|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
AGINTG, EMOTIONAL MEMORY AND THE HIPPOCAMPUS
ANN ELIZABETH MIKOS
A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE
UNIVERSITY OF FLORIDA
Ann E. Mikos
This research was supported by a grant from the McKnight Brain Institute. I am
grateful to my mentor, Dr. Dawn Bowers, for her constant availability and guidance on
this project. I would like to extend my appreciation to members of the Cognitive
Neuroscience Laboratory and others within the department of Clinical and Health
Psychology, including Dr. Catherine Price and Dr. Mabel Lopez, for their assistance in
running this project. Additionally, I would like to acknowledge Dr. Michael Marsiske of
the Department of Clinical and Health Psychology for his consultation on statistical
analyses. Finally, I am ever grateful to my parents for their encouragement and support.
TABLE OF CONTENTS
ACKNOWLEDGMENT S ............ ....._._. ............... iii...
LIST OF TABLES ........._.._.. ...._... ..............vi....
LIST OF FIGURES .............. ....................vii
AB S TRAC T ......_ ................. ..........._..._ viii..
1 INTRODUCTION ................. ...............1.......... ......
Mild Cognitive Impairment ................. ...............2.................
Characterization ................. ...............2............ ....
Neuropathology .......................... .. .. .. .................3
Issues in the Classification of Mild Cognitive Impairment ........._...... .........._....4
Emotional M emory ................. .. ..... .. ... ... ... ..... ..... ..........5
The Modulation Hypothesis: Involvement of the Amygdala ............... ... ............5
Evidence Supporting the Modulation Hypothesis in Humans .............. ................7
Emotional Memory in Older Adults ...._ ......_____ .......___ ............
Emotional Memory in Alzheimer's Disease ........._.._.. ......._ ................10
2 STATEMENT OF THE PROBLEM ................. ...............13........... ...
Specific Aim I............... ...............15...
Specific Aim II .............. ...............16....
Specific Aim III ................. ...............16................
3 M ETHODS ................. ...............18.......... .....
Subj ect Recruitment and Screening ................. ...............18........... ...
Neuropsychological/Mood Testing ............... ...............18....
Intellectual Assessment and Dementia Screening ................. ............ .........19
M em ory .............. ...............19....
Language .............. ........ .. .. ... .......2
Attention, Psychomotor Speed, and Executive Function .................. ...............21
S elf-Report M measures .............. ...............22....
Consensus Conference............... ...............2
M RI ............ _... ......_ _..... .............2
Emotional Memory Task ............ ..... .__ ...............24...
Study Phase .............. ...............24....
Test Phases .............. ...............25....
Picture Stimuli .............. .......... ...............2
Skin Conductance Response (SCR)............... ...............26.
Statistical Analyses ................. ...............27.................
4 RE SULT S .............. ...............3 0....
Neuropsychological and Mood Performance: MCI versus Control Groups ..............30
Aim 1: Comparison of Overall Rate of Forgetting ................. ............. ...............3 1
Aim 2: Effects of Emotional Arousal on Recognition Memory Performance ...........32
Emotional Memory Performance: MCI versus Control Group ................... ........32
Arousal Ratings .............. ....... ..............3
Skin Conductance Response (SCR) ................. ... ...............33
Emotional Memory Performance: Spectrum of Aging ................. .................. ..34
Aim 3: Relationship Between Hippocampus Volumes and Recognition Memory
Perform ance .............. .. .. ....... .......... .. ..........3
Hippocampal Volume and Overall Rate of Forgetting. .................. ................ .34
Hippocampal Volume and Rate of Forgetting for Emotional Material ...............3 5
5 DI SCUS SSION .....___................. ........___.........4
Summary and Interpretation of the Findings ......_. ..........._. .............. .....42
Interpretation and Relationship to the Literature ....__. ................. .........._.._... 47
Limitations of the Present Study ........................_. ...............50.....
Directions for Future Research ................. ....__. ...._.._ ............5
LIST OF REFERENCES ....__. ................. ........_.._.........5
BIOGRAPHICAL SKETCH .............. ...............62....
LIST OF TABLES
4-1 Demographic characteristics and performance on neuropsychological tests.
Means (+standard deviation) are shown. ........._..._.._ .........._. ....... 36..._ ...
4-2 Regression coefficients (R2) for individual regression equations with memory
status as the predictor of recognition memory performance for the low, medium,
and high arousal conditions (N=25). .............. ...............37....
4-3 Correlations between hippocampal volume and recognition memory
performance for each arousal condition at each testing session (N=13). .................37
LIST OF FIGURES
3-1 Study design flowchart ................. ...............29................
4-1 Recognition performance at the 4 testing sessions for the MCI (N=11) and
control groups (N=13) ................. ...............38........... ...
4-2 Percent correct picture recognition for low, medium, and high arousal pictures
over the four testing sessions for the MCI (N=11) and control groups (N=13)......39
4-3 Mean arousal ratings (+95% confidence limits) for low, medium and high
arousal pictures by control subj ects (N=13) and MCI subj ects (N=-12). .................40
Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science
AGINTG, EMOTIONAL MEMORY AND THE HIPPOCAMPUS
Ann Elizabeth Mikos
Chair: Dawn Bowers
Major Department: Clinical and Health Psychology
Purpose: The purpose of the present study was to investigate the effects of
emotional arousal on memory in individuals with amnestic mild cognitive impairment
and healthy controls. Additionally, this study sought to examine the relationship between
hippocampal volume and memory for emotionally arousing material.
Background and hypotheses: Emotional arousal is known to enhance memory,
presumably due to the influence of the amygdala, an important emotional region of the
brain, on the hippocampus, a brain region involved in memory consolidation. The effects
of emotional arousal on memory have not yet been investigated in mild cognitive
impairment (MCI), a condition characterized by impaired memory in the context of
preserved general cognition and normal activities of daily living. We presented MCI
patients and normal controls with a set of emotional and neutral photographs, followed by
four recognition tests at increasing time delays. We hypothesized that a) patients with
MCI would demonstrate poorer performance than controls on the recognition tests, b)
controls, and to a lesser extent MCI patients, would have better recognition memory for
arousing than neutral pictures, and c) hippocampus volumes would be positively
correlated with memory for arousing but not neutral photographs. Due to concerns about
accurate group classification, we also examined the first two hypotheses using a
continuous indicator of memory status rather than the between-groups classification.
Methods: Twelve MCI patients and 13 age and education-matched healthy
controls were presented with a set of emotional and neutral photographs, followed by
four recognition tests 10 minutes, 1 hour, 2 weeks, and 3 months later. The percentage of
correctly recognized pictures was calculated for pictures classified in low, medium, and
high arousal categories.
Results: The control and MCI groups did not differ in their overall picture
recognition performance and neither group showed enhanced recognition performance
for emotionally arousing pictures. However, upon removal of the between-groups
classification, memory status was significantly associated with 2-week recognition
memory performance for high arousing pictures. Additionally, hippocampal volume was
positively correlated with recognition memory performance for arousing but not neutral
Conclusions: This study suggests that individuals with better memory status
benefit more from emotional arousal on memory than those with poorer memory status.
Additionally, the hippocampus appears to be involved in this enhanced consolidation of
emotional information. Therefore, the use of emotionally arousing stimuli may represent
a promising memory aid for older individuals, particularly those with good memory
status and larger hippocampal volumes.
Recently, there has been increasing interest in a subgroup of older individuals who
present with mild memory complaints but are not demented. Amnestic mild cognitive
impairment, a variant of this condition characterized by impaired memory in the context
of preserved general cognition and normal activities of daily living, may represent a
transitional state between normal aging and Alzheimer' s disease (Petersen et al., 2001).
Although a multitude of investigations have examined performance on tests of memory
and of other cognitive functions in amnestic MCI, there have been no studies of
emotional memory in MCI. Memory for emotional material is often enhanced in healthy
individuals and this effect has been attributed to the response of the amygdala, a small
almond-shaped structure within the brain, to emotionally arousing stimuli (McGaugh,
2004). However, it is not clear whether the enhancement of memory by emotion is
preserved in individuals with amnestic MCI. Determining whether emotional arousal
enhances memory for individuals with amnestic MCI will have tremendous theoretical
and practical importance. Regardless of the outcome, this investigation will help to
elucidate the influence of emotional content on memory in older adults with and without
memory impairments as well as to provide greater insight into the role of certain brain
structures in emotional memory. If emotional arousal is shown to enhance memory for
individuals who have amnestic MCI, it may be possible to design memory improvement
strategies based on this form of memory. Before presenting the specific aims and
hypotheses of this study, I will review the following aspects of mild cognitive
impairment: (1) characterization, (2) neuropathology, and (3) issues in the classification
of MCI. Additionally, I will review the following issues pertaining to emotional
memory: (1) the predominant current theory explaining emotional memory, (2) evidence
supporting this theory in human studies, (3) studies of emotional memory in normal older
adults, and (4) studies of emotional memory in Alzheimer' s disease.
Mild Cognitive Impairment
Mild cognitive impairment (MCI) may represent a transitional state between
normal aging and dementia (Petersen et al., 2001). However, not all cases of MCI
necessarily progress to dementia, and MCI may be caused by heterogeneous conditions.
For example, certain medications and other medical conditions can lead to reversible
MCI (Morris, 2005). Given the variability of the very early clinical signs and their
overlap with the normal changes of aging, various subtypes of MCI have been identified
(Petersen et al., 2001). The present study focuses on the subtype of amnestic MCI, which
is characterized by slightly abnormal memory compared to same-aged peers in the
context of in tact general cognitive abilities and activities of daily living (Petersen et al.,
2001; Petersen et al., 1999). The syndrome of amnestic MCI is of great theoretical
interest, and it may be relevant to the study of early detection of Alzheimer' s disease
(AD) because of the high rates at which people who present with the syndrome go on to
develop AD. Subj ects who present with amnestic MCI have an increased risk of
developing diagnosable AD at rates of 12 to 15% per year, in contrast to 1 to 2% per year
in age-matched normals (Petersen et al., 2001).
The medial temporal lobe limbic structures, particularly the hippocampus, play a
central role in memory formation. Indeed, the hippocampus is one of the earliest brain
regions affected by neurofibrillary pathology in Alzheimer' s disease (Arriagada,
Marzloff, & Hyman, 1992). Studies of normal older adults, individuals with mild
cognitive impairment and patients with AD have shown correlations between memory
performance and size of the hippocampus, with smaller hippocampal volumes associated
with worse memory scores (Golomb et al., 1993; Petersen et al., 2000; Soininen et al.,
1994). Additionally, hippocampal pathology has been observed in MCI. Reduced
hippocampal glucose metabolism and hippocampal atrophy have been reported in
individuals with MCI compared to normal controls (Convit et al., 1995; Convit et al.,
1997; Du et al., 2001; Mosconi et al., 2005). Furthermore, premorbid MRI-based volume
measurements of hippocampal atrophy in individuals with MCI have been shown to be
predictive of subsequent conversion to AD (de Leon et al., 1997; Jack et al., 1999; Visser,
Verhey, Hofman, Scheltens, & Jolles, 2002). Rate of change of hippocampal volume
measurements is also significantly associated with conversion from a stable control group
to MCI status and from MCI status to AD diagnosis over the course of one to 5 years
(Jack et al., 2004). Thus, the hippocampus, a structure crucially involved in memory
formation, undergoes significant atrophy and metabolic changes in MCI and these
changes may be associated with subsequent conversion to AD.
Like the hippocampus, the amygdala is also known to be atrophic in AD. The
amygdala is a limbic structure located anterior to the hippocampus, which appears to play
a critical role in the emotional enhancement of memory. MRI investigations have
demonstrated smaller amygdala volumes in Alzheimer' s patients compared to controls
(Callen, Black, Gao, Caldwell, & Szalai, 2001; Hensel et al., 2005; Horinek et al., 2006;
Ishii et al., 2006; Krasuski et al., 1998). Although few studies have examined the
amygdala in MCI, one such investigation found that the volume of the left amygdala was
significantly smaller in individuals with mild cognitive impairment than in normal
controls (Bottino et al., 2002). Thus, it is not entirely clear at what stage of AD amygdala
atrophy develops or whether amygdala pathology is consistently present in MCI.
Issues in the Classification of Mild Cognitive Impairment
Classification of MCI, while supported by neuropsychological data, ultimately
rests on clinical judgment. Given the heterogeneity of MCI, there exists the possibility of
"contaminating" the MCI group with healthy individuals, or vice versa, when conducting
research studies comparing these two groups. Some subjects classified as having MCI
may actually have had long-standing poor memory function that may not progress to
dementia (Petersen et al., 1999). Additionally, healthy subjects experiencing transient
memory problems due to medication or other factors may be inaccurately classified in the
MCI group (Morris, 2005). Accurate group classification is vital for the study ofMCI;
however, there is evidence that some studies have significant levels of misclassification.
For instance, Visser and colleagues (Visser, Scheltens, & Verhey, 2005) retrospectively
applied the MCI diagnostic criteria used by six clinical drug trials in a cohort of 150 non-
demented subj ects from a memory clinic. They found that the diagnostic accuracy of
MCI criteria used in these trials was low to moderate, suggesting the possibility of
inclusion of many patients who do not have MCI or to exclusion of many who do.
Some scholars have questioned the use of the MCI classification, suggesting that
this classification "pathologizes" features of normal aging and creates
"artificial.. .boundaries" within the aging process (Whitehouse & Juengst, 2005, p. 1419).
These authors argue instead for consideration of "a continuous spectrum of normal
aging" (p. 1419).
In summary, the syndrome of amnestic mild cognitive impairment, characterized
by impaired memory in the context of preserved general cognitive abilities and activities
of daily living, has generated much research interest. Although some investigators have
argued against the use of this classification, others have found it theoretically useful.
MCI is associated with pathology of the hippocampus, a vital medial temporal lobe
memory structure. However, it is not known for certain whether the amygdala, an
important structure for the formation of emotional memory, is affected in MCI.
Most people vividly remember their first kiss and what they were doing when they
first heard about Princess Diana' s death or the Challenger disaster. Such events are
thought to be vividly preserved in memory because of their emotionally arousing nature.
A large body of research suggests that moderate levels of emotional arousal enhance
memory consolidation, the process by which short-term memories are strengthened and
maintained as long-term memories.
The Modulation Hypothesis: Involvement of the Amygdala
One of the most clearly articulated models of the biological basis of emotional
memory is the "modulation" hypothesis of McGaugh and colleagues (McGaugh, 2004;
McGaugh, Cahill, & Roozendaal, 1996; Roozendaal, Nguyen, Power, & McGaugh,
1999). According to this hypothesis, the amygdala plays a critical role in the
enhancement of memory for emotional material. First explicated using animal research,
this model specifies that (a) stress hormones and certain neurotransmitters released in
response to emotional stimuli enhance long-term memory formation, (b) the amygdala
mediates the memory-enhancing effects of these substances, and (c) the amygdala
modulates memory consolidation via its influence on other brain regions that are
intrinsically involved in learning and maintenance of memories (e.g., hippocampus,
caudate nucleus, and nucleus accumbens).
The premises outlined by the modulation model are supported by several lines of
research. First, post training injections of the rat adrenal stress hormones, epinephrine
and corticosterone, produce dose- and time-dependent enhancement of memory in rats
(McGaugh, 2004). Second, the amygdala appears necessary for mediating the effects of
these hormones, since lesions of the amygdala effectively block their memory enhancing
effect (Cahill & McGaugh, 1998). Similarly, the amygdala appears to mediate the
memory-modulating effects of several classes of neurotransmitters, including glutamate
and acetylcholine (McGaugh, 2004). Interestingly, several findings have indicated that
intra-amygdala infusions of drugs enhance memory retention in rats tested after 24 hours
but not in those tested within a few hours of the intervention (Barros, Pereira, Medina, &
Izquierdo, 2002; Bianchin, Mello e Souza, Medina, & Izquierdo, 1999; Schafe &
LeDoux, 2000). This suggests that these hormones and neurotransmitters selectively
affect the consolidation of long-term memory.
Third, although the amygdala is an important site for bringing together memory-
modulating influences of stress hormones and certain neurotransmitters, the actual
memory encoding and consolidation takes place elsewhere in the brain. One key memory
area that receives direct input from the amygdala is the hippocampus (Petrovich,
Canteras, & Swanson, 2001; Pitkanen, Pikkarainen, Nurminen, & Ylinen, 2000).
Electrophysiological evidence strongly suggests that influences from the amygdala
modulate memory consolidation processes in the hippocampus (Cahill & McGaugh,
1998; Frey, Bergado-Rosado, Seidenbecher, Pape, & Frey, 2001). Thus, extensive
research with rats provides a model for emotional memory formation that implicates the
amygdala in this process. Once activated by stress-related hormones and
neurotransmitters, the amygdala enhances long-term memory consolidation in other brain
regions, particularly the hippocampus.
Evidence Supporting the Modulation Hypothesis in Humans
Findings from memory studies in humans provide additional support for the
modulation hypothesis of emotional memory, forming parallels with the animal research.
First, the results of several human studies indicate a critical role of stress hormones in
memory consolidation. Some investigators have found that the administration of the
stress hormone cortisol prior to presentation of words or pictures results in improved
recall performance (Abercrombie, Kalin, Thurow, Rosenkranz, & Davidson, 2003).
Additionally, epinephrine administered after subjects viewed emotionally arousing slides
enhanced their long-term memory of the slides (Cahill & Alkire, 2003).
Consistent with the animal research, the amygdala also appears to be critical in
mediating the memory-enhancing effects of these hormones in humans. Human subj ects
with selective bilateral lesions of the amygdala do not demonstrate the enhanced memory
for emotionally arousing material that normal subjects do (Adolphs, Cahill, Schul, &
Babinsky, 1997; Cahill, Babinsky, Markowitsch, & McGaugh, 1995). Functional brain
imaging studies provide additional evidence that the influence of emotional arousal on
memory involves amygdala activation. For instance, amygdala activity during encoding
is correlated with long-term memory of emotionally arousing but not neutral material
(Cahill et al., 1996). This relationship between amygdala activity at encoding and long-
term memory performance appears to be greatest for stimuli rated as most emotionally
intense (Canli, Zhao, Brewer, Gabrieli, & Cahill, 2000).
Finally, functional neuroimaging studies provide evidence that amygdala activation
in humans influences memory consolidation in other brain regions, particularly the
hippocampus. Hamann and colleagues (Hamann, Ely, Grafton, & Kilts, 1999) presented
emotional and neutral photographs to normal participants while they underwent position
emission tomography (PET) scanning. Subj ects had better memory for emotional than
neutral pictures upon 4-week recognition testing. This enhanced emotional recognition
memory was correlated with bilateral activity in the amygdala and hippocampus
measured at encoding. Additionally, amygdala activity was significantly correlated with
hippocampal activity for emotional pictures. Similar results were obtained in a functional
magnetic resonance imaging (fMRI) study by Dolcos, LaBar and Cabeza (2004).
Consistent with the animal research, the results of these studies suggest that the
amygdala's influence on other brain regions particularly the hippocampus is critical
for creating lasting memories in humans.
Emotional Memory in Older Adults
Because no studies have yet examined emotional memory in amnestic MCI, it is
helpful to first review emotional memory in the context of normal aging and Alzheimer' s
disease. To some extent, the demonstration of the emotional enhancement effect in older
adults depends on the type of memory test used, i.e. recognition versus recall. There is
general support for the preservation of emotional enhancement of memory in aging with
the use of recall tests, and some support for the preservation of this effect with use of
Studies employing recall tests have demonstrated emotional enhancement of
memory in older adults tested at various time intervals. For instance, Kensinger and
colleagues (Kensinger, Brierley, Medford, Growdon, & Corkin, 2002) reported that older
adults showed a similar degree of emotional memory enhancement for words and pictures
upon immediate recall testing as did younger adults. According to Charles and
colleagues (Charles, Mather, & Carstensen, 2003), middle-aged adults (41-53 years) and
older adults (65-80 years) remembered more positive than negative or neutral
photographs when tested with a recall probe 15 minutes after stimulus presentation.
Hamann, Monarch and Goldstein (2000) found that older adults had better recall for
negative and positive compared to neutral pictures one minute and two weeks after
stimulus presentation. Finally, Denburg and colleagues (Denburg, Buchanan, Tranel, &
Adolphs, 2003) presented young (3 5-5 1 years), middle-aged (52-69 years), and older (70-
85 years) adults with positive, negative, and neutral photographs accompanied by a one-
sentence narrative description. The authors found that although older adults showed a
decline in memory relative to young adults, the groups showed similar patterns of
emotional influence on memory, including enhanced free recall of negative and positive
compared to neutral pictures. This pattern of emotional enhancement was observed at 24
hours and 8 months after stimulus presentation.
Several studies employing recognition tests have demonstrated enhanced emotional
memory in older adults, although there have been some studies that have not shown this
effect. Abrisqueta-Gomez and colleagues (Abrisqueta-Gomez, Bueno, Oliveira, &
Bertolucci, 2002) found that upon 30-minute recognition testing, older adults recognized
significantly more emotional than neutral pictures. According to Hamann et al. (2000),
two-week recognition discrimination ability for a group of older adults was enhanced for
negative compared to positive and neutral photographs. On the other hand, Charles et al.
(2003) did not find an emotional enhancement effect with the use of recognition testing.
In their study, young adults correctly recognized a greater proportion of negative than
positive or neutral pictures, while middle and older adults recognized equal proportions
across emotional valence categories. Thus, most studies employing recall tests have
demonstrated emotional enhancement of memory in older adults while some studies
employing recognition tests have demonstrated this effect.
Emotional Memory in Alzheimer's Disease
Findings from the Alzheimer' s disease literature on the effects of emotionally
arousing material on memory have been contradictory. While some studies have
demonstrated a preserved emotional enhancement effect on memory in Alzheimer's
disease (Kazui et al., 2000; Moayeri, Cahill, Jin, & Potkin, 2000), others have not
(Abrisqueta-Gomez et al., 2002; Hamann et al., 2000). Differences in the types of
emotional stimuli used and the types of memory tests used may help to explain this
In general, emotional enhancement of memory is more likely to be preserved in
Alzheimer' s disease when stimuli are presented within a semantic context. For instance,
Alzheimer' s patients have demonstrated better memory for an emotional compared to a
neutral passage in a story when memory was tested by recall (Kazui, Mori, Hashimoto, &
Hirono, 2003; Kazui et al., 2000) or by recognition (Moayeri et al., 2000). Although
overall memory was worse for Alzheimer' s patients than controls, the magnitude of
enhancement by emotional arousal was similar in the Alzheimer' s patients and controls
(Kazui et al., 2000). However, semantic context does not always allow for the
preservation of emotional enhancement in individuals with Alzheimer' s disease. In a
study by Kensinger and colleagues (Kensinger, Anderson, Growdon, & Corkin, 2004),
older adults had higher recognition scores for the negative portion of a story than the
neutral portion after 10 minutes, while individuals with Alzheimer's disease showed no
effect of emotion.
On the other hand, presentation of stimuli in the absence of a semantic context (i.e.,
a series of unrelated pictures) has been less likely to be associated with preservation of
the emotional enhancement effect in Alzheimer' s disease. For instance, Abrisqueta-
Gomez et al. (2002) presented AD patients and older controls with emotional and neutral
photographs. Upon 30 minute recognition testing, controls recognized significantly more
emotional than neutral pictures, but AD patients showed no benefit for emotional picture
As with normal older adults, the type of memory test used affects the pattern of
memory performance for AD patients presented with a series of unrelated stimuli. When
tested with a one minute free recall test following the presentation of 60 positive,
negative, and neutral photographs, patients with Alzheimer' s disease had enhanced
memory for positive compared to negative and neutral pictures. When tested with a
recognition test, Alzheimer's disease patients demonstrated no emotional enhancement of
memory (Hamann et al., 2000). However, recall tests have not always yielded an
emotional enhancement effect in AD patients who are presented with unrelated stimuli.
In a study by Kensinger et al. (2002), AD patients did not show memory enhancement for
emotional words or pictures on immediate recall tasks, although young and older controls
did show emotional enhancement for these items.
Therefore, there is some evidence for emotional enhancement of memory in AD
when stimuli are presented within a semantic context but little evidence of this effect
when unrelated stimuli are presented in the absence of a semantic context. When patients
with AD are presented with unrelated stimuli, they may be more likely to show an
emotional enhancement effect if memory is tested using a recall rather than a recognition
STATEMENT OF THE PROBLEM
Memory problems are among the most prominent age-related cognitive changes,
possibly due to their influence on activities of daily living. Not all aspects of memory are
equally affected by normal aging. Retrieval of old remote memories remains relatively
stable well into the 70's and 80's. In contrast, the ability to learn and retain new episodic
memories is particularly susceptible to aging and is the primary symptom of the memory
variant of amnestic MCI. Many researchers have attempted to characterize the types of
memory disturbances associated with aging in an effort to design effective rehabilitation
techniques and memory management strategies.
The overall focus of the present study is to determine whether the memory
problems displayed by older individuals, particularly those with amnestic MCI, extend to
emotionally arousing materials. Specifically, does the memory performance of
individuals with amnestic MCI benefit from emotionally arousing information? If so, is
this emotional enhancement effect similar in magnitude to that of healthy peers?
Moreover, is there any relationship between volumetric measures of medial temporal
regions involved in memory (i.e., hippocampus) and this enhancement effect?
It is vital to address these questions for two primary reasons. First, a wealth of
literature indicates that emotional memory modulation is mediated by the very brain
regions (i.e., amygdala, hippocampus) that seem to be involved early in Alzheimer' s
disease. To date, findings from the Alzheimer' s literature on the effects of emotionally
arousing materials in memory have been contradictory. Virtually no studies have
examined emotional memory among patients with amnestic MCI, who are at particular
risk for transitioning into Alzheimer' s disease. Based on the pathophysiology of brain
changes in MCI and early Alzheimer' s disease, one might hypothesize that MCI patients
will show less robust effects of emotional arousal on memory performance than normal
age peers. Second, if emotional arousal is shown to enhance memory for individuals who
have amnestic MCI, it may be possible to design memory improvement strategies based
on this form of memory. Even if the enhancement is reduced relative to peers, this might
nevertheless present an opportunity to structure memory tactics and strategies for
remembering salient events.
Research into the modulation hypothesis of emotional memory suggests that the
amygdala exerts its modulatory effects on memory consolidation over long-term
intervals. While most studies investigating emotional memory have been short-term, we
chose to examine emotional memory in the present study using a "rate of forgetting" task
that probed memory over extended periods of time. Specifically, recognition testing
occurred at 10 minutes, 1 hour, 2 weeks, and 3 months after initial stimulus presentation.
The rate of forgetting procedure used in our study is modeled after that of Huppert and
Piercy (1979) who found that the famous bitemporal amnesic H.M. exhibited an
abnormally rapid forgetting rate, relative to controls, over a 7 day period. In contrast,
patients experiencing amnesia due to alcoholic Korsakoff' s disease (i.e., diencephalic
amnesics) showed a normal forgetting rate when initial level of encoding was equated
with that of controls. These findings illustrate the utility of rate of forgetting as a
sensitive measure of medial temporal memory dysfunction in humans. The use of
extended testing intervals in the context of the rate of forgetting paradigm therefore
allows us to test the long-term emotional memory consolidation processes thought to be
mediated by the amygdala and hippocampus.
In order to examine the effects of emotional arousal on memory, pictures from the
International Affective Picture Set were used as stimuli in the rate of forgetting memory
task. While most studies of emotional memory have classified pictures according to
valence categories (positive, negative, neutral), we expected that arousal categories
would be more helpful for investigation of emotional memory, given that amygdala
activity is related more to emotional arousal than to valence (Kensinger & Corkin, 2004).
We therefore selected pictures according to low, medium, and high arousal categories
based on normative arousal ratings. The specific aims and hypotheses of the study are
Specific Aim I
To test the hypothesis that individuals with MCI will exhibit greater forgetting for
all picture stimuli (regardless of emotional content) than normal age-matched controls.
To examine this hypothesis, we compared information loss between the MCI and control
groups across each of the memory time delays (10 minutes, 1 hour, 2 weeks, 3 months).
Additionally, given the difficulty of accurately identifying individuals with MCI and the
suggestion by some investigators that the MCI classification be replaced by a "spectrum
of aging," we next examined picture recognition performance as an outcome of a
continuous indicator of memory status. We hypothesized that individuals with better
memory status would perform better on the picture recognition task at all time points.
Specific Aim II
To examine the effects of emotional arousal on memory performance in individuals
with amnestic MCI and normal peers. Given the relatively well-established emotional
enhancement effect, we hypothesized that controls will show better recognition memory
performance for high arousing stimuli compared to medium and low arousing stimuli.
While emotional memory has not yet been investigated in MCI, studies of suggest that
emotional enhancement of memory is preserved in some instances for individuals with
Alzheimer's disease, although not as frequently as controls. It is therefore hypothesized
that MCI patients will also benefit from arousal but to a lesser extent than controls.
We also examined self-reported and physiological reactivity to emotional stimuli of
MCI patients and controls in the context of this aim. This was done in order to determine
whether emotional reactivity at the time of initial presentation (encoding) of the pictures
was similar for the MCI and control groups.
Finally, as with specific aim I, we examined picture recognition performance for
the low, medium, and high arousal conditions as an outcome of a continuous memory
status variable. We hypothesized that individuals with better memory status would
benefit more from emotionally arousing pictures than people with poorer memory status.
Specific Aim III
To investigate the relationship between hippocampus volume and (1) overall
memory performance on the rate of forgetting task and (2) memory performance for
emotional items on the rate of forgetting task. Because the rate of forgetting task has
been shown to be sensitive to hippocampal pathology, we hypothesized that hippocampal
volume would be positively correlated with overall picture recognition memory
performance on this task. Additionally, because the hippocampus interacts intimately
with the amygdala in enhancing memory for emotional material, we hypothesized that
hippocampal volume would be correlated with recognition memory for the medium and
high arousal conditions but not the low arousal conditions.
Figure 3-1 displays a flowchart of the study design. In the following pages, each
element will be discussed in detail.
Subject Recruitment and Screening
Potential participants were initially recruited from advertisements, from previous
MCI studies conducted by colleagues at the University of Florida, and from an existing
research registry for older adults. All underwent an initial screening evaluation by
telephone in order to address inclusion/exclusion criteria. To be included, participants
had to be over 55 years of age. Exclusion criteria included history of a recent heart
attack, presence of neurological disease, history of head injury or stroke, psychiatric
disturbance sufficient to warrant hospitalization, substance abuse, and adverse reaction to
prescription medication. A total of 58 individuals underwent screening. Of these, 32
individuals did not meet initial eligibility criteria or were no longer interested in
participating after learning more about the study. Reasons for exclusion included being
below the minimum age, history of stroke, and presence of neurological conditions.
Twenty-six subj ects passed initial screening and underwent additional testing of
neuropsychological and mood status, including tests of memory, intellectual functioning,
language, attention, and executive functioning.
All participants underwent neuropsychological testing in order to rule out dementia
and to obtain data which would assist in group classification at the consensus conference.
Intellectual Assessment and Dementia Screening
Mini-Mental State Examination (MMSE). The MMSE (Folstein, Folstein, &
McHugh, 1975) has been used for detecting dementia for over 25 years. The maximum
score on the MMSE is 30 points. We employed a cut-off score of 24, considering scores
of 24 or below to be indicative of probable dementia. The MMSE includes measures of
memory, attention, formation, orientation, figure copying, reading and writing. Factor
analysis of the MMSE administered to a community-dwelling population revealed
dimensions that corresponded directly to the MMSE sections articulated by the
developers of the instrument, supporting the construct validity of the MMSE (Jones &
Wechsler Abbreviated Scale of Intelligence (WASI). The Vocabulary and
Matrix Reasoning subtests from the WASI were administered in order to obtain an
estimate of intellectual functioning (Psychological Corporation, 1999) yielded an
estimated two-scale IQ. The Vocabulary subtest assesses the ability to define words and
the Matrix Reasoning subtest is a measure of non-verbal abstract reasoning.
North American Adult Reading Test (NAART). We administered the NAART
in order to obtain an estimate of premorbid intellectual ability. The NAART (Blair &
Spreen, 1989) is a word reading test used for this purpose. The NAART is thought to
primarily index prior (rather than current) intellectual ability (Crawford, Deary, Starr, &
Whalley, 2001). Additionally, Uttl (2002) demonstrated that the NAART is a reliable
and valid measure of verbal intelligence.
Hopkins Verbal Learning Test-Revised (HVLT-R). The HVLT-R (Benedict,
Schretlen, Groninger, & Brandt, 1998) is a three-trial list learning and free recall task
comprising 12 words, 4 words from each of three semantic categories. The Total Recall
score refers to the total number of words recalled after all three list presentations. The
delayed recall test is administered 20 minutes after completion of the first three recall
trials. The Percent Retention score is determined by dividing the total number of correct
items recalled after the delay by the number of items recalled on the third learning trial
prior to the delay. The HVLT-R also includes a 24-item delayed recognition test, from
which a discrimination index can be calculated. All raw scores obtained on the HVLT-R
were converted to z-scores based on the age-based norms of Benedict et al. (1998).
Test-retest correlations of the HVLT are similar to those of other verbal memory
tests (Rasmusson et al., 1995). Other studies of the HVLT support its alternate form and
test-retest reliability (Benedict et al., 1998) and its construct and content validity
(Shapiro, Benedict, Schretlen, & Brandt, 1999).
Boston Naming Test-Short Form. We administered the 30-item short form of
the Boston Naming Test (Fastenau, Denburg, & Mauer, 1998). In this version, either odd
or even items from the standard 60-item test (Kaplan, Goodglass, & Weintraub, 1983) are
given. Fastenau et al. (1998) demonstrated adequate reliability and validity for each
form. Fisher, Tierney, Snow, and Szalai (1999) found support for criterion-related
validity of the short forms. For the present study, raw scores obtained on the short form
of the Boston Naming Test were converted to z-scores based on age-based norms
(Fastenau et al., 1998).
Phonemic fluency. The total number of words generated in 1 minute for the
letters F, A, and S was obtained. The instructions were identical to those used by Spreen
and Benton (1977). Participants were instructed that proper nouns and multiple words
using the same stem with a different suffix (e.g., friend, friends, friendly) were not
acceptable. The total number of words generated in all three trials was converted to a
scaled score according to age-based norms (Ivnik, Malec, Smith, Tangalos, & Petersen,
1996). Scaled scores were then converted to z-scores for the purposes of consistency
throughout the neuropsychological test battery.
Semantic fluency. Following the instructions of Rosen (1980), participants were
asked to say the names of as many animals that they could think of in a one-minute
period. The total number of animals named was converted to a z-score based on age and
education-based norms (Tombaugh, Kozak, & Rees, 1999).
Attention, Psychomotor Speed, and Executive Function
Stroop Color and Word Test. The Stroop Color and Word Test (Golden, 1978)
consists of three trials, one in which the participant is asked to read color words printed in
black ink (Word Reading), one requiring the naming of printed colors (Color Naming),
and one in which reading focuses on color words printed in ink of incongruent colors
(Interference). The Stroop technique has satisfactory reliability (Franzen, Tishelman,
Sharp, & Friedman, 1987; Spreen & Strauss, 1991). Scores obtained on the Stroop test
for each of the three trials were converted to scaled scores according to age-based norms
(Ivnik et al., 1996). Scaled scores were then converted to z-scores for the purposes of
consistency throughout the neuropsychological test battery.
Trail Making Test. The Trail Making Test (Army Individual Test Battery, 1944)
is given in two parts. In Part A the subj ect must first draw lines to connect consecutively
numbered circles on a worksheet. In Part B, the subject must connect consecutively
numbered and lettered circles by alternating between letters and numbers. Part A is a
measure of visuospatial attention, motor speed, and numerical sequencing, while Part B
has the additional requirement of mental flexibility. The Trail Making Test generally
yields reliability coefficients above 0.60, with most in the 0.80s (Spreen & Strauss,
1991). Scores obtained on the Trail Making Test were converted to scaled scores
according to age-based norms (Ivnik et al., 1996). Scaled scores were then converted to
z-scores for the purposes of consistency throughout the neuropsychological test battery.
Geriatric Depression Scale (GDS-15). The GDS-15 (Sheikh & Yesavage, 1986)
is a 15-item self-report instrument designed specifically for older adults that assesses
depressive symptoms without focusing on physical complaints. It is scored
dichotomously (yes/no) and inquires into subj ective depression experienced during the
prior week. The GDS-15 has been widely recommended as a brief screening instrument
for late-life depression and has been found to be useful in detecting late-life maj or
depression in primary care settings (Lyness et al., 1997; Watson & Pignone, 2003).
State Trait Anxiety Inventory (STAI). The STAI was developed to gauge two
measures of anxiety: state anxiety, defined as a transitional emotional response, and trait
anxiety, referring to enduring personality differences in anxiety. Stanley, Beck and Zebb
(1996) found strong test-retest reliability for the STAl-trait and slightly weaker reliability
for the STAl-state, as might be expected. Additionally, they found adequate convergent
validity with other anxiety measures.
The Instrumental Activities of Daily Living Scale (IADL). The IADL scale
(Lawton & Brody, 1969) assesses functional ability to perform several activities,
including using the telephone, managing personal finances, and preparing food. The
score ranges from 15 to 45, with a score of 15 representing no impairment.
Memory Functioning Questionnaire (MFQ). The MFQ (Gilewski, Zelinski, &
Schaie, 1990) assesses the self-appraisal of everyday memory functioning in adults. It
consists of 64 items rated on 7-point scales. For the purposes of this study, we report the
response to the following question: "How would you rate your memory in terms of the
kinds of problems that you have?" Answers are given on a 7-point scale, with "1"
representing "maj or problems" and "7" representing "no problems."
After completion of the neuropsychological and mood testing, participant profies
were discussed at a case conference and classified into MCI and control groups.
Members of the consensus conference committee included Ann Mikos (author), Dr.
Dawn Bowers, PhD (neuropsychologi st), Dr. Catherine Price, PhD (neuropsychologi st),
and Ida Kellison (graduate student). Participants were classified as belonging to the MCI
group if they had a memory impairment of at least 1.5 standard deviations below the
mean on the Hopkins Verbal Learning Test-Revised, intact performance on other
cognitive tests, and in tact activities of daily living.
Twenty-six participants passed the initial screening, underwent neuropsychological
and mood assessment, and were discussed at a consensus conference meeting. A
decision was made to exclude one of these participants, because this individual scored
below the MMSE cut-off of 24 (MMSE = 23) and displayed impairment in several other
cognitive domains besides memory (e.g., language, executive function, attention).
According to consensus conference classifications, 12 of the remaining individuals were
assigned to the MCI group and 13 to the normal control group.
A subset of 13 participants underwent MRI scanning at the University of Florida
McKnight Brain Institute. Of the 25 participants that were enrolled in the study, 12 had
contraindications for MRI (claustrophobia, presence of metal implants in the body) or
were unwilling to complete the MRI scan. MRI scans were obtained for 7 individuals in
the MCI group and 6 controls. Images were obtained on a Siemens 3T Allegra scanner.
The imaging protocol consists of a series of gradient echo scans (MPR3D: TR = 10 ms,
TE = 4 ms, 10 degree flip angle, matrix = 130 by 256, 160 mm volume, and section
thickness = 1.25 to 1.40 mm), producing a gapless series of high quality images. Volume
of the hippocampus was traced in the program "Measure" by a trained and blinded rater
who previously attained inter-rater reliability of >0.90.
Emotional Memory Task
Participants first viewed a set of 120 emotional and neutral target pictures (study
phase). Pictures were presented in two blocks of 60 pictures each, with a short break in
between blocks. Participants then underwent several recognition tasks at increasing time
delays (test phases), modeled after the rate of forgetting task of Huppert and Piercy
(1979) in order to assess their rate of forgetting over time.
During the study phase of the picture task, subj ects sat quietly in a darkened
electrically and sound attenuated chamber where they viewed a series of emotional and
neutral pictures. This testing took place in the Cognitive Neuroscience Laboratory at the
McKnight Brain Institute. Visual stimuli were presented on a 21" computer monitor.
During testing each picture was shown for 6 seconds, followed by an intertrial interval of
Following presentation of each picture, participants verbally rated each picture
using two independent 1-9 ordinal rating scales (one for valence and one for arousal).
The scales were vertically arrayed on a slide as cartoon figures and were presented on the
computer screen following each picture.
Subj ects participated in four test phases in order to evaluate rate of forgetting over
time. Recognition testing occurred 10 minutes, 1 hour, 2 weeks, and 3 months after
completion of the study phase. During each test phase, a unique set of 30 target pictures
from the initial presentation were shown in conjunction with an equal number of
di stracters. Subj ects were asked to indicate whether or not each picture was one of the
The stimuli presented in the study phase of the emotional memory task consisted of
120 pictures selected from the International Affective Picture Set (Lang, Bradley, &
Cuthbert, 2001) on the basis of normative valence and arousal ratings. For the purpose of
analysis, pictures were divided into low, medium, and high arousal categories.
Normative arousal ratings differed significantly among the low, medium, and high
arousal categories. Low arousal pictures had a mean normative arousal rating of 3.08,
medium arousal pictures had a mean normative arousal rating of 4.79, and high arousal
pictures had a mean normative arousal rating of 6. 14. For each recognition memory test,
target pictures were matched to distracter pictures on the basis of content and of
normative valence and arousal ratings. Low, medium, and high arousal targets did not
differ significantly from their distracters in terms of valence or arousal. One-way
ANOVAs also revealed that the stimuli did not differ within the four testing blocks in
terms of valence or arousal.
Skin Conductance Response (SCR)
The skin conductance response was assessed in this study because it is an index of
sympathetic arousal, correlates with self-reports of emotional arousal, and is relatively
independent of valence (Bradley & Lang, 2000). The SCR was measured from
electrodes attached to the palms with adhesive collars. Prior to beginning the study phase
of the emotional memory task, the skin from both hands was cleaned and dried
thoroughly. Two 12 mm Ag/AgCl were filled with conducting gel (K-Y Brand Jelly,
McNeil-PPC, Inc.) and were attached adj acently using electrode collars on the thenar and
hypothenar surfaces of each palm in order to obtain measures of skin conductance during
Skin conductance data were sampled at 20 Hz using two Colbourn Isolated Skin
Conductance Couplers. The SCR was defined as the difference between the peak
conductance during the 6-second viewing period and the mean conductance achieved
during the last pre-stimulus second, derived independently for each hand. SCR was
represented in microsiemens (CIS) units. The SCR was averaged for both hands, unless
data from one hand contained excessive artifact or did not contain enough valid trials
(i.e., less than 3 out of 20) to generate acceptable averages. In these cases, the data from
the other hand were used in place of the composite data.
Due to concerns about habituation, only skin conductance data obtained during
presentation of the first block of 60 pictures were analyzed. Because skin conductance
habituates relatively quickly, data from the first 30 trials of this block were analyzed
To test the first aim (comparison of rate of forgetting for pictures between the MCI
and control groups), a mixed ANOVA consisting of a between-subj ects factor of group
(control, MCI) and a within-subj ects factor of recognition test delay (10 minutes, 1 hour,
2 weeks, 3 months) was conducted. The dependent variable was the percentage of
correct responses on the picture recognition tests.
Given the concerns discussed earlier about the heterogeneity of MCI and potentials
for misclassification, data were also examined using a continuous indicator of memory
status as a predictor of performance on the picture recognition task. The percent
retention z-score on the HVLT-R was used as the continuous variable, because it
represents loss of information over time. Next, a series of regression analyses was
conducted with memory status (as indicated by the HVLT-R retention z-score) as the
predictor and recognition memory performance as the dependent variable. Separate
regression analyses were conducted for each of the four recognition testing sessions.
In order to test the second aim (examination of the effect of emotional arousal on
recognition memory performance at increasing time delays in MCI patients compared to
controls), a mixed factorial ANOVA was conducted. The ANOVA consisted of a
between-subj ects factor of group (control, MCI) and within-subj ects factors of arousal
(low, medium, high) and recognition test delay (10 minutes, 1 hour, 2 weeks, 3 months).
Within this second aim, a comparison of the MCI and control groups' ratings and
skin conductance responses to emotional stimuli was carried out using two separate
mixed ANOVAs. The ANOVAs consisted of a between-subj ects factor of group
(control, MCI) and a within-subj ects factor of arousal (low, medium, high). For one
ANOVA, the dependent variable was self-reported picture ratings. For the second
ANOVA, the dependent variable was skin conductance response.
As with the previous aim, the data for aim 2 were also examined using a continuous
indicator of memory status instead of the between-groups classification. We conducted a
series of regression analyses with memory status (as indicated by the HVLT-R retention
z-score) as the predictor and recognition memory performance as the dependent variable.
Separate regression analyses were conducted for the low, medium, and high arousal
conditions at each of the four recognition testing sessions.
In order to test the third aim (examination of the relationship between hippocampus
volume and recognition memory performance), two-tailed Pearson correlations were
conducted between hippocampal volumes and recognition scores on the picture test.
Correlations were conducted both for overall recognition scores at each of the four testing
sessions and for recognition scores within the low, medium, and high arousal categories
for each of the four testing session.
Figure 3-1. Study design flowchart.
Table 4-1 displays the demographic characteristics of the control and MCI groups.
As a whole, participants ranged in age from 55 to 88 (mean = 75.6), were well educated
(mean = 15.9 years, range 12-20) and included approximately twice as many males as
females (ratio of 16:9). As shown in Table 4-1, the two groups did not differ
significantly in terms of age, education, or ratio of males to females.
Neuropsychological and Mood Performance: MCI versus Control Groups
Results of the neuropsychological and mood evaluation are shown in Table 4-1. As
indicated in the table, the MCI group attained significantly lower scores across the
various indices of recent memory from the Hopkins Verbal Learning Test-Revised. This
included measures of overall learning, delayed recall after 20 minutes, percent retention,
and the recognition discrimination index. On average, these memory scores ranged from
1.6 to 2.2 standard deviations below normative standards and are in line with those
reported in the literature for MCI classifications. The MCI group did not differ from the
control group on any other measures, including tasks of language (Boston Naming,
phonemic and semantic fluency), attention and processing speed (Trailmaking and Stroop
Color and Word Naming), and executive functioning (Stroop interference). The MCI
group attained a mean IQ estimate of 113 (high average) and performed in the
nondemented range on the MMSE. Taken together, performance by the MCI group
across the neurocognitive screening measures was largely intact, with the exception of
relatively impaired scores on tests of recent memory.
In terms of mood, the MCI group did not differ from the control group on the
Geriatric Depression Scale or the State Trait Anxiety Inventory. Additionally, the two
groups did not differ in their abilities to perform activities of daily living. However,
individuals with MCI did report significantly more memory problems than did controls.
Aim 1: Comparison of Overall Rate of Forgetting
One participant from the MCI group had missing data for the one hour picture
recognition task and was therefore not included in the rate of forgetting analyses. Figure
4-1 shows the overall recognition performance at the 4 testing sessions for each group.
To test the first hypothesis that MCI patients would show an increased rate of forgetting
over time for all stimuli, we conducted a repeated measures ANOVA consisting of a
between-subj ects factor of group (control, MCI) and a within-subj ects factor of
recognition test delay (10 minutes, 1 hour, 2 weeks, and 3 months). There was a
significant effect of delay, F(3,66) = 201.056, p < 0.001. Bonferroni-adjusted post hoc
tests revealed that subj ects correctly recognized significantly more pictures at 1 hour
(mean = 93.05%) than at 2 weeks (mean = 72. 11%; p < 0.001), and at 2 weeks than at 3
months (mean = 59.83%; p < 0.001). There was no significant difference between the
percentage of correctly recognized items at 10 minutes (mean = 93.32%) and 1 hour
(mean = 93.05%; p = 1.0). There was no significant group by delay interaction [F(3,66)
= 1.934, p = 0. 133], and no main effect of group [F(1,22) = 0.528, p = 0.475].
We then ran a series of regressions with memory status (as indicated by the percent
retention z-score on the HVLT-R) as the predictor variable and overall picture
recognition memory performance as the dependent variable. Memory status was a
significant predictor of picture recognition performance at 10 minutes (R2 = 0.377, p =
0.001) and 2 weeks (R2 = 0.214, p = 0.012), but not at I hour (R2 = 0.018, p = 0.210) or at
3 months (R2 = 0.032, p = 0. 193).
Aim 2: Effects of Emotional Arousal on Recognition Memory Performance
Emotional Memory Performance: MCI versus Control Group
In order to investigate the effects of arousal on memory performance we conducted
a repeated measures ANOVA consisting of a between-subj ects factor of group (control,
MCI) and within-subjects factors of recognition test delay (10 minutes, 1 hour, 2 weeks,
3 months) and arousal category (low, medium, high). Figure 4-2 shows the rate of
forgetting for low, medium, and high arousal pictures over the four testing sessions for
the two groups. There was no significant main effect of arousal [F(2,44) = 0.902, p =
0.413], no significant main effect of group F(1,22) = 0.597, p = 0.448], no significant
group by arousal interaction [F(2,44) = 0.686, p = 0.509], and no significant arousal by
delay interaction [F(6,132) = 0.715, p = 0.509].
After examining memory performance for the various arousal conditions, we
wanted to verify whether participants' subj ective arousal ratings at the time of encoding
were consistent with our predetermined arousal categories (low, medium, high). Figure
4-3 shows the mean picture arousal ratings given by participants in both groups as a
function of predetermined arousal category. To determine whether individuals in the two
groups rated stimuli differently, a 2 X 3 mixed factorial ANOVA was conducted. Group
(MCI, control) was the between-subj ects factor, arousal (low, medium, high) was the
within-subj ects variable, and mean picture arousal ratings was the dependant variable.
There was a significant main effect of arousal, F(1.5 59,3 5.85 8) = 170.050, p<0.001.
Bonferroni-adjusted post hoc tests revealed that pictures in the low arousal category
(mean rating = 6.0) were rated as significantly less arousing than pictures in the medium
arousal category (mean rating = 5.0; p < 0.001), and that pictures in the medium arousal
category were rated as significantly less arousing than pictures in the high arousal
category (mean rating = 3.2; p < 0.001). There was no significant arousal by group
interaction [F(1.559,35.858) = 1.752, p = 0. 193] and no main effect of group [F(1,23) =
0.997, p = 0.328]. In summary, there were no differences between the MCI and control
groups in their arousal ratings of emotional pictures at the time of encoding.
Skin Conductance Response (SCR)
Next, we evaluated the skin conductance response, a measure of sympathetic
arousal, to determine whether subj ects reacted physiologically to arousing stimuli.
Because skin conductance habituates relatively quickly, data from only the first 30 trials
of this block were analyzed first. The data from 11 participants were excluded from the
analysis of the first 30 trials because they had too few valid trials (i.e., less than 3 out of
20) to generate valid averages. To assess whether the control and MCI groups had
similar physiological responses to the photographic stimuli, a 2 X 3 mixed factorial
ANOVA was conducted with group (control, MCI) as the between-subj ects factor and
arousal (low, medium, high) as the within-subj ects factor. There was a significant main
effect of arousal, F(2,24) = 7.480, p = 0.003. Bonferroni-adjusted post hoc tests revealed
that SCRs were significantly lower for pictures in the low arousal category than pictures
in the high arousal category (p = 0.010), and significantly lower for pictures in the
medium arousal category than pictures in the high arousal category (p = 0.050). There
was no significant arousal X group interaction [F(2,24) = 0. 154, p = 0.858], and no main
effect of group [F(1,12) = 0.503, p = 0.492].
A second similar mixed factorial ANOVA was conducted in order to analyze skin
conductance data from the entire first block of 60 trials. The data from 4 participants
were excluded from the analysis because they had too few valid trials to generate valid
averages. For this analysis, there was no significant main effect of arousal [F(2,38) =
2.539, p=0.092], no significant arousal X group interaction [F(2,3 8) = 2.078, p = 0. 139],
and no main effect of group [F(1,19) = 0.867, p = 0.363].
Emotional Memory Performance: Spectrum of Aging
Next, we ran a series of regressions with memory status (as indicated by the percent
retention z-score on the HVLT-R) as the predictor and recognition memory performance
as the dependent variable for the low, medium, and high arousal conditions at each of the
four testing sessions. Table 4-2 shows the regression coefficients for each condition. At
the 10 minute picture recognition testing session, memory status was a significant
predictor of picture recognition performance for all arousal conditions. Memory status
was not a significant predictor of recognition performance at the 1 hour session for any of
the arousal conditions. At the 2-week picture recognition testing session, memory status
was a significant predictor of picture recognition performance for only the high arousal
condition. At the 3-month testing session, memory status was not a significant predictor
for any of the arousal conditions.
Aim 3: Relationship Between Hippocampus Volumes and Recognition Memory
Hippocampal Volume and Overall Rate of Forgetting
To test the third hypothesis that hippocampal volume would be positively
correlated with picture recognition memory performance, Pearson correlations were
conducted between hippocampal volume and overall recognition performance at each of
the four testing sessions. Hippocampal volume was significantly correlated with picture
recognition performance at 10 minutes (R2 = 0.715, p = 0.006), at 1 hour (R2 = 0.637, p =
0.026), at 2 weeks (R2 = 0.67,2 p = 0.012), and at 3 months (R2 = 0.690, p = 0.009).
Hippocampal Volume and Rate of Forgetting for Emotional Material
Additionally, to test the hypothesis that hippocampal volume would be correlated
with recognition memory for the medium and high arousal conditions but not the low
arousal condition, Pearson correlations were conducted between hippocampal volume
and recognition memory test performance for each arousal condition at each testing
session. As shown in Table 4-3, hippocampal volume was significantly correlated with
recognition memory performance for high arousal pictures at 10 minutes, 1 hour, and 2
weeks. Additionally, hippocampal volume was significantly correlated with recognition
memory performance for medium arousal pictures at 10 minutes, 1 hour, and 3 months.
4.58 (1.17) 3.60 (0.97)
Table 4-1. Demographic characteristics and performance on neuropsychological tests.
Means (+standard deviation) are shown.
77.92 (7. 11)
WASI 2-scale IQ
NAART estimated IQ:
121.92 (11.59) 113.42 (10.39)
ry (Hopkins Verbal Learning Test-Revised; z-scores shown)
0.75 (0.97) -1.81 (0.70) 3.122
-0.30 (0.98) -2.23 (0.69) 5.630
-0.066 (0.89) -1.64 (1.51) 3.142
-0.16 (1.0) -2.12 (0.87) 5.237
STAI Standard Sco
Language (z-scores displayed)
-0.57 (1.64) 0.243
0.53 (0.83) 0.48 (0.86) 0.152
0.20 (1.28) -0.28 (1.15) 0.965
Psychomotor Speed / Executive Function (z-scores displayed)
0.30 (0.77) 0.28 (0.82) 0.064
0.64 (0.87) 0.11 (0.55) 1.805
0.069 (0.78) -0.09 (1.03) 0.421
0.27 (1.09) -0.06 (0.81) 0.817
0.47 (1.01) 0.30 (0.88) 0.431
2.31 (2.56) 3.67 (2.67) -1.298
Table 4-3. Correlations between hippocampal volume and recognition memory
performance for each arousal condition at each testing session (N=13).
*Indicates significance at the p < 0.001 level. 'Indicates significance at the p
< 0.05 level.
Table 4-2. Regression coefficients (R2) for individual regression equations with memory
status as the predictor of recognition memory performance for the low,
medium, and high arousal conditions (N=25). *Indicates significance at the p
< 0.001 level. +Indicates significance at the p < 0.05 level.
3 months 0.174
10 minutes 1 hour 2 weeks 3 months
Figure 4-1. Recognition performance at the 4 testing sessions for the MCI (N=11) and
control groups (N=13).
2 weeks 3 months
Figure 4-2. Percent correct picture recognition for low, medium, and high arousal
pictures over the four testing sessions for the MCI (N=11) and control groups
Figure 4-3. Mean arousal ratings (+95% confidence limits) for low, medium and high
arousal pictures by control subj ects (N=13) and MCI subj ects (N =12).
The present study examined three maj or aims. The first hypothesis was that
patients with mild cognitive impairment would demonstrate increased rates of forgetting
for a picture recognition task compared to controls, as demonstrated by their recognition
memory performance at increasing time delays. This prediction was based on the
Endings that performance on memory retention tasks at increasing time delays is
dependent on an in tact hippocampus (Huppert & Piercy, 1979) and that the hippocampus
is pathologically involved in MCI. In follow-up analyses for this aim, the MCI versus
control group classification was removed and memory status was entered as continuous
predictor of picture recognition performance in a series of regression equations. These
analyses were conducted because of increasing concerns with the accuracy and utility of
the MCI group classification.
The second aim examined whether the picture recognition memory performance of
controls and MCI subjects benefited from emotional arousal. Based upon a review of the
available emotional memory literature in aging and in Alzheimer's disease, we
hypothesized that controls would show better recognition memory performance for high
arousing stimuli compared to medium and low arousing stimuli and that MCI patients
would also benefit from arousal but to a lesser extent than controls. In assessing the
second aim, we next investigated whether MCI patients differed from controls in their
arousal ratings of neutral and emotional stimuli and in their physiological reactivity (as
assessed by skin conductance responses) to these stimuli. These analyses were necessary
in order to verify whether individuals in both groups had similar levels of arousal (both
subjective and physiological) to emotional stimuli at the time of encoding. Finally, we
examined this aim using memory status as a continuous predictor of picture recognition
performance for the various arousal conditions.
The third hypothesis was that hippocampus volumes would be positively correlated
with overall recognition memory performance at each of the four testing sessions and that
hippocampus volume would be more strongly related to memory of medium and high
arousal pictures than of low arousal pictures. This prediction was based on the
hypothesized role of the hippocampus in processes mediated by the amygdala that lead to
enhanced memory for emotionally arousing materials.
Summary and Interpretation of the Findings
The first hypothesis was not supported when the data were examined using the
dichotomous group classification (MCI versus control). That is, there were no group
differences in recognition memory performance at any of the four recognition testing
sessions. There are several potential explanations for this finding. It is possible that a
ceiling effect at the first two testing sessions (10 minutes and 1 hour) and a floor effect at
the last testing session (3 months) may have masked group differences. Additionally, our
MCI group may have contained individuals who performed poorly on memory testing at
screening but later reverted to normal. Thus, contamination of our MCI group with
normal controls may have confounded our results. Several factors led us to question the
consistency and accuracy of consensus conference classifications, lending support to this
explanation. For instance, a linear discriminant function analysis with measures from the
HVLT-R (total recall, delayed recall, and percent retention) as the independent variables
classified two cases (one MCI and one control) differently from the consensus conference
classifications. Additionally, one of the participants that we classified as a control
participant had been referred to us by another investigator who had previously classified
that person as an MCI participant. Finally, two participants who were classified by us as
MCI cases were later classified as controls by another clinician who was conducting a
separate research study.
Given the difficulty of accurately identifying individuals with MCI, we next
examined the rate of forgetting data using a continuous indicator of memory status rather
than group classifications. Using this approach, we found a relationship between
memory status and performance on the rate of forgetting task at some of the testing
intervals. That is, memory status predicted picture recognition performance at 10
minutes and 2 weeks, but not at I hour or 3 months. It is unclear why memory status was
related to recognition memory performance at 10 minutes and 2 weeks but not I hour.
However, it is possible that memory status was not correlated with recognition memory
performance at 3 months due to a floor effect in picture recognition performance at this
In summary, the regression analyses revealed some significant relationships
between memory status and picture recognition performance on the rate of forgetting task
while the between-groups analyses revealed no relationship between group status and
performance on the rate of forgetting task. This indicates that using the dichotomous
group classification masked an important relationship between memory status and
performance on the picture rate of forgetting task.
The second aim examined the effect of emotional arousal on memory performance.
The hypothesis that control subj ects, and MCI subjects to a lesser extent, would
demonstrate better recognition memory performance for highly arousing pictures was not
supported when data were examined using a dichotomous MCI versus control
classification. In fact, neither controls nor MCI patients showed an emotional
enhancement effect for memory at any of the four recognition testing sessions. The
Ending that controls did not show an emotional enhancement effect for memory is
somewhat surprising given the reasonably well-established enhancing effect of emotion
on memory. Potential explanations are discussed below.
Next, we examined subj ective arousal ratings and physiological arousal levels at
encoding in order to determine whether both groups responded similarly to the emotional
stimuli. Across all subjects, high arousal pictures were rated as significantly more
arousing than medium arousal pictures, and medium arousal pictures were rated as
significantly more arousing than low arousal pictures. There was no significant group
difference or group by arousal interaction, suggesting that MCI patients and controls had
similar arousal ratings of the low, medium, and high arousal pictures.
The skin conductance data were generally consistent with this pattern of arousal
ratings. Due to concerns about habituation, skin conductance data were examined for
only the first 30 pictures and then for the entire first block of 60 pictures. When data
from the first 30 trials were examined, all subj ects showed significantly greater
physiological reactivity to high than to low arousal pictures. There were no group
differences, indicating that both groups showed this pattern of enhanced physiological
reactivity to high arousal pictures. However, when data from the first 60 trials were
examined, there were no differences in physiological reactivity across arousal categories
for the entire set of subjects. There was no significant group effect, suggesting that both
groups showed a similar pattern of habituation when all 60 trials were considered
together. Thus, there do not appear to be differences in self-reported arousal rating or in
physiological reactivity to arousing stimuli between MCI patients and controls. Both
groups rated high arousal pictures as more arousing than low arousal pictures and had
higher levels of physiological reactivity to pictures in the high arousal than the low
arousal categories. This suggests that the absence of a demonstrated emotional memory
enhancement effect in the previous analysis cannot be explained by a lack of emotional
arousal at encoding.
However, when the group classification was removed and memory status was used
as a continuous predictor of recognition memory performance, we found evidence
indicating that higher memory status is associated with more of a benefit from emotional
arousal on subsequent memory performance. Memory status predicted picture
recognition performance for all arousal categories at the first recognition session (10
minutes). At I hour, memory status was not a significant predictor of recognition
performance for any arousal categories, but at 2 weeks memory status was a significant
predictor of recognition performance for the high arousal category. At 3 months,
memory status was not a significant predictor of recognition for any of the arousal
categories. Thus, the relationship observed at 10 minutes between memory status and
picture recognition memory performance was not specific to emotional arousal category.
Because the literature on the modulation hypothesis of emotional memory suggests that
the amygdala's enhancement of memory consolidation for emotional material is a long-
term process, this result is not surprising. However, at 2 weeks, people with better
general memory status benefited more from arousal on picture recognition performance
than did people with poorer memory status. The fact that this emotional enhancement
effect was demonstrated at 2 weeks but not at I hour is also consistent with the notion
that the amygdala's enhancement of emotional memory consolidation in the hippocampus
is a long-term process. In fact, research from the animal literature suggests that this
process takes place over the course of days and weeks, and that emotional enhancement
effects on memory are not yet evident after a few hours (Barros et al., 2002; Bianchin et
al., 1999; Schafe & LeDoux, 2000). The lack of relationship between memory status and
picture recognition performance at 3 months may be due to a floor effect.
The third aim investigated the relationship between hippocampus volume and rate
of forgetting performance on the picture recognition task. Our hypothesis that
hippocampal volume would be positively correlated with overall recognition memory
performance at each of the four testing sessions was supported by the data. Specifically,
hippocampal volume was significantly correlated with picture recognition performance at
each of the four testing sessions.
Additionally, our hypothesis that hippocampus volume would be more strongly
related to recognition memory for medium and high arousal pictures than for low arousal
pictures was generally supported by the data. Hippocampal volume was not correlated
with recognition of low arousal pictures at any of the testing sessions. Hippocampal
volume was significantly correlated with recognition memory performance for medium
arousal pictures at 10 minutes, 1 hour, and 3 months, and hippocampal volume was
significantly correlated with memory for high arousal pictures at 10 minutes, 1 hour, and
Interpretation and Relationship to the Literature
Huppert and Piercy (1979) demonstrated that a patient suffering from amnesia due
to bilateral hippocampal ablation had impaired memory performance as assessed by his
rate of forgetting at increasing time delays. Our study adds to this observation by
demonstrating that rate of forgetting over time is sensitive to hippocampal pathology in
older adults. While performance on the rate of forgetting task, assessed at four time
points, did not differ between individuals classified as belonging to an MCI group and
those classified as controls, the task was sensitive to a continuous indicator of memory
status as well as to hippocampal volume. That is, memory status and hippocampal
volume were each significantly associated with picture recognition performance at some
of the recognition testing sessions. Interestingly, hippocampal volume was significantly
correlated with picture recognition performance at all four testing sessions, while
memory status was significantly associated with picture recognition performance at only
the 10 minute and 2 weeks testing sessions. This suggests that the rate of forgetting may
be more sensitive to hippocampal pathology than to a neuropsychological memory
Our finding that normal older controls did not demonstrate enhanced recognition
memory performance for highly arousing pictures is not entirely unprecedented in the
literature. While some investigators have demonstrated enhanced recognition
performance for emotional pictures in older adults (Abrisqueta-Gomez et al., 2002;
Hamann, Monarch, & Goldstein, 2002), others have not (Charles et al., 2003). In
general, tests of recall have been more likely to elicit emotional enhancement effects for
normal older adults than have recognition tests (Denburg et al., 2003; Kensinger et al.,
2002). However, the decision to use recognition tests in the present study was based on
the risk of encountering floor effects with the use of recall tests at the long-term testing
The dual process theory of memory (Jacoby, 1991) may help to explain the lack of
observed emotional enhancement for memory when comparing MCI and control groups.
This theory posits that recollection involves a 'controlled' process in which there is
conscious retrieval of a prior learning episode, while familiarity is a fast and automatic
process requiring few cognitive resources which occurs when prior exposure or
processing lead to a feeling of familiarity or 'perceptual fluency.' Interestingly,
recollection and familiarity have been found to be differentially affected by emotional
arousal in both laboratory (Ochsner, 2000) and autobiographical memory studies
(Reisberg, Heuer, McLean, & O'Shaughnessy, 1988). Ochsner (2000) found that
recollection, but not familiarity, was boosted for negative or highly arousing and, to a
lesser extent, positive stimuli. The authors concluded that greater recollection for
affective events leads them to be more richly experienced in memory.
In an elegant study employing the dual process theory, Dolcos LaBar, and Cabeza
(2005) examined memory for emotional and neutral pictures one year after encoding, and
they used fMRI to measure neural activity during the recognition test. Participants
included nine healthy young adult females. The authors found that memory performance
was better for emotional than for neutral pictures, but that this emotional enhancement
was limited to recollection-based memory. Successful retrieval of emotional pictures
elicited greater activity than successful retrieval of neutral pictures in the amygdala,
entorhinal cortex, and hippocampus. Moreover, in the amygdala and hippocampus, the
emotion effect was greater for recollection than for familiarity, whereas in the entorhinal
cortex, it was similar for both forms of retrieval. Thus, emotion selectively enhanced
recollection-based activity in both the amygdala, a prototypical emotion region, and the
hippocampus, a prototypical memory region thought to be involved in recollection. The
co-activation of the amygdala and hippocampus in the recollection of emotional items is
consistent with the modulation hypothesis (McGaugh, 2004). Importantly, the Eindings
of Dolcos et al. (2005) and Ochsner (2000) suggest that emotional arousal enhances
recollection but not familiarity. Therefore, recognition tests that rely on familiarity
process might not detect any emotional enhancement effect on memory.
Our recognition task likely involved a substantial familiarity component, because
subj ects were asked to make quick decisions about whether or not they had seen each
stimulus but were not asked to effortfully retrieve details of the original learning
experience. Because emotional arousal appears to have negligible effects on familiarity-
based memory, the fact that we did not observe an emotional enhancement effect may be
due to the reliance of our recognition tests on familiarity rather than recollection.
However, we did Eind a relationship between emotional arousal and picture recognition
performance when we used a continuous predictor of memory performance rather than
the between-groups classification. This suggests that our group classification may be a
more important factor than the issue of recollection versus familiarity.
When we used a continuous predictor of memory performance rather than the
between-groups classification, memory status (represented by percent retention z-score
on the HVLT-R) significantly predicted picture recognition performance at 10 minutes
for all arousal categories, and at 2 weeks only for high arousal categories. This indicates
that memory status seems to be associated with the degree to which people benefit from
emotional arousal at long-term delays (2 weeks). This relationship did not hold at three
months, perhaps due to a floor effect. These Eindings suggest the importance of
examining a "spectrum of aging," because the relationship between emotional arousal
and memory performance was masked when the data were examined using a between-
groups classification. This may have been due to inaccurate classifications ofMCI
and/or control subj ects, which is difficult to avoid given the heterogeneity of MCI and its
overlap with memory changes associated with normal aging.
Finally, the Einding that hippocampal volume was associated with memory
performance for medium and high arousal pictures but not for low arousal pictures is
consistent with the modulation hypothesis of emotional memory. This hypothesis posits
that the amygdala enhances memory consolidation for emotionally arousing stimuli
through its modulatory effects on other brain structures, including the hippocampus. It
appears that larger hippocampal volumes were associated with a greater memory benefit
from emotionally arousing stimuli. Therefore, having an in tact hippocampus may be
necessary for exhibiting emotional enhancement of memory.
Limitations of the Present Study
One limitation of the study to which I have already referred relates to the difficulty
distinguishing MCI subj ects from control subj ects. Our examination of emotional
memory performance using a between-groups classification (MCI versus control)
indicated that neither group benefited from emotional arousal. On the other hand, our
examination of emotional memory performance using a continuous predictor of memory
status revealed a relationship between memory status and the degree to which individuals
benefit from emotional arousal. This discrepancy indicates that our classification of MCI
and control subj ects may not have been entirely accurate. This may be due to the fact
that group classifications were based on performance at a single testing session.
Classification accuracy may have been improved by basing the consensus conference
judgments on performance at multiple testing sessions rather than a single session and by
usmng more memory measures.
It is also possible that the three month testing interval was too long. Some
participants, particularly those with poorer memory status, may have demonstrated a
floor effect at this time. It may have been more useful to perform the final recognition
test at one month. This would have allowed us to examine long-term consolidation
processes before the likelihood of encountering floor effects became too great.
Additional limitations of the study relate to the emotional stimuli employed. The
static photographs may be of limited ecological validity. It will be important to
investigate memory for real-life emotional stimuli and events that have personal
Finally, this study included no imaging measure of amygdala pathology (either
structural or functional). This makes it difficult to draw definitive conclusions about the
role of the amygdala in the emotional enhancement effects observed, and prevents us
from investigating the status of the amygdala in MCI or in individuals with poor general
Directions for Future Research
It appears likely that the recognition format of the picture memory tests employed
in this study required familiarity processes rather than recollective processes. The
findings in the literature that normal controls and Alzheimer' s patients are more likely to
demonstrate emotional enhancement of memory with the use of recall tests than the use
of recognition tests suggests that emotional arousal serves to enhance recollective rather
than familiarity processes. Thus, future studies should investigate emotional memory in
MCI using tasks that tap recollective processes. In fact, we intend to assess the one-year
recognition memory performance of subj ects from this study using a paradigm that
distinguishes between recollection and familiarity. Based on the previously discussed
literature, we expect that these individuals will be more likely to demonstrate emotional
memory enhancement for items of which they have a conscious "recollection" rather than
for items that they perceive as "familiar." We will investigate whether MCI diagnosis or
memory status has any influence on the effect of emotional enhancement of recollective
Future studies should continue to address the difficulty of accurately distinguishing
MCI subjects from controls. Longitudinal studies may examine the fluidity of the MCI
classification over time. Additionally, studies should investigate the utility of examining
performance along a "spectrum of aging" rather than relying solely on the MCI versus
Given the limited ecological validity of emotional and neutral photographs,
emotional memory should be examined further in MCI using other tasks, such as
emotional narratives, or using other paradigms, such as fear conditioning.
Because it is not known with certainty whether or not the amygdala undergoes
significant pathology or atrophy in MCI, future studies should investigate the status of the
amygdala in this condition. Studies investigating amygdala volumetrics and functional
activation are especially relevant for studies of emotional memory in MCI, given the
central role of the amygdala in the emotional enhancement of memory.
In conclusion, this study sought to investigate the effects of emotional arousal on
memory for individuals with mild cognitive impairment and normal controls and to
examine the relationship between hippocampal volume and this enhancement effect.
Although we did not demonstrate the expected emotional enhancement effect with the
use of an MCI versus control group classification, we did show that emotional
enhancement of memory is related to memory status as well as to hippocampal volume.
That is, individuals with better memory status benefited more from emotional arousal at a
long-term memory testing delay (2 weeks), and larger hippocampal volume was
associated with better memory recognition performance for arousing stimuli. This study
suggests that memory function in older adults should be assessed not only according to
MCI versus normal classifications but also along a "spectrum of aging". Additionally,
this study suggests that the use of emotionally arousing stimuli may be a promising
memory aid for older individuals at least for those with good memory status and larger
hippocampal volumes. However, the emotional stimuli employed in this study included
static photographs which may have had limited ecological validity and little personal
relevance to the participants. Therefore, it is entirely possible that emotionally
meaningful stimuli with autobiographical relevance could also be helpful for individuals
who have isolated memory impairments.
LIST OF REFERENCES
Abercrombie, H. C., Kalin, N. H., Thurow, M. E., Rosenkranz, M. A., & Davidson, R. J.
(2003). Cortisol variation in humans affects memory for emotionally laden and
neutral information. Behavioral Neuroscience, 11 7(3), 505-516.
Abrisqueta-Gomez, J., Bueno, O. F., Oliveira, M. G., & Bertolucci, P. H. (2002).
Recognition memory for emotional pictures in Alzheimer's patients. Acta
Neurologica Scand'inavica, 105(1), 51-54.
Adolphs, R., Cahill, L., Schul, R., & Babinsky, R. (1997). Impaired declarative memory
for emotional material following bilateral amygdala damage in humans. Learning
and Memory, 4(3), 291-300.
Army Individual Test Battery (1944). Manual of directions and' scoring. Washington,
D.C.: War Department, Adjutant General's Office.
Arriagada, P. V., Marzloff, K., & Hyman, B. T. (1992). Distribution of Alzheimer-type
pathologic changes in nondemented elderly individuals matches the pattern in
Alzheimer's disease. Neurology, 42(9), 1681-1688.
Barros, D. M., Pereira, P., Medina, J. H., & Izquierdo, I. (2002). Modulation of working
memory and of long- but not short-term memory by cholinergic mechanisms in
the basolateral amygdala. Behavioral Pharmacology, 13(2), 163-167.
Benedict, R. H. B., Schretlen, D., Groninger, L., & Brandt, J. (1998). Hopkins Verbal
Learning Test--Revised: Normative data and analysis of inter-form and test-
retest reliability. The Clinical Neuropsychologist, 12, 43-55.
Bianchin, M., Mello e Souza, T., Medina, J. H., & Izquierdo, I. (1999). The amygdala is
involved in the modulation of long-term memory, but not in working or short-
term memory. Neurobiology of~earning and Memory, 71(2), 127-13 1.
Blair, J. R., & Spreen, O. (1989). Predicting premorbid IQ: A Revision of the National
Adult Reading Test. The Clinical Neuropsychologist, 3, 129-136.
Bottino, C. M., Castro, C. C., Gomes, R. L., Buchpiguel, C. A., Marchetti, R. L., & Neto,
M. R. (2002). Volumetric MRI measurements can differentiate Alzheimer's
disease, mild cognitive impairment, and normal aging. International
Psychogeriatrics, 14(1), 59-72.
Bradley, M. M., & Lang, P. J. (2000). Affective reactions to acoustic stimuli.
Psychophysiology, 37(2), 204-215.
Cahill, L., & Alkire, M. T. (2003). Epinephrine enhancement of human memory
consolidation: interaction with arousal at encoding. Neurobiology ofLearning and
Memory, 79(2), 194-198.
Cahill, L., Babinsky, R., Markowitsch, H. J., & McGaugh, J. L. (1995). The amygdala
and emotional memory. Nature, 377(6547), 295-296.
Cahill, L., Haier, R. J., Fallon, J., Alkire, M. T., Tang, C., Keator, D., et al. (1996).
Amygdala activity at encoding correlated with long-term, free recall of emotional
information. Proceedings of the National Academy of Sciences of the United
States ofAmerica, 93(15), 8016-8021.
Cahill, L., & McGaugh, J. L. (1998). Mechanisms of emotional arousal and lasting
declarative memory. Trends in Neuroscience, 21(7), 294-299.
Callen, D. J., Black, S. E., Gao, F., Caldwell, C. B., & Szalai, J. P. (2001). Beyond the
hippocampus: MRI volumetry confirms widespread limbic atrophy in AD.
Neurology, 57(9), 1669-1674.
Canli, T., Zhao, Z., Brewer, J., Gabrieli, J. D., & Cahill, L. (2000). Event-related
activation in the human amygdala associates with later memory for individual
emotional experience. Journal ofNeuroscience, 20(19), RC99.
Charles, S. T., Mather, M., & Carstensen, L. L. (2003). Aging and emotional memory:
the forgettable nature of negative images for older adults. Journal of Experimental
Psychology: General, 132(2), 310-324.
Convit, A., de Leon, M. J., Tarshish, C., De Santi, S., Kluger, A., Rusinek, H., et al.
(1995). Hippocampal volume losses in minimally impaired elderly. Lancet,
Convit, A., De Leon, M. J., Tarshish, C., De Santi, S., Tsui, W., Rusinek, H., et al.
(1997). Specific hippocampal volume reductions in individuals at risk for
Alzheimer's disease. Neurobiology ofdging. 18(2), 131-138.
Crawford, J. R., Deary, I. J., Starr, J., & Whalley, L. J. (2001). The NART as an index of
prior intellectual functioning: a retrospective validity study covering a 66-year
interval. Psychological M~edicine, 31(3), 45 1-45 8.
de Leon, M. J., Convit, A., DeSanti, S., Bobinski, M., George, A. E., Wisniewski, H. M.,
et al. (1997). Contribution of structural neuroimaging to the early diagnosis of
Alzheimer's disease. InternationalPsychogeriatrics, 9(Supplementl), 183-190;
Denburg, N. L., Buchanan, T. W., Tranel, D., & Adolphs, R. (2003). Evidence for
preserved emotional memory in normal older persons. Emotion, 3(3), 239-253.
Dolcos, F., LaBar, K. S., & Cabeza, R. (2004). Interaction between the amygdala and the
medial temporal lobe memory system predicts better memory for emotional
events. Neuron, 42(5), 855-863.
Dolcos, F., LaBar, K. S., & Cabeza, R. (2005). Remembering one year later: role of the
amygdala and the medial temporal lobe memory system in retrieving emotional
memories. Proceedings of the National Academy of Sciences of the United States
ofAmerica, 102(7), 2626-263 1.
Du, A. T., Schuff, N., Amend, D., Laakso, M. P., Hsu, Y. Y., Jagust, W. J., et al. (2001).
Magnetic resonance imaging of the entorhinal cortex and hippocampus in mild
cognitive impairment and Alzheimer's disease. Journal ofNeurology,
Neurosurgery, and Psychiatry, 71(4), 441-447.
Fastenau, P. S., Denburg, N. L., & Mauer, B. A. (1998). Parallel short forms for the
Boston Naming Test: psychometric properties and norms for older adults. Journal
of Clinical and Experimental Neuropsychology, 20(6), 828-834.
Fisher, N. J., Tierney, M. C., Snow, W. G., & Szalai, J. P. (1999). Odd/Even short forms
of the Boston Naming Test: preliminary geriatric norms. The Clinical
Neuropsychologist, 13(3), 359-364.
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(3), 189-198.
Franzen, M. D., Tishelman, A. C., Sharp, B. H., & Friedman, A. G. (1987). An
investigation of the test-retest reliability of the Stroop Color-Word Test across
two interval s. Archives of Clinical Neuropsychology, 2(3), 265-272.
Frey, S., Bergado-Rosado, J., Seidenbecher, T., Pape, H. C., & Frey, J. U. (2001).
Reinforcement of early long-term potentiation (early-LTP) in dentate gyrus by
stimulation of the basolateral amygdala: heterosynaptic induction mechanisms of
late-LTP. Journal ofNeuroscience, 21(10), 3 697-3 703 .
Gilewski, M. J., Zelinski, E. M., & Schaie, K. W. (1990). The Memory Functioning
Questionnaire for assessment of memory complaints in adulthood and old age.
Psychology and A ing. 5(4), 482-490.
Golden, C. J. (1978). Stroop Color and Word Test. Chicago, 1L: Stoelting.
Golomb, J., de Leon, M. J., Kluger, A., George, A. E., Tarshish, C., & Ferris, S. H.
(1993). Hippocampal atrophy in normal aging. An association with recent
memory impairment. Archives ofNeurology, 50(9), 967-973.
Hamann, S. B., Ely, T. D., Grafton, S. T., & Kilts, C. D. (1999). Amygdala activity
related to enhanced memory for pleasant and aversive stimuli. Nature
Neuroscience, 2(3), 289-293.
Hamann, S. B., Monarch, E. S., & Goldstein, F. C. (2000). Memory enhancement for
emotional stimuli is impaired in early Alzheimer's disease. Neuropsychology,
Hensel, A., Wolf, H., Dieterlen, T., Riedel-Heller, S., Arendt, T., & Gertz, H. J. (2005).
Morphometry of the amygdala in patients with questionable dementia and mild
dementia. Journal of the Neurological Sciences, 238(1-2), 7 1-74.
Horinek, D., Petrovicky, P., Hort, J., Krasensky, J., Brabec, J., Bojar, M., et al. (2006).
Amygdalar volume and psychiatric symptoms in Alzheimer's disease: an MRI
analy sis. Acta Neurologica Scand'inavica, 113(1), 40-45.
Huppert, F. A., & Piercy, M. (1979). Normal and abnormal forgetting in organic amnesia:
effect of locus of lesion. Cortex, 15(3), 385-390.
Ishii, H., Meguro, K., Yamaguchi, S., Hirayama, K., Tabuchi, M., Mori, E., et al. (2006).
Different MRI findings for normal elderly and very mild Alzheimer's disease in a
community: Implications for clinical practice The Tajiri Proj ect. Archives of
Gerontology and' Geriatrics, 42(1), 59-71.
Ivnik, R. J., Malec, J. F., Smith, G. E., Tangalos, E. G., & Petersen, R. C. (1996).
Neuropsychological Tests' Norms Above Age 55: COWAT, BNT, MAE Token,
WRAT-R Reading, AMNART, Stroop, TMT, and JLO. The Clinical
Neuropsychologist, 10(3), 262-278.
Jack, C. R., Jr., Petersen, R. C., Xu, Y. C., O'Brien, P. C., Smith, G. E., Ivnik, R. J., et al.
(1999). Prediction of AD with MRI-based hippocampal volume in mild cognitive
impairment. Neurology, 52(7), 1397-1403.
Jack, C. R., Jr., Shiung, M. M., Gunter, J. L., O'Brien, P. C., Weigand, S. D., Knopman,
D. S., et al. (2004). Comparison of different MRI brain atrophy rate measures
with clinical disease progression in AD. Neurology, 62(4), 591-600.
Jacoby, L. L. (1991). A process dissociation framework: Separating automatic from
intentional uses of memory. Journal of2~emory and Language, 30(5), 513-541.
Jones, R. N., & Gallo, J. J. (2000). Dimensions of the Mini-Mental State Examination
among community dwelling older adults. Psychological M~edicine, 30(3), 605-
Kaplan, E., Goodglass, H., & Weintraub, S. (1983). The Boston Naming Test. USA: Lea
Kazui, H., Mori, E., Hashimoto, M., & Hirono, N. (2003). Enhancement of declarative
memory by emotional arousal and visual memory function in Alzheimer's disease.
Journal ofNeuropsychiatry and Clinical Neuroscience, 15(2), 221-226.
Kazui, H., Mori, E., Hashimoto, M., Hirono, N., Imamura, T., Tanimukai, S., et al.
(2000). Impact of emotion on memory. Controlled study of the influence of
emotionally charged material on declarative memory in Alzheimer's disease.
British Journal ofPsychiatry, 177, 343-347.
Kensinger, E. A., Anderson, A., Growdon, J. H., & Corkin, S. (2004). Effects of
Alzheimer disease on memory for verbal emotional information.
Neuropsychologia, 42(6), 791-800.
Kensinger, E. A., Brierley, B., Medford, N., Growdon, J. H., & Corkin, S. (2002). Effects
of normal aging and Alzheimer's disease on emotional memory. Emotion, 2(2),
Kensinger, E. A., & Corkin, S. (2004). Two routes to emotional memory: distinct neural
processes for valence and arousal. Proceedings of the National Academy of
Sciences of the United States ofAmerica, 101(9), 3310-3315.
Krasuski, J. S., Alexander, G. E., Horwitz, B., Daly, E. M., Murphy, D. G., Rapoport, S.
I., et al. (1998). Volumes of medial temporal lobe structures in patients with
Alzheimer's disease and mild cognitive impairment (and in healthy controls).
Biological Psychiatry, 43(1), 60-68.
Lang, P. J., Bradley, M. M., & Cuthbert, B. (2001). The international affective picture
system (photographic slides). Gainesville, FL: The Center for Research in
Psychophysiology, University of Florida.
Lawton, M. P., & Brody, E. M. (1969). Assessment of older people: self-maintaining and
instrumental activities of daily living. Gerontologist, 9(3), 179-186.
Lyness, J. M., Noel, T. K., Cox, C., King, D. A., Conwell, Y., & Caine, E. D. (1997).
Screening for depression in elderly primary care patients. A comparison of the
Center for Epidemiologic Studies-Depression Scale and the Geriatric Depression
Scale. Archives oflnternal M~edicine, 157(4), 449-454.
McGaugh, J. L. (2004). The amygdala modulates the consolidation of memories of
emotionally arousing experiences. AnnualReview ofNeuroscience, 27(9), 1-28.
McGaugh, J. L., Cahill, L., & Roozendaal, B. (1996). Involvement of the amygdala in
memory storage: interaction with other brain systems. Proceedings of the
National Academy of Sciences of the thrited States ofAnzerica, 93(24), 1 3508-
Moayeri, S. E., Cahill, L., Jin, Y., & Potkin, S. G. (2000). Relative sparing of emotionally
influenced memory in Alzheimer's disease. Neuroreport, 11(4), 653-655.
Morris, J. C. (2005). Mild cognitive impairment and preclinical Alzheimer's disease.
Geriatrics, Supplement, 9-14.
Mosconi, L., Tsui, W. H., De Santi, S., Li, J., Rusinek, H., Convit, A., et al. (2005).
Reduced hippocampal metabolism in MCI and AD: automated FDG-PET image
analysis. Neurology, 64(11), 1860-1867.
Ochsner, K. N. (2000). Are affective events richly recollected or simply familiar? The
experience and process of recognizing feelings past. Journal ofExperintental
Psychology: General, 129(2), 242-261.
Petersen, R. C., Doody, R., Kurz, A., Mohs, R. C., Morris, J. C., Rabins, P. V., et al.
(2001). Current concepts in mild cognitive impairment. Archives ofNeurology,
Petersen, R. C., Jack, C. R., Jr., Xu, Y. C., Waring, S. C., O'Brien, P. C., Smith, G. E., et
al. (2000). Memory and MRI-based hippocampal volumes in aging and AD.
Neurology, 54(3), 581-587.
Petersen, R. C., Smith, G. E., Waring, S. C., Ivnik, R. J., Tangalos, E. G., & Kokmen, E.
(1999). Mild cognitive impairment: clinical characterization and outcome.
Archives ofNeurology, 56(3), 303-308.
Petrovich, G. D., Canteras, N. S., & Swanson, L. W. (2001). Combinatorial amygdalar
inputs to hippocampal domains and hypothalamic behavior systems. Brain
Research Reviews, 38(1-2), 247-289.
Pitkanen, A., Pikkarainen, M., Nurminen, N., & Ylinen, A. (2000). Reciprocal
connections between the amygdala and the hippocampal formation, perirhinal
cortex, and postrhinal cortex in rat. A review. Annals of the New York Academy of
Sciences, 911, 369-391.
Psychological Corporation (1999). Wechsler Abbreviated Scale oflntelligence. San
Antonio, TX: Harcourt Brace & Company.
Rasmussen, D.X., Bylsma, F.W., & Brandt, J. (1995). Stability of performance on the
Hopkins Verbal Learning Test. Archives of Clinical Neuropsychology, 10, 21-26.
Reisberg, D., Heuer, F., McLean, J., & O'Shaughnessy, M. (1988). The quantity, not the
quality, of affect predicts memory vividness. Bulletin of the Psychonomic Society,
Roozendaal, B., Nguyen, B. T., Power, A. E., & McGaugh, J. L. (1999). Basolateral
amygdala noradrenergic influence enables enhancement of memory consolidation
induced by hippocampal glucocorticoid receptor activation. Proceedings of the
National Academy of Sciences of the United States of America, 96(20), 1 1642-
Rosen, W. G. (1980). Verbal fluency in aging and dementia. Journal of Clinical
Neuropsychology, 2, 135-146.
Schafe, G. E., & LeDoux, J. E. (2000). Memory consolidation of auditory pavlovian fear
conditioning requires protein synthesis and protein kinase A in the amygdala.
Journal ofNeuroscience, 20(18), RC96.
Shapiro, A. M., Benedict, R. H., Schretlen, D., & Brandt, J. (1999). Construct and
concurrent validity of the Hopkins Verbal Learning Test-revised. The Clinical
Neuropsychologist, 13(3), 348-358.
Sheikh, J. I., & Yesavage, J. A. (1986). Geriatric Depression Scale (GDS): recent
evidence and development of a shorter version. In T. L. Brink (Ed.), Clinical
gerontology: A guide to assessment and intervention (pp. 165-173). New York:
The Haworth Press.
Soininen, H. S., Partanen, K., Pitkanen, A., Vainio, P., Hanninen, T., Hallikainen, M., et
al. (1994). Volumetric MRI analysis of the amygdala and the hippocampus in
subjects with age-associated memory impairment: correlation to visual and verbal
memory. Neurology, 44(9), 1660-1668.
Spreen, O., & Benton, A. L. (1977). Neurosensory center comprehensive examination for
aphasia: Manual of instructions. Victoria, BC: University of Victoria.
Spreen, O., & Strauss, E. (1991). A compendium ofneuropsychological tests:
Administration, norms and commentary. New York: Oxford University Press.
Stanley, M. A., Beck, J. G., & Zebb, B. J. (1996). Psychometric properties of four anxiety
measures in older adults. Behavior Research and Therapy, 34(10), 827-83 8.
Tombaugh, T. N., Kozak, J., & Rees, L. (1999). Normative data stratified by age and
education for two measures of verbal fluency: FAS and animal naming. Archives
of Clinical Neuropsychology, 14(2), 1 67- 177.
Uttl, B. (2002). North American Adult Reading Test: age norms, reliability, and validity.
Journal of Clinical and Experimental Neuropsychology, 24(8), 1 123-1 137.
Visser, P. J., Scheltens, P., & Verhey, F. R. (2005). Do MCI criteria in drug trials
accurately identify subjects with predementia Alzheimer's disease? Journal of
Neurology, Neurosurgery and Psychiatry, 76(10), 1348-1354.
Visser, P. J., Verhey, F. R., Hofman, P. A., Scheltens, P., & Jolles, J. (2002). Medial
temporal lobe atrophy predicts Alzheimer's disease in patients with minor
cognitive impairment. Journal ofNeurology, Neurosurgery and Psychiatry, 72(4),
Watson, L. C., & Pignone, M. P. (2003). Screening accuracy for late-life depression in
primary care: a systematic review. Journal ofFamnily Practice, 52(12), 956-964.
Whitehouse, P. J., & Juengst, E. T. (2005). Antiaging medicine and mild cognitive
impairment: practice and policy issues for geriatrics. Journal of the American
Geriatric Society, 53(8), 1417-1422.
Ann Mikos was born in Milwaukee, WI, and received her B.A. in psychology from
Grinnell College. She obtained subsequent research experience at the University of lowa
Huntington's Disease Center of Excellence. She is currently pursuing her doctorate in
clinical psychology, with a specialty in neuropsychology, at the University of Florida.
Current research interests include aging and mild cognitive impairment, emotional
memory, facial expressivity, and Parkinson's disease.