THESIS

i

IIIIIIIIIIIIIIIIII

LIBRARY IIIIIIIIIIIILIIIIIIIIIIIIIIIII'IIII III IZIII‘IIIIIIIIII

Michigan State
University

 

 

 

 

This is to certify that the

thesis entitled

A Study of Attention
and Depression in the Able Elderly

presented by

Angela Mae McBride

has been accepted towards fulfillment
of the requirements for

M-A- degree in_E$JLC_thQQ.Y

III. III;

Major professor

 

 

Date ‘43qu

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINE return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mo mu

A STUDY OF ATTENTION AND DEPRESSION IN THE ABLE ELDERLY
By

Angela Mae McBride

A THESIS

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

MASTER OF ARTS
Department of Psychology

1998

ABSTRACT
A STUDY OF ATTENTION AND DEPRESSION IN THE ABLE ELDERLY
By

Angela Mae McBride

Although depression is second only to dementia in prevalence among older adults
(Kessler, McGonagle, Zhao, et al., 1994), it is unclear how factors such as age, gender,
and level of cognitive functioning may be related to an individual’s feelings of depression.
This study addressed the association between symptoms of depression and cognitive
fimctioning among 50 community-dwelling older adults (Mean age = 71), as well as
investigating the relationship between gender and types of depressive symptoms reported.
Finally, the effects of memory and attention training were measured. Significant
correlations were found between both the Beck Depression Inventory and the Geriatric
Depression Scale and one measure of attention, Trails B (r = .434, p < .01; r = .358, p <
.05, respectively). Symptoms of depression accounted for 18% of the variance on this
measure (2 < .01), and symptoms related to motivational problems explained more of this
variance (Beta = .38, p < .05) than did those related to mood (Beta = .08, p_ < .59, ns).
Contrary to expectations, there were no difl‘erences between men and women in the types
of depressive symptoms reported. Attention scores as measured by the Paced Auditory
Serial Addition Task improved from baseline levels after participants completed memory
and attention training workshops (1(49) = -4.02, p_ < .001). Implications for the diagnosis

of older adults with depression and cognitive impairment are discussed.

ACKNOWLEDGMENTS

The completion of this work would not have possible without the assistance of a
number of other people. I would like to first thank my committee chairperson and mentor,
Dr. Norman Abeles, for helping me to develop an area of interest which I could
enthusiastically pursue, and providing me with the guidance and advice to see it through.
Thanks are also due to Dr. Ray Frankmann, whose statistical knowledge and insights were
invaluable, and Dr. Dozier Thornton, who provided helpful comments to improve the
breadth and clarity of my writing. My committee members all gave freely of their time to
meet and talk with me, for which I am very appreciative.

I would also like to express thanks to my fiiends, in particular Scott Moss, who
provided immeasurable support and enouragement over the miles, and Jodi
Levy-Cushman, who was always there to lifi my spirits when the going got tough.

I feel extremely lucky to have been blessed with such a generous, supportive, and
loving family. My grandparents, Ed and Ann Mikovich, deserve a special “thank you” for
all they have done for me over the years. And to my parents, Dennis and Ellen McBride,
and my brother Denny-J couldn’t ask for a more caring family. Because of you, I have
been able to achieve what I have today. Your support on every level has made my life

richer in so many ways.

iii

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................... v
INTRODUCTION ........................................................................................................... l
Depressive Symptoms in Older and Younger Adults ............................................. 2
Normative Age-Related Changes in Attention ....................................................... 4
Relationship between Depression and Cognition ................................................... 6
Symptoms of Depression and Their Relation to Attention ..................................... 9
Gender Differences in Depression ....................................................................... 12
Clinical Depression vs. Depressive Symptomatology ........................................... 14
Effects of Memory Training on Cognitive Performance ....................................... 14
Aims of Current Study ........................................................................................ 18
Hypotheses ......................................................................................................... 19
METHOD ...................................................................................................................... 20
Participants and Procedure .................................................................................. 20
Measures ............................................................................................................ 22
RESULTS ..................................................................................................................... 27
Hypothesis I ....................................................................................................... 27
Hypothesis II ...................................................................................................... 27
Hypothesis III ..................................................................................................... 28
Hypothesis IV .................................................................................................... 30
DISCUSSION ............................................................................................................... 31
Depressive Symptoms and Performance on Attention Tests ................................ 31
Mood, Memory, and Attention Training ............................................................. 32
Mood vs. Motivational Symptoms of Depression ................................................ 33
Gender and Self-Reported Symptoms of Depression ........................................... 34
Limitations of Study ........................................................................................... 3 5
Implications ........................................................................................................ 36
Directions for Further Research .......................................................................... 36
APPENDIX A Items Comprising Mood and Motivation Factors ................................... 39
APPENDIX B Demographic Information for the Sample ............................................... 41
LIST OF REFERENCES ............................................................................................... 44

iv

LIST OF TABLES

TABLE 1 Education ..................................................................................................... 41
TABLE 2 Marital Status ............................................................................................... 41
TABLE 3 Occupation ................................................................................................... 42
TABLE 4 Participants’ Self-Ratings of Physical Health ................................................. 42

INTRODUCTION

Depression is one of the most common and serious conditions listed in the
Diagnostic and Statistical Manual of Mental Disorders (DSM-IV). Its general features
include a depressed mood or loss of interest and pleasure in nearly all activities. An
individual with depression may also experience changes in appetite or weight, disturbances
in sleep or psychomotor activity, feelings of guilt or hopelessness, difficulty concentrating,
or thoughts of suicide. This type of mood disorder can manifest in a major depressive
episode, or may occur in a less severe, yet chronic pattern known as dysthymic disorder
(American Psychiatric Association, 1994).

Lifetime prevalences of depressive disorders range from up to 15% for men to as
high as 24% for women in the United States (Kessler, McGonagle, Zhao, et al., 1994). In
older populations, it is second only to dementia in prevalence. These rates of depression
in the elderly are comparable to those'in younger populations, although there may be a
trend among older adults to experience milder forms of depression than do younger adults
(LaRue, 1992). For example, while symptoms of depression can be found in about 15%
of community-dwelling older adults, major depression is diagnosed in only 3% of this
population. Rates of depression reach levels of 15% to 25% among nursing home
residents when both major depression and dysthymic disorder are included in the number
(F riedhoff, 1991).

Less severe depressive disorders affect older populations in somewhat difl‘erent
ways than younger cohorts, including increased cognitive impairment (LaRue, Swan, &

Carmelli, 1995; Sweeney, Wetzler, Stokes, & Kocsis, 1989) and higher rates of somatic

complaints (LaRue, 1992). This must be considered when diagnoses are being made, as
individuals diagnosed with depression share many of the cognitive deficits seen in mild
dementia. In addition, it is important to analyze the construct of ‘depression’ to determine
if these different patterns of symptoms are indeed indicators of the same underlying
disorder. For example, will two people who indicate the same number or severity of
depressive symptoms demonstrate the same types of deficits of memory and attention,
regardless of the particular symptoms reported? If not, it will likely be useful for both
treatment and prevention efforts to recognize which types of depressive symptoms are
associated with cognitive impairment. To understand more fiilly the factors which relate
to cognitive problems in older adults, however, one would also need to consider each
individual’ s unique environment, including the amount of social support they receive, as
well as life events such as the loss of a spouse or close friend. These factors may also

create large individual differences in the effects of depression on cognitive fimctioning.

D r iv tms'n lder YunerA ts

One way in which depression often appears to differ between older and younger
people is in the patterns of symptoms each group experiences. While acknowledging that
there is wide individual variation in the experience of depression, some researchers have
found that older adults tend to have fewer feelings of guilt and self-blame and less
dysphoric mood than do younger adults. Although they do not as often endorse feelings
of sadness, older adults report feeling a greater sense of hopelessness and lack of interest
in the world around them, as well as a sense of worthlessness (Newmann, Jensen, &

Engel, 1991). Newmann and her colleagues suggest that we call this feeling of

hopelessness, worthlessness, and lack of interest a “depletion syndrome.” They comment
that in some older adults with this depletion syndrome there is insuflicient presence of
dysphoric mood to make a diagnosis of depression. Depressive diagnoses such as
dysthymic disorder which require the presence of dysphoric mood will tend to exclude
those older adults who lack this symptom. However, older adults may simply be
experiencing an equally distressing, yet different type of depression. It is therefore
important to determine if this lack of interest in one’s surroundings and depletion of
energy can be suficient indicators of depression in older adults even if they do not report
experiencing sad moods.

This pattern of differences between depressive symptoms in older and younger
adults has not been found in other studies, however. Musetti, Pemgi, Soriani, Rossi,
Cassano, and Akiskal (1989) examined difl‘erences in symptomatology between adults
under and over age 65 (N = 400) with primary major depression, including both
outpatients and inpatients. The older group was found to have higher levels of weight loss
and psychomotor retardation, while the younger group had more cognitive symptoms,
such as self-blame and suicidal ideation. In general, however, both groups showed a
similarity among symptoms, suggesting a core symptomatology of depression that exists
across the range of ages.

A study of patients hospitalized for major depression (N = 257) produced similar
results. For both men and women, cognitive symptoms such as feelings of guilt and
failure, self-accusations and suicidal ideation showed a negative relationship with age

(Wallace & Pfth 1995). However, it is important to note that clinically depressed

individuals may show different patterns of symptoms than those who do not meet a
diagnosis of major depression.

In a 1989 review by McNeil and Harsany, the authors argue for an “age
difi‘erence” view of depression. They provide evidence for a discontinuity between older
and younger adults in the types of symptoms they experience; namely, that somatic
symptoms are more common among older adults, and that older women report more
somatic symptoms than older men. This gender difference was not found to be present
among younger adults. The authors argue that somatic symptoms, rather than being the

result of normal aging or simply the presence of physical illness, are valid indicators of

depression in older adults.
Nmative AggRegted Qhaggg in AttQtion

Everyone knows what attention is. It is the taking possession by the mind,

in clear and vivid form, of one out of what seems several simultaneously

possible objects or trains of thought. Focalization, concentration, of

consciousness are its essence. It implies withdrawal from some things in

order to deal better with others (James, 1890, pp. 403-404).

William James’ definition of attention dates back to 1890, yet it continues to be
quoted and cited as an apt description of this complex process. Attention has been
conceptualized in many ways, but perhaps the most important feature of attention is its
multifacetedness: attention cannot be characterized as a unified trait or ability. Van
Zomeren and Brouwer (1994) point out that to describe somebody as having poor
attention is of little use unless one can explain in more detail exactly what ability they are

referring to, and on what measure this judgment is based. Attention can include,

“. . .selectively preparing for, maintaining the preparation for, and processing certain

aspects of experience. Attention is also responsible for the coordination of multiple
simultaneous tasks” (Hartley, 1992, pp. 4-5). Stankov (1988) discusses six basic types of
attentional processes, each of which may be altered to varying degrees as an individual
ages. These processes include concentration, search, divided attention, selective attention,
attention switching, and vigilance.

General theories of attention point to a need for the selection of particular stimuli
to the exclusion of others because people have a finite capacity for processing information.
Although the exact mechanisms of the process are not known, it has been demonstrated
that in general, older adults show slower reaction times and slower speeds Of task
performance than do younger adults (Tarbuck & Paykel, 1995). Changes in the brain that
normally occur as one ages may account for some of this slower processing in older
adults. More specifically, there is a loss of neurons in certain areas of the brain, as well as
fewer dendrites, which allow for fewer connections between neurons, and problems in
synaptic transmission such as a decline in the levels of neurotransmitters. Hartley writes,
“Given a wide variety of changes at many difi‘erent levels, it is unlikely that age changes in
the modules that together constitute attention will be uniform in their magnitude, time
course, or qualitative efl‘ects” (1992, p. 10). It is therefore difficult to predict the manner
in which individual changes in attentional processes will occur.

Thus, the capacity for attention, as measured by tests of processing speed and
response time, does generally decline with age. However, because it is such a complex
process, it becomes necessary to determine the many components comprising the process

of attention, and investigate how these are afi‘ected by aging. Many clinical tests have

been designed to capture different aspects of attention, and can be used to determine more

specifically which aspects of attention change as one ages.

Relationship between Depression and Cogm'tion

When an older adult presents with memory complaints, difficulty concentrating,
and related cognitive deficits, one of the most challenging diagnostic decisions involves
determining the origin of these symptoms. Specifically, are the reported problems a result
of dementia, which indicates a continual decline in cognitive functioning, or rather are they
due to depression, which can often be successfully treated, resulting in the elimination of
these deficits? A term that has often been used to describe these cognitive problems is
“pseudodementia,” so named because of the similarities between cognitive symptoms
associated with depression and those of dementia.

Salzrnan and Gutfreund (1986) define pseudodementia as, “...true impairment of
memory processes secondary to depression but without impairment of other aspects of the
mental status” (p. 260). There are several features of this condition which difi‘er between
individuals with dementia and those with depression. In true dementia, symptoms will
develop over time, and depression will follow the decline in cognitive firnction. In
contrast, in people who are depressed and show pseudodementia, the depression appears
before a loss of cognitive function, and the onset of symptoms takes place over a relatively
short span of time. Because of their similar presentation, the distinction between dementia
and pseudodementia is often extremely difficult to determine; however, it is of great

importance in determining an appropriate intervention and treatment plan In

pseudodementia it is vital that depression be treated aggressively since doing so may
reverse cognitive impairments (Blazer, 1993; Salzman & Gutfreund, 1986).

Although certain types of problems are more common among older adults with
depressive disorders, the literature is unclear as to how the variables of age, gender, and
level of cognitive fimctioning may be related in a diagnosis of depression (Tarbuck &
Paykel, 1995). Lichtenberg, Ross, Millis, and Manning (1995) propose that the
inconsistent findings in the literature may be attributed in part to differing methodologies,
variable ages of the subjects, and the large number of different measures of cognition that
are used. In addition, they advocate the use of depression and cognition as continuous
variables as opposed to discrete (and often extreme) groups. Their study investigated the
question of whether depression can be an independent predictor of cognition, using a
sample of 220 geriatric medical patients. They found that Geriatric Depression Scale
scores accounted for roughly 8% of the unique variance for measures of cognition, which
consisted of the Dementia Rating Scale and Logical Memory measures, and concluded
that increased levels of depression are associated with lower cognitive fimctioning.

Tarbuck and Paykel (1995) also looked at the cognitive fimctioning of people
hospitalized for major depression to determine if older and younger individuals would be
afi‘ected differently. They divided subjects into those who were under 60 years of age
(N = 18) and those who were older (N = 19), and found differences between the two
groups in their performance on cognitive tasks, both while depressed and after they had
recovered. The older group did not perform as well as the younger group, particularly on
tasks which were timed and required rapid response time. For both age groups,

depression was associated with lowered performance on all of the measures of cognition

except for simple reaction time. The authors interpret these results in terms of task
complexity; depression had a more pronounced effect on tasks requiring greater cognitive
effort, and little to no effect on those tasks which called for minimal information
processing.

LaRue, Swan, and Carmelli (1995) used a sample of male subjects (N = 1217) who
were participating in a study of risk factors for heart disease. Only 5% of their sample was
depressed, as opposed to the inpatient populations used in many studies. However, their
results were similar; they found that younger subjects performed better than older ones on
cognitive measures, and self-reported symptoms of depression were found to predict
scores on these measures, although to only a modest degree. The authors propose that
the absence of positive afi‘ect, rather than depression per se, may be most responsible for
difi‘erences in psychomotor speed and memory scores. This is similar to the depletion
syndrome discussed earlier; older adults may not feel dysphoric, yet still may lack a
positive, hopefirl view of themselves and the fixture.

Sweeney, Stokes, Wetzler, and Kocsis (1989) used a sample of psychiatric
inpatients who had been diagnosed with major depression (N = 23). Using the Rey
Auditory Verbal Learning Test (RAVLT) and the Famous Faces Test (FFT) as their
measures of cognitive function, they found that depression was highly associated with
subjects’ performance on the initial trial of the RAVLT but not on subsequent trials. This
is presented as supporting the hypothesis that problems with attention and initial encoding
of stimuli are most affected by depression, and that the effects of depression dissipate
when the stimuli are repeated over time. The authors suggest that to clarify the

relationship between attention and depression, “. . .it may be important to utilize a task that

strains the limits of attentional capacity” (p. 841). This may be an important component
Of the equivocal findings regarding attention and depression; Tarbuck and Paykel (1995)
also suggested that differences in cognitive function become more pronounced with tasks
requiring more cognitive effort.

Rabbitt, Donlan, Watson, McInnes, and Bent (1995) found that individuals in their
sample (N = 4243) with elevated scores on the Beck Depression Inventory performed
worse on all of the cognitive tests they administered than did those with lower BDI scores,
providing additional support for the link between depression and cognitive fiinctioning.
Their results also indicated that advancing age was associated with a decline in
performance; however, these two variables did not interact. That is, aging per se does not
appear to be a risk factor for the negative association between cognitive function and
depression.

The link between depressive symptoms and cognitive impairment suggests that
techniques designed to improve cognitive abilities, such as visual scanning, shifting
attention, and short-term memory, may have the additional effect of alleviating some
depressive symptoms. Past studies (e.g., Bulbena & Berrios, 1993) have shown that
affective disorders can cause temporary problems with attention and memory, but that

these impairments are reversible.

Symptoms of Depression gd Their Relation to Attention

A program of research by Forsell and colleagues proposes that mood and
motivation may be differentially related to cognitive firnctioning. One study by these

researchers (N =1304) found that the construct of depression may be separated into

symptoms related to mood and motivation, and that motivational symptoms ‘(e.g.,
difiiculties concentrating or thinking; loss Of energy and interest) are increasingly
associated with cognitive deficits in old age, while affective symptoms (e. g. dysphoria,
feelings of guilt, suicidal ideation) increase but then gradually decrease with continued
cognitive impairment. In addition, women had significantly more mood-related symptoms
of depression than did men, while men had somewhat more motivational symptoms
(Forsell, Jorm, & Winblad, 1994).

Backman, Hill, and Forsell (1996) asked older community-dwelling adults
(N = 303) to perform a variety of word recall tasks, as well as administering the
Comprehensive Psychopathological Rating Scale. Specific items on this scale were
grouped separately into symptoms of mood and motivation, and then each group of
symptoms was compared with subjects’ performance on the recall tasks. Motivational
symptoms were found to be negatively associated with the subjects’ performance on all
four tasks, while mood-related symptoms were not associated with level of performance.
Looking more closely at the specific symptoms, it appeared that symptoms which were
most closely associated with the ability to focus and sustain attention had the most
pronounced impact on performance. These symptoms included difficulty concentrating
and lack of interest. The authors also emphasize the importance of examining attentional
performance separately from memory performance, as these factors are often related, but
may difi‘er in key ways that can provide insight into the particular ways that depression
impacts various cognitive functions.

Studies by F orsell, Jorm, Fratiglioni, Grut, & Winblad (1993) and Forsell, Jorm, &

Winblad (1994) employed a factor analysis of DSM-III-R criteria for major depression.

10

Their subjects included adults 75 years of age and older (N = 643 and 1304, respectively);
and consisted of both community-dwelling as well as institutionalized individuals. The
factor analysis for both studies showed that the two factors accounting for the most
variance were dimensions labeled “mood” and “motivation” by the authors. The mood
factor (accounting for 15% and 34.9% of the variance, respectively) included symptoms of
dysphoria, sleep and appetite disturbance, feelings of guilt, and suicidal ideation. The
motivation factor (accounting for 29% and 13.7% of the variance, respectively) included
symptoms such as loss of interest, psychomotor change, loss of energy, and disturbance in
concentration. The authors conclude that “. . .dividing the symptoms of major depressive
syndrome into mood disturbance and motivation disturbance factors might have potential
for explaining some of the perplexing results in previous studies of depression in the
elderly” (Forsell, Jorm, & Winblad, 1994, p. 1603).

Self-report instruments commonly used to assess depressive symptoms in older
adults include the Geriatric Depression Scale (GDS) and the Beck Depression Inventory
(BDI). Several studies have investigated the factor structure of the GDS using difi‘erent
populations, including psychiatric outpatients, medical inpatients, and non-referred
community-dwelling older adults. Sheikh, Yesavage, Brooks, Friedman, Gratzinger, Hill,
Zadeik, and Crook (1991) studied psychometric properties of the GDS. Their factor
analysis produced five factors, including sad mood, lack of energy, positive mood,
agitation, and social withdrawal. They found that the sad mood factor accounted for 11%
of the variance, and the lack of physical and mental energy factor accounted for 9.7% of
the variance. Other studies (Abraham, Wofford, Lichtenberg, & Holroyd, 1994; Salamero

& Marcos, 1992) have found the factor structure to vary somewhat depending on the

11

population being investigated. However, these studies indicate that it is possible to
produce similar factors comprised of mood and motivational symptoms.

Many factor analytic studies have also been conducted using the Beck Depression
Inventory. Tanaka and Huba (1984) performed a confirmatory factor analysis using two
sets of data from previous studies. Both analyses indicated support for three dimensions:
negative attitudes/ suicide, performance difficulty, and physiological. Excluding
physiological symptoms of depression which indicate somatic disturbances, these first two
factors correspond with the concept of separate mood and motivation factors. A review
of factor analytic studies by Startup, Rees, & Barkham (1992) indicates that three general
factors compose the BDI. These include mood/motivation, self-denigration, and
vegetative disturbances. The mood/motivation component particularly reflects DSM-IV
criteria for depression, which require a sad mood or lack of interest or pleasure in most
activities. The authors note that this component appears to be the focus of most current

theories of depression.

(finer Difi‘erepces in Depression

After about age 13, females are diagnosed with depression twice as often as men
(N olen-Hoeksema, 1990). A variety of explanations have been proposed as to why this
difference occurs, including biological challenges such as hormonal changes, social
challenges such as sexual abuse, parental and peer expectations, and personality factors
(for a review, see Nolen-Hoeksema, 1994). It is important to see if this pattern also exists

in old age, and if so, whether men and women report different clusters of symptoms.

12

In a sample of clinically depressed individuals ranging in age fi'om 21 to 65 years
(N = 294), Komstein, Schatzberg, Yonkers, Thase, Keitner, Ryan, and Schlager (1995)
found that women experienced more severe depression than men, and their symptoms
included higher levels of psychomotor retardation. In addition, there was a trend toward
increased anxiety and somatization in women. However, there was no gender difi‘erence
in overall number of depressive symptoms, using DSM-III-R criteria for major depression.

Rabbitt, Donlan, Watson, McInnes, and Bent (1995) found that elevated Beck
Depression Inventory scores were associated with decreased performance on cognitive
tasks to a lesser degree for women than men; they propose that perhaps women are less
affected by low levels of depression than are men. However, the authors also note that
this finding may be due simply to a difference in symptom reporting: men may be more
hesitant to indicate severe depression than are women, so that equivalent scores on the
BDI may not indicate equal levels of depression.

A review of studies investigating gender difi‘erences in depressive symptomatology
among older adults (F opma-Loy, 1988) suggests that difi‘erences may exist, although
research in this area is inconclusive. In one sample of 106 individuals ranging in age from
63 to 84, women were found to have higher levels of anxiety and interpersonal sensitivity,
while men showed a greater amount of paranoid ideation (Hale & Cochran, 1983).
Another study found no differences in the depressive syndrome experienced by older men
and women (Oltman, Michals, & Steer, 1980). These equivocal findings point to a need

for further investigation of gender differences in depression among older adults.

13

Clinical Depression vs. Depressive Smptomatology

A 1991 study by Newmann, Engel, and Jensen explored the patterns of symptoms
reported by women who were divided into a younger (ages 51 to 65, N = 184) and Older
(ages 66 to 92, N = 184) cohort. In explaining the implications of their results, they state:
“A central assumption guiding the analysis was that it is important to distinguish between
symptoms suggestive of a clinical syndrome and other, more delimited, forms of distress.
Further, we assumed that an adequate understanding of age differences in depressive
symptom experiences required attention to the different forms that a depressive syndrome
may take in the population” (p. 23 2). The distinction between depressive symptoms, such
as those indicated by a self-report measure, and less common clinical depressive
syndromes, is not always clarified in the literature. Thus, it is important to assess whether
findings that occur in clinically depressed populations are also true for individuals who are
experiencing distress that has not reached clinical levels.

While there is a large body of research indicating an association between
depression and cognitive functioning (see review by Burt, Zembar, & Niederehe, 1995),
these studies mainly involve samples of clinically depressed individuals, or those who show
a serious decline in cognitive functioning. Thus, it is dificult to tell whether depressive
symptoms that do not compose a clinical diagnosis are associated with cognitive decline

(Rabbitt, Donlan, Watson, McInnes, & Bent, 1995).

E f Mem T ' 'n on five Performance
Because attention and memory have such a close and reciprocal relationship to one

another (Cowan, 1995), even strategies which do not focus exclusively on attention

14

training may help to improve one’s attentional ability. Well-established memory training
techniques may change one’s attentional strategies, perhaps leading to more effective
encoding of stimuli and increasing the likelihood that information will be retained in
memory.

In his review of the memory training literature, Poon (1984) concludes that
training in memory strategies can improve task performance in both younger and older
adults. He also states that older adults generally show even more of an improvement from
baseline levels than do younger adults, thus supporting the use of memory instruction as a
way to improve memory in older adults. As Smith (1980) points out, older adults have
more difficulty than younger adults with the encoding and retrieval of information;
however, storage of information remains relatively stable as one ages. Thus, many
techniques designed to teach strategies for improving memory and attention to older
adults focus on the processes of encoding and retrieval. A variety of techniques are often
used to teach these skills, including the method of loci, peg-word system, use of cues, and
visual imagery.

Imagery has been shown to be a powerful technique for remembering lists of
words. In one study (Hulicka & Grossman, 1967), older adults who were asked to form
their own visual images to associate pairs of words performed better than those who were
told what images to use; this group in turn recalled more words than those who were
given no mnemonic instruction. This technique appears to be helpful for both encoding
and retrievaL as it requires the individual to put effort into making an association; this
image is then personally salient, which increases the likelihood that the words will be

remembered.

15

Another important component of many memory and attention workshops is an
emphasis on the development of a sense of efficacy in performing tasks requiring these
skills. Woods and Bernard (1987) hypothesized that older adults who read a statement
explaining that one’s attitude can affect memory performance would remember more
information than those who did not read the statement. They believed that the statement
would affect the subjects’ metamemory, or knowledge about their memory processes, thus
leading to an increased sense of efiicacy. Results indicated a tendency for those in the
group who read the attitude statement as well as learning memory strategies to
outperform those who learned the same strategies but did not read the attitude statement,
although this difi‘erence was not statistically significant.

Memory and attention training may also afi‘ord an opportunity for discussion
among the participants. They may be encouraged to bring up concerns related to aging,
memory decline, the mnemonic techniques themselves, mood, and stress, as well as having
some less structured social time. Flynn and Storandt (1990) compared three groups of
older adults, ranging in age from 60 to 82. The first group (N = 18) read a
self-instructional manual explaining memory techniques and containing exercises related to
these techniques. The second group (N = 21) also used the manuals, but in addition
attended four supplementary discussion groups. These groups were designed to allow the
participants an opportunity to discuss concerns related to aging and memory decline, and
possible coping strategies. A third control group (N = 19) who did not receive
instruction, was also part of the study.

Results at posttest indicated that individuals in the discussion groups performed

significantly better than the control group on most of the five memory tests. In addition,

16

they significantly outperformed the manual-only group on two of the tasks. The
manual-only group was better than the control group on only one task, indicating that the
role of discussion groups served an important purpose in teaching the memory techniques.
Dellefield and McDougall (1996) used a sample of community dwelling older

adults (N = 145) with an average age of 71 to test an intervention designed to increase
self-efficacy regarding memory skills. They found that after the four-session treatment [
intervention participants significantly increased both their memory self-efficacy as well as ‘
objective memory performance. By contrast, the no-treatment control group experienced

a decline in memory self-efficacy. Participants who were depressed, as measured by the

 

Geriatric Depression Scale-Short Form, indicated lower levels of memory self-efficacy, yet
performed equally well as the nondepressed subjects. Depressed subjects showed a
decrease in their memory self-eflicacy between the posttest and 2-week follow-up, while
nondepressed subjects showed no change. The authors hypothesize that depressed
subjects experienced deficits in the areas of attention, initiation, effort, and efficiency,
which were alleviated during the intervention, thus leading to improved performance.
However, possibly these deficits returned when the intervention was over, and therefore
the depressed subjects could not maintain their previous gains.

Research has also been done examining attention training specifically. Willis,
Cornelius, Blow, and Baltes (1983) performed an experiment designed to investigate four
dimensions of attention processes: discrimination, selective attention, attention switching,
and concentration-vigilance. Their sample consisted of 73 older adults with an average
age of 70.5 who were divided into three groups: attention training, social contact, and

no-contact control. The attention training group received five one-hour sessions in which

17

they learned techniques to improve performance on attention tasks. The social contact
group had an equal amount of time in which to discuss issues related to fiiendship in old
age but did not receive attention training; the control group merely completed pre- and
post-test measures.

The authors of this study were interested in subjects’ improvements on attention
tasks, as well as transfer of these skills to other domains, such as fluid-crystallized
intelligence, perceptual speed, and memory span. They found that the group that received
attention training performed significantly better at post-test and follow-up than at pre-test;
this group also performed better than the social-contact and control groups. However,
their hypothesis that attentional training effects would transfer to other domains was not
supported. This study indicates, therefore, that the benefits of attention training would

appear to be limited to the specific tasks for which it is given.

Aims of Current Study

The aim of this study is to investigate the relationship between attention and
depressive symptoms in older adults, as well as the role of gender in this relationship.
Symptoms of depression related to motivation are expected to be closely related to deficits
in cognitive fimctioning, as assessed by measures of attention. This motivational factor
includes such symptoms as difficulty in making decisions, feeling low in energy, difficulty
concentrating, and somatic preoccupation. Mood-related symptoms are also expected to
be related to cognitive problems; however, these types of symptoms are not expected to
have as strong a relationship with cognitive deficits as will motivational symptoms. This

mood factor includes feelings of sadness, crying, irritability, and withdrawal from others.

18

 

Hypotheses

1. Higher levels of depressive symptomatology will be associated with poorer
performance on tasks requiring a large attentional component. Scores on the Beck
Depression Inventory (BDI) and Geriatric Depression Scale (GDS) will correlate
positively with scores on Trails B, and negatively with scores on Digit Span and the Paced
Auditory Serial Addition Task (PASAT).

2. Motivational symptoms of depression will predict performance on attentional tasks
to a greater extent than will mood symptoms. The factors comprised of items related to
motivation on the BDI and GDS will account for more of the total score variance than
items related to mood, on each of the three attention measures.

3. There will be a gender difference in the types of depressive symptoms reported;
specifically, women will endorse more mood-related symptoms of depression than men,
and men will endorse more motivational symptoms of depression than women.

4. Participation in workshops designed to teach attention and memory strategies will
improve performance on tasks requiring a large attentional component, and lower
participants’ scores on measures of depression. Specifically, scores on the BDI and GDS
are expected to decrease from pretest to posttest. Scores on Trails B are expected to
decrease (indicating improved performance), and scores on Digit Span and the PASAT are

expected to increase (also indicating improved performance).

19

METHOD

Participants

Fifty adults over the age of 55 were recruited from the East Lansing area and its
surrounding communities, as part of the Michigan State University (MSU) Psychological
Clinic Aging Research Project. F lyers were posted in prominent areas throughout the
community, which provided basic information about the mood and memory workshops.
In addition, former employees of Michigan State University were contacted by mail; they
received letters inviting their participation in the workshops and explaining the testing
procedures. This sample included 23 males and 27 females, ranging from 56 to 84 years of
age (M = 71.06; SD = 6.94). The group had a mean education of 16.24 years

(SD = 2.92), and all participants had completed at least 12 years of school.

Pr re

Participants were assessed prior to their enrollment in the attention and memory
workshops, and again after the completion of the workshops. Each assessment lasted for
approximately 2 1/2 hours, and consisted of a variety of tests designed to measure severity
of depression, as well as memory and attentional functioning. The dimensions of attention
assessed in this study were selective attention, attention switching, and
concentration-vigilance; measured by the Paced Auditory Serial Addition Task, Trails B,
and Digit Span Subtest, respectively. Depression was assessed through scores on the
Beck Depression Inventory and the Geriatric Depression Scale. The measures employed
in this study were a subset of those administered during the full assessment. Any

individuals showing more than mild dementia as indicated by scores lower than 24 on the

20

Mini-Mental State Examination (Folstein, F olstein, & McHugh, 1975) were excluded from
the sample.

After completing the pretesting, participants attended a series of workshops with
six to ten other people. Each workshop was approximately one hour and fifteen minutes
in length; each group of participants attended a total of seven sessions over a period of
three-and-one-half weeks. Participants were allowed to continue in the groups, but their
data was not used, if they missed more than two sessions. These workshops were
designed to teach strategies to improve memory as well to discuss the ways that mood
relates to these factors (Lewinsohn, Antonuccio, Breckenridge, & Teri, 1984).

Two types of workshops were conducted; one focusing more on development of
attention skills, and the other teaching strategies for relaxation. However, both types
cover similar material, and both include information about attention and its relationship to
memory. For example, clinicians discuss handouts which deal with stress and attention,
myths of aging, the relationship between mood and memory, and assertiveness. Memory
techniques covered include the method of loci, how to remember names and faces through
creative associations, and the PQRST method for text recall. There were equal numbers of
participants in each type of workshop (n = 25). The manual for this workshop was
designed by the Andrus Foundation, and previous work has shown it to be effective in
teaching these skills. After the last workshop, participants were tested again; the posttest

measures were identical to those used for the pretest.

21

Measures of Depression
Beck Depression Inventory (BDI; Beck, Ward, Mendelson, Mock, & Erbaugh, 1961;
Beck, Rush, Shaw, & Emery, 1979)

The BDI is a self-report measure, developed to assess the intensity of depression.
It consists of 21 items which are rated on a 4-point scale. More than one response to each
question is allowed; however, the response indicating the highest level of depressive
symptoms is scored. These items assess symptoms and attitudes, including guilt, feelings
of failure, mood, motivation, and somatic complaints. Reliability figures for the BDI are
usually over .90, although it must be noted that most reliability studies have been done on
psychiatric inpatients rather than the general population. It has been found to correlate
moderately highly with psychiatric ratings of depression, in the range of 0.65 to 0.77
(Stehouwer, 1987). The BDI also has correlations of 0.50 to 0.80 with other depression
scales, and has a high test-retest stability (Steer, Beck, & Garrison, 1986).

Originally published in 1961, and revised in 1979, the BDI had been used in over
1,000 research studies by 1988 (Beck, Steer, & Garbin, 1988) and continues to be widely
used for research purposes, particularly as a method of screening for depression. One
study by Olin, Schneider, Eaton, Zemansky, and Pollock (1992) found the BDI to be an
efiicient instrument for screening depression when the cutoff score was set at 10, as
recommended by Beck et al. (1979). They found a sensitivity of 100% and specificity of
96%. While other studies have found lower values, the subjects used in these studies are
medical and psychiatric inpatients, who are more likely to have health problems than
outpatients and control subjects. The authors conclude that the law specificity and

sensitivity found in these studies are due to the BDI’s assessment of somatic complaints.

22

Thus, the BDI is an appropriate measure for use with subjects who are not ill or whose

difficulties are less severe.

Geriatric Depression sge (GDS; Yesavage, Brink, Rose, Lum, Huang, Adey, & Leirer,
1983)

This depression scale is also a self-report instrument, which was specifically
designed for use with older populations. It consists of 30 items, to which the individual
responds yes or no. This scale taps symptoms of mood disturbance, as well as problems
with social behavior and cognitive performance; thus, it will be useful in assessing
motivational symptoms of depression.

The GDS was developed to address three major areas of concern when assessing
depression in older adults (Yesavage et al., 1983). The first is that the somatic symptoms
which are frequently found on many depression inventories may not be as appropriate as
indicators of depression in older adults. Therefore, the GDS does not assess somatic
complaints. Second, questions that are designed to be answered by young people may not
be interpreted the same way by older adults. The authors attempted to delete questions
that may make older people uncomfortable or unaccepting of the measure itself. Finally,
the need for an easily understood measure led the authors to develop the yes/no format of
the instrument. This was done to eliminate confusion stemming fi'om questions which may
elicit multiple responses.

Yesavage et al. (1983) report a one-week test-retest reliability of 0.85, and internal
consistency of 0.94. The mean inter-item correlation was 0.36, and the mean item-total

correlation was 0.56. In addition, the GDS shows good convergent validity; correlations

23

with other measures of depression were in the range of 0.80 to 0.84. This measure has
also been found to have good reliability and sensitivity for identifying depressed
individuals. As used in this study, the sensitivity of a measure refers to how many
individuals with depression are classified correctly as meeting criteria for the diagnosis,
while the specificity of a measure refers to its ability to correctly identify which people do
not meet criteria for the diagnosis. A 1981 study by Brink et al. found a sensitivity of
84% and a specificity of 95% when a cutoff score of 11 was used; when the cutoff score
was raised to 14, the figures were 80% and 100%, respectively. Salamero and Marcos
(1992) found a high internal consistency of the GDS, with an alpha of 0.87, and a
correlation of 0.7 7 with the Hamilton Rating Scale for Depression, indicating moderately
high convergent validity.

A factor analysis of the GDS (Salamero & Marcos, 1992) indicated a high degree
of homogeneity among the items, making it difiicult to obtain factors which clearly difl‘er
from one another. The first factor to emerge was interpreted by the authors as a general
factor, consisting of variable types of items. The second factor consisted of problems with
attention, concentration, decision-making, and memory, while the third factor is mainly
composed of items regarding social withdrawal and decreased activity level. The
remaining items are varied and create several dimensions which do not have a discernible
theme. Thus, the GDS is best viewed as a unifactorial scale, but one which can produce

three dimensions with items loading on these factors to a moderate degree.

24

 

W of Attention
Pam Auditopy Serial Addition Task (PASAT; Gronwall, 1977).

The PASAT is a measure of sustained attention and rate of information processing
which consists of four tape-recorded 50-digit trials. This task requires the individual to
listen to a series of digits and add each set of two successive digits. Thus, after producing
a response, the individual must immediately forget this number and focus on attending to I
the next digit presented. The interval between the presentation of each number decreases
with each trial (2.4, 2.0, 1.6, and 1.2 seconds), thus increasing the processing demands

placed upon the subject. O’Donnell, MacGregor, Dabrowski, Oestreicher, and Romero

 

(1994) found that this task requires focused mental processing speed as well as focused
attention; thus it is an appropriate measure of attentional performance. In addition, it is a
sensitive indicator of performance which can detect subtle differences in functioning
between individuals that may not be as easily assessed through other attentional measures.
Egan (1988) found the split-half reliability of the PASAT to be 0.96, indicating a
very high internal consistency. In addition, this study found the PASAT to have a
moderate correlation with general intelligence and numerical ability (0.68). Gronwall
(1977) reports significant practice effects from the first administration to the second;
however, this has only been shown when the test was given at one-week intervals. Adult

norms have been established for this measure (D’Elia, Boone, & Mitrushina, 1995).

Trail M' g Test; Pm B (TMT-B; Reitan & Wolfson, 1985).
The Trail Making Test is part of the Halstead-Reitan Battery (Reitan & Wolfson,

1985), and consists of two different tasks. In Part A of the Trail Making Test, the task is

25

to draw a line connecting numbers in sequence, while in Part B the task is to connect
numbers and letters. The individual must alternate between the two, keeping both in their
correct sequence (numerically and alphabetically). The score for this test consists of the
total time taken to successfully connect all of the numbers and letters, including any time
spent correcting errors.

This test assesses mental flexibility and motor function, and Part B is more useful n.
than Part A for measuring information processing (Spreen & Strauss, 1991). In addition,
as with the PASAT, this test has a strong attentional component, particularly in relation to

visual scanning (O’Donnell, MacGregor, Dabrowski, Oestreicher, & Romero, 1994).

 

Lezak (1983) reported a reliability coeflicient of 0.67 for this measure, while Goldstein
and Watson (1989) found test-retest reliability to be in the range of 0.66 to 0.86.
Normative data is available for adults up to the age of 79 (D’Elia, Boone, & Mitrushina,

1995)..

 

Wechsler Adult Intelligence Scale-Revised, Digit Span Spbtest (Digit Span; Wechsler,

1981)

This subtest of the WAIS-R is the most commonly used clinical test of attention
(LaRue, 1992). The Digit Span task consists of listening to and then repeating back of a
string of numbers. Two sets of trials are completed; on the first the numbers are repeated
forward and on the second, they are repeated backward. The main advantages of this test
are that it has a small number of trials and is therefore brief; in addition, there are age
norms available for adults up to age 74. A study by Larrabee and Curtiss (1995) found

evidence supporting the use of this subtest to measure attention and information

26

processing. Kaufman (1990) also describes this test as useful because of its sensitivity to
inattention, distractibility, and lack of concentration. Test-retest reliability of Digit Span is

0.83 (Kaufman, 1990).

RESULTS

The first three hypotheses were tested using data from Time 1 only.

Hypothesis 1: The first hypothesis predicted that higher levels of depressive
symptomatology would be associated with poorer performance on tasks requiring a large
attentional component. We expected that scores on the Beck Depression Inventory (BDI)
and Geriatric Depression Scale (GDS) would correlate positively with scores on Trails B,
and negatively with scores on Digit Span and the Paced Auditory Serial Addition Task
(PASAT). Recall that higher scores on Trails B indicate poorer performance, while high
scores on Digit Span and the PASAT indicate better performance on these tests. This
hypothesis was partially supported.

Pearson _r_ correlations were computed between total scores on each of the
depression measures (Beck Depression Inventory and Geriatric Depression Scale) and
scores on each of the attention measures (Trails B, Digit Span, and PASAT). Significant
correlations were found in the predicted direction between scores on Trails B and the BDI
(g = .434, p < .01) as well as between Trails B and the GDS (; = .358, p < .05). All other
correlations between depression and attention measures were nonsignificant, although in

all cases the correlations were in the predicted direction.

Hypothesis 2: The second hypothesis predicted that motivational symptoms of depression

would predict performance on attentional tasks to a greater extent than would mood

27

symptoms. We expected that the factors comprised of items related to motivation on the
BDI and GDS would account for more of the total score variance than would items
related to mood on each of the three attention measures. This hypothesis was partially
supported.

Those symptoms of depression which load onto the factors of mood and
motivation were selected based on previous factor analytic studies of the BDI and GDS
(Shaikh et al., 1991; Startup, Rees, & Barkham, 1992; Tanaka & Huba, 1984). Those
items were used which corresponded to those comprising the mood and motivation factors
found by these researchers. Participants’ scores were Obtained by adding the items of each
factor separately (see Appendix A for lists of items). These scores were then entered into
a regression equation to see which factor (mood or motivation) accounted for more of the
variance in the scores on the attention measures, which were scored according to the
instructions provided in the testing manuals. Separate regressions were performed for
each of the three measures of attention, first with the BDI and then with the GDS.

Overall, total score on the BDI accounted for 18% of the variance on Trails B

(R2 = .18, p < .01). When entered into a regression model, the motivation factor

explained more of this variance (Beta = .38, p < .05) than did the mood factor (Beta = .08,
p < .59, ns). No other differences were found in the amount of variance accounted for by

the mood and motivation factors on the remaining attention measures.

Hymthesis 3: The third hypothesis predicted that there would be a gender difference in

the types of depressive symptoms reported; specifically that women would endorse more

28

mood-related symptoms of depression than would men, and men would endorse more
motivational symptoms than would women. This hypothesis was not supported.

Separate raw scores for both the BDI and GDS were obtained by adding the total
number of symptoms contained within the mood factor for each individual participant.
These scores were then summed across all female participants and divided by the number
of female participants to produce a mean score for number of symptoms. This same
procedure was followed for male participants. A t-test was conducted to compare the
mean number of mood symptoms reported by men and those reported by women. On the
BDI, the mean mood score for women was 3.63 (S_Q = 5.25), which was not significantly
different fiom the mean score of 3.52 (SD = 3.33) for men. On the GDS, the mean score
for women was 1.33 (SQ= 2.18), and the mean score for men was 1.00 (S_D= 1.51); this
difi‘erence was also not significant.

Next, separate raw scores for both the BDI and GDS were obtained by adding the
total number of symptoms contained within the motivation factor for each individual
participant. Group means for each gender were derived using the method described above.
A t-test was conducted to compare the number of motivation symptoms reported by men
and those reported by women. On the BDI, the mean motivation score for women was
4.15 (SD = 3.37), and the mean score for men was 3.83 (_S_Q = 2.08). A t-test to compare
these means revealed that this difference was not statistically significant. On the GD S, the
mean score for women was 2.93 (S2 = 1.94), which was not significantly different from

the mean score of 3.00 (SJ; = 1.78) for men.

29

Hypothesis 4: The fourth hypothesis predicted that participation in workshops designed
to teach attention and memory strategies would improve performance on tasks requiring a
large attentional component and lower participants’ scores on measures of depression.
Specifically, scores on the BDI and GDS were expected to decrease from pretest to
posttest. Scores on Trails B were expected to decrease (indicating improved
performance), and scores on Digit Span and the PASAT were expected to increase (also A
indicating improved performance). This hypothesis was partially supported.

A multivariate analysis of variance (MANOVA) was first performed to determine

the efi’ects of group membership, if any, on treatment effects. In this analysis, the

 

dependent variables were the attention and depression scores, and the within-subjects
independent variable was time (pretest vs. posttest). The between-subjects independent
variable was group membership. Results of this analysis showed no significant efi’ects of
group membership on posttest scores for any measures. Thus, the two groups were
combined for the remainder of the analysis.

A paired-differences t-test was performed for each of the five measures to
investigate change in scores fi'om pretest to postest. Mean scores on the PASAT were
97.48 (SD: 51.52) at pretest and 121.273 (_S_Q = 39.69) at posttest. Thus, a significant
change was observed in the predicted direction for scores on this measure (t(49) = -4.024,
p < .001). Scores on Digit Span tended to improve from pretest (M = 14.82, _S_I_)_ = 4.2) to
posttest (M = 15.45, S1; = 3.71); however, this change approached, but did not meet,
statistical significance (t(49) = -2.00, p < .051). No other significant differences between

pre- and posttest were found on the GDS, BDI, or Trails B.-

30

DISCUSSION

Depressive Symptoms and Performance on Attention Tests

Higher levels of depressive symptomatology were expected to be associated with
poorer performance on tasks requiring a large attentional component. This hypothesis was
partially supported. The three measures of attention used in this study were each designed
to test a particular aspect of the construct of attention. Performance on Trails B only was
significantly correlated with level of depression. Trails B is primarily designed to measure
mental flexibility, visual scanning, and attention-switching ability, but is a good indicator
of overall brain functioning (Reitan & Wolfson, 1985). Thus, low levels of depression
may be linked to more global brain impairment rather than to specific attention tasks.
Also, even at low levels of depression, it may be difficult for individuals to shift fiom one
mental “set” to another. The Digit Span subtest is useful for measuring difficulty
concentrating, and distractability; however, it is possible that an individual could perform
adequately on a task which does not require him or her to shift attention, yet show
decreased ability when called upon to demonstrate mental flexibility. Originally designed
to track recovery from patients who had sufi’ered concussions, the PASAT is a measure of
sustained, focused attention. It has been shown to be a sensitive measure of attentional
abilities which can detect subtle changes between individual performance (O’Donnell,
MacGregor, Dabrowski, Oestreicher, & Romero, 1994). It is somewhat surprising that
this measure was not significantly correlated with levels of depression. It is possible that
this finding is due in part to the modality in which attention was assessed, as the

Trailmaking Test requires visual scanning while the PASAT is presented aurally.

31

While the expected correlation between depressive symptoms and all three
attention measures was not found, these results are important because of the information
they provide regarding different types of attentional abilities. This study allows a more
detailed look at the concept of attention than is often reported. It suggests that some
types of attentional performance, such as tasks requiring attention-snatching and visual
scanning, may be associated with depression, while others, such as less complex tasks
requiring sustained attention, may not be. Thus, an individual who is experiencing
depression may find himself or herself quite capable of a task involving rote memorization,
while having noticeable difficulty attempting to engage in more than one activity at a time.
However, if this person is aware of the types of cognitive deficits that are commonly
associated with depression, he or she may be able to focus instead on tasks requiring a less

complex attentional component and perhaps lessen feelings of fiustration and distress.

M M m A n ' Tr ' '

It was expected that after participating in workshops designed to teach memory
and attention strategies, subjects would demonstrate improved performance on tasks
requiring a large attentional component. In addition, participants’ scores on measures of
depression were expected to decrease after completion of the workshOps. This hypothesis
received partial support. Of the three measures of attention, subjects showed significantly
improved performance only on the PASAT. However, this improvement was very large.
The reasons for dramatic improvement on this measure to the exclusion of others are

unclear. Practice efi‘ects with the PASAT are notable; however, these effects have not

32

been reported for intervals longer than one week between pre- and posttest (Gronwall,
1977), and in our study posttesting occurred eight weeks after initial testing. .

Some study participants have stated that they found the PASAT to be the most
difficult of all the attention tasks. Perhaps the PASAT is a somewhat more intimidating
task which participants approach with more confidence the second time. While possibly
attributable to practice efl‘ects, this is a positive sign if we relate this to tasks found in
everyday life. That is, attention training may help these individuals to approach difficult
tasks with the belief that their newly acquired knowledge will improve their ability to carry
out these tasks.

The expected decline in depressive symptoms was not found on either the BDI or
the GDS. It is likely that since levels of depression were low to begin with, it was dificult

to detect a further decline in symptoms.

M vs Motivati nal S toms of De ression

It was expected that motivational symptoms of depression would predict
performance on attentional tasks to a greater extent than would mood symptoms. This
hypothesis received partial support, as this was found to be true only for the relationship
between the BDI and the Trailmaking Test. That is, BDI scores accounted for 18% of the
variance in performance on Trails B, and symptoms related to motivation accounted for
significantly more of this variance than did symptoms related to mood. This could be due
in part to the inclusion of somatic symptoms on the BDI which are not found on the GDS.
Perhaps for older adults somatic symptoms of depression have a greater association with

attentional problems than do more cognitive types of motivational problems. It is unclear

33

why this pattern was not found for the other two measures of attention, Digit Span and
the PASAT. As discussed earlier, it is possible that the more global sensitivity of Trails B
allows this measure to demonstrate more subtle links between motivational symptoms and

problems with attention.

Capder and Self-Reported Smptoms of Depression

It was expected that there would be differences between men and women on the
types of symptoms reported. Specifically, women were expected to endorse more
mood-related symptoms, and men were expected to endorse more motivation-related
symptoms of depression. This hypothesis was not supported. Previous studies and
reviews (Fopma-Loy, 1988; Hale & Cochran, 1983; Komstein et al., 1995) have
suggested that females are more likely to experience difi‘erent clusters of depressive
symptoms than are men. Other studies (Oltman, Michals, & Steer, 1980) have found no
difi‘erences in the depressive syndrome experienced by men and women

This study found no gender difi’erences in the reporting of mood and motivational
clusters of symptoms. Therefore, data for males and females were combined for all other
analyses. It is possible that differences in reported symptoms decline with increasing age.
In addition, this study may be showing efi‘ects of a slight selection bias, in that the male
participants may have been more cognitively intact and less depressed than their female
counterparts. VVrth a smaller number of men than women over the age of 55, it is possible
that the men who do survive into old age function at a higher level than women in the

same age cohort.

34

Limitations of Study

The results of this study are limited in the extent to which they may generalize to
other populations. This is because the sample of older adults used in this study was of
above-average intelligence, well-educated, and generally physically healthy. In addition,
participants were self-selected, so it is likely that these individuals were motivated to
attend the groups and to learn the attention and memory strategies presented to them.
Thus, it is not possible to draw conclusions about the link between depressive symptoms
and attention in older adults who are in poor health, less educated, and less motivated to
improve their attentional abilities.

This study also lacked a control group, which would be necessary in order to
determine whether the effects of attention training were due to participation in the groups
rather than some other factor, or combination of factors. Verhaegen, Marcoen, and
Goossens (1992) adressed this issue in their meta-analysis of the efi‘ects of mnemonic
training. These authors point out that the main emphasis in these types of programs is on
plasticity, or intraindividual change rather than on interindividual comparisons. In
addition, 70% of their representative sample of research findings does not include
no-treatment control groups; thus, it is common in the empirical literature to find studies
which compare only pre- and post-treatment data rather than using subjects who did not
undergo the memory training.

As noted earlier, participants in this study were high-fimctioning,
community-dwelling older adults who were self-referred into the study. Thus, it is not
surprising that their levels of depression were low overall at pre-test. With a more

depressed sample, changes over time may have been more dramatic, in both a reduction in

35

depressive symptoms, as well as improved attentional performance. Similarly, individuals
who performed less well initially on measures of attention may have improved more

noticeably over time than those in the current sample.

Implications

Results of this study suggest that memory and attention training groups can be
effective with older adults. Additionally, even low levels of depression may be associated
with attention problems, and symptoms related to motivation may be more strongly
associated with attention problems than are symptoms related to mood. This indicates
that mental health professionals who are working with older clients should attempt to
identify whether there are specific clusters of depressive symptoms experienced by their
clients. prarticular symptoms, such as lack of motivation, are more closely linked to
attention and memory problems, this is potentially usefirl information when making a
differential diagnosis of dementia vs. depression. Recognition of “pseudodementia” can
lead to more approriate and efi’ective treatment for older clients who are experiencing

depression, and not a permanent decline in their cognitive functioning.

Directions for Future Reswch

Use of a control group in future studies would provide more information about
whether changes over time could be attributed to the memory and attention training only.
With the current design, it is not possible to conclude that group participation alone was

responsible for the significant improvements on the PASAT fi'om pre-to posttest.

36

The current sample was fairly homogeneous in the level of functioning of study
participants, as these individuals generally reported few symptoms of depression. It would
seem to be wothwhile for future studies to compare older adults experiencing varying
levels of depression, to see if the hypothesis would be supported for individuals meeting

criteria for diagnosable levels of depression.

37

APPENDIX A

38

APPENDIX A

ITEMS COMPRISING MOOD AND MOTIVATION FACTORS

Questions from the GDS - Mood F Eton

8.

6.

23.
l3.
16.
18.
10.
25.
22.

Are you afraid that something bad is going to happen to you?
Are you bothered by thoughts you can’t get out of your head?
Do you think that most people are better off than you are?

Do you fi'equently worry about the future?

Do you often feel downhearted and blue?

Do you worry a lot about the past?

DO you feel helpless?

DO you fiequently feel like crying?

Do you feel that your situation is hopeless?

Shestions fi'om the GDS - Motivation Factor:

 

 

29. Is it easy for you to make decisions?

20. Is it hard for you to get started on new projects?

21. Do you feel fiill of energy?

30. Is your mind as clear as it used to be?

26. Do you have trouble concentrating?

2. Have you dropped many of your activities and interests?

11mg frpm the BDI - Mood Factor: Items from the BDI - Motivation Factor
1. Sadness l3. Indecisiveness

2. Pessimism/Discouragement 15. Work retardation

3. Sense of failure 16. Insomnia

4. Dissatisfaction 17. Fatiguability

5. Guilt 20. Somatic preoccupation
6. Expectation of Punishment 21. Loss of libido

7. Self-disfike

8. Self-accusations

9. Suicidal ideation

10. Crying

11. Irritability

12. Withdrawal

14. Body image distortion

39

 

APPENDIX B

40

APPENDIX B

DEMOGRAPHIC INFORMATION FOR THE SAMPLE

Table 1. Education
Education ears Number of

12
l3
14
15

16
l7
18
19
20

 

Table 2. Marital Status

I Marital Status I Number of Participants 5%) !
N 1 (2)

 

 

 

 

 

 

ever Married
Married 29 (5 8)
Widowed 12 (24)
Divorced 8 (16)

 

41

 

Table 3. Occupation
T e of Occu ation

Number ofPartici ants %

 

 

 

 

 

 

 

Unskilled 1(2)
Semi-Skilled 1 (2)
Skilled 4(8)
Managerial 16 (32)
Professional 27 (54)
None 1 (2)

 

Table 4. Participants’ Self-Rams of Physical Health

 

 

 

 

Health Ratin Number of Partici ants %
Excellent 13 (26)
Good 29 Q8)
Fair 6 Q2)
Poor 2 (4)

 

 

 

42

 

 

LIST OF REFERENCES

43

LIST OF REFERENCES

Abraham, 1. L., Wofford, A. B., Lichtenberg, P. A, & Holroyd, S. (1994). Factor
structure of the Geriatric Depression Scale in a cohort Of depressed nursing home

residents. International Journal of Geriatric Psychiatgy, 9, 611-617.

American Psychiatric Association (1994). Diagnostic and statistical manual of
mental disorders (4th ed.). Washington, DC: Author.

Backman, L., Hill, R. D., & Forsell, Y. (1996). The influence of depressive
symptomatology on episodic memory functioning among clinically nondepressed older
adults. Journal of Abnormal Psychology, 105(1), 97-105.

Beck, A T., Rush, A. 1, Shaw, B. F ., & Emery, G. (1979). Cogm'tive therapy of
depression. New York: Guilford Press.

Beck, A. T., Steer, R. A., & Garbin, M. G. (1988). Psychometric properties of the
Beck Depression Inventory: Twenty-five years of evaluation. Clinical Psychology Review,
8, 77-100.

Beck, A T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An
inventory for measuring depression. Archives pf M thidtg, 4, 561-571.

Blazer, D. (1993). Depression in lpte life (2nd ed.). St. Louis: Mosby-Year Book.

Brink, T. A, Yesavage, J. A, Lum, 0., Heersema, P., Adey, M., & Rose, T. L.
(1981). Screening tests for geriatric depression. Wm 37-44.

Bulbena, A, & Berrios, G. E. (1993). Cognitive function in the afi‘ective disorders:
A prospective study. Pflphppatholdgy, 2d 6-12.

Burt, D. B., Zembar, M. J ., & Niederehe, G. (1995). Depression and memory

impairment: A meta-analysis of the association, its pattern, and specificity. Psychplogical
Bullgg’ ma), 285-305.

Cowan, N. (1995). Attention and memogz: An intemted fi'amework. New York:
Oxford University Press.

D’Elia, L. R, Boone, K. B., &.Mitrushina, A M. (1995). Hgdbppk of normative
f r n o chol 'cal smen . New York: Oxford University Press.

Dellefield, K. S., & McDougall, G. J. (1996). Increasing metarnemory in older
adults. nging Rmcd 45, 284-290.

44

 

Egan, V. (1988). PASAT: Observed correlations with 1Q. Personality and
Individupl Differences,_9_, 179-1 80.

Flynn, T. M., & Storandt, M. (1990). Supplemental group discussions in memory
training for older adults. Psychology and Aging, 5, 178-181.

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. qumfl of

Psychiatrip Resgrgh 12, 189-198.

Fopma-Loy, J. (1988). The prevalence and phenomenology of depression in
elderly women: A review of the literature. Archives of Psychiatric Nigsing, 2, 74-80.

Forsell, Y., Jorm, A. F ., Fratiglioni, L., Grut, M., & Winblad, B. (1993).
Application of DSM-III-R criteria for mq'or depressive episode to elderly subjects with
and without dementia. American Journal of Psychiatg, 150, 1199-1202.

Forsell, Y., Jorm, A. F., & Winblad, B. (1994). Association of age, sex, cognitive
dysfunction, and disability with major depressive symptoms in an elderly sample.
American 1;;qu of Psychiatry, L§_l_(11), 1600-1604.

Friedhofl‘, A. J. (1991). Consensus development statement: Diagnosis and
treatment of depression in late life. In L. S. Schneider, C. F. Reynolds, B. D. Lebowitz, &

A J. Friedhofi‘ (Eds). Di osis trnent of r ion inl te lif : R
NIH Cppmsps Development Confereng; (pp. 493-511). Washington, D. C.: American
Psychiatric Press.

Goldstein, G., & Watson, J. R. (1989). Test-retest reliability of the
Halstead-Reitan Battery and the WAIS in a neuropsychiatric population Clinidal

Neurdpsyphdlggist, 3, 265-272.

Gronwall, D. M. A. (1977). Paced Auditory Serial-Additions Task: A measure of
recovery from concussion. Percep_tda_l and Mqtpr Skills, 44, 367-373.

Hale, W., & Cochran, C. D. (1983). Sex difi‘erences in patterns of self-reported
psychopathology in the manied elderly. mm of Clinical Psychology, 38, 253-257.

Hartley, A A. (1992). Attention. In F. I. M. Craik & T. A Salthouse (Eds) IE
handpopk of agrp' g apd cognition (pp. 3-49). Hillsdale, NJ: Lawrence Erlbaum.

Hulicka, I. M., & Grossman, J. L. (1967). Age group comparisons for the use of
mediators in paired-associate learning. 0 of rontolo 22 46-51.

James, W. (1890). Principles of psychology, (pp. 403-404). New York: Holt.

45

Kaufinan, A. S. (1990). Assessing adolescent and adult intelligence (pp. 66;
394-396). Boston: Allyn and Bacon.

Kessler, R. C., McGonagle, K. A, Zhao, et al. (1994). Lifetime and 12-month
prevalence of DSM-III-R psychiatric disorders in the United States: Results from the

National Comorbidity Survey. Arphivps pf (finerat Psychiatg, 51, 8-19.

Komstein, S. G., Schatzberg, A F ., Yonkers, K. A, Thase, M. E., Keitner, G. 1.,
Ryan, C. E., & Schlager, D. (1995). Gender difi‘erences in presentation of chronic major

depression. Pychppharmacolpgy Buflgpt' 31, 711-718.

Larrabee, G. J., & Curtiss, G. (1995). Construct validity of various verbal and

visual memory tests. oumal of Clini d E ' tal Neur cholo 11(4),
536-547.

LaRue, A. (1992). Agpt' g and neurppsycholpgical assessment New York: Plenum

Press.

LaRue, A, Swan, G. E., & Carmelli, D. (1995). Cognition and depression in a
cohort of aging men: Results from the Western Collaborative Group Study. Psychology

MEG), 30-33.

Lewinsohn, P. M., Antonuccio, D. O., Breckenridge, J. S., & Teri, L. (1984). I_h_e_
mping with dgpression coutsp. Eugene, OR Castalia.

Lezak, M. (1995). Neuropsycholpgig gsessment (3rd ed.) New York: Oxford

University Press.

Lichtenberg, P. A, Ross, T., Millis, S. R., & Manning, C. A. (1995). The
relationship between depression and cognition in older adults: A cross-validation study.
f Geront lo ' Ps h l ' 'en 50B P25-P32.

McNeil, J. K., & Harsany, M. (1989). An age difference view of depression.
Cmdian Psychology, 30, 608-615.

Musetti, L., Perugi, G., Soriani, A, Rossi, V. M., Cassano, G. B., & Akiskal, H.
S. (1989). Depression before and after age 65: A reexamination. British meal pf
Pyphiatpy, 155, 330-336.

Newmann, J. P., Engel, R. J ., & Jensen, J. E. (1991). Age differences in depressive
symptom experiences. Joumgt pf (figutplpgy: chhplogical Sciences, 46, 224-235.

Nolen-Hoeksema, S. (1990). Sex differengs in depression. Stanford, CA:
Stanford University Press.

46

Nolen-Hoeksema, S., & Girgus, J. S. (1994). The emergence of gender differences
in depression during adolescence. Psychological Bulletip, 115, 424-443.

O’Donnell, J. P., MacGregor, L. A, Dabrowski, J. J ., Oestreicher, J. M., &
Romero, J. J. (1994). Construct validity of neuropsychological tests of conceptual and
attentional abilities. Journal of Clinical Psychology, 50, 596-600.

Olin, J. T., Schneider, L. S. ,Eaton, E. M ,,Zemansky M. F. ,& Pollock, V. E.
(1992). The Geriatric Depression Scale and the Beck Depression Inventory as screening
instruments in an older adult population. Psychological Assessment, 4, 190-192

Oltman, A M., Michals, T., & Steer, R. A. (1980). Structure of depression in
older men and women. Journal of Clinical Psychology, 36, 672-674.

Poon, L. W. (1984). Memory training for older adults. In J. P. Abrahams & V.
Crooks (Eds). @riatric Mental Health (pp. 135-151). Orlando, FL: Grune & Stratton.

Rabbitt, P., Donlan, 0, Watson, P., McInnes, L., & Bent, N. (1995). Unique and
interactive effects of depression, age, socioeconomic advantage, and gender on cognitive
performance of normal healthy older people. Psychology and Aging, 10, 307-313.

Reitan, R. M., & Wolfson, D. (1985). Th H -Reit Neuro ch lo ’cal
Tat Batteg. Tempe, AZ: Neuropsychology Press.

Salamero, M., & Marcos, T. (1992). Factor study of the Geriatric Depression
Scale. Acta Pychiatrica Scandinavipa, 86, 283-286.

Salzman, C. & Gutfreund, M. J. (1986). Clinical techniques and research strategies

for studying depression and memory. In L. W. Poon (Ed), Clinig memogt assessment pf
plder adults (pp. 257-267). Washington, D. C.: American Psychological Association.

Sheikh, J. 1., Yesavage, J. A, Brooks, J. O., Friedman, L., Gratzinger, P., Hill, R.
D., Zadeik, A, & Crook, T. (1991). Proposed factor structure of the Geriatric Depression
Scale. mmfional Psychoggriatrics, 3, 23-28.

Smith, A. D. (1980). Age difl'erences in encoding, storage, and retrieval. In L. W.

Poon, J. L. Fozard, L. S. Cermak et al. (Eds), New dirfi'og in memog and agpt’ g:

PrMings pf the @rge A. Talland mmrial gpnference. Hillsdale, NJ: Lawrence
Erlbaum Associates.

Spreen, O., & Strauss, E. (1991). A compendium pf ngopsycholpgical tests:
W New York: Oxford University Press.

Stankov, L. (1988). Aging, attention, and intelligence. Puchology and Agjpg, 3,
59-74.

47

Startup, M., Rees, A, & Barkham, M. (1992). Components of major depression
examined via the Beck Depression Inventory. Joumpl~ of Affective Disorders, 26, 251-260.

Steer, R. A, Beck, A. T., & Garrison, B. (1986). Applications of the Beck
Depression Inventory. In N. Sartorius & T. Ban (Eds), Assessment of depression (pp.
1 23-142). Berlin: Springer-Verlag.

Stehouwer, R. S. (1987). Beck Depression Inventory. In D. J. Keyser & R C.
Sweetland (Eds). Test critiques compendium. Kansas City: Test Corporation of America.

Sweeney, J. A, Wetzler, S., Stokes, P., & Kocsis, J. (1989). Cognitive firnctioning
in depression. toms! of Clinical Psychology, §_5_(6), 836-842.

Tanaka, J. S., & Huba, G. J. (1984). Confirrnatory hierarchical factor analyses of

psychological distress measures. Journal of Personalig and Social chhology, 46,
621-635.

Tarbuck, A. F., & Paykel, E. S. (1995). Effects of major depression on the
cognitive firnction of younger and older subjects. Psychological Medicine, 2;, 285-296.

VanZomeren, A. H., & Brouwer, W. H. (1994). Clinical neuropgchology of
attention. New York: Oxford University Press.

Verhaeghen, P., Marcoen, A, & Goossens, L. (1992). Improving memory
performance in the aged through mnemonic training: A meta-analytic study. Pychelogy

Ed Ag 3, 7, 242-251.

Wallace, J ., & Pfohl, B. (1995). Age-related differences in the symptomatic
expression of major depression. Jetpn_a1_ of Nervous and Mental Disease, 183, 99-102.

Wechsler, D. (1981). Weehpler Mult Intelligence Scale-Revised. New York:
Psychological Corporation.

Willis, S. L., Cornelius, S. W., Blow, F. C., & Baltes, P. B. (1983). Training
research in aging: Attentional processes. qumal of Educational Psyehelegy, 75, 257-270.

Woods, J. H., & Bernard, R. M. (1987). Improving older adults’ retention of text:
A test of an instructional strategy. Edueetienal Gerontolegy, 13, 107-120.

Yesavage, J. A, Brink, T. L., Rose, T. L., Lum, 0., Huang, V., Adey, M., &
Leirer, V. O. (1983). Development and validation of a geriatric depression screening scale:

A preliminary report. Jewel. pf Pachietdc Kemp, 17, 37-49.

48

      

"IIIIIIIIIIIIIIIIII III III III ..

7662473