link between age and frontal lobe dysfunction has been demonstrated not
only by biological and anatomical research, but also in studies involving cognitive
performance. These studies have often pointed to the similarities between the elderly and
those with frontal lobe brain damage (Mittenberg et al., 1989). In fact, the researchers
who developed the frontal lobe hypothesis (Albert & Kaplan, 1980) did so after noting
that their subjects illustrated age-related differences in neuropsychological performance

and appeared similar to those with frontal lobe damage.

Several investigations involving comparisons of neuropsychological test
performance of healthy young and older adults have provided additional support for
Albert and Kaplan’s frontal hypothesis. Among the most widely used measures of
executive functioning are the Stroop Color-Naming Task and the Wisconsin Card Sorting
Test (WCST). Both of these measures have been valuable tools in the assessment of the
aging brain and its relationship to frontal lobe dysfunction.

Stroop Color-Naming Task

Research incorporating the Stroop Color-Naming Task has provided a number of
findings related to the effects of increasing age and executive functioning. In the Stroop
Task, an individual is asked to perform a number of reading tasks of both word and color.
During the word-reading portion of the test, subjects are asked to read a series of color
names printed in black ink as quickly as possible. The individual is then asked to name
the color of a series of colored x’s. The naming of the color is typically slower than the
previous reading of words task. Finally, in the color-word naming portion, the individual
is asked to name the color of a series of words printed in an incongruent color. For
example, the word ‘red’ would be printed in blue ink. The naming of colors in this
condition is generally slower than the previous tasks.

The effects of increasing age on this task have been well documented, with older
adults showing significantly more slowing during the color-word interaction portion of
the test (Cohn et al, 1984). These age-related differences begin to appear in individuals
60 and 70 years old and continue to increase into old age. The Stroop effect is thus

consistent with the frontal lobe hypothesis.

In addition to the age-related Stroop effect, a number of variables have been
shown to interact with the effects of increasing age, including health status, education,
and the spatial location of the color and word information. The interaction between
health status and age suggest that there is an accelerated increase of the Stroop effect with
age for individuals with even relatively minor health problems. In addition, higher levels
of education have been found to provide some protection against the cognitive slowing
seen with aging (Houx et al, 1993).

Over the years, a number of theoretical explanations of the Stroop effect have
been proposed and tested in relation to the effects of increasing age. The Stoop effect
appears to reflect the difficulty of inhibitory responses on the brain to process relevant
and irrelevant information (Cohen et al, 1990) and has been shown to be sensitive to
frontal lobe damage. Functional brain imaging using positron emission tomography
(PET) has shown that a number of frontal lobe brain structures are active during the
Stroop task, validating the fact that it is, in fact, a good measure of frontal lobe ability
(Bench et a1, 1993).

Wisconsin Card Sorting Test ( W CS T)

Age group differences in performance on the Wisconsin Card Sorting Task have
also been shown to be consistent with the frontal lobe hypothesis. The WCST requires
the subject to sort cards in a number of different ways according to either color, number
or symbol. The individual receives the cards in consecutive order and must deduce from
the yes/no feedback of the examiner how to complete the task. After correctly sorting ten

cards in a row, the principle is shifted and the subjects must be able to adjust their

strategy in order to meet the changing demands of the examiner. Thus, the test involves
higher order cognitive abilities including planning, reasoning and flexibility.

The Wisconsin Card Sorting Task has also been found to be particularly sensitive
to frontal lobe damage (Nelson, 1976). Age associated decline of WCST performance
has been shown in many studies. In one such study, Isingrini (1997) reported that
significant age differences in favor of the young were found in adults ranging in age from
70 to 99 years of age. Thus, it appears that the age-related differences associated with
this task result fi'om at least some degree of frontal lobe dysfunction in older adults.

As illustrated above, many studies have shown that changes in the aging brain do
not occur in a uniform manner, but rather vary by region. According to recent studies, it
seems that the frontal lobe is a primary target for many of these changes which include
neuronal shrinkage, reduction of mass, neurochemical changes, increases in pathological
structures, and decreased cerebral blood flow. Indeed, the research has shown that
changes in the frontal region of the aging brain are some of the most dramatic.

Although the effect of these changes is not fully understood, researchers have
associated the above findings with the cognitive decline often seen in old age. A great
deal of evidence from neuropsychological instruments which test frontal lobe functioning
has been overwhelmingly in favor of the frontal lobe hypothesis. The two examples that
were illustrated here, the Stroop Color-Naming Task, and the Wisconsin Card Sorting
Task, have shown a great deal of age-related decline in executive functioning. In fact, the
decline has been so remarkable, that many researchers have made a theoretical

connection between the performance of the elderly and those with frontal lobe damage.

Age-Related Decline in Speed of Processing

Slowing with age is often considered one of the best documented and least
controversial behavioral phenomena of aging (Birren and Fisher, 1995). Cognitive aging
researchers generally agree that performance on many information-processing tasks
which assess cognitive or perceptual processes is slowed with increased age (Hertzog,
1991; Salthouse, 1985). Proponents of the general slowing hypothesis of aging argue that
this phenomenon is caused by a general decrease in processing rate with age (N ettelbeck
and Rabbitt, 1992) and this reduction in rate is believed to affect most perceptual and
cognitive processes and is thus regarded as the prime determining factor of cognitive
deterioration associated with late adulthood (Salthouse, 1985; Welford, 1984). In fact,
Salthouse has demonstrated that the median correlation between age and measures of
speed across a wide range of behavioral activities is .45 (1985).

The evidence supporting the role of processing speed as a mediator of cognitive
decline has come from cross-sectional studies in which the performance of individuals of
different ages on cognitive variables was compared. During middle and late adulthood,
measures of intellectual speed tend to show the largest negative age differences, followed
by fluid abilities such as spatial reasoning, which in turn are followed by more
knowledge-dependent measures (Salthouse, 1982). For example, the Digit Symbol
Substitution subtest of the Wechsler Adult Intelligence Scale (WAIS), a typical measure
of cognitive-perceptual speed, shows very pronounced age decrements, whereas the
Vocabulary subtest of the WAIS, a measure of knowledge, remains stable through late

adulthood (Salthouse, 1982).

Cross-sectional data reported by Hertzog (1989) and Hultsch et al. (1992) also
support the importance of speed in the cognitive fimctioning of older adults. Hertzog
investigated the influence of speed measures on other intellectual abilities in a sample of
622 adults with an age range of 43 to 89 years. Significant negative age trends were
observed for all abilities that were studied, however, afier controlling for speed of
processing, the age trends explained only between 1-3% of the total variance in the other
ability measures. Similarly, an important longitudinal study conducted by Hultsch et al.
(1992) examined the changes in performance on memory, intellectual ability and
information processing tasks over a three year period in 328 older adults. They found
significant declines in functioning on measures of working memory, verbal fluency, and
world knowledge, however, these findings were reduced or completely eliminated when
they controlled for processing speed.

Salthouse and his colleagues (1991) also conducted a series of studies on the
extent to which measures of perceptual speed and working memory account for age
differences in complex intellectual tasks. They reported the results of three independent
studies which covered an age range of 20 to 84 years. After controlling for speed, the
age-related variance in cognitive tasks and working memory measures was reduced fi'om
a range of 17-3 1% to a range of l-5%. In line with Hertzog’s results, these findings show
the significance of age-related decreases in speed of processing and demonstrate that this
variable can explain a large portion of the variance associated with age-related cognitive
decline.

The empirical work that has been conducted on older adults thus far has led

researchers to suggest that the cognitive processes associated with speed of information

processing are among the first to decline with old age. Although a definite slowing effect
has been shown to characterize the cognitive functioning of older adults, it most likely
does not affect all individuals equally. However, the extent to which the decrease in
processing speed can be predicted by individual premorbid variables is not yet known.
Summary

Although it is suspected that a number of risk factors are directly related to the
development of late-life cognitive decline, there exists little information with respect to
the influence of demographic, health, and psychological factors on neuropsychological
test performance (Boone et al., 1993; Golden, 1981; Reynolds, 1997). Research has, in
large part, neglected the influence of demographic variables on neuropsychological test
results. Demographic variables, such as gender, premorbid intelligence, level of
education, type of occupation, and other factors such as health status, can have an impact
on the integrity of cognitive functions. Knowledge about the relative influence of these
variables is critical for our understanding of the individual differences that are associated
with intellectual decline in old age (Boone et al., 1993). Therefore, the present study will
significantly contribute to the literature on aging and cognition by investigating the
potential impact of those demographic variables which have thus far been neglected.

Hypotheses

Based on the research reviewed above, the specific hypotheses of the present
study include the expectation of the following results:
1. Confirmation of an age-related decline in measures of executive functioning and speed

of processing.

2. A significant relationship will exist between age-associated cognitive decline in
executive functioning and individual demographic variables. That is, fewer years of
education, lower levels of occupational attainment, poor health, lower levels of
premorbid intelligence, and female gender will be predictive of greater decline in
executive functioning.

3. A significant relationship will exist between age-associated cognitive decline in speed
of processing and individual demographic variables. That is, fewer years of education,
lower levels of occupational attainment, poor health, lower levels of premorbid
intelligence, and female gender will be predictive of greater decline in speed of
processing.

It is expected that this sample of healthy, community-dwelling older adults will
show individual variability in cognitive decline in the areas of executive functioning and
speed of processing. However, it is also predicted that the level of deterioration will be
predicted, at least in part, by demographic variables including premorbid intelligence,
level of education, type of occupation, gender, or health status. Since the empirical work
on risk and protective factors of adult cognition is at a surprisingly early stage of
development, it is not known which of these factors will be the strongest in predicting the

rate of cognitive decline.

METHOD
Participants
The present investigation included 234 community dwelling older adults between

the ages of 54 and 87 (M = 69.11, S1; = 7.86 years). Public advertisements in local
newspapers and other mailings were used to solicit willing participants. Subjects who
expressed an interest in the study were screened using the Mini Mental Status Exam
(MMSE) as well as measures of depression, including the Beck Depression Inventory
[(BDI); Beck, 1987] and the Geriatric Depression Scale [(GDS); Brink et al., 1982].
Individuals who demonstrated cognitive impairment with a score below 24 on the MMSE
or those with depression as indicated by scores of 30 or higher on the Beck Depression
Inventory and/or scores of 20 or higher on the Geriatric Depression Scale were excluded
from the sample in order to ensure that all participants were healthy and functioning at a
level consistent with normal aging. Given the exclusionary criteria used in this study, the
actual sample size employed in the data analyses was 203. There were 90 males and 113
females. Subjects who participated in the study were given the opportunity to attend
mood and memory training workshops at Michigan State University. Finally, informed
consent was obtained from each participant prior to the completion of study procedures.

Procedures

Participants completed detailed questionnaires yielding demographic and life

history information before completing a set of neuropsychological tests. The total length
of testing for each individual was approximately 2 hours. The order of measures was
systematically varied across participants. All testing sessions were conducted

individually. The administered procedures assessed, among other things, the

individual’s mood, memory, executive functioning, speed of processing and general
cognitive ability. For the purposes of this study, however, only data relevant to measures
of executive functioning and speed of processing were analyzed and reported.
Instruments
The measures used in this investigation will include the following:
Screening Measures
Beck Depression Inventory (BDD

The BDI (Beck, 1987) is a self-report measure consisting of 21 multiple-choice
statements presented on four pages of paper. Each item is concerned with a particular
aspect or symptom of depression and asks subject to rate their experience on a scale of
graded severity. The approximate time required to complete the inventory is 5-10
minutes. A total score is obtained by adding the highest score circled for each of the 21
items. The total score can range from 0-63, with higher scores indicating more severe
levels of depression. The instrument has classified depression severity in the following
manner: 0-9 = normal range; 10-15 = mild depression; 16-19 = mild/moderate
depression; 20-29 = moderate/severe depression ; 30+ = severe depression (Spreen &
Strauss, 1998).

The BDI is a well researched instrument with strong psychometric properties. In
addition, it has been used with a wide range of populations due to the ease of
administration and scoring. In the original norming sample, Beck (1970) reported that
test-retest reliability was above .90 and that change scores tended to parallel changes in
depression severity for individual subjects. Reynolds and Gould (1981) have reported

Spearman-Brown reliability to be .93 and internal consistency of test items to be .86.

30

Other authors have reported a coefficient alpha of .88 (Steer et al., 1989). Furthermore,
the BDI is known to have good concurrent validity with instruments such as the MMPI
Depression Scale (.75; Reynolds and Gould, 1981) and the Hamilton Rating Scale (.85;
Brown et al., 1995) as well as clinical ratings of depression (.66; Schaefer et al., 1985).
Geriatric Depression Scale (GDS)

The GDS (Brink et al., 1982) is a self-report measure consisting of 30 yes/no
questions pertaining to potential symptoms of depression. The instrument was designed
specifically for the elderly population and is recommended for use by Spreen and Strauss
(1998) due to the simplicity of administration and the fact that it contains somatic items
which are more suitable for older adults than those presented on the BDI. The GDS
requires subjects to read statements and answer each by circling the yes/no response
which most accurately reflects their experience. Alternatively, the items can be read to
subjects if there is any concern regarding their ability to read or respond using paper-and-
pencil. The estimated time required for completion of the GDS is approximately 5-10
minutes. A total score is obtained on the instrument by adding the point values assigned
to each response (0 or 1). The following cutoff points are used to determine level of
depression: 0-9 normal range; 10-19 = mild depression; 20-30 = moderate/severe
depression (Spreen and Strauss, 1998).

The psychometric properties of the GDS warrant the instrument’s wide-spread use
among older adult populations. Brink et al. (1982) reported that in the original norming
sample, internal consistency (alpha) and split-half reliability were both .94. Retest
reliability over a one-week time span has been reported to be .85 (Koenig et al., 1988).

In addition, concurrent validity has been established with other self-report measures

including the BDI (.73; Hyer &Blount, 1984), Hamilton Rating Scale (.83; Yesavage et
al., 1986), MMPI Depression Scale (.72; Bielauskas & Lamberty, 1992), and the DMS-
based Symptom Checklist for Major Depressive Disorders (.77; Bielauskas & Lamberty,
1992). Criterion validity as measured against the Research Diagnostic Criteria has been
reported as .82 (Yesavage & Brink, 1983).

Mini-Mental State Exam (MMSE)

The MMSE (F olstein and McHugh, 1975) is a 30-item measure used to screen for
cognitive impairment and document changes in intellectual functioning over time. The
measure is easily administered and scored in a brief 5-10 minutes. The MMSE requires
the examiner to ask questions and record responses given by the subject. The test
consists of items that assess orientation to time and place, attention and concentration,
language, constructional ability, and immediate and delayed memory recall. Each correct
response is awarded one point and the total score is the number correct out of 30. Scores
below 24 are considered abnormal when screening for dementia or delirium and this cut-
off is recommended with most populations (Lezak, 1995).

The psychometric properties of the MMSE are well documented in the literature.
Inter-rater reliability has been shown to be above .65 (Foster et al., 1988) and test-retest
reliability estimates for intervals of less than two months generally fall between .80 and
.95 (O’Connor et al., 1989). Most studies report that the MMSE has adequate specificity
and sensitivity for detecting the presence of dementia, particularly in cases where there
exists moderate to severe forms of cognitive impairment (Spreen & Strauss, 1998). The
MMSE also shows modest to high correlations with measures of intelligence, memory,

attention and concentration, and executive functions (Axelrod et al., 1992). In addition,

Crum et al. (1993) has recently published extensive norms for the MMSE by age (18-85)
and education based on probability sampling of more than 18,000 community-dwelling
individuals.
Demographic Questionnaire

Information on age, gender, level of education, occupation, and health status were
obtained by a self-report style demographic questionnaire. Level of education was
measured in the conventional manner and based on responses to the following questions:
a.) ‘What is the highest grade of school you completed?’ and b.) ‘If you attended college,
please specify how many years and what degree you obtained’. Occupational level was
assessed by asking participants, ‘What kind of work did you do most of your working
life?’. Answers to this question were then coded according to the Eppinger Demographic
Formula (Eppinger et al., 1987) in which unskilled labor = 1; semi-skilled labor = 2; not
in labor force (including homemakers) = 3; skilled labor = 4;
managerial/official/clerical/sales/ self-employed = 5; professional/technical/artistic = 6.
Finally, health status was measured using the following questions: a.) ‘How would you
rate your overall health at the present time?’ in which excellent = 1; good = 2; fair = 3;
poor = 4 and b.) ‘Do your health problems stand in the way of doing things you want to
do?’ in which not at all = l; a little = 2; a great deal = 3.

Premorbid Intelligence

American New Adult Reading Test (AMNART)
The AMNART (Grober & Sliwinski, 1991) is a reading test involving 45

irregularly spelled words (e.g. na‘r‘ve, subtle, facade) in which pronunciation is used to

predict premorbid IQ. The subject is presented with a piece of paper containing two

b)
L»)

columns of printed words and is asked to “Look at each word carefully and say it aloud”.
The examiner scores each pronunciation based on the phonetic spelling guidelines
provided with the instrument, which are based on the common English language
dictionary. Each correct response is awarded one point and the maximum number of
correct responses is 45. The estimated time needed to complete the test is approximately
2-5 minutes. Once the AMN ART has been scored, a formula is used to obtain the
subject’s premorbid IQ score based on the number of AMN ART errors and level of
education. The formula used most commonly is that by Grober and Sliwinski (1991):
Estimated VIQ = 118.2 - .89(AMNART errors) + .64(Years of Education).

The AMN ART formula is based on a sample of 230 nonimpaired older adults
(Grober & Sliwinski, 1991). The sample included approximately equal numbers of males
and females with a mean age of 75 (SD = 4) and mean education of 12.5 years (SD = 3).
The correlation between actual WAIS-R VIQ and estimated VIQ based on this regression
formula was .76. 1

Speed of Process_ing
Svmbol Digit Modalities Test (SDMT)

The SDMT (Smith, 1982) is a measure of speed of processing, complex scanning
and visual tracking and allows comparison between oral and written performance (Lezak,
1995). The oral and written portions of the test each have a time-limit of 90 seconds.
During the first trial, subjects are presented with a sheet of paper that has a code table
displaying pairs of digits and symbols. Beneath the code table are rows of double boxes
with the symbol in the top box and nothing in the bottom box. The participants are then

asked to use the code table to determine which number is associated with each symbol

and to write as many nrunbers as possible in the empty boxes in the 90 second period.
During the second trial, the examiner records the numbers while they are given orally by
the subject. The score obtained from this test is the number of correctly “substituted”
numbers in each portion of the test with a maximum score of 110 on each. The norms for
the SDMT are well established and reliability is satisfactory (Spreen & Strauss, 1998).
More specifically, test-retest reliability correlations have been reported to be .80 for the
written portion and .76 for the oral portion (Spreen & Strauss, 1998).
Executive Functioning

Wiscopsin Card Sorting Test (WCSfl

The WCST (Heaton, 1993) assesses the ability of an individual to use abstraction
and to shift cognitive strategies in response to changing environmental cues (Lezak,
1995). It is widely accepted as a measure of executive function as individuals must
employ strategic planning, organized searching, the ability to use environmental cues to
shift cognitive sets and must act in a way that is goal-oriented (Spreen & Strauss, 1998).
Subjects are asked to match a set of cards, which vary according to the number and color
of different shapes, with one of 4 “key” cards. The individual is asked to complete the
task in the absence of explicit matching guidelines. Instead, they must rely solely on the
feedback given after each trial indicating whether they have made a correct or incorrect
match. In addition, once the subject correctly matches 10 cards consecutively, the sorting
principle is altered — subjects are not aware of this change and must modify their
responses accordingly. A great deal of empirical work on executive functioning has been

done using the WCST as both the construct validity and interrater reliability of the

£1]

LIJ

instrument have been well established (Lezak, 1995). In fact, interrater reliability
estimates have been reported to range from .88 to .96 (Lezak, 1995).
Trail Making Test (TMT), Part B

Part B of the TMT (Reitan, 195 8) is used to assess aspects of executive
functioning including mental flexibility, attention and sequencing (Spreen & Strauss,
1998). The measure has enjoyed a long history of use as it is easy to administer and takes
little time to complete (Lezak, 1995). Subjects are asked to connect a series of circles in
a “connect-the-dot” fashion, alternating between a sequence of numbers and letters. The
instructions ask participants to complete the task as fast as they can without removing the
pencil from the paper. Interrater reliability of the test has been reported to be .90 (Spreen
& Strauss, 1998).
Stroop Tesft ( ST)

This test measures components of executive processes including concentration,
the ability to shift attentional resources and level of response inhibition (Golden, 1978).
There are 3 trials administered to the subject and each has a 45 second time-limit. During
the first trial, subjects read words aloud (either “red”, “blue” or “green”) printed in black
ink. Twenty words are printed in each of 5 columns on the page and appear in random
order. In the second trial, these words are replaced with X’s which are printed in
different color inks (either red, blue or green) and the subject is asked to name these
randomly listed colors as quickly as possible. The final trial combines both of the
preceding stimuli and requires that the subject read the color of ink in which the words
“red”, “blue” and “green” are printed. This last trial is known as the interference

condition because individuals must suppress one percept in favor of another. Lezak

(1995) reports that the ST has satisfactory reliability and more specific test-retest
reliabilities are reported by Spreen & Strauss (1998) to be .90, .83, and .91 for the three

portions of the ST.

()0
\1

RESULTS
Descriptive Statistics

Means and standard deviations for age, education, occupational level, premorbid
IQ, health status and scores on the BDI, GDS and MMSE for this sample are displayed in
Table 1. In all, a total of 203 of the 234 participants met screening criteria by
demonstrating adequate levels of cognitive and emotional functioning as measured by the
MMSE, EDI and GDS. This group of 203 subjects was designated as the final sample for
the present study. The average age of the participants was 69.11 (S_D = 7.86 years) and
ranged from 54 to 87. The sample was well educated as the average level of educational
attainment was 15.86 years (fl = 3.02). In addition, on a scale of 1 to 6, mean level of
occupation was reported to be 5.07 (E = 1.18), indicating that most subjects held jobs
categorized above that of a skilled laborer (e.g. managerial, official, clerical, sales, self-
employment). Finally, on a scale of 1 to 4, average self-reported health status was 1.85
(SD = 0.63), indicating that most individuals rated their overall health as ‘good’ (as
compared to ‘excellent’, ‘fair’, or ‘poor’).

Means and standard deviations for subjects’ performance on all dependent
measures, including WCST, ST, TMT and SDMT appear in Table 2. A correlation
matrix with all measured variables is presented in Table 3.

Power Analysis

A power analysis was conducting to ensure adequate statistical power in detecting
small, medium and large effect sizes with the present sample. For significance tests of
Pearson Product Moment Correlations, Cohen (1992) defines small, medium, and large

effect sizes as .10, .30, and .50 respectively. For a desired power of .80 and with alpha

L1)
00

set at .05, Cohen indicates that 783 individuals are needed to detect small effect sizes, 85
to detect medium effect sizes, and 28 to detect large effect sizes of .50 and above.
Therefore, the present study has adequate power to detect medium and large effects for
correlational analyses but is inadequate for effects smaller than .30.

For significance tests involving multiple regression analyses with 2 independent
variables, a desired level of power at .80, and with alpha set at .05, Cohen (1992)
indicates that 481 individuals are needed to detect small effect sizes, 67 to detect medium
effect sizes, and 30 to detect large effect sizes of .50 and above. Therefore, the present
study has adequate power to detect medium and large effects for multiple regression
analyses but is inadequate for effects smaller than .30.

Hypothesis 1: Age, Executive Functioning and Speed of Processing

The first hypothesis of the present study sought a confirmation of an age-related
decline in measures of executive functioning and speed of processing. As predicted, age
was correlated with measures of executive functioning. More specifically, increasing age
was associated with poorer performance on both the Trailrnaking Test (r=.274, p<.01)
and the Wisconsin Card Sorting Test (g=.309, p<.01). However, a significant relationship
between age and executive functioning was not clearly demonstrated with the use of the
Stroop Test (ST), as this measure did not significantly correlate with age (r=.066). Due
to the lack of association between age and the Stroop Test, no further analyses included
this particular measure of executive functioning. Therefore, the operationalization of
executive functioning in later analyses was limited to the Trailmaking Test (TMT) and

the Wisconsin Card Sorting Test (WCST).

39

As predicted, age was also related to processing speed. More specifically,
increasing age was associated with slower performance on the Symbol Digit Modalities
Test (r=-.437, p<.01). Pearson Product Moment correlations between all measures of
executive fimctioning, speed of processing and age are displayed in Table 4.

Composite Scores

Based on the correlations between various measures of executive functioning, a
composite score was formed in order to simplify analyses and to avoid overmodelling the
data. As displayed in Table 4, the correlation between the two remaining measures of
executive functioning, the Trail Making Test (TMT) and Wisconsin Card Sorting Task
(WCST), was positive and significant (r: .21, p<.01). Therefore, a composite score that
combined these measures was used as an overall indicator of executive functioning, a
technique endorsed by Salthouse and Bobcock (1991). The composite score was derived
by converting individual scores from each test to z-scores and then summing them to
form a single composite measure of executive fimctioning. As expected, this new
composite score correlated significantly with age such that increasing age was associated
with poorer performance (r=.38; p<.01).

Hypothesis 2: Age, Demographic Variables, and Executive Functioning
Tests for Linearig

Prior to conducting the individual regression analyses to test this hypothesis, the
relationship between age and the executive functioning composite score was examined to
determine whether or not there were any non-linear components (e. g. quadratic
functions). The relationship between age and each of the proposed moderating variables,

including education, occupation, health status, premorbid intelligence and gender, was

also examined. First, the executive functioning composite score and all demographic
variables were regressed on age. Subsequent additions of a quadratic function to the
regression models did not produce a significant increase in R2. Thus, none of the
relationships deviated from linearity which simplified all later analyses.
Regression Analyses

Multiple regression analyses were used to test hypothesis 2. These analyses
assessed whether individual differences in level of education, occupation, premorbid IQ,
gender, or health status would moderate the relationship between age and executive
functioning. Separate analyses tested main and interaction effects for each of the
demographic variables . The dependent variable for all regression models was the
executive functioning composite score. In addition, because regression models
necessitate working solely with continuous variables, it was necessary to “dummy code”
gender, a categorical variable, by assigning values of 0 and 1. This technique allows the
model to properly accept categorical variables for use in the regression equation.

Hypothesis 2 stated that a significant relationship would exist between age-
associated cognitive decline in executive functioning and individual demographic
variables. That is, lower levels of occupational attainment, fewer years of education,
poor health, female gender, and lower levels of premorbid intelligence would be
predictive of greater decline in executive functioning with age. Results from individual
regression analyses indicate that this hypothesis was not supported, as the identified
demographic variables were not shown to be significant moderators in the relationship

between age and executive functioning.

41

The results of the multiple regression analyses appear in Table 5. When age was
entered as the sole predictor in Model 1, results indicated that age accounted for 15% of
the variance in executive functioning (p<.01). Subsequent analyses tested the main
effects and interaction effects for education, occupation, gender, health status, and
premorbid IQ using 2 models for each variable. In the first model, the demographic
variable was entered in combination with age to examine the main effects of the two
predictors. In model 2, the interaction variable (e.g. education by age) was entered as the
sole predictor of executive functioning in order to test the interaction effect and
determine whether the specified demographic variable significantly moderated the
relationship between age and executive functioning.

As can be seen in Table 5, none of the interaction effects were significant. R-
squared change for the demographic variables failed to reach significance for education
(R2 Change= .010, p=.112), occupation (R2 Chang = .004, p=.3 59), gender (R2 Change=
.005, p=.283), health status (R2 Chang = .001, p=.722) and premorbid IQ (R2 Change=
.003, p=.388).

Hypothesis 3: Age, Demographic Variables, and Speed of Processing
Tests for Linem

Prior to conducting the individual regression analyses to test this hypothesis, the
relationship between age and speed of processing was examined to determine whether or
not there were any non-linear components (e.g. quadratic functions). The relationship
between age and each of the proposed moderating variables, including education,
occupation, health status, premorbid intelligence and gender, was also examined. First,

the processing speed score and all demographic variables were regressed on age.

Subsequent additions of a quadratic function to the regression models did not produce a
significant increase in R2. Thus, none of the relationships deviated from linearity which
simplified all later analyses.

Regression Analyses

Multiple regression analyses were used to test hypothesis 3. These analyses
assessed whether individual differences in level of education, occupation, premorbid IQ,
gender, or health status would moderate the relationship between age and speed of
processing. Separate analyses tested main and interaction effects for each of the
demographic variables . The dependent variable for all regression models was the
processing speed score from the Symbol Digit Modalities Test (SDMT).

Hypothesis 3 stated that a significant relationship would exist between age-
associated cognitive decline in speed of processing and individual demographic variables.
That is, lower levels of occupational attainment, fewer years of education, poor health,
female gender, and lower levels of premorbid intelligence would be predictivetof greater
decline in speed of processing with age. Results from individual regression analyses
indicate that this hypothesis was not supported, as the identified demographic variables
were not shown to be significant moderators in the relationship between age and speed of
processing.

The results of the multiple regression analyses appear in Table 6.’ When age was
entered as the sole predictor in Model 1, results indicated that age accounted for 22% of
the variance in speed of processing (p<.01). Subsequent analyses tested the main effects
and interaction effects for education, occupation, gender, health status, and premorbid IQ

using 2 models for each variable. In the first model, the demographic variable was

43

entered in combination with age to examine the main effects of the two predictors. In
model 2, the interaction variable (e. g. education by age) was entered as the sole predictor
of processing speed in order to test the interaction effect and determine whether the
specified demographic variable significantly moderated the relationship between age and
speed of processing.

As can be seen in Table 6, none of the interaction effects were significant. R-
squared change for the demographic variables failed to reach significance for education
(R2 Change= .004, 12=-313), gender (R2 Chang = .010, n=.120), health status (32
(Sh—ange= .000 p=. 979) and premorbid IQ (R hange= 0.03, p=.365). However, it
should be noted that while the interaction effect for occupation also failed to show
significance (R ange= .014, p=.059), the trend of this relationship was in the predicted
direction. That is, individuals with lower levels of occupational attainment showed
somewhat greater decline in speed of processing than individuals with greater levels of

occupational attainment.

44

DISCUSSION

Hypotheses 1 of the present study predicted that age would be significantly
associated with decline in measures of speed of processing and executive functioning.
This hypothesis was supported. ”As predicted, age was correlated with a measure of speed
of processing, the Symbol Digit Modalities Test (SDMT), such that increasing age was
associated with slower performance. This finding is consistent with other research
(Cerella, 1990; Hartley, 1986, 1993; Hultsch, Hertzog, & Dixon, 1990; Kail & Salthouse,
1994; Salthouse 1985a, 1992; Schaie, 1989, 1994; Van Gorp, Satz, & Mitrushina, 1990)
and demonstrates that normal aging is indeed associated with a general slowing of
cognitive processing. This finding lends support to proponents of the general slowing
hypothesis (N ettelbeck and Rabbitt, 1992) who argue that the decrease in performance
demonstrated by older adults on a variety of cognitive tasks is caused by a general
decrease in processing rate with age. In addition, it is important to note that the size of
the correlation between age and speed of processing in the present study is consistent
with other research. Salthouse (1985) has demonstrated that the median correlation
between age and measures of speed across a wide range of behavioral activities is .45.
The correlation between age and speed of processing in the present study was .466
(p<.01).

As predicted, age was also associated with measures of executive functioning,
including the Wisconsin Card Sorting Test (WCST) and the Trail Making Test (TMT).
This is consistent with a vast amount of research indicating that executive abilities
deteriorate with age (Haaland et al, 1987; Daigneault, Braun, & Whitaker, 1992; Ardila

& Rosselli, 1989; Whelihan & Lesher, 1985; Shimamura & Jurica, 1994; Salthouse &

45

Mitchell, 1990; Burke, 1972; Barr & Giambra, 1990; Pierce et al, 1989). However, one
measure of executive functioning used in this study, the Str00p Test (ST), did not
significantly correlate with age. This is surprising given the extensive psychological
literature indicating a decline in performance on the Stroop Test in older adults (Boone et
al., 1990; Cohn et al., 1984; Diagneault et al., 1992; Houx et al., 1993; Macleod, 1991;
Ober & Albert, 1985; Spreen & Strauss, 1991; Whelihan & Lesher, 1985). This may be
an important finding as it may indicate that performance on this measure does not decline
in samples of competent older adults. An alternative explanation may be that decline in
the abilities measured by the Stroop Test have already taken place by the time individuals
have entered older adulthood, and therefore the test may not be as sensitive to change in
older cohorts.

Hypotheses 2 and 3 predicted that a significant relationship would exist between
individual demographic variables and age-associated cognitive decline in the areas of
executive functioning and speed of processing. That is, lower levels of occupational
attainment, fewer years of education, poor health, female gender, and lower levels of
premorbid intelligence would be predictive of greater decline with age in measures of
executive functioning and speed of processing. These hypotheses were not supported.
More specifically, none of the identified demographic variables had a significant
moderating effect on the relationship between age and executive functioning or on the
relationship between age and speed of processing. Although it was hypothesized that
these demographic variables would moderate age-associated cognitive decline, and
therefore be identified as risk or protective factors in the aging process, these results were

not all-together surprising given the conflicting findings with respect to these variables in

46

the literature. There have been very few studies which have investigated the impact of
contextual or demographic variables on age and neuropsychological test performance and
those that have been conducted reveal conflicting results with respect to the impact of
education (Colsher and Wallace, 1991; Evans et al., 1993; Cobb etal., 1995, Beard et al.,
1992), occupation (Brayne and Calloway, 1990; O’Connor et al., 1991; Evans et al.,
1997; Bonaiuto et al., 1995), premorbid IQ (Plassman et al., 1995; Snowdon et al.1996),
health status (Arbuckle, 1992; Perlmutter and Nyquist, 1990), and gender (Launer et al.,
1999; Rocca et al., 1990; Fratiglioni et al., 1991; Bowler et al., 1998). In addition, since
the majority of research finding a relationship between these variables and cognitive
functioning has been conducted using pathological samples, for example, those with
Alzheimer’s disease or other dementias, it may be that these variables are in fact
moderators of cognitive decline with respect to the development of a dementing illness
but do not moderate the decline seen in normal aging. Therefore, it would appear that, at
least for some, normal aging can be viewed as a benign process as suggested by Rowe
and Kahn (1997).

Another likely explanation for the lack of association between demographic
variables and cognitive functioning in the present study may be related to restricted
variance among the independent variables. Exclusions fi'om the sample were made when
subjects indicated clinical levels of depression and/or exhibited impaired performance on
a dementia screening instrument. This was to ensure that the final sample would consist
only of those individuals experiencing the normal aging process. However, this was a
fairly well-functioning sample and may therefore have encompassed selection biases not

controlled for in this study. For example. the sample was a fairly homogeneous group of

47

highly functioning older adults with higher than average education (M=15.86), IQ scores
(M=118.83), and occupational attainment. This lack of variation in the sample
presumably contributed to an attenuation of the correlations between the variables,
making it difficult to detect any moderating effect in the regression analyses. Therefore,
future research that investigates the relationship between contextual variables and
cognitive functioning in older adults should make attempts to include a more diverse
sample.

A third explanation for the lack of association between demographic variables and
cognitive functioning may have to do with inadequate power. A power analysis was
conducted prior to examining the data. This analyses revealed that the present sample
size of 203 subjects was sufficient for detecting medium and large effect sizes, .30 and
.50 respectively (Cohen, 1992), but was inadequate for effects smaller than .30. In order
to detect smaller effects, a sample size of 481 subjects would be needed (Cohen, 1992).
Therefore, it may be that demographic variables indeed have a significant, but relatively
small, impact on cognitive functioning with age but the present sample size was
inadequately powerfirl to detect such an effect.

Methodological Considerations
In assessing the quality of empirical work related to age and neuropsychological
test performance, it is clear that a number of methodological considerations are needed.
Those most relevant to the present study include limitations involved in measuring
intended constructs, generalizability of the findings, and the constraints of correlational

research.

First, the use of different measures or operational definitions of intended
constructs across individual studies has created difficulty in interpretation. Operationally
defining the constructs of interest poses an exceptionally troublesome limitation in the
research of neuropsychological test performance. What is the most accurate way of
assessing constructs such as executive functioning or speed of processing?
Unfortunately, experts in the field disagree as to what constitutes an accurate or
appropriate definition of such cognitive abilities and often debate about which
instruments should be used to adequately measure them. For example, the on-going
dispute between Parkin (1998) and Baddeley (1998) regarding the construct of executive
functioning poses empirical questions that need to considered when designing any study
involving the measurement of executive abilities. Such disagreements highlight the need
to use multiple measures when attempting to assess constructs of the sort examined in the
present study. Indeed, one strength of this study is that three instruments were used to
measure executive functioning, including the Wisconsin Card Sorting Test (WCST), Trail
Making Test (TMT), and the Stroop Test (ST). However, it should also be noted that a
limitation exists with regard to the measurement of processing speed, as only one
instrument was utilized to assess this construct in the present study. Future research
should make every attempt possible to include more than one measure of each variable.
While the use of multiple measures does not completely solve the issue of construct
validity, it does contribute to a more accurate estimation of the construct of interest.

A second limitation in this study is concerned with the generalizability of the
research findings. The sample consisted of fairly well-functioning, normal, older adults

and was not particularly representative of the socioeconomic, educational, or cultural

49

variation in the general population. While there is undoubtedly interest in understanding
the processes of normal aging as it pertains to such individuals, it is important to point
out that the conclusions drawn from the present study may not generalize to the larger
population of older individuals.

A third limitation is concerned with problems more generally attributed to the
nature of correlational and cross-sectional research. Correlational data indicates
associations between the variables, but offers no explanation as to why these
relationships exist. For example, the present research findings indicate that age is
associated with cognitive decline in executive fimctioning and speed of processing, but
causal inferences concerning these variables cannot be made. Although it is necessary to
estimate significant relationships, as was done in the present study, it is also necessary to
define the mechanisms through which the relationships exist. Such information can only
be obtained by conducting non-correlational or experimental research, however, even
when using an experimental research design one is not guaranteed the necessary
requisites for making causal inferences. Thus, the limitations of correlational research
create obstacles for the investigation of many empirical questions and should be carefully
considered when designing studies in the field of cognitive aging in the future.

Future Directions

As already mentioned, future research would do well to include samples of
individuals which are more heterogeneous with respect to socioeconomic, educational,
and cultural characteristics and therefore more representative of the population as a

whole. A diverse sample not only helps to avoid the problems associated with restricted

variance and attenuated correlations, it also increases the likelihood that results will be
generalizable to a more extensive population of older adults.

In addition, future research would likely benefit from the inclusion of a younger
comparison group. The comparison of extreme age groups, as opposed to examining data
solely from a group of older adults, would be more likely to pick up on and emphasize
age-related variance in cognitive performance. Including a younger comparison group
would serve to increase the variation in performance assessed and, in turn, more clearly
highlight age-related differences in the areas of executive functioning and speed of
processing.

Finally, future research may consider the inclusion of both normal, healthy adults
as well as those who are functioning at a level inconsistent with normal aging, such as
those with Alzheimer’s disease or other dementing illnesses. This would not only allow
an investigation of the relative impact of demographic variables on the processes of
normal aging, but could also assess their impact in the development of pathological
diseases processes in later life. As the present study points out, although variables such
as education, occupation, premorbid IQ, health status, and gender, were not significant
predictors of cognitive decline associated with normal aging, they may significantly
moderate one’s chance of developing a dementing illness. Such empirical questions
could not be answered by this study and warrant further investigation.

Summary

It appears that age is significantly associated with cognitive decline in the areas of

executive functioning and speed of processing. However, the demographic variables

investigated, including gender, premorbid intelligence, level of education, type of

LII
p...

occupation, and health status did not significantly moderate the relationship between age
and executive functioning or the relationship between age and speed of processing.
These findings emphasize the need for future research in this area in order to more clearly

elucidate the impact of such variables on the process of normal aging.

Ill
IJ

APPENDIX

APPENDIX

Table 1
Means and standard deviations for subject’s age, education, occupational level,

premorbid IQ, health status and scores on the BDI, GDS and MMSE.

 

 

 

 

 

 

 

 

 

 

 

Participant Variable N Mean (SD)
Valid Missing

Age (years) 203 0 69.11 (7.86)
Years of Education 203 0 15.86 (3.02)
Occupational Level 202 1 5.07 (1.18)
Premorbid IQ 203 0 118.83 (8.48)
Health Status 203 0 1.85 (0.63)
BDI Total Score 203 0 6.33 (4.73)
GDS Total Score 203 0 5.88 (4.92)
MMSE Total Score 203 0 28.37 (1.45)

 

 

 

 

 

(,II

4.3

Table 2
Means and standard deviations for subjects’ performance on all dependent measures,

including WCST, ST, TMT and SDMT.

 

 

 

 

 

 

 

Dependent Measure N Mean (SD)
Valid Missing

WCST (Total Number of Perseverative 196 7 16.25 (12.70)
Errors)
ST (Interference Score) 195 8 4.70 (9.35)
TMT (Difference Score in Seconds) 203 0 56.73 (33.69)
SDMT (Total Number Correct, Oral 196 7 52.61 (12.46)
Portion)

 

 

 

 

55

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

.5. v 03? a I. ”3. v 2:? a a. .NSHZ cemummEoQ 333m: .3 macaw—2.50 EquE 8.605 :85?— $82

co; haw—mm.- nnmomr imom. $2. m2. *NS. .35ch 58.- 1%ri 3.23. 828958 mm
a... _ mm; 8.— .....NR. $5.- 95.. v3.- **mmm.- ......mom. for NS. .333.- $355 w;
Imam.- .INBN. co; mm? 26.. So. .52.- **mom. .2. 3:. 92.- mmZE
*amom. $9- m2; co; INK. mwo. Iowa. 91.- *mm _ .- $2.. 08.. mQO
awe _. 95.- 30.. int. co; 3 _. 15v. mm_.. mm _ .- * 3:.- Sc. RE
2:. Eco.- So. 98. 3 _. co; mmo. 3o: *3? Immmr v8. SEED
*NE. Immmr *mm _.- love. .1. av. mmo. co; Immmr _m _ .- *cc _ .- Re. .355. 5.8:
1.0ch Imom. Imam. 91.. m2; wwor 1.8.0..- oo; Immm. INS. 05.- ? $582.2;
>8.- er .2. *mm_.. mm_.- $2.- 22.. *nmmm. co; 1.2%. M22. :EEESUO
133.. NE. 3:. *mm_.- a. 5 _.- 13m.- ..ofr *nmmc. 1.8%. cc; Sc. :ccnoscm
imam. Icon.- ..am _.- omor moo. ewe. Eb. 05.- wm _. :5. cc; om<

am ma mmzz moo Em $50 5.8: 9 5:355 aosaaém a?

 

 

 

 

 

 

 

 

 

 

 

 

4.33%.; vubaaoE =m c8 55:: 5:20:30

m 92.5.

56

 

Table 4

Correlation matrix between age and measures of executive functioning and speed of

 

 

 

 

 

 

 

 

 

 

processing.
Age WCST TMT STROOP SDMT

Age 1.00 .309" .274" .066 -.437**
Executive Functioning:

WCST .309" 1.00 .207" -.189* -.312**

TMT .274" .207M 1.00 -.196** -.367**

STROOP .066 -.189* -. 196* * 1.00 .090
Speed of Processing:

SDMT -.437** -.312** -.367"‘* .090 1.00

 

 

 

 

 

 

Note. Pearson Product Moment Correlations by listwise comparison N=182.

* p value < .05; ** p value < .01

57

 

Eva ..... 032

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

an. 43.- as... «.8. . EN. E... a o: N

:25. a2. .NmN: BEN 2N. 2N. 0?. .0. _ 0. 2.22.2;
NNN. meN. E... :5. NS. 33 a 5.85 N

.58. 4mm. .5. SEN N2. NE. a? £3: _ .28:
SN. 3e. $2 2... 8.. an... a .859 N

:25. am. as. 8N2 N2. NE. a? .528 . 550
awn. meat 32. 25. N2. 3.... a 5:252: N

.38. SN. .3? 93: ”2. NE. a? 52.8380 _ 5:235
N_ _. 23. eva S... NNN. a»... a 5:336 N

:80. ca. 52.- 48.3 NNN. NNN. a? 52.838 _ 5:83a—

....ooo. am. $4.2 o2. o2. am< _ a?

$520 r.— .wfi 59m “cw—:30 r.— owcazu NM— Nm 9.3953.— .252

 

 

.chotocé 0350on can own 52.33 3:20:22

05 :o 9 298805 was £38m 5:8: .eovcow .coumanooo acumen—co .03 mo 88:: 2t @5695 mombmcn Samoa“: 29:32

m 2.5

58

Eva t as...

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mom. mgr Na. 3... .wwN. .&< x O: N

.550... m9..- .wVN. minim mwN. 3N. ow< .O. . 20.8.5...—
03. as. 2...... cc... va. Aow< x 5.3... N

:8... mmvr .ww..- SmNm EN. VmN. 0m< A23: . 5.3:
cN.. wmm. vva S... mmN. Gut. 0» .3150. N

Iooo. NS..- .woor hmhsN MNN. MNN. 03.. .5050 . .50—SO
awe. MAN..- vmm...” 3... SN. A0m< 0» 53.2.3ro N

Iooo. etc.- ..vo. NomNN MNN. mNN. 0m< 50.89.80 . 5.22.500
2m. .50.- .N... 2.... 3N. Aow< a 5:855: N

:80. m;- .0o.. www.cN mmN. mmN. 0w< 50:80.0: . 5.30:3.

:80. ms... 032 NNN. NNN. aw... _ an...

09.5.0 r.— .w.m 50m angsu a. 09.2.0 3. NM. 9.33.5... .0005.

 

 

.mEmmouoE .0 000% 0.5 own 503.2. £522.22

2.. :0 0. 05.0820 can .336 5.8: .50—Sm £222.38 £038.60 .owm .0 SEE. o... 9.8.020 mumbmcm sew-mew“: 0.05.23.

0 035...

59

LIST OF REFERENCES

60

LIST OF REFERENCES

Abate, G., Ferrari-Ramondo, V. & Di Iorio, A. (1998). Risk Factors For
Cognitive Disorders in the Elderly: A Review. Archives of Gerontology and Geriatrics

Suppl. 6, 7-15.

 

Albert, M. 8., Kaplan, E. (1980). Organic implications of neuropsychological
deficits in the elderly. In L. W. Poon, J. L. F ozard, L.S. Cermak, D. Arenberg, L. W.
Thompson (Eds), New directions in memory and aging. Hillsdale, New Jersey:
Erlbaum, 403- 432.

Alexander, G. E. & Moeller, J. R. (1994). Application of the Scaled Subprofile
Model to Functional Imaging in Neuropsychiatric Disorders: A Principal Component
Approach to Modeling Brain Function in Disease. Human Brain Mapping, 2, 79-94.

Anthony, J. C., LeResche, L. & Niaz, U. (1982). Limits of the “Mini-Mental
State” as a Screening Test for Dementia and Delirium Among Hospital Patients.
Psychological Medicinafl 397-408.

Arbuckle, Tannis Y., Gold, Dolores, P., Andres, David, Schwartzrnan, Alex &
Chaikelson, June (1992). The Role of Psychosocial Context, Age, and Intelligence in
Memory Performance of Older Men. Psychology & Aging, 711 1, 25-36.

Ardila, A., & Rosselli, M. (1989). Neuropsychological characteristics of normal
aging. Developmental Neuropsychology, 5, 307-320.

Axelrod, B. N., Goldman, R. S. & Henry, R. R. (1992). Sensitivity of the Mini-
Mental State Examination to Frontal Lobe Dysfimction in Normal Aging. Journal of
Clinical PsychologY. 48. 68-71.

Baddeley, Alan. (1998). The central executive: A concept and some
misconceptions. Journal of the International Neuropsychological Sociefl, 4, 523-526.

 

Baltes, P. B. (1987). Theoretical Propositions of Life-Span Developmental
Psychology: On the Dynamics Between Growth and Decline. Developmental

Psychology, 23, 611-626.

Barr, R. A., & Giambra, L. M. (1990). Age-related decrement in auditory
selective attention. Psychology and Aging, 5, 597-599.

Beard, C. M., Kokmen, E., Offord, K. & Kurland, L. T. (1992). Lack of
Association Between Alzheimer’s Disease and Education, Occupation, Marital Status or
Living Arrangement. Neurology, 42, 2063-2068.

61

Beck, A. T. (1970). Depression: Causes and Treatment. Philadelphia: University
of Pennsylvania Press.

Beck, A. T. (1987). Beck Depression Inventory. San Antonio, TX: The
Psychological Corporation.

Bench, C. J ., F rith, C. D., Grasby, P. M., F riston, K. J ., Paulesu, E., F rackowick,
R. S. J ., Dolan, R. J. (1993). Investigations of the functional anatomy of attention using
the Stroop task. Neuropsychologv. 31. 907-922.

Bielauskas, L. A. & Lamberty, G. J. (1992). Assessment of Depression in Elderly
Patients. Clinical Neuropsychologist, 6, 322.

Birren, James E. & Fisher, Laurel, M. (1995). Aging and speed of behavior:
Possible consequences for psychological fimctioning. Annual Review of Psychology, 46,
329-353.

Bonaiuto, S., Rocca, W. A., Lippi, A., Giannandrea, E., Mele, M., Cavarzeran, F.
& Amaducci, L. (1995). Education and Occupation as Risk Factors for Dementia: A
Population-Based Control Study. Neuroepidemiology, 14, 101-109.

Boone, K. B., Ghaffarian, S., Lesser, I. M., Hill-Gutierrez, E. & Berman, N. G.
(1993). Wisconsin Card Sorting Test Performance in Healthy, Older Adults:
Relationship to Age, Sex, Education, and IQ. Journal of Clinical Psychology, 49, 54-60.

Bomstein, R. A. & Suga, L. J. (1988). Educational Level and «
Neuropsychological Performance in Healthy Elderly Subjects. Developmental
Neuropsychology, 4(1 1, 17- 22.

Bowler, John V., Munoz, David G., Merskey, Harold & Hachinski, Vladimir
(1998). Factors Affecting the Age of Onset and Rate of Progression of Alzheimer’s
Disease. Jouflal of Neurology, Neurosurgery aLnd Psychiag, 65, 184-190.

Brayne, C. & Calloway, C. (1990). The Association of Education and
Socioeconomic Status with the Mini Mental Sate Examination and the Clinical Diagnosis
of Dementia in Elderly People. A e and A ein 19 91-96.

Brink, T. L., Yesavage, J. A., Lum, O., Heersema, P. H., Adey, M. & Rose, T. S.
(1982). Screening Tests for Geriatric Depression. Clinical Gerontoloiist. 1. 37-43.

Brown, C., Schulberg, H. C., & Madonia, M. J. (1995). Assessing Depression in
Primary Care Practice with the Beck Depression Inventory and the Hamilton Rating
Scale for Depression. Psychological Assessment, 7, 59-65.

Buell, S. J. & Coleman, P. D. (1979). Dendritic Growth in the Aged Human
Brain and Failure of Growth in Senile Dementia. Science 206 854-856.

 

62

Burke, H. R. (1972). Raven’s Progressive Matrices: Validity, reliability and
norms. Journal of Psychology, 82, 253-257.

Callahan, Christopher M., Hall, Kathleen, S., Hui, Siu L., Musick, Beverly S.,
Unverzagt, F. W. & Hendrie, H. C. (1996). Relationship of age, education, and
occupation with dementia among a community-based sample of African Americans.

Archives of Neurology, 53(21, 134-140.

Cerella, J. (1990). Aging and information-processing rate. In J. E. Birren & K.
W. Shaie (Eds. ), Handbook of the Psychology of Aging (3rd ed., pp. 201-221). New
York: Academic.

Christensen, H., Mackinnon, A., Jorrn, A. F ., Henderson, A. S., Scott, L. R. &
Korten, A. E. (1994). Age differences and inter individual variation in cognition in
community-dwelling elderly. Psychology and Aging, 9, 381-390.

Cobb, J. L., Wolf, P. A., Au, R., White, R. & D’Agostino, R. B. ( 1995). The
Effect of Education on the Incidence of Dementia and Alzheimer’s Disease in the
Framingham Study. Neurolo 45 1707-1712.

Cohen, J. (1992). A Power Primer. Psychological Bulletin, 112111, 155-159.

Cohen J. D., Dunbar, K., McClelland, J. L. (1990). On the control of automatic
processes. A parallel distributed processing account of the Stroop effect. Psychological
Review 97 332- 361. .

 

Cohn, N. B., Dustman, R. E., Bradford, D. C. (1984). Age-related decrements in
Stroop color test performance. Jouml of Clinical Psychology, 40, 1244-1250.

Colsher, P. L. & Wallace, R. B. (1991). Longitudinal Application of Cognitive
Function Measures in a Defined Population of Community-Dwelling Elders. Annals of

Epidemiology, 1, 215-230.

Corder, E. H., Saunders, A. M., Strittmatter, W. J ., Schmechel, D. E., Gaskell Jr.,
P. C., Rimmler, J. B., Locke, P. A., Conneally, P. M., Schmader, K. E., Tanzi, R. E.,
Gusella, J. F., Small, G. W., Roses, A. D., Pericak-Vance, M. A. & Haines, J. L. (1995).
Apolipoprotein E, Survival in Alzheimer’s Disease Patients and the Competing Risks of
Death and Alzheimer’s Disease. W 1323-1328.

Crum, R. M., Anthony, J. C., Bassett, S. S. & F olstein, M. F. (1993). Population-
Based Norms for the Mini-Mental State Examination by Age and Education Level.
Journal of the American Medical Association 269 23 86-2391.

 

63

Diagneault, S., Braun, C. M. J ., Witaker, H. A. (1992). Early effects of normal
aging on perseverative and non-perseverative prefrontal measures. Developmental

Neuropsychology, 8, 99-114.

Diamond, M. C. (1990). Morphological Cortical Changes as a Consequence of
Learning and Experience. In Scheibel, A. B. & Wechsler, A. F. (Eds), Neurobiology of
Higher Cognitive Function. New York: Guilford Press, 1-12.

Eppinger, M. G., Craig, P. L., Adams, R. L. & Parsons, O. A. (1987). The WAIS-
R Index for Estimating Premorbid Intelligence: Cross-Validation and Clinical Utility.

Journg of Consulting and Clinical Psychology, 55(11, 86-90.

Escobar, J. I., Burnam, A. & Kamo, M. (1986). Use of the Mini-Mental State
Examination (MMSE) in a Community Population of Mixed Ethnicity: Cultural and
Linguistic Artifacts. Journal of Nervous Mental Disorders 174 607-614.

 

 

Evans, D. A., Beckett, L. A., Albert, M. S., Hebert, L. E., Scherr, P. A.,
Funkenstein, H. H. & Taylor, J. O. (1993). Level of Education and Change in Cognitive
Function in a Community Population of Older Persons. Annals of E idemiolo 3 71-
77.

Evans, D. A., Hebert, Liesi E., Beckett, Laurel A., Scherr, Paul A., Albert,
Marilyn S., Chown, Marilyn J ., Pilgrim, David M. & Taylor, James O. (1997). Education
and Other Measures of Socioeconomic Status and Risk of Incident Alzheimer Disease in
a Defined Population of Older Persons. Archives of Neurology, 54, 1399-1405.

Farmer, M. E., Kittner, S. J., Rae, D. S., Bartko, J. J., & Regier, D. A. (1995).
Education and Change in Cognitive Function: The Epidemiological Catchment Area

Study. Annals of Epidemiology, 5, 1-7.

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. Journal of

Psychiatric Research, 12, 189-198.

Foster, J. R., Sclan, S., Welkowitz, J ., Boksay, I. & Seeland, I. (1988).
Psychiatric Assessment in Medical Long-Tenn Care Facilities: Reliability of Commonly

Used Rating Scales. International Journal of Geriatric Psychiagy, 3, 229-233.
Fratiglioni, L., Grut, M., Forsell, Y., Viitanen, M., Grafstrom, M., Homen, K.,
Ericsson, K., Bakman, L., Ahlbom, A. & Winblad, B. (1991). Prevalence of Alzheimer’s

Disease and Other Dementias in an Elderly Urban Population: Relationship to Age, Sex
and Education. Neurology, 41, 1886-1892.

Golden, C. J. (1978). Stroop Color and Word Test. Chicago, IL: Stoelting Press.

64

Golden, C. J. (1981). Diagnosis and Rehabilitation in Clinical Neuropsychology
(2“l ed.). Springfield, IL: Charles C. Thomas.

Greenough, W. T. & Green, E. J. (1981). Experience and the Changing Brain. In
McGaugh, J. & Kiesler, S. (Eds.), Aging, Biology, and Behavior. New York: Academic
Press, 159-193.

Grober, E. & Sliwinski, M. (1991). Development and Validation of a Model for
Estimating Premorbid Verbal Intelligence in the Elderly. Journal of Clinical and

Experiemental Neuropsychology, 13(61, 933-949.

Gur, R. C., Gur, R. E., Orbist, W. D., Skolnick, B. E., Reivich, M. (1987). Age
and regional cerebral blood flow at rest and during cognitive activity. Archives of
General Ps chia 44 617-621.

Haaland, K. Y., Vranes, L. F., Goodwin, J. S., & Garry, P. J. (1987). Wisconsin
Card Sort test performance in a healthy elderly population. Journal of Gerontology, 42,
345-346.

Hartley, J. T. (1986). Reader and text variables as determinants of discourse
memory in adulthood. Psychology and Aging, 1, 150-158.

Hartley, J. T. (1993). Aging and prose memory: Tests of the resource-deficit
hypothesis. Psychology and Aging, 8, 538-551.

Haug, H., Eggers, R. (1991). Morphometry of the human cortex cerebri and
corpus striatum during aging. Neurobiology of Aging, 12, 336-33 8.

Heaton, R. K., Chelune, G. J ., Talley, J. L., Kay, G. G., & Curtiss, G. (1993).
Wisconsin Card Sorting Test Manual: Revised and expanded. Odessa, FL:
Psychological Assessment Resources.

Hertzog, C. (1989). Influences of Cognitive Slowing on Age Differences in
Intelligence. Developmental Psychology, 25, 636-651.

Hertzog, C. & Dixon R. (1990). Aging, Information Processing Speed, and
Intelligence. In K. W. Schaie & M. P. Lawton (Eds.1, Annual Review of Gerontology
and Geriatrics 11 55-79.

 

Hertzog, C. (1991). Aging, Information Processing Speed, and Intelligence. In
K. W. Schaie & M. P. Lawton (Eds.), Annual Review of Gerontology and Geriatrics, 11,
55-79.

Hollingworth, H. L. (1928). Mental growth and decline. New York: Appleton.

65

Houx, P. J., Jolles, J. Vreeling, F. W. (1993). Stroop interference: Aging effects
assessed with the Stroop Color-Word Test. Experimental Aging Research, 19, 209-224.

Hultsch, D. F., Hertzog, C., & Dixon, R. A. (1990). Ability correlates of memory
performance in adulthood and aging. Psychology and Aging, 5, 356-368.

Hultsch, D. F ., Hertzog, C., Small, B. J ., McDonald-Miszczk, L. & Dixon, R A.
(1992). Short-Tenn Longitudinal Change in Cognitive Performance in Later Life.

Psychology and Aging, 7, 571-584.

Huttenlocher, P. R. (1997). Synaptic density in human frontal cortex-
developmental changes and effects of aging. Brain Research 163 195-205.

 

Hyer, L. & Blount, J. (1984). Concurrent and Discriminant Validities of the GDS
with Older Psychiatric Patients. Psychological Remus, 54, 611-616.

Isingrini, Michel. (1997). Relation between fluid intelligence and fiontal lobe
functioning in older adults. International Journ_al of Aging and Human Development, 45
(2139-109.

Kail, R. & Salthouse, T. A. (1994). Processing speed as a mental capacity. Acta
Psychologica, 86, 199-225.

Katzman, R. (1993). Education and the Prevalence of Dementia and Alzheimer’s
Disease. Neurology, 43, 13-20.

Kittner, S. J., White, L. R. & Farmer, M. E. (1986). Methodological Issues in
Screening for Dementia: The Problem of Education Adjustment. Journal of Chronic
Disorders 39 163-170.

 

Koenig, H. G., Meador, K. G., Cohen, H. J. & Blazer, D. G. (1988). Self-Rated
Depression Scales and Screening for Major Depression in Older Hospitalized Patients

with Medical Illnesses. Journal of the American Geriatrics Socieg, 36, 699-796.

Korczyn, A. D., Kahana, E. & Galper, Y. (1991). Epidemiology of Dementia in
Ashkelon, Israel. Neuroepidemiology, 10, 100.

Launer, L. J ., Andersen, K., Dewey, M. E., Letenneur, L., Ott, A., Amaducci, L.
A., Brayne, C., Copeland, J. R. M., Dartigues, J. F ., Kragh-Sorensen, P., Lobo, A.,
Martinez-Lage, J. M., Stijnen, T. & Hofman, A. (1999). Rates and Risk Factors for
Dementia and Alzheimer’s Disease: Results from EURODEM Pooled Analyses.

Neurology, 52, 78-84.

Letenneur, L., Commenges, D., Dartigues, J. F. & Barberger-Gateau, P. (1994).
Incidence of Dementia and Alzheimer’s Disease in Elderly Community Residents of
South-Western France. International Journal of Epidemiology, 23, 1256-1261.

66

Lezak, Muriel D. (1987). Norms for Growing Older. Developmental
Neuropsychology, 3( 1 1, 1-12.

Lezak, Muriel D. (1993). Newer contributions to the neuropsychological
assessment of executive functions. Journg of Head Trauma Rehabilitation 8 24-31.

 

Lezak, Muriel D. (1995). Neuropsychological Assessment (3rd Ed.). New York:
Oxford University Press.

Lowe, Patricia A. & Reynolds, Cecil R. (1999). Age, Gender, and Education May
Have Little Influence on Error Patterns in the Assessment of Set-Shifting and Rule

Induction Among Normal Elderly. Archives of Clinical NeurOpsychology, 14131, 303-
315.

Melamed, E., Lavy, S., Shlomo, 3., Cooper, G., Rinot, Y. (1980). Reduction in
regional cerebral blood flow during normal aging in man. Stroke, ll (11, 31-34.

Mittenberg, W., Seidenberg, M., O’Leary, D. S. & DiGuilo, D. V. (1989).
Changes in cerebral functioning associated with normal aging. Journal of Clinical and

Exgrimental Neuropsychology, 11, 918-932.

Morse, C. K. (1993). Does variability increase with age? An archival study of
cognitive measures. Psychology and Aging, 8, 156-164.

Moscovitch, M., Winocur, G. (1992). The Neuropsychology of Memory and
Aging. The Handbook of Aging and Cognition. F .I.M. Craik and TA. Salthouse (eds.)
Hillsdale, New Jersey: Erlbaum.

Nelson, H. E. (1976). A modified card sorting test sensitive to frontal lobe
defects. Cortex 12 313-324.

 

Nettelbeck, Ted & Rabbitt, Patrick M. A. (1992). Aging, Cognitive Performance,
and Mental Speed. Intelligence, 16, 189-205.

O’Connor, D. W., Pollitt, P. A. & Treasure, F. P. (1991). The Influence of
Education and Social Class on the Diagnosis of Dementia in a Community Population.

Psychological Medicine, 21, 219-224.

O’Connor, D. W., Pollitt, P. A., Treasure, F. P., Brook, C. P. B. & Reiss, B. B.
(1989). The Influence of Education, Social Class and Sex on Mini-Mental State Scores.

Psychological Medicine, 19, 771-776.

Parkin, A. J. (1998). The central executive does not exist. Journal of the
International Neuropsychological Sociefl, 4, 518-522.

67

Paykel, E. S., Brayne, C., Huppert, F. A., Gill, C., Barkley, C., Gehlhaar, E.,
Beardsall, L., Girling, D. M., Pollitt, P. & O’Connor, D. (1994). Incidence of Dementia
in a Population Older Than 75 Years in the United Kingdom. Archives of General

Psychiag, 54, 325-332.

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-P155.

Perls, T. T., Morris, J. N., Ooi, W. L. & Lipsitz, L. A. (1993). The Relationship
Between Age, Gender and Cognitive Performance in the Very Old: The Effect of

Selective Survival. Journal of the American Geriatric Socieg, 41, 1193-1201.

Pierce, T. W., Elias, M. W., Keohane, P. J ., Podraza, A. M., Robbins, M. A., &
Schultz, N. R. (1989). Validity of a short form of the Category Test in relation to age,
education, and gender. Experimental Aging Research, 15, 137-141.

Plassman, B. L., Welsh, K. A., Helms, M., Brandt, J ., Page, W. F. & Breitner, J.
C. S. (1995). Intelligence and Education as Predictors of Cognitive State in Late Life: A
50-Year Follow-Up. Neurolo 45 1446-1450.

Powell, D. H., & Whitla, D. K. (1994). Normal cognitive aging: Toward
empirical perspectives. Current Directions in Psychological Science, 3, 27-31.

Rabbitt, P. (1993). Does it all go together when it goes? The nineteenth Bartlett
Memorial Lecture. The Quarterly Journal of Exmrimental Pachology, 46A, 385-434.

Reitan, R. M. (1958). Validity of the Trail Making Test as an Indicator of
Organic Brain Damage. Perceptual and Motor Skills, 8, 271-276.

Reynolds, C. R. (1997). Measurement and Statistical Problems in
Neuropsychological Assessment of Children. In C. R. Reynolds & E. Fletcher-Janzen
(Eds.), Handbook of Clinical Child Neuropsychology (2Ind ed.). New York: Plenum
Press, 296-319.

Reynolds, W. M. & Gould, J. W. (1981). A Psychometric Investigation of the
Standard and Short form Beck Depression Inventory. Journal of Consulting and Clinical

Psychology, 49, 306-307.

Rocca, W. A., Bonaiuto, S., Lippi, A., Luciani, P., Turtu, F ., Cavarzeran, F. &
Amaducci, L. (1990). Prevalence of Clinically Diagnosed Alzheimer’s Disease and
Other Dementing Disorders: A door-to-door Survey in Appignano, Macerata Province,
Italy. Neurology, 40, 626-631.

Rowe, John W. & Kahn, Robert L. (1997). Successful aging. Gerontologist,
37(41, 433-440.

68

Salthouse, T. A. (1982). Adult Cognition: An Experimental Psychology of
Human Aging. New York: Springer.

Salthouse, T. A. (1985a). Speed of behavior and its implications for cognition. In
J. E. Birren & K. w. Schaie (Eds.), Handbook of the Psychology of Aging (2"d ed., pp.
400-426). New York: Van Nostrand Reinhold.

Salthouse, T. A. (1985b). A Theory of Cognitive Aging. Amsterdam: North-
Holland.

Salthouse, T. A. (1991). Mediation of Adult Age Differences in Cognition by
Reductions in Working Memory and Speed of Processing. Psychological Science, 2,
179-183.

Salthouse, T. A. (1992). What do adult age differences in the Digit Symbol
Substitution Test reflect? Journal of Gerontology: Psychological Sciences, 47, P121-l28.

Salthouse, Timothy A. & Babcock, Renee L. (1991). Decomposing Adult Age
Differences in Working Memory. Developmental Psychology, 27(51, 763-776.

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

Schaefer, A., Brown, J. & Watson, C. G. (1985). Comparison of the Validities of
the Beck, lung, and MMPI Depression Scales. Journal of Consulting and Clinical

Psychology, 53, 415-418.

Schaie, K. W. (1983). The Seattle Longitudinal Study: A 21-Year Exploration of
Psychometric Intelligence in Adulthood. In K. W. Schaie (Ed.), Longitudinal Studies of
Adult Cognitive Development. New York: Guilford Press, 64-135.

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

Schaie, K. W. (1994). The course of adult intellectual development. American
Psychologisg, 49, 304-313.

Schmand, B., Srnit, J. H., Geerlings, M. I. & Lindeboom, J. (1997). The Effects
of Intelligence and Education on the Development of Dementia: A Test of the Brain
Reserve Hypothesis. Psychological Medicine, 27, 1337-1344.

Shimamura. A. P., & Jurica, P. J. (1994). Memory interference effects and aging:
findings from a test of frontal lobe function. Neuropsychology, 8, 408-412.

69

Smith, A. (1982). Symbol Digit Modalities Test (SMDT). Manual (Revised).
Los Angeles: Western Psychological Services.

Snowdon, D. A., Kemper, S. J. & Mortimer, J. A. (1996). Linguistic Ability in
Early Life and Cognitive Function and Alzheimer’s Disease in Late Life: Findings From
the Nun Study. Jomal of the American Medical Association 275 528-532.

 

Spreen, Otfried & Strauss, Esther. (1998). A Compendium of
Neuropsychological Tests: Administration, Norms, and Commentary (2Ml Ed.). Oxford
University Press: New York, NY.

Steer, R. A., Beck, A. T. & Brown, G. (1989). Sex Differences on the Revised
Beck Depression Inventory for Outpatients with Affective Disorders. Journal of

Personalig Assessment, 53, 693-702.

Stern, Y., Gurland, B., Tatemichi, T. K., Tang, M. X., Wilder, D. & Mayeux, R.
(1994). Influence of Education and Occupation on the Incidence of Alzheimer’s Disease.
Journal of the American Medical Association 271 1004-1010.

 

Stern, Yaakov, Alzexander, G. E., Prohovnik, I., Stricks, L., Link, B., Lennon,
M. C. & Mayeux, R. (1995). Relationship between lifetime occupation and parietal flow:
Implications for a reserve against Alzheimer’s pathology. Neurology, 45(11, 55-60.

Struble, R. G., Price, D. L., Cork, L. A. (1985). Senile plaques in cortex of aged
normal monkeys. Brain Research 361 267-275.

 

Sulkava, R., Wikstrom, J ., Aromaa, A., Raitasalo, R., Lahtinen, V., Lahtela, K. &
Palo, J. (1985). Prevalence of Severe Dementia in Finland. Neurology, 35, 1025-1029.

Swaab, D. F. (1991). Brain Aging and Alzheimer’s Disease: “Wear and Tear”
versus “Use it or Lose it”. Neurobiological Aging, 12, 317-324.

Tranel, Daniel, Benton, Arthur, & Olson, Kerri (1997). A lO-Year Longitudinal
Study of Cognitive Changes in Elderly Persons. Developmental Neuropsychology,

13(11, 87-96.

Van Gorp, Wilfred, G., Satz, Paul, Mitrushina, Maura. (1990). ,
Neuropsychological processes associated with normal aging. Developmental
Neuropsychology, 6(41, 279-290.

Wechsler, D. (1956). Manual for the Weschler Adult Intelligence Scale. New
York: Psychological Corporation.

Welford, A. T. (1984). Between Bodily Changes and Performance: Some
Possible Reasons for Slowing with Age. Experimental Aging Research, 10, 73-88.

70

Wheliham, W. M., Lescher, E. L. (1985). Neuropsychological changes in fi'ontal
functions with aging. Developmental Neuropsychology, 1, 371-380.

White, L., Katzman, R., Losonczy, K., Salive, M., Wallace, R., Berkman, L.,
Taylor, J ., F illenbaum, G. & Havlik, R. (1994). Association of Education with Incidence
of Cognitive Impairment in Three Established Populations for Epidemiological Studies of
the Elderly. Journal of Clinical Epidemiology. 47. 363-3 74.

Wiederholt, W. C., Cahn, Deborah, Butters, Nelson M., Salmon, David P., Kritz-
Silverstein, Donna & Barrett-Connor, Elizabeth (1993). Effects of Age, Gender and
Education on Selected Neuropsychological Tests in an Elderly Community Cohort.

Journal of the American Geriatric Socieg, 41, 639-647.

Yesavage, J. A., Brink, T. L, Rose, T. L. & Adey, M. (1986). The Geriatric
Depression Rating Scale: Comparison with Other Self-Report and Psychiatric Rating
Scales. In L. W. Poon (Ed.), Handbook of Clinical Memory Assessment of Older Adults.
Washington, D. C.: American Psychological Association.

Yesavage, J. A., Brink, T. L., Rose, T. L., Lum, 0., Huang V., Adey, M. B. &
Leirer, V. O. (1983). Development and Validation of a Geriatric Depression Rating

Scale: A Preliminary Report. Journal of Psychiatric Research, 17, 37-49.

Zhang, M., Katzman, R., Salmon, D., Jin, H., Cai, G., Wang, Z., Qu, G., Grant, I.,
Yu, E. Levy, P., Klauber, M. R. & Liu, W. T. (1990). The Prevalence of Dementia and
Alzheimer’s Disease 1n Shanghai, China: Impact of Age, Gender and Education. Annal___s

f Neurology, 27, 428-437.

71

l+Illill‘llllllilll