University of Southern California Dissertations and Theses

Effects of estradiol on cortisol response, working memory, and emotional memory during stress in young and post-menopausal women

(USC Thesis Other)

Effects of estradiol on cortisol response, working memory, and emotional memory during stress in young and post-menopausal women

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Effects of estradiol on cortisol response, working memory, and emotional memory during stress
in young and post-menopausal women

By

Alexandra E. Ycaza

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PSYCHOLOGY)

May 2014

Copyright 2014 Alexandra E. Ycaza
i
Dedication
This dissertation is dedicated to my parents, Maria and Tim Noone, for teaching me that
anything worth doing requires hard work and dedication. Thank you for your unwavering
support in all I do. Also to my fiancé, Alex Herrera. Your work ethic and dedication inspire me
every day, and remind me that everything I strive for is within my reach.

ii
Acknowledgements
First, I would like to thank my committee members, Mara Mather, Margaret Gatz,
Christian Pike, and David Walsh. My thanks to David Walsh for his guidance with statistical
analyses, particularly when I transitioned from animal models to human-subjects research, and to
Christian Pike for being a sounding board for my understanding of hormone systems and
mechanisms, as well as for organizing an initial meeting with the ELITE trial, without which
Study 1 would not have been possible. I would like to thank Margaret Gatz for her unwavering
support, assistance, and belief in me throughout my time in graduate school. Most of all, I would
like to thank my advisor, Mara Mather, for her complete support of my research interests. Her
guidance and trust in my abilities have shaped who am I, and have to yet to become, as a
scientist.
Second, I would like to thank Dr. Howard Hodis and Dr. Wendy Mack for making the
ELITE trial available for Study 1. Similarly, I would like to thank all of the wonderful women
enrolled in the ELITE trial who so graciously participated in this stress study. I also would like to
thank all the young women who participated in Study 2. This dissertation work was supported by
the National Institute on Aging R01AG038043.

iii
Table of Contents

List of Figures ........................................................................................................................ix

Abstract ..................................................................................................................................1

Chapter 1. Estrogen And Stress. Interactions And Implications For Postmenopausal
Women ...................................................................................................................................3

1.1 Introduction ......................................................................................................................3

1.2 Overviews Of The Hypothalamic-Pituitary-Adrenal And
Hypothalamic-Pituitary-Gonadal Axes And Their Interactions ............................................4

1.2.1 The Hypothalamic-Pituitary-Adrenal Axis: The Stress System ...................................4

1.2.2 The Hypothalamic-Pituitary-Gonadal Axis: The Estradiol System ..............................5

1.2.3 HPA And HPG Cross-Talk: Basis For Stress And Estradiol Interactions ....................7

1.3 The Competing Effects Of Stress And Estrogens On Neuronal Morphology
And Function .........................................................................................................................8

1.3.1 Glucocorticoids Are Necessary For Normal Neuronal Function ..................................8

1.3.2 Exposure To Glucocorticoids Exacerbates Damage By Other Insults .........................9

1.3.3 Excess Exposure To Glucocorticoids Alone Is Dangerous To Neural Tissue ..............10

1.3.4 The Effects Of Estrogens On Neurogenesis And Neural Maintenance ........................11

1.3.5 Estrogens As A Neuroprotectant ..................................................................................12

1.3.6 Estrogens And Neurodegenerative Diseases ................................................................13

1.4 The Competing Effects Of Stress And Estrogens On Cognition .....................................14

1.4.1 Stress Can Benefit Cognitive Function .........................................................................14

1.4.2 Associations Between Stressful Life Events And Cognition In Humans .....................16

1.4.3 Negative Effects Of Glucocorticoids On Cognition In Animals ..................................17

1.4.4 Negative Effects Of Stress Application On Cognition In Animals ..............................18

1.4.5 Negative Effects Of Controlled Non-Laboratory Stressors On Cognition
iv
In Humans ..............................................................................................................................18

1.4.6 Negative Effects Of Laboratory Stressors On Cognition In Humans ...........................19

1.4.7 How Are Glucocorticoids And Stress Affecting Cognition .........................................20

1.4.8 Estrogen Can Impair Cognitive Function .....................................................................22

1.4.9 Positive Effects Of Estradiol On Cognition In Young-Adult Female Animals ............25

1.4.10 Effects Of Menstrual Cycle On Cognition In Young-Adult Human Females ............26

1.4.11 Positive Effects Of Estrogen On Cognition In Aging Female Animals .....................27

1.4.12 Positive Effects Of Estrogens On Cognition In Aging Human Females ....................28

1.4.13 How Is Estradiol Affecting Cognition? ......................................................................30

1.5 Estradiol-Stress Interactions ............................................................................................31

1.5.1 Estrogenic Protection Against Glucocorticoid Insults ..................................................31

1.5.2 Estrogenic Modulation Of The Stress Response ..........................................................32

1.5.3 Implications For Young-Adult Women ........................................................................34

1.5.4 Implications For Aging Women ...................................................................................35

1.6 Conclusions ......................................................................................................................37

Chapter 2. Study 1: Influence Of Estradiol On The Stress Response, And Stress Effects
On Working Memory And Emotional Memory ....................................................................40

2.1 Introduction ......................................................................................................................40

2.2 Methods............................................................................................................................43

2.2.1 Participants ....................................................................................................................43

2.2.2 Inclusionary And Exclusionary Criteria .......................................................................43

2.2.3 Sessions .........................................................................................................................44

2.2.4 Hormone Sampling .......................................................................................................44

2.2.5 Stress Manipulation ......................................................................................................45
v

2.2.6 Psychological Measures ................................................................................................45

2.2.7 Behavioral Tasks ...........................................................................................................46

2.2.8 Emotional Memory Task: Encoding Phase ...................................................................46

2.2.9 Working Memory Task .................................................................................................47

2.2.10 Emotional Memory Task: Recall Phase ......................................................................49

2.2.11 Emotional Memory Task: Association Test Phase .....................................................49

2.2.12 Estradiol Condition And Analyses ..............................................................................49

2.2.13 Statistics ......................................................................................................................50

2.3 Results ..............................................................................................................................50

2.3.1 Participants: Demographic Information For Low E2

Versus High E2

Women ............50

2.3.2 Correlation Analyses Between Estradiol Levels And Questionnaire Responses .........51

2.3.3 Subjective Ratings Of Stress And Pain Immediately Before And After
Stress Exposure ......................................................................................................................51

2.3.4 Cortisol Response For Low E2

Versus High E2

Women During
The Stress And Control Sessions ...........................................................................................53

2.3.5 Working Memory: Word Recall ...................................................................................54

2.3.6 Emotional Memory: Free Recall Of Emotional And Neutral Pictures .........................55

2.3.7 Emotional Memory: Memory For Picture Location .....................................................56

2.4 Discussion ........................................................................................................................57

2.4.1 Differences In Overall Health Ratings And Education Between
HE And LE Women ...............................................................................................................57

2.4.2 Subjective Ratings Of Pre-Water Stress Levels Between LE And HE Women ...........58

2.4.3 Baseline Cortisol Levels In LE And HE Women .........................................................59

2.4.4 Cortisol Response In LE And HE Women ...................................................................60

vi
2.4.5 Effects Of Stress On Working Memory And Emotional Memory
In LE And HE Women ..........................................................................................................62

2.4.6 Conclusion ....................................................................................................................64

Chapter 3. Study 2: Influence Of Estradiol Fluctuations During The Menstrual Cycle On The
Stress Response, And Stress Effects On Working Memory And Emotional Memory ..........66

3.1 Introduction ......................................................................................................................66

3.2 Methods............................................................................................................................67

3.2.1 Participants ....................................................................................................................67

3.2.2 Inclusionary And Exclusionary Criteria .......................................................................67

3.2.3 Sessions .........................................................................................................................68

3.2.4 Hormone Sampling .......................................................................................................69

3.2.5 Stress Manipulation ......................................................................................................70

3.2.6 Psychological Measures ................................................................................................70

3.2.7 Behavioral Tasks ...........................................................................................................71

3.2.8 Emotional Memory Task: Encoding Phase ...................................................................71

3.2.9 Working Memory Task .................................................................................................72

3.2.10 Emotional Memory Task: Recall Phase ......................................................................73

3.2.11 Emotional Memory Task: Association Test Phase .....................................................74

3.2.12 Statistics ......................................................................................................................74

3.3 Results ..............................................................................................................................75

3.3.1 Participants: Demographic Information ........................................................................75

3.3.2 Correlation Analyses Between Cortisol Levels And Questionnaire Responses ...........75

3.3.3 T-Tests Between EF And LF Phases On Questionnaire Responses .............................77

3.3.4 Subjective Ratings Of Stress And Pain Immediately Before
And After Stress Exposure .....................................................................................................77
vii

3.3.5 Cortisol Response During The EF And LF Phases .......................................................77

3.3.6 Working Memory: Word Recall ...................................................................................81

3.3.7 Emotional Memory: Free Recall Of Emotional And Neutral Pictures .........................81

3.3.8 Emotional Memory: Memory For Picture Location .....................................................82

3.3.9 Cortisol Response Between Responders And Nonresponders During The EF
And LF Phases .......................................................................................................................83

3.3.10 Working Memory: Word Recall In Responders Versus Nonresponders During The
EF And LF Phases .................................................................................................................86

3.3.11 Emotional Memory: Free Recall Of Emotional And Neutral Pictures In Responders
Versus Nonresponders During The EF And LF Phases .........................................................88

3.3.12 Emotional Memory: Memory For Picture Location In Responders Versus
Nonresponders During The EF And LF Phases .....................................................................89

3.4 Discussion ........................................................................................................................91

3.4.1 Differences In Subjective Ratings Of Pain ...................................................................92

3.4.2 Cortisol Response To A Stressful Event During The EF And LF Phases ....................92

3.4.3 Relationship Between Stress And Working Memory During The EF And
LF Phases ...............................................................................................................................94

3.4.4 Relationship Between Stress And Emotional Memory During The EF And
LF Phases ...............................................................................................................................96

3.4.5 Conclusion ....................................................................................................................100

Chapter 4. Estradiol Differentially Alters The Effects Of Stress In Post-Menopausal
And Young Spontaneously Cycling Women .........................................................................102

4.1 Introduction ......................................................................................................................102

4.2 Differences In Cortisol Response To An Ice Water Stressor In Young
Spontaneously Cycling And Post-Menopausal Women ........................................................104

4.3 How Estradiol Is Preventing Stress From Interfering With Working Memory In Young
Spontaneously Cycling And Post-Menopausal Women ........................................................106

viii
4.4 Differences In The Effects Of Stress On Emotional Memory In Young Spontaneously
Cycling And Post-Menopausal Women .................................................................................107

4.5 Closing Remarks ..............................................................................................................109

References ..............................................................................................................................110

Appendix A: Tables ...............................................................................................................132

Appendix B: Figures ..............................................................................................................137

ix
List of Figures

Table 2.1: Average Timeline For Study 1 Sessions ...............................................................132

Table 2.2: Sex Hormone Levels, Demographics, Emotional State, And Mood ....................133

Table 3.1: Average Timeline For Study 2 Sessions ...............................................................135

Table 3.2: Emotional State, Mood, And PMS Symptoms During First Session
Of Each Phase ........................................................................................................................136

Figure 1.1: The Hypothalamic Pituitary Adrenal Axis ..........................................................137

Figure 1.2: The Hypothalamic Pituitary Gonadal Axis .........................................................138

Figure 2.1: Subjective Ratings Of Stress Before And After Water Exposure .......................139

Figure 2.2: Subjective Ratings Of Pain Before And After Water Exposure .........................140

Figure 2.3: Cortisol During The Stress And Control Sessions Collapsed Across
HE And LE Women ...............................................................................................................141

Figure 2.4: Cortisol Levels In HE And LE Women Collapsed Across Sessions ..................142

Figure 2.5: Cortisol Levels In HE And LE Women During The Stress Session Only ..........143

Figure 2.6: Working Memory Performance In HE And LE Women ....................................144

Figure 2.7: Picture Recall Collapsed Across Session And Estradiol: Emotional Versus
Neutral Pictures ......................................................................................................................145

Figure 2.8: Emotional Memory Picture Recall During The Stress And Control Sessions ....146

Figure 2.9: Recall Of Emotional Pictures By All Women During The Stress Session .........147

Figure 2.10: Recall Of Emotional Pictures By LE Women In The Stress Session ...............148

Figure 3.1: Subjective Ratings Of Stress Before And After Water Exposure .......................149

Figure 3.2: Subjective Ratings Of Pain Before And After Water Exposure .........................150

Figure 3.3: Cortisol Levels Collapsed Across Phase And All 3 Time Points ........................151

Figure 3.4: Cortisol Levels Collapsed Across Session And Phase ........................................152

Figure 3.5: Cortisol Levels Collapsed Across Phase .............................................................153
x

Figure 3.6: Cortisol Levels During The EF Phase: Collapsed Across Time Points ..............154

Figure 3.7: Cortisol Levels During The EF Phase: Baseline Versus 15m Post Onset ...........155

Figure 3.8: Cortisol Levels During The LF Phase: Collapsed Across Time Points ..............156

Figure 3.9: Cortisol Levels During The LF Phase: Baseline Versus 15m Post Onset ...........157

Figure 3.10: Picture Recall Across Stress And Valence: Emotional Versus Neutral
Pictures ...................................................................................................................................158

Figure 3.11: Picture Recall Across Stress And Valence: Negative Versus Neutral
Pictures ...................................................................................................................................159

Figure 3.12: Picture Recall Across Stress: Positive Versus Neutral Pictures ........................160

Figure 3.13: Picture Recall Across Stress: Negative Versus Positive Pictures .....................161

Figure 3.14: Picture Location Memory Across Phase: Emotional Versus Neutral
Pictures ...................................................................................................................................162

Figure 3.15: Picture Location Memory Across Stress And Valence: Negative Versus
Neutral Pictures ......................................................................................................................163

Figure 3.16: Picture Location Memory Across Phase: Negative Versus Neutral
Pictures ...................................................................................................................................164

Figure 3.17: Cortisol Levels Between Responders And Nonresponders During The EF
Phase: Stress Versus Control Sessions ..................................................................................165

Figure 3.18: Cortisol Levels Between Responders And Nonresponders During The LF
Phase: Stress Versus Control Sessions ..................................................................................166

Figure 3.19: Cortisol Levels In Responders And Nonresponders During The Stress
Session In The EF Phase ........................................................................................................167

Figure 3.20: Cortisol Levels In Responders And Nonresponders During The Stress
Session In The LF Phase ........................................................................................................168

Figure 3.21: Cortisol Levels In Responders In The EF Phase: Stress Versus Control
Sessions ..................................................................................................................................169

Figure 3.22: Cortisol Levels In Responders In The LF Phase: Stress Versus Control
Sessions ..................................................................................................................................170

xi
Figure 3.23: Working Memory Performance In The EF Phase: Stress Versus Control
Session ...................................................................................................................................171

Figure 3.24: Working Memory Performance In Responders During The EF Phase .............172

Figure 3.25: Working Memory Performance In Responders During The LF Phase .............173

Figure 3.26: Picture Location Memory In The LF Phase: Positive Versus
Neutral Pictures ......................................................................................................................174
1

Abstract
Estradiol and the class of stress hormones called glucocorticoids exert contrasting effects
on various systems throughout the body, including neural tissue and cognition. Evidence also
exists showing that estradiol can mitigate the damaging effects of excessive glucocorticoid levels
on neural tissue. Given the sharp decline in estradiol levels that characterize the menopausal
transition in human females, it is important to understand if the loss of estradiol leaves post-
menopausal women at a higher risk of the negative effects of stress on neural tissue, and by
extension the negative effects of stress on cognition.
Studies do show that estradiol treatment after menopause can dampen the physiological
stress response to a stressful event; however, it is less clear whether estradiol can dampen the
effects of stress on cognition. The general aim of this dissertation was to examine whether, and
to what extent, estradiol could blunt the stress response and the effects of stress on working
memory and emotional memory in women. This was examined in a population of post-
menopausal women taking estradiol or placebo through the ELITE trial (Study 1) and in a
population of young, spontaneously cycling, women during the low-estradiol and high-estradiol
phases of the menstrual cycle (Study 2).
Study 1 investigated the effects of estradiol treatment after menopause on the cortisol
response to the cold pressor task and the effects of stress on working memory and emotional
memory. It was revealed that higher estradiol levels, as a result of estradiol treatment after
menopause, were associated with a blunted cortisol response to ice water exposure and
protection against stress-induced impairment of working memory. Although no effects of stress,
estradiol, or interactions were found for emotional memory.
2
Study 2 investigated the effects of estradiol fluctuations during the menstrual cycle in
young, spontaneously cycling women on the cortisol response to the cold pressor task and the
effects of stress on working memory and emotional memory. It was revealed that the late
follicular, higher estradiol, phase of the menstrual cycle was associated with a larger cortisol
response to ice water exposure, but still provided protection against stress-induced impairment of
working memory. Although no effects of stress, estradiol, or interactions were found for
emotional memory. The potential mechanisms involved leading to the different patterns of
results are discussed.
Overall, this dissertation provides evidence that estradiol treatment after menopause can
mitigate the effects of stress on cortisol release and working memory, and that the pattern of
estradiol effects on stress may differ in young premenopausal women. The mechanisms that
contribute to the differences may be related to the different physiological relationships between
the stress response system and the estradiol system across the two age groups, and how this may
influence the cortisol response and how estradiol and cortisol interact. These results can further
inform the medical field on the effects of estradiol treatment after menopause, as well as help
women understand their vulnerabilities to stress depending on their internal hormone milieu.

3
CHAPTER 1
Estrogen and Stress. Interactions and implications for postmenopausal women.
1.1 Introduction
Estrogen and stress, particularly the class of stress hormones known as glucocorticoids
(e.g. corticosterone in rodents and cortisol in humans), are two hormone systems that may not be
considered to influence one another. However, cross-communication between these two systems
is well documented. Yet, despite this documentation, little work examines the vastly contrasting
effects the two systems individually exert on the body, brain, and cognition. First, stress
hormones are associated with numerous maladaptive effects. For example, long-term stress
hormone exposure is linked to development of the metabolic syndrome (Pasquali, Vicennati,
Cacciari, & Pagotto, 2006; Rosmond, 2005), unhealthy alterations in fat distribution (Rebuffe-
Scrive, Walsh, McEwen, & Rodin, 1992), promotion of hyperglycemia and hyperinsulinemia
(McGuinness et al., 1993; Rebuffe-Scrive et al., 1992), promotion of bone resorption (O'Brien et
al., 2004), and maintenance of bone degrading osteoclasts (Jia, O'Brien, Stewart, Manolagas, &
Weinstein, 2006). In contrast, the primary estrogen in women, estradiol, is linked to less
unhealthy fat distribution (Green et al., 2004; Musatov et al., 2007), lesser occurrence of
hyperglycemia and hyperinsulinemia (Krotkiewski, Bjorntorp, Sjostrom, & Smith, 1983;
Musatov et al., 2007), and promotion and maintenance of bone mineral density (Delmas et al.,
1997; Felson et al., 1993; Sowers et al., 1998). Further, chronic or extreme stress results in cell
damage or death and impairments in cognitive performance, while estradiol promotes neural
growth and protection, and improvement of cognitive function.
Because stress and estradiol are present in the everyday lives of women, it is important to
consider the dramatically different effects they have on neural tissue and cognition. Stress, or any
4
experience that is emotionally or physiologically challenging (McEwen, 2007), is present
throughout the lifespan, while estradiol presence changes across the lifespan. Women experience
monthly fluctuations of estradiol during their reproductive years, until they reach menopause, at
which time the ovaries stop producing estradiol. Given that the stress and estradiol hormone
systems interact, the dramatic decrease in estradiol levels during menopause may alter
functioning of the stress system. This alteration may leave post-menopause women more
vulnerable to the above-described negative effects of stress exposure.
In the following sections, I will briefly review the stress and estradiol hormone systems,
then discuss the effects of glucocorticoids, stress, and estrogen on neural tissue and cognition,
and how these two hormone systems interact. I will show that these factors exert dramatically
different effects in both the brain and cognition, and the interaction of the two hormone systems
may have important day-to-day implications for both pre- and post-menopausal women.
1.2 Overviews of the Hypothalamic-Pituitary-Adrenal and Hypothalamic-Pituitary-
Gonadal Axes and their Interactions
The hormone systems governing the stress response and reproductive function are
complicated and include many different components. The following sections will provide brief
overviews of the major components of each system. These brief overviews will provide the basic
understanding of the systems needed to review this work, but are by no means comprehensive
discussions of the many intricacies of these two systems or their interactions.
1.2.1 The Hypothalamic-Pituitary-Adrenal Axis: The Stress System (Figure 1.1)
The hypothalamic-pituitary-adrenal (HPA) axis is the primary response system to a
stressor. When an emotional or physiological challenge is detected, the paraventricular nucleus
of the hypothalamus releases corticotropin-releasing hormone into the portal system (CRH; also
5
known as corticotropin-releasing factor) where the neuropeptide is carried to, and acts on, the
anterior pituitary. Delivery of CRH to the anterior pituitary causes release of
andrenocorticotropic hormone (ACTH). ACTH then enters the bloodstream and travels to the
adrenal glands, which sit atop the kidneys. ACTH causes the adrenal cortex (comprised of the
outer layers of the adrenal gland) to release glucocorticoids and the adrenal medulla (the most
medial layer of the adrenal gland) to the release the catecholamines epinephrine and
norepinephrine (Lupien, McEwen, Gunnar, & Heim, 2009).
This work is interested in the effects of glucocorticoids on the system. In humans, the
primary glucocorticoid released during the stress response is cortisol. Cortisol provides access to
immediate energy by tapping into energy stores, which prepares the individual for a flight-or-
flight response. For instance, cortisol helps break down protein for conversion to glucose, assists
in converting fat to usable energy, increases blood to skeletal muscles for energy, and decreases
blood flow to immediately nonessential systems, like the gut (Carlson, 2010).
Importantly, cortisol also is responsible for terminating the HPA-axis-initiated stress
response via negative feedback. In addition to exerting inhibitory action on the hypothalamus
and pituitary gland, cortisol also acts on brain regions outside of the HPA axis to shut down the
stress response. Of the regions containing glucocorticoid receptors capable of initiating
inhibition, the hippocampus contains the highest concentrations and is considered to be the major
source of HPA axis inhibition.
1.2.2 The Hypothalamic-Pituitary-Gonadal Axis: The Estradiol System (Figure 1.2)
Unlike the HPA axis, which is anatomically and mechanistically the same regardless of
sex, the hypothalamic-pituitary-gonadal (HPG) axis differs for males and females, with the final
6
target of the axis being the testes or ovaries, respectively. Because this work is concerned with
females, only the female HPG axis will be discussed.
The HPA and HPG axes follow the same general pathway before reaching their terminal
tissues (i.e., the adrenal gland or gonads, respectively). The HPG axis also begins in the
hypothalamus, where neurons that produce and release gonadotropin-releasing hormone (GnRH)
reside. However, these neurons are located in multiple subnuclei, including the arcuate nucleus
and the sexually dimorphic preoptic area, rather than just the paraventricular nucleus. GnRH-
producing neurons project to the median eminence where GnRH is released into the portal
system and carried to the anterior pituitary. Delivery of GnRH to the anterior pituitary causes
release of luteinizing hormone (LH) and follicle stimulating hormone (FSH), which travel
through the bloodstream and act on the ovaries to cause release of estrogens and progestins.
There are three major forms of estrogens: estrone, estradiol, and estriol. Estradiol (E2) is the
primary estrogen and is predominantly responsible for the estrogenic effects observed in the
brain (Purves et al., 2004).
Estrogens, particularly E2, exert both stimulating and inhibiting effects on the HPG axis
(Herbison, 1998; Jones, 2009). Estradiol levels maintained during most of the menstrual cycle
exert a constant inhibitory effect on the hypothalamus, limiting elevations in LH and FSH.
However, large increases in estradiol lasting for 2 or more days, such as the pre-ovulatory
estradiol increase observed during the menstrual cycle, increases the sensitivity of LH and FSH
neurons to GnRH leading to LH and FSH release. Reductions in E2 after this surge return GnRH,
LH, and FSH to more common moderate-to-low levels (Jones, 2009).

7
1.2.3 HPA and HPG Cross-talk: Basis for Stress and Estradiol Interactions
Cross-talk between the HPA and HPG axes is well documented. The most commonly
discussed functional cross-communication between the two systems is the ability of stress to
interfere with normal reproductive function (Selye, 1939). Particularly, glucocorticoids appear to
be necessary for HPG interference (Breen & Karsch, 2006), to reduce sensitivity of the ovaries to
LH, suppress estradiol release by the ovaries (Carlson, 2010), and block ovulation (Rivier &
Rivest, 1991).
Activation of the HPA axis is capable of interfering with the HPG axis at every level,
particularly at the level of the hypothalamus and pituitary (Tilbrook, Turner, & Clarke, 2000),
but also the ovaries (Tetsuka, 2007) where glucocorticoid exposure can stop follicle maturation.
Further, HPA interference differs depending on whether stress exposure is acute or chronic. In
rodents, acute exposure to a synthetic glucocorticoid decreased FSH levels, but not GnRH
mRNA, suggesting acute stress hormone exposure reduces sensitivity of the pituitary to GnRH
rather than suppress GnRH secretion (Gore, Attardi, & DeFranco, 2006). However, multiple
days of glucocorticoid exposure resulted in decreased GnRH mRNA, suggesting longer exposure
to stress hormones may act on the hypothalamus to suppress GnRH secretion thereby dampening
overall HPG axis activity (Gore et al., 2006).
Cross-talk between the axes is not unidirectional, however. Estradiol can indirectly limit
glucocorticoid availability (Brien, 1981). A large percentage of circulating glucocorticoid is
often bound to a binding globulin, making only 5-10% of the hormone available to enter and act
on cells (Burke & Roulet, 1970). Estradiol can upregulate corticosteroid binding globulin,
thereby decreasing levels of biologically available glucocorticoids (Brien, 1981). That HPA-
HPG interactions are bidirectional, it is important to understand whether, and how, these
8
interactions may change when one system is dramatically altered, such as when women go
though the menopause transition. The decrease in estradiol levels accompanying menopause may
alter how HPG activation affects HPA function. For instance, the decreased estradiol levels may
lead to increased stress responses as a result of no longer limiting the level of unbound,
biologically active cortisol.
This work is focused on investigating how changes in estradiol levels may increase HPA
activity in response to stress and how such an increase may facilitate or attenuate effects of stress
on different memory processes. In the following sections I will review the effects of estradiol,
stress, and stress hormones on neural morphology, function, and cognition.
1.3 The Competing Effects of Stress and Estrogens on Neuronal Morphology and Function
1.3.1 Glucocorticoids Are Necessary For Normal Neuronal Function
Exposure to stress and glucocorticoids (GC) are documented to negatively impact the
brain, however, GCs also are necessary for normal brain functioning (Doi, Miyahara, & Hori,
1991; Nadeau & Rivest, 2003; Sapolsky, Stein-Behrens, & Armanini, 1991). In rodents, loss of
glucocorticoids by adrenalectomy (removal of the adrenal gland) decreases hippocampal CA1
excitability relative to CA1 excitability from adrenal-intact control animals; while corticosterone
treatment (the primary rodent GC) returned reactivity to the normal amplitude (Doi et al., 1991).
Further, rats receiving a GC antagonist (i.e. blocking the effects of GC) experienced potentiated
inflammatory responses following lipopolysaccharide infusion resulting in more prominent
neural damage compared with control animals (Nadeau & Rivest, 2003). Results such as these
suggest GC is necessary for proper stimulation of neurons in the hippocampus, and likely
elsewhere in the brain. Necessity of GC for proper neuronal stimulation is likely part and parcel
to the necessity of GC for long-term potentiation (Pavlides, Watanabe, Magariños, & McEwen,
9
1995). Additionally, GC also is necessary for shutting down the innate inflammatory response
that follows insults to tissue, and protecting the central nervous system from increased damage as
a result of prolonged inflammation. Yet, despite GC necessity for normal function, the hormone
can impede the system when simultaneously present with other insults (e.g. beta-amyloid), or
when present in extreme levels or for prolonged periods (i.e. extreme or chronic stress).
1.3.2 Exposure to Glucocorticoids Exacerbates Damage by Other Insults
The most deleterious effects of GC exposure occur in the hippocampal formation, a
region containing a high concentration of glucocorticoid receptors (Gerlach & McEwen, 1972;
McEwen, Weiss, & Schwartz, 1968). One such effect is the ability of GC to exacerbate damage
induced by other neuronal insults (Behl, 1998; Sapolsky, 1990, 1999, 2000). For instance,
neuron cultures pretreated with dexamethasone (a synthetic GC) or corticosterone for 24 hours
experienced greater rates of cell death from either beta-amyloid (Aβ) or glutamate exposure than
cultures not pretreated with GC (Behl et al., 1997). Corticosterone preexposure has the same
effects in cultures exposed to hypoxic or hypoglycemic environments (Tombaugh, Yang,
Swanson, & Sapolsky, 1992) and kainic acid (Stein-Behrens et al., 1992). The ability of GC to
increase glutamate and aspartate levels (Stein-Behrens et al., 1992) and concomitantly inhibit
excess glutamate uptake from the synaptic cleft by astrocytes (Virgin et al., 1991) may be one
mechanism for the heightened damage induced by GC. Increased glutamate levels coupled with a
decrease in glutamate clearance from the cleft would leave neurons susceptible to a level of
excitotoxic cell death beyond what is observed when exposed to the other neuronal insults alone.
1.3.3 Excess Exposure to Glucocorticoids Alone is Dangerous to Neural Tissue
Chronic glucocorticoid exposure also damages hippocampal neurons in the absence of
other toxic insults. Illustrating this are reports countering the anti-inflammatory effects
10
mentioned above (Nadeau & Rivest, 2003). First, cell cultures chronically exposed to GC
showed upregulation of the immune factors interlukin-1-beta and tumor necrosis factor-alpha
1

suggesting chronic GC exposure is pro-inflammatory, not anti-inflammatory (MacPherson,
Dinkel, & Sapolsky, 2005). Second, chronic restraint stress in rats increased the presence of
microglia markers in areas involved in the immune response, such as the medial prefrontal
cortex, anterior cingulate, nucleus accumbens, dorsal bed nucleus of stria terminalis, CA3 region
of the hippocampus, and the lateral periaqueductal gray (Tynan et al., 2010). Increases in
inflammatory and microglia markers are indicative of an increased immune response, the
combination of which suggests chronic exposure to GC induces a neuroinflammatory response,
rather than an anti-inflammatory response.
Chronic GC and stress exposure also affect neuronal morphology and function in the
brain. Twenty-one days of corticosterone injections decreased dendrite length and branch points
in the CA3 region of rats compared with injections of a control substance (Woolley, Gould, &
McEwen, 1990), while dexamethasone administration induced apoptosis and cell damage in the
dentate gyrus, CA1, and CA3 regions of the rat hippocampus and striatum (Haynes, Lendon,
Barber, & Mitchell, 2003). Further, prolonged exposure or exposure to high levels of
corticosterone or a GC agonist has the potential to interfere with processing and learning by
decreasing long-term potentiation (Pavlides et al., 1995; Pavlides, Watanabe, & McEwen, 1993).
Induction of GC release by stress exposure results in a similar pattern of neuronal damage
as administration of exogenous GC. Three weeks of daily restraint stress decreased apical
dendrite length and branch points in CA3 pyramidal neurons of rats (McLaughlin et al., 2010;
Watanabe, Gould, & McEwen, 1992), as well as in vervet monkeys exposed to social stress.

1
Interlukin-1-beta and tumor necrosis factor-alpha are cytokines involved in the inflammatory response of the immune system. Both
cytokines have involvement in apoptosis.
11
Chronic social stress in these primates was associated with a reduction of CA3 hippocampal
pyramidal cells and atrophy of dendritic branches of the surviving cells (Uno, Tarara, Else,
Suleman, & Sapolsky, 1989). The same effects have been observed in layers II and III of the
prefrontal cortex, where repeated acute and chronic restraint stress decreased apical dendrite
branch number and length in male rats (Brown, Henning, & Wellman, 2005; Cook & Wellman,
2004; Wellman, 2001). Thus, GC exposure damages prominent brain regions, whether studied in
vitro or in vivo or as a result of physical, psychological, or social stress.
1.3.4 The Effects of Estrogens On Neurogenesis and Neural Maintenance
Analogous to experiments examining the effects of stress hormones on neuronal viability,
many experiments have examined if, and to what extent, estrogen might affect neuronal integrity.
In contrast to the damaging effects of GC and stress on neuronal integrity, estrogens, particularly
estradiol, show neurogenic properties. First, fetal primary hippocampal neurons exposed to
conjugated-equine estrogens (Brinton et al., 2000) or 17β-estradiol (Chen, Nilsen, & Brinton,
2006) showed increased neurite number and length, as well as increased secondary branching
and branch length when compared with cultures free of any estrogenic compounds. The growth
promoting effect of conjugated-equine estrogens (CEE) was most robust in hippocampal
neurons, but also promoted neurite outgrowth in basal forebrain and cortical tissue (Brinton et
al., 2000). Estradiol also protected oligodendrocytes, the glial cells responsible for myelinating
axons within the central nervous system, from hyperoxic insults in vitro, indicating estradiol may
promote proper myelination later in development (Gerstner et al., 2007).
Estradiol also promotes neurogenesis in mature animals. Treatment with E2

increased cell
proliferation and decreased cell death in the dentate gyrus of middle-aged males (Saravia,
Beauquis, Pietranera, & De Nicola, 2007), essentially slowing down the rate of age-related
12
declines in cell proliferation of this hippocampal region. In adult animals, mature granule cells
appear to only respond to specific stimuli or stimuli ranges, while new granule cells have a lower
threshold for response to a wider range of stimuli (Marin-Burgin, Mongiat, Pardi, & Schinder,
2012) possibly allowing them to form new connections as learning occurs. Thus, estrogenic
protection of this process should aid maintenance of hippocampal-dependent learning and
memory processes.
1.3.5 Estrogens as a Neuroprotectant
In addition to promoting neurogenesis and maintaining neural integrity, estradiol also acts
as a neuroprotectant. In in vitro studies, E2 (Chen et al., 2006) or CEE (Brinton et al., 2000)
pretreatment of hippocampal neurons blocked Aβ-induced neurodegeneration, decreased the
degree of apoptosis, and attenuated the Aβ-induced decrease in ATP levels. Estradiol exerted a
similar protective effect against hydrogen peroxide challenge and glutamate excitotoxicity
(Brinton et al., 2000). These results extend the pattern of in vitro promotion of neural growth and
development, to in vitro protection against deleterious agents.
In vivo experiments also report estrogenic neuroprotection. Ovariectomy (removal of the
ovaries) followed by cyclic estradiol replacement increased dendrite spine density in layer III of
dorsolateral prefrontal cortex in young-adult rhesus monkeys and restored density in aged
monkeys, compared with rhesus monkeys not treated with estradiol (Hao et al., 2007). Estradiol
also protects against the excitotoxicity typically accompanying severe seizure activity.
Ovariectomized (OVX) female rats injected with kainic acid, an animal model for seizures,
showed marked dose-dependent neuron loss in the entorhinal cortex and hippocampus; an effect
blocked by E2 pretreatment (Hoffman, Moore, Fiksum, & Murphy, 2003). Estradiol pretreatment
also protected male mice from the stereotypical motor deficits accompanying methylmercury
13
exposure and reduced the amount of lipid peroxidation in the cerebellum (Malagutti et al., 2009),
suggested to be a result of E2 protection of the antioxidant glutathione (Malagutti et al., 2009).
These in vivo studies suggest estrogenic protection is partially achieved by reducing excitatory
ion influx and subsequent excitotoxicity, and by promoting other endogenous neuroprotective
factors, such as antioxidant systems.
1.3.6 Estrogens and Neurodegenerative Diseases
Estrogens also protect against neurodegenerative diseases. In animal models, endogenous
E2 levels were associated with delayed progression of neuropathology in a transgenic rat model
of Huntington’s disease (Bode et al., 2008), and E2 treatment prevented the progression of motor
dysfunction in a rodent model of amyotrophic lateral sclerosis (Choi et al., 2008). In these cases,
estradiol attenuated degeneration in the brain regions most targeted by these diseases. Estradiol
also exerts protective effects over the brain regions most affected by the Aβ plaques
characteristic of Alzheimer’s disease (Brinton, 2001, 2008; Pike, Carroll, Rosario, & Barron,
2009). In vitro, pretreatment with CEE (Brinton et al., 2000) or E2 decreased cell death (Chen et
al., 2006; Hosoda, Nakajima, & Honjo, 2001; Pike, 1999), attenuated the accompanying rise in
intracellular calcium concentrations (Chen et al., 2006; Hosoda et al., 2001), and decreased
presence of apoptotic markers in cells exposed to Aβ (Brinton et al., 2000; Hosoda et al., 2001;
Sharma & Mehra, 2008). Estradiol pretreatment also significantly increased the level of Bcl-x
L
and Bcl-2, anti-apoptotic proteins, in cultured fetal hippocampal neurons (Pike, 1999) and in
OVX adult female rats (Sharma & Mehra, 2008), suggesting estradiol’s neuroprotective effects
also may include a combination of inhibiting proteins that promote apoptosis while
simultaneously upregulating proteins that defend against apoptosis, potentially offering dual
protection from the cell death typical of neurodegenerative diseases.
14
1.4 The Competing Effects of Stress and Estrogens on Cognition
1.4.1 Stress Can Benefit Cognitive Function
The pattern of glucocorticoid exposure on neural tissue extends to cognitive functioning,
such that GC is required for optimal functioning, but is deleterious in excess levels or over
prolonged exposure. In their classic 1908 work, Yerkes and Dodson reported that learning of a
shock-motivated avoidance task followed an inverted-U shaped function, such that insufficient
arousal (too weak a shock) led to low learning rates, as did excess arousal (too strong a shock).
However, moderate stimulation resulted in the best learning rates (Yerkes & Dodson, 1908).
Given that shock elicits a strong corticosterone response in rats (S. B. Friedman, Ader, Grota, &
Larson, 1967), the results indicate that some amount of stress is required for peak performance to
be attained, and deviations from this optimal level result in impaired learning and memory. More
contemporary works suggest a similar inverted-U shaped pattern for stress-cognition interactions
in different domains, including the radial arm water maze (Salehi, Cordero, & Sandi, 2010),
spatial learning and associative learning (Mateo, 2008), and corticosterone-primed burst
potentiation interactions (Diamond, Bennett, Fleshner, & Rose, 1992). Importantly, the inverted-
U shaped function appears to vary by task difficulty, with moderate arousal beneficial for
difficult tasks and stronger arousal beneficial for easier tasks (Dodson, 1915). In this work,
Dodson (1915) showed kittens learned an avoidance task faster with moderate shock when visual
discrimination of the shock and no-shock chambers was difficult, but that stronger shock was
required to aid learning rates as visual discrimination between the chambers became easier. This
suggests that as tasks become increasingly easy, more stress is required for more rapid learning.
However, Dodson utilized the same “strong” shock for all three difficulties of discrimination,
therefore, the inverted-U shape could still hold, but the “strong” shock has now become the
15
optimal “moderate” shock required to aid learning. In this case had the “strong” shock been
made stronger on the easier versions of the task, the kittens’ learning may have been impaired,
again showing an inverted-U function. Such a shift of the inverted U with task difficulty has
been observed and supported more recently (Anderson, 1994; Salehi et al., 2010; for a brief
discussion see Shors, 2004).
While stress may generally benefit learning, stress reactivity may actually be necessary
for the type of avoidance learning discussed above, and others. In the conditioned taste aversion
literature, learning is based on the introduction of a physical stressor (the illness agent), such that
animals must be made ill in order to learn to avoid the recently consumed novel substance
(Hintiryan, Foster, & Chambers, 2009; Miele, Rosellini, & Svare, 1988; Nachman & Ashe,
1973). Arguably, this would be the case for most types of avoidance learning. In these instances,
activation of the stress response would signal to the animal that the stimulus most temporally and
spatially contingent to stress onset should be avoided.
The ability of these stimuli to elicit physiological responses likely makes them easy to
remember and contributes to the relative ease of avoidance learning. If the innate ability to elicit
physiological responses contributes to learning and memory for these stimuli, then stress
application prior to exposure of these negative stimuli should enhance encoding as a function of
the stress response plus the unconditioned response being interpreted as a stronger unconditioned
response to the stimulus. This seems to be the case with memory for emotional items. For
instance, exposure to stress 30 minutes before encoding impaired recall of neutral words, but had
no effect on the recall accuracy of emotional words, relative to controls (Smeets, Jelicic, &
Merckelbach, 2006). Stress experience also can enhance memory for emotional items. Stress
applied before encoding resulted in better memory for an emotional story and worse memory for
16
a neutral story after a one-week delay interval, compared with comparable memory for both
stories in control subjects (Payne et al., 2007). Similar enhancement has been reported in
pharmacological studies, where administration of 20mg hydrocortisone resulted in better cued
recall of highly arousing emotional stimuli than cued recall after placebo administration
(Buchanan and Lovallo, 2001). However, stress can impair recall of emotional and neutral words
when the stressor is applied after encoding and before retrieval (Kuhlmann, Piel, & Wolf, 2005),
versus only prior to encoding as in the aforementioned studies. Together, the literature on
avoidance learning and emotional memory suggest stress benefits learning, and may be
necessary for certain types of learning, just as GC is necessary for proper neural function.
However, as reviewed in the section on glucocorticoid effects on neural tissue, although stress
hormones are required and beneficial for some forms of learning and memory, excess or
prolonged exposure negatively impacts many other cognitive domains.
1.4.2 Associations Between Stressful Life Events and Cognition in Humans
Experiencing highly stressful events or a greater number of life stressors is associated
with cognitive impairment. Individuals residing within 5 miles of the Three Mile Island nuclear
meltdown had higher levels of epinephrine and norepinephrine in their urine, more somatic
complaints, higher anxiety and isolation, and worse performance on an embedded figure and
proofreading task, 17 months after the nuclear meltdown, compared with residents near an
undamaged nuclear plant, a coal-burning plant, or no plant (Baum, Gatchel, & Schaeffer, 1983).
Likewise, individuals scoring higher on the Life Experiences Scale (e.g. more stressful events)
displayed impaired word recall and greater intrusion of irrelevant thoughts during an operation
span task assessing working memory (Klein & Boals, 2001). However, these non-experimental
studies cannot conclude that the cognitive deficits are a result of the stressors measured. In the
17
case of the Baum et al. (1983) study, participants may have suffered minor radiation exposure
contributing to the differences in physiology and cognition, and the impaired performance of
college students reporting a higher number of life stressors could result from distraction due to
these stressors rather than the effects of cortisol. However, experimental studies in animals and
humans report similar relationships between stress and cognitive performance.
1.4.3 Negative Effects of Glucocorticoids on Cognition in Animals
The animal literature is rife with reports of stress-induced cognitive impairment.
Discussion of these works, however, first requires a brief description of the tasks and apparatus
commonly employed. One such apparatus is the radial arm maze. Radial arm mazes can have a
variable number of arms radiating out from a central starting box. Some arms are baited with
food, and the animal is trained to traverse each baited arm only once and never enter the never-
baited arms. A similar procedure is used in the Y-maze and T-maze, except these mazes consist
of one starting alley and two arms shaped like a “Y” or a “T”, respectively. These tasks measure
working memory in different ways. The radial arm maze assesses working memory because the
animal must remember which arms contain food, which arms have already been entered, and
which arms still need to be entered. The Y-maze and T-maze can measure short-term memory,
long-term memory, and working memory, but specific means of doing so vary depending on the
task employed within the maze.
Chronic GC administration impairs cognition in these animal models of working
memory. Nine weeks of corticosterone administration increased the number of working memory
errors in a radial arm maze (Arbel, Kadar, Silberman, & Levy, 1994), and eight weeks and three
weeks of corticosterone injections resulted in increased working memory and reference memory
errors in a Y-maze, respectively (Coburn-Litvak, Pothakos, Tata, McCloskey, & Anderson,
18
2003). These pharmacological experiments extend the detrimental effects of GC exposure on
neural tissue to detrimental effects in animal models of working memory.
1.4.4 Negative Effects of Stress Application on Cognition in Animals
Importantly, chronic stress results in the same cognitive deficits as chronic administration
of exogenous GCs. Subordinate male tree shrews, whose stress hormone levels were elevated
compared with dominant males, were unable to remember which holes contained and did not
contain a food reward in a holeboard task, suggesting stress exposure leads to greater reference
memory errors (Ohl & Fuchs, 1999). In another social stressor, rats chronically exposed to a cat
showed impaired working memory on a radial arm water maze – a radial arm water maze
requires animals to swim to a platform that allows them to rest rather than running to a food
reward as seen in a standard radial arm maze (Park, Campbell, & Diamond, 2001). Similarly,
repeated exposure to a novel environment or the Morris water maze resulted in more working
memory errors in a radial arm maze compared with animals in the control condition (Diamond,
Fleshner, Ingersoll, & Rose, 1996; Diamond & Rose, 1994), and rats exposed to chronic restraint
stress showed worse retention for a familiar versus novel arm in a Y-maze (Kleen, Sitomer,
Killeen, & Conrad, 2006). Thus, stress induces declines in cognitive function, whether the stress
treatment involves exposure to exogenous stress hormones or to various physical or social
stressors.
1.4.5 Negative Effects of Controlled Non-Laboratory Stressors on Cognition in Humans
Overall, the animal literature suggests that chronic glucocorticoid and stress exposure
negatively affect cognitive processes, and working memory in particular. This pattern implies
humans should show impaired cognitive function in the face of various stressors. In order to
examine the effects of extreme stress on cognition, researchers made use of a unique situation:
19
Survival School. Survival School trains and prepares soldiers for becoming trapped behind
enemy lines, pursued by enemy soldiers, and captured and interrogated as prisoners of war.
Soldiers tested using the Rey Osterrieth Complex Figure drawing task showed atypical strategy
for copying the figure and poor recall of figure details when tested immediately after an
interrogation session, compared with soldiers tested prior to commencement of Survival School
or after release from the mock prisoner of war camp (Morgan, Doran, Steffian, Hazlett, &
Southwick, 2006). Furthermore, soldiers provided with sources of misinformation, such as being
shown a photograph of another person during their interrogation, were more likely to falsely
identify their interrogator than soldiers not provided with potential sources of misinformation
(Morgan, Southwick, Steffian, Hazlett, & Loftus, 2013). These results suggest that even
individuals trained to withstand stressful situations are susceptible to the negative effects of
highly stressful experiences. These studies also provide a more plausible relationship between
stress exposure in the world and declines in cognitive performance in humans than did Baum et
al. (1983) and Klein and Boals (2001), although they still fail to provide measurement of the
stress response and the relation between stress hormone levels and cognitive performance.
1.4.6 Negative Effects of Laboratory Stressors on Cognition in Humans
The discussion thus far provides a connection between chronic stress and cognitive
interference; however, most laboratory research with humans focuses on the effects of acute
stress on cognitive function. As shown by Yerkes and Dodson (1908) and Dodson (1915) such
acute manipulations may lead to mixed results depending on how much arousal is induced by the
stressor or how difficult the target task is relative to the amount of stress induced. One frequently
used acute stressor is the Trier Social Stress Test (TSST). The TSST reliably elicits a stress
response from participants by requiring them to give a speech and perform mental arithmetic in
20
front of an audience (Oei, Everaerd, Elzinga, van Well, & Bermond, 2006; Schoofs, Preuß, &
Wolf, 2008; Wolf, Minnebusch, & Daum, 2009). Performance of this task prior to testing
resulted in increased reaction times and working memory error rates in the Sternberg item
recognition task (Oei et al., 2006), increased reaction times and decreased accuracy in the N-
back task (Schoofs et al., 2008), and interference with acquisition of a classically conditioned
eye-blink task (Wolf et al., 2009).
Another commonly employed acute laboratory stressor is the cold pressor task (CPT).
The CPT reliably induces a stress response by requiring participants to hold one of their hands in
ice water (Bullinger et al., 1984; Edelson & Robertson, 1986; Lighthall, Mather, & Gorlick,
2009; Lighthall et al., 2011; Mather, Lighthall, Nga, & Gorlick, 2010). Like the TSST,
completion of the CPT can impair cognitive function. CPT exposure impaired performance on
working memory tasks such as the digit span backward (Schoofs, Wolf, & Smeets, 2009) and the
Sternberg item recognition task (Duncko, Johnson, Merikangas, & Grillon, 2009). Thus,
although the effects of acute stress might lead to variable effects on cognition, these stressors
must have been sufficient to induce interference in working memory. Based on the above-
discussed works for Yerkes and Dodson, these stressors either induced a level of stress that
surpassed the optimal level of arousal for optimal performance and/or the difficulty of working
memory tasks makes this type of cognition relatively sensitive to the effects of stress. In either
case, the literature shows that stress exerts a parallel pattern in neural tissue and cognition,
namely, that overexposure of stress detrimentally affects both systems.
1.4.7 How Are Glucocorticoids and Stress Affecting Cognition?
Thus far the evidence demonstrates that exposure to stress or stress hormones can
enhance memory for emotional items, but also negatively affects neuronal health and LTP, both
21
essential components in learning and memory. The different pattern of stress effects on
emotional versus non-emotional memory is likely a result of different neural involvement – with
the amygdala more involved in emotional memory and the prefrontal cortex and hippocampus
more involved in non-emotional memory – and the ways stress effects these regions. In humans
acute stress is associated with reduced activity in the dorsolateral prefrontal cortex (Cabeza,
Dolcos, Graham, & Nyberg, 2002) during performance of a non-emotional working memory task
(Qin, Hermans, van Marle, Luo, & Fernández, 2009). Furthermore, the most deleterious effects
of glucocorticoids and stress on neural tissue are observed in the hippocampus and prefrontal
cortex, regions involved in the cognitive domains impaired by stress and its hormones, like short-
term memory (Alonso et al., 2002; Cabeza et al., 2002), long-term memory (Alonso et al., 2002),
and working memory (Cabeza et al., 2002; Curtis & D'Esposito, 2003; Laroche, Davis, & Jay,
2000).
The neural underpinnings for learning and memory aided by stress application are
different, and mainly rely on the amygdala. The amygdala is considered of utmost importance in
recognition and processing of innately arousing stimuli (Zald, 2003) as well as in stimuli that
were once neutral but have since obtained a negative association (Gallo, Roldan, & Bures, 1992;
Lamprecht & Dudai, 1996; Navarro, Spray, Cubero, Thiele, & Bernstein, 2000; Roldan & Bures,
1994). Importantly, glucocorticoids can modulate amygdalar activation during memory
formation of emotional items. Glucocorticoid application resulted in better retention of a one-
trial avoidance learning task after a 48-hour delay when infused into the basolateral nucleus of
the amygdala immediately after training (Roozendaal & McGaugh, 1997). Inactivation of GC
resulted in the opposite effect, however; 24-hour retention of a contextual fear response was
impaired when a glucocorticoid antagonist was infused into the basolateral amygdala 10 minutes
22
prior to training (Donley, Schulkin, & Rosen, 2005). That glucocorticoid or stress exposure
doesn’t interfere with amygdala-dependent learning could be a result of the structure’s
imperviousness to glucocorticoid damage (Morales-Medina, Sanchez, Flores, Dumont, &
Quirion, 2009), while stress-induced enhancement of memory consolidation may be a result of
GC ability to increase kinase activity related to learning (pErk1/2) when infused in the
basolateral amygdala (Roozendaal et al., 2009).
1.4.8 Estrogen Can Impair Cognitive Function
Similar to the effects of glucocorticoids on cognition, estrogen is associated with both
negative and positive effects on cognitive function. The variability is largely observed in studies
of estrogen effects on post-menopausal women, with some reporting positive effects (Baker et
al., 2012; Duff & Hampson, 2000; Maki et al., 2011; Maki, Zonderman, & Resnick, 2001;
Miller, Conney, Rasgon, Fairbanks, & Small, 2002; Smith, Giordani, Lajiness-O'Neill, &
Zubieta, 2001; Wolf & Kirschbaum, 2002; Wolf et al., 1999), no effect (Maki et al., 2001;
Pefanco et al., 2007; Yaffe et al., 2006), and negative effects (Espeland et al., 2004; Mulnard et
al., 2000; S. R. Rapp et al., 2003; Resnick et al., 2006; Shumaker et al., 2003).
The Women’s Health Initiative (WHI) studies comprise the most widely reported
negative findings. At the time of inception, the Women’s Health Initiative Memory Study
(WHIMS) and Women’s Health Initiative Study of Cognitive Aging (WHISCA) were the largest
randomized studies examining prospective observations that post-menopausal hormone treatment
maintained cognitive function otherwise observed to decline after menopause. Rather than
supporting these observations of estrogenic protection, the WHISCA and WHIMS studies found
that CEEs negatively affected performance on various cognitive measures, such as the California
Verbal Learning Test (Resnick et al., 2006) and the Modified Mini Mental State Exam (a
23
measure of global cognitive function) whether administered unopposed (Espeland et al., 2004),
or opposed with medroxyprogesterone acetate (S. R. Rapp et al., 2003), as well as an increased
risk of developing dementia (Shumaker et al., 2003) and a faster rate of Alzheimer’s progression
in older women with preexisting Alzheimer’s disease (Mulnard et al., 2000).
The results were surprising because of the reports of estrogenic protection of cognitive
function (Baker et al., 2012; Duff & Hampson, 2000; Maki et al., 2001; Miller et al., 2002; Smith
et al., 2001; Wolf & Kirschbaum, 2002; Wolf et al., 1999) and the basic science reports of
estrogenic fortification of the central nervous system (Brinton et al., 2000; Chen et al., 2006;
Hosoda et al., 2001; Pike, 1999). To account for the disparity between basic science experiments
and human studies, Brinton (2005) proposed a healthy cell bias of estrogen action, stating
estrogen would be advantageous to a population of healthy cells, but would be detrimental to a
population of declining or already injured cells. To test this hypothesis, Brinton and colleagues
compared treatment and prevention models in vitro. In treatment models, estradiol is applied to
neurons either at the time of, or after, application of the deleterious agent; this differs from the
prevention model, which pretreats cells with estradiol before exposure to the agent. Consistent
with the hypothesis, neurons simultaneously treated with estradiol and Aβ protein did not receive
any estrogenic protection, nor did neurons exposed to the toxic protein for 1 to 2 days prior to
estradiol exposure. However, introduction of estradiol to the cultures after 5 days of Aβ exposure
resulted in greater cell death than the Aβ-alone cultures (Chen et al., 2006). Given that the
women in the WHI studies were over 65 years of age, the negative findings may have been a
result of estrogenic exacerbation of preexisting age-related tissue damage, particularly in the
those women with preexisting Alzheimer’s disease.
24
Emotional memory is another cognitive domain negatively associated with estrogen. A
recent review (Sakaki & Mather, 2012) proposed that decreased neural activation of the
amygdala during high estradiol phases of the menstrual cycle, relative to low estradiol phases,
should translate into worse memory for negative stimuli during the high estrogen phases; a
prediction supported in a study examining the effects of menstrual cycle phase on memory for
emotionally valenced pictures. The study found free recall for emotional photographs was worse
during the high estradiol phase of the menstrual cycle than during the low estradiol phase of the
cycle (Mather, Cowell, Whiteside, Ledger, & Mangold, under review). The same pattern was
reported in an emotional working memory task. Performance of a delayed-match-to-sample task
using “disgust” and “sadness” as the target facial expressions to remember was worse in women
tested later in the follicular phase of the menstrual cycle (higher estradiol) than in women tested
during menses when estradiol levels are lowest (Gasbarri et al., 2008). In contrast to this,
however, are reports of better reward learning when estradiol levels are high (Sakaki & Mather,
2012). Blunted responses to negative imagery and increased proclivity for learning positive
associations occurring together during the late follicular phase (i.e. preovulatory) may have an
evolutionary underpinning. A female less sensitive to detecting negative stimuli and more
susceptible to recognizing positive associations during the time-sensitive ovulatory phase may be
more receptive to copulating with available mates, increasing the likelihood of passing on her
genes.
Although the estradiol findings discussed here are opposite the effects on the central
nervous system, these negative effects are limited to specific circumstances, such as time of
initiation of post-menopausal hormone treatment, and specific cognitive domains during certain
25
limited days on the menstrual cycle. When considering broader effects of estrogens on cognition,
however, the literature largely supports a beneficial role for the hormone.
1.4.9 Positive Effects of Estradiol on Cognition in Young-adult Female Animals
Although estradiol is associated with impaired performance in some domains, the
evidence largely supports a beneficial effect of the hormone. Ovariectomized animals exhibited
fewer working memory errors (Bimonte & Denenberg, 1999), a higher rate of correct choices
(Daniel, Fader, Spencer, & Dohanich, 1997; Fader, Johnson, & Dohanich, 1999), and a greater
number of consecutive correct choices (Wilson, Puolivali, Heikkinen, & Riekkinen Jr., 1999) if
treated with estradiol after ovariectomy compared with animals not treated with estradiol.
Similar results have also been reported in other paradigms. Ovariectomized animals treated with
estradiol benzoate displayed better retention for which arm to enter in a delayed match-to-sample
task across 10-, 30-, and 100-minute delay intervals than did OVX animals not receiving
estradiol treatment (Sandstrom & Williams, 2001). Estradiol treated animals also required less
time to learn a delayed non-match-to-sample task (learning the reward was in the non-baited arm
during training) than the non-estradiol treated OVX animals (Gibbs, 1999). The same pattern
was observed in an object recognition task. In this task animals were exposed to a total of 4
novel objects. Animals first explored a pair of 2 novel objects, after some delay they were
exposed to another 2 objects (one object from pair #1 and the third novel object); after a second
delay period the animals were exposed to 2 additional objects (one being the second object from
pair #1 and the fourth novel object). The animal is said to recognize an object if it spends less
time exploring that object compared with time spent exploring the novel object. Ovariectomy
followed by 21 days of E2 treatment enhanced retention for the first-tested object after a 3-hour
delay and of the second-tested object after a 6-hour delay, compared with non-estradiol treated
26
mice (Vaucher et al., 2002). Thus, estradiol aids in maintaining or enhancing working memory,
short-term memory, and long-term memory in young-adult female animals.
1.4.10 Effects of Menstrual Cycle on Cognition in Young-adult Human Females
The effect of estradiol on cognition in young-adult human females is not as consistent as
the reports in young-adult female animals. The examination of estradiol effects typically looks
for differences in cognition between the different menstrual cycle phases, when estradiol levels
are low versus high, but have failed to find consistent effects across the hormone fluctuations of
the menstrual cycle. The preovulatory phase (high estrogen and low progesterone) has been
associated with increased creativity and decreased motor perseveration, compared with menses
(low estrogen and low progesterone) and the midluteal phase when estradiol levels are moderate
and progesterone levels are high (Rosemarie Krug, Stamm, Pietrowsky, Fehm, & Born, 1994).
However, no effect of cycle phase has been observed in other cognitive domains. Females did
not differ in their reaction times or accuracy in a semantic decision task asking participants to
press a button when shown a pair of synonyms or when two strings of consonants were identical,
when tested during menses or the midluteal phase (Fernandez et al., 2003). Correlation analysis
also failed to find a relationship between estradiol levels during the whole follicular phase
(consistently low progesterone and rising estradiol levels) and recognition of previously seen
words mixed in a list of previously seen and never seen words (Craig et al., 2008). Still others
reported no differences in performance of the Iowa Card Task or the Weather Task between
menses and the midluteal phase (Reavis & Overman, 2001). Similar mixed results were also
reported in a study that administered estradiol to young spontaneously cycling females; while
estradiol enhanced accuracy on a 1-back task, a measure of working memory, treatment had no
27
effect on declarative learning and memory, verbal fluency, attention, cognitive flexibility, or
psychomotor function and information processing (Bartholomeusz et al., 2008).
Failure of the young-adult human female literature to be as consistent as the animal
literature is not surprising. Animal studies allow for the complete control of all gonadal
hormones by ovariectomy and hormone administration. In humans, however, studies must search
for differences between the naturally fluctuating hormone levels of reproductively intact females.
Thus, when searching for the true effects of estrogen on cognition in humans we may need to
focus on populations of women who no longer endogenously produce estrogen, such as
postmenopausal women. In this regard, much work has been completed investigating the effects
of estrogen on cognition in aged animal and human females.
1.4.11 Positive Effects of Estrogen on Cognition in Aging Female Animals
Aging female rodents do not experience menopause, instead they experience
“estropause”. This phase of the reproductive lifecycle is characterized by irregular cycling until a
stable constant state of moderate estradiol is reached coinciding with reproductive senescence. In
the rat, the average age the estrous cycle becomes irregular ranges from 10 to 12 months, slowly
progressing to a state of anestrous at approximately 19 months of age (Lu, Hopper, Vargo, &
Yen, 1979; Matt, Sarver, & Lu, 1987). Little work has focused on the effects of estradiol
administration in intact middle-aged or aged female rodents, although studies that have find
beneficial effects of the hormone. For instance, intact aged mice reached asymptotic performance
of the Morris water maze task faster if treated with a moderate daily dose of estradiol benzoate,
compared with animals receiving a lower dose or no estradiol treatments (Frick, Fernandez, &
Bulinski, 2002), suggesting that moderate levels of estradiol aid in the recovery of working
memory capabilities of aged animals.
28
Like studies using younger female animals, however, studies of estradiol effects on
cognition in older females typically employ ovariectomy and find estradiol replacement
beneficial to cognitive function. Aged animals ovariectomized 8- to 12-months prior to testing
displayed lower error rates on a delayed match-to-sample task if treated with estradiol either
immediately or starting 3 months after ovariectomy, compared with animals not treated with
estradiol (Gibbs, 2000). Similarly, animals ovariectomized in middle age performed better on a
delayed non-match-to-sample task if estradiol administration mimicked the estrous cycle
compared with animals not treated with estradiol (Markowska & Savonenko, 2002). Similar
results have been reported in primates. Twenty-year-old OVX rhesus monkeys learned a delayed
response task and delayed non-match-to-sample task faster when treated with estradiol than non-
estradiol-treated primates, and performed comparably to young, intact, females (Rapp, Morrison,
& Roberts, 2003). Together, the literature indicates that estrogen replacement exerts beneficial
effects on working memory processes of aging female animals.
1.4.12 Positive Effects of Estrogens on Cognition in Aging Human Females
Aging human females experience menopause, a state of reproductive senescence
accompanied by dramatic declines in estradiol production and estradiol levels. Few studies have
examined the cognitive effects of naturally declining estradiol levels in the absence of hormone
replacement. One such study found endogenous levels of estradiol in postmenopausal women not
taking any hormone supplements was positively correlated with performance on a verbal fluency
task (Wolf & Kirschbaum, 2002). A comparable pattern was also observed in females receiving
2 weeks of estradiol treatment; those who experienced higher endogenous levels of estrogen in
response to the dosage performed better on the immediate and delayed recall portion of a verbal
memory test compared with women experiencing lower endogenous estradiol levels in response
29
to the dosage (Wolf et al., 1999). In accordance with the reviewed animal literature, the human
literature to this point supports the putative beneficial effects of estradiol on cognition.
In addition to exerting beneficial effects on verbal memory and executive function, other
prospective studies have found estrogen therapy to benefit cognition. Women on a hormone
replacement regimen made fewer errors on a non-spatial working memory task, spatial working
memory task, and had a longer memory span during the digit span backward task (Duff &
Hampson, 2000), performed better on measures of semantic fluency, attention, and working
memory (Miller et al., 2002), and displayed better performance on a non-verbal memory task
(Smith et al., 2001) compared with women not taking any hormone supplements. The difference
in results between these studies and the negative results on the WHI studies may be that, on
average, the women in these studies began treatment closer to the age of menopause than the
women in the WHI studies. For example, estrogen-associated enhancements in working memory
and spatial working memory were observed in women who began hormone treatment during
peri-menopause or soon after menopause (Duff & Hampson, 2000), those showing benefits in
semantic fluency, attention, and working memory were on average 63 years old but had been
taking estrogen supplements for an average of 12 years (Miller et al., 2002), and those displaying
better performance on a non-verbal memory task initiated estrogen replacement within two years
of menopause (Smith et al., 2001).
This pattern of estrogenic benefit accords with the basic science findings suggesting
estradiol must be present before damaging agents are introduced in order to prove protective or
beneficial, otherwise the hormone will prove detrimental (Chen et al., 2006). Taken together the
results indicate that initiating estradiol treatment during the peri-menopausal phase and
30
prolonging estradiol exposure may prove beneficial to the maintenance of neural function and
cognitive capabilities in postmenopausal females.
1.4.13 How is Estradiol Affecting Cognition?
Estrogenic effects on cognition in younger adult females may occur as neural
concentrations of E2 fluctuate, coinciding with the hormone fluctuations of the menstrual cycle.
Neural levels of estradiol have been linked to circulating levels of the hormone. Rodents either
left intact or gonadectomized followed by estradiol replacement or vehicle showed treatment-
dependent levels of estradiol in the hippocampus (Barker & Galea, 2009). Given that the
hippocampus produces estradiol locally (Mukai et al., 2006), the Barker and Galea (2009) results
suggest that local production of estradiol in the hippocampus may depend on circulating plasma
levels of the hormone. Further, hippocampal dendrite spine density (Gould, Woolley, Frankfurt,
& McEwen, 1990; Woolley, Gould, Frankfurt, & McEwen, 1990; Woolley & McEwen, 1993)
and synapse density (Woolley & McEwen, 1992) are modulated by estradiol levels, as a function
of NMDA receptor activation (Woolley & McEwen, 1994), suggesting that the ability of the
hippocampus to locally produce estradiol may affect cognition by modulating the ability of the
hippocampus to exhibit long-term potentiation.
Older females may be affected through the same mechanism only with more drastic
effects. If the hippocampus locally produces estradiol relative to the circulating plasma levels,
then older females would produce very little estradiol in the structure as result of the low plasma
estradiol levels after menopause. Combined with the above-reviewed protective effects of the
hormone on the hippocampus, absence of circulating and locally produced estradiol in the
hippocampus may make the structure more vulnerable to degeneration, resulting in cognitive
decline. Application of estradiol close to the time of menopause, however, may help maintain
31
estradiol production in the hippocampus, thereby prolonging the integrity of the structure and
aiding in maintenance of cognitive function.
As for the inhibitory effect of estrogen on emotional memory, estrogenic modulation of
the amygdala may be responsible. Unlike in the hippocampus, estradiol decreases excitability in
the amygdala (Womble, Andrew, & Crook, 2002) by decreasing the amplitude of excitatory
post-synaptic potentials (EPSPs). Decreasing EPSPs in the amygdala could lead to failure of an
emotional stimulus to elicit enough EPSPs to reach threshold and cause activation of a cell. The
failure to activate neurons in response to the stimulus would then leave the nucleus unable to
communicate with other brain regions that there is a stimulus present requiring attention, thereby
impairing learning about the stimulus and memory formation for such events.
1.5 Estradiol-Stress Interactions
1.5.1 Estrogenic Protection Against Glucocorticoid Insults
In addition to the countervailing effects of estradiol and glucocorticoids on neural tissue
and cognition, estrogen can directly protect neural tissue from glucocorticoid exposure.
Administration of dexamethasone (DEX) to male rats increased apoptosis and cell damage in the
striatum and hippocampus, an effect attenuated by estradiol pretreatment (Haynes et al., 2003),
and hippocampal slices from males exposed to a stressor prior to termination and tissue
collection displayed greater long-term potentiation when bathed in a medium containing E2

than
slices bathed in artificial cerebrospinal fluid (Foy, Baudry, Foy, & Thompson, 2008). Similar
patterns are observed between OVX and intact female rodents. Chronic restraint stress decreased
apical dendrite complexity in the CA3 region of the hippocampus (McLaughlin et al., 2010) and
layers II and III of the medial prefrontal cortex (Garrett & Wellman, 2009) in OVX females,
32
effects prevented by E2 replacement after OVX (Garrett & Wellman, 2009; McLaughlin et al.,
2010).
A possible mechanism for estradiol protection against glucocorticoid overexposure is
estrogenic modulation of the GC receptor. Typically, chronic stress or stress hormone
administration results in downregulation of the GC receptor (Ferrini, Lima, & De Nicola, 1995;
Herman, Patel, Akil, & Watson, 1989; Sapolsky, Krey, & McEwen, 1984; Seckl & Olsson,
1995). Such downregulation likely acts as a protective factor. Reducing the number of receptors
should minimize the number of cells vulnerable to damage from glucocorticoid overexposure.
However, the nature of a chronic stressor is continual presence and continued release of GC. The
reduced number of receptors induced by GC presence would already be filled to capacity and
unable to detect the continually elevated glucocorticoid levels, hindering the effectiveness of the
HPA axis’ negative feedback system. However, estrogen administration upregulates
glucocorticoid receptors whether chronic stress is simulated by drug administration (Ferrini et
al., 1995) or directly applied (Ferrini & De Nicola, 1991). Upregulation of GC receptors by
estrogen would allow the brain to better detect lower levels of the stress hormone. The increased
sensitivity to GC detection would make the negative feedback system of the HPA axis more
responsive leading to more rapid shut down the stress response. These basic science findings
indicate estradiol provides physical protection from the damaging effects of glucocorticoids and
may directly modulate HPA axis response to a stressor by modulating GC receptor density.
1.5.2 Estrogenic Modulation of the Stress Response
Sex differences provide one line of evidence for estrogenic influence on HPA axis
activation. Young-adult females show lower adrenocorticotropic hormone (Kirschbaum,
Kudielka, Gaab, Schommer, & Hellhammer, 1999; Kudielka et al., 1998) and biologically active
33
free cortisol (Davis & Emory, 1995; Kirschbaum et al., 1999; Kirschbaum, Wust, &
Hellhammer, 1992; Kudielka, Buske-Kirschbaum, Hellhammer, & Kirschbaum, 2004; Kudielka
et al., 1998) responses compared with young-adult males. Likewise, although brain activation
while viewing negative visual stimuli did not differ between young men and young women in the
low estradiol phase of the menstrual cycle, women tested during the high estradiol phase of the
menstrual cycle showed lower stress-related activation of the anterior cingulate gyrus,
orbitofrontal cortex, medial and ventromedial prefrontal cortex, amygdala, and hippocampus
than men (Goldstein, Jerram, Abbs, Whitfield-Gabrieli, & Makris, 2010), with similar
activational differences in women tested during the low estradiol versus higher estradiol phases
of the menstrual cycle (Goldstein et al., 2005). Further, 1-day estradiol treatment in young-adult
males resulted in blunted systolic blood pressure, pulse rate, epinephrine, and norepinephrine
responses to a mental stressor (Del Rio et al., 1994).
Similar effects have been observed in post-menopausal women on estrogen treatment.
Transdermal or oral E2 interventions spanning 1 day (Del Rio et al., 1998), 3 weeks (Ceresini et
al., 2000), 1 month (Puder, Freda, Goland, & Wardlaw, 2001), 6 weeks (Lindheim et al., 1992),
and 8 weeks (Komesaroff, Elser, & Sudhir, 1999) reduced the epinephrine (Ceresini et al., 2000;
Del Rio et al., 1998), norepinephrine (Ceresini et al., 2000), diastolic blood pressure (Ceresini et
al., 2000; Komesaroff et al., 1999; Lindheim et al., 1992), systolic blood pressure (Komesaroff et
al., 1999; Lindheim et al., 1992), ACTH (Puder et al., 2001), and cortisol (Puder et al., 2001)
responses to various stressors, such as mental stressors (Ceresini et al., 2000; Del Rio et al.,
1998; Lindheim et al., 1992), the cold pressor task (Lindheim et al., 1992), and an endotoxin
challenge (Puder et al., 2001).
34
In all, these findings indicate that estradiol blunts the HPA axis response. When
considered with the above-discussed effects of stress on cognition, this pattern of attenuated
HPA axis reactivity suggest that: 1) the high estradiol phase of the menstrual cycle should be less
susceptible to stress-induced alterations in cognitive function and 2) post-menopausal women
using hormone supplements should be less likely to experience stress-induced alterations in
cognitive function.
1.5.3 Implications for Young-Adult Women
Based on the findings thus far, younger women should show reduced HPA reactivity to
stressors during the high estradiol phase of the menstrual cycle compared with the low estradiol
phase. However, the literature examining the effects of cycle phase on stress reactivity is not so
clear-cut. In fact, many studies fail to find any difference in stress reactivity across the different
hormone phases of the cycle (Galliven et al., 1997; Morimoto et al., 2008), or report greater
reactivity during higher estradiol phases (Altemus et al., 1997; Altemus, Roca, Galliven,
Romanos, & Deuster, 2001; Andreano, Arjomandi, & Cahill, 2008; Kirschbaum et al., 1999;
Roca et al., 2003).
This counterintuitive pattern of results may be related to in which phases young-adult
women undergo testing. In order to maximize the difference in hormone profile, and ensure cycle
phase accuracy, most studies examine females during the early follicular phase when estradiol
and progesterone levels are at the lowest concentrations, and the luteal phase when both estradiol
and progesterone are at significantly higher concentrations. Thus, it is important to note that the
findings reported in these studies are examining the combined effects of estradiol and
progesterone on the stress response, not the effects of estradiol on the stress response. The
inclusion of progesterone is important since this hormone counteracts some of estrogen’s
35
beneficial effects (Woolley & McEwen, 1993). This may also be the reason for the lack of clear
differences in cognitive performance across the menstrual cycle. Recall that most of the studies
investigating cognition across the menstrual cycle tested women during menses and the mid-
luteal phase. If the inconsistent reports of menstrual cycle effects on cognition and the stress
response are a result of progesterone presence, it may still be the case that the estradiol-only
fluctuations characteristic of the first half of the menstrual cycle can independently affect
cognition and the stress response, thereby modulating stress effects on cognition.
1.5.4 Implications for Aging Women
Estrogenic protection against glucocorticoid-induced damage and the putative attenuation
of stress reactivity by estradiol have important implications for the aging female. Based on the
evidence reviewed herein, the loss of estradiol after menopause, and therefore loss of the
hormone’s protective effects, may lead to potentiated age-related changes in HPA function.
Typical aging is accompanied by a host of changes in the body. Included in these changes
are alterations in the function of the stress response system, particularly the HPA axis. Animal
studies have found aged rodents show smaller DEX-induced corticosterone suppression to a
stressor, a treatment that significantly reduces the corticosterone response in young-adult animals
(Ferrini et al., 1999; Riegle & Hess, 1972), because of activation of the system’s negative
feedback mechanism. This has also been observed in baboons and rhesus monkeys subjected to
CRH, ACTH, and DEX tests (Goncharova & Lapin, 2002) and in a CRH test performed in dogs
(Reul, Rothuizen, & de Kloet, 1991). Aged rats also displayed prolonged corticosterone
secretion after termination of an acute stressor compared with young animals (Ferrini et al.,
1999; Sapolsky, Krey, & McEwen, 1983); with similar results observed in rodents exposed to
chronic stress (Riegle, 1973; Sapolsky et al., 1983).
36
Additionally, aged animals experience significant downregulation of glucocorticoid
receptors, particularly in the hippocampus (Ferrini et al., 1999; Mizoguchi et al., 2009; Sapolsky,
Krey, & McEwen, 1986). Loss of glucocorticoid receptors in the aged hippocampus leads to a
failure to accurately detect circulating glucocorticoid levels and thus failure to terminate HPA
axis reactivity. The resulting excess GC exposure leads to hippocampal damage and subsequent
loss of more GC receptors. This cycle of receptor loss, tissue damage, and HPA hyperactivity
was first presented as the Glucocorticoid Cascade Hypothesis (Sapolsky et al., 1986) and is a
possible model to account for the hippocampal and cognitive decline observed in aging.
Consistent with the glucocorticoid-cascade hypothesis, some report that older human
adults also exhibit higher basal cortisol, potentiated and prolonged cortisol secretion to stress
(Almela et al., 2011; Dodt et al., 1991), and reduced levels of GC receptor mRNA in plasma
compared with younger adults (Grasso, Lodi, Lupo, & Muscettola, 1997). Reduced levels of GC
receptor mRNA in plasma may indicate that receptor density in the brain also declines, leading to
failures to downregulate the HPA axis after a brief stressful event.
The previously described estradiol-related protection against glucocorticoid-induced
damage, and the putative attenuation of stress reactivity by estradiol, have important implications
for the aging female. The loss of estradiol, and its protective effects, during the post-menopause
period is likely to lead to potentiated age-related changes in HPA axis function, whereas
prolonging estradiol exposure past menopause is likely to maintain proper HPA axis function.
For instance, recall that estradiol administration upregulated GC receptors in the hippocampus of
older male rats, compared with older animals not treated with estradiol (Ferrini et al., 1999). This
type of estrogenic upregulation of GC receptors in the hippocampus could help reverse the age-
37
related hyperactivity of the HPA axis as more receptors would be available to detect circulating
GC levels and spare the hippocampus from further damage.
Importantly for human females, estradiol has been shown to counter each of the age-
related dysfunctions (e.g., hippocampal damage, receptor downregulation, and HPA axis
hyperactivity) described above. As such, estradiol supplementation past menopause should
mitigate the extent or postpone the onset of HPA axis malfunctions. Insofar as estradiol protects
the hippocampus and consequently the negative feedback system of the HPA axis, then extension
of estradiol exposure past the time of menopause may be a possible mechanism for the relatively
enhanced cognitive function observed in post-menopausal women who initiated hormone
treatment close to the time of menopause.
1.6 Conclusions
Estrogens, glucocorticoids, and stress each impact daily female functioning. Depending
on the circumstances surrounding their presence and action in the female body, each factor can
exert positive or negative effects on the system. Furthermore, estradiol can alter how
glucocorticoids and stress affect the brain and cognition. Understanding how these factors
influence the adult female can be useful in determining how one should handle certain life
events; for example, scheduling stressful events to fall on high estradiol phases of the menstrual
cycle, or initiating estradiol replacement close to the time of menopause to maximize benefits
and minimize risks.
It is important, however, not to speak in absolutes when discussing the effects of these
factors, as each are associated with positives and negatives. For instance, glucocorticoids are
necessary for normal neural stimulation (Doi et al., 1991) and can be necessary or beneficial for
some types of learning, like avoidance learning (Yerkes & Dodson, 1908) and emotional
38
memory (Buchanan & Lovallo, 2001; Payne et al., 2007; Smeets et al., 2006). This is in stark
contrast to the general trend of neural damage and cognitive interference resulting from
excessive or chronic glucocorticoid application (Behl et al., 1997; Coburn-Litvak et al., 2003;
Haynes et al., 2003; MacPherson et al., 2005; Stein-Behrens et al., 1992; Tombaugh et al., 1992;
Virgin et al., 1991; Woolley, Gould, & McEwen, 1990) or stress exposure (Baum et al., 1983;
Behl et al., 1997; Brown et al., 2005; Cook & Wellman, 2004; Diamond et al., 1996; Diamond &
Rose, 1994; Haynes et al., 2003; Klein & Boals, 2001; McLaughlin et al., 2010; Ohl & Fuchs,
1999; Park et al., 2001; Tynan et al., 2010; Uno et al., 1989; Watanabe et al., 1992; Wellman,
2001). This same pattern is observed with estradiol. While most research suggests the hormone is
beneficial to neural tissue (Brinton et al., 2000; Chen et al., 2006; Gerstner et al., 2007; Hao et
al., 2007; Hoffman et al., 2003; Hosoda et al., 2001; Malagutti et al., 2009; Pike, 1999; Saravia et
al., 2007; Sharma & Mehra, 2008) and cognition (Duff & Hampson, 2000; Maki et al., 2001;
Miller et al., 2002; Smith et al., 2001; Wolf & Kirschbaum, 2002; Wolf et al., 1999), estradiol
can be detrimental in other regards – such as when administered long after menopause (Espeland
et al., 2004; Mulnard et al., 2000; S. R. Rapp et al., 2003; Resnick et al., 2006; Shumaker et al.,
2003) and for emotional memory (Gasbarri et al., 2008; Mather et al., under review; Sakaki &
Mather, 2012). Thus while each of these factors can be associated with a typical outcome, they
each have exceptions where they exert an influence opposite to what is considered within their
range of effects. It also is important that these factors work in concert to affect females. Not only
do estrogen and stress result in dramatically different effects in the brain and cognition, but
estrogen can act directly on the stress response to modulate the magnitude of response to a
stressor (Ceresini et al., 2000; Del Rio et al., 1998; Komesaroff et al., 1999; Lindheim et al.,
1992; Puder et al., 2001).
39
The cortisol and estradiol hormone systems show opposing effects both in isolation and
in their interactions. We see that chronic stress exposure can be dangerous, estradiol can be
protective, and estradiol can act to make stress less detrimental. Each point is interesting on its
face, but the real value of the existing research is in putting these puzzle pieces together in order
to understand how such an estradiol-stress interaction affects behavior and health. In particular,
this interaction affects women in their daily lives via its influences on cognitive function. The
dampening effect of estradiol on the stress response can affect the day-to-day cognitive functions
of peri- and post-menopausal women who experience elevated cortisol levels in response to
perceived family and work stressors. For instance, women who feel under pressure to complete
family oriented tasks, and those who feel unrewarded for the amount of effort given to tasks,
experience higher cortisol levels throughout the day (Eller, Netterstrøm, & Hansen, 2006), while
those who feel work pressures are generally too high due to time pressures at work and having
too many projects to simultaneously attend to also show elevated cortisol levels (Evolahti,
Hultcrantz, & Collins, 2006). Given the potentially negative effects of these types of on-going
stress in the everyday lives of women, it is critical to further investigate the possibility of
estrogenic protection against stress-induced cognitive interference and its underlying
mechanisms. The studies presented and discussed here aim to determine if estradiol does in fact
protect women from the deleterious effects of stress on higher order cognitive functions, such as
emotional memory and working memory.

40
CHAPTER 2
Study 1: Influence of estradiol on the stress response, and stress effects on working
memory and emotional memory.
2.1 Introduction
Although the HPA and HPG systems are known to communicate and influence one
another, less attention is paid to countervailing effects the two systems exert on the body, brain,
and cognition. For example, long-term glucocorticoid exposure is linked to development of the
metabolic syndrome (Pasquali et al., 2006; Rosmond, 2005), unhealthy alterations in fat
distribution (Rebuffe-Scrive et al., 1992), promotion of hyperglycemia and hyperinsulinemia
(McGuinness et al., 1993; Rebuffe-Scrive et al., 1992), promotion of bone resorption (O'Brien et
al., 2004), and maintenance of bone degrading osteoclasts (Jia et al., 2006). In contrast,
estrogens, the primary female gonadal hormone, are linked to less unhealthy fat distribution
(Green et al., 2004; Musatov et al., 2007), lesser occurrence of hyperglycemia and
hyperinsulinemia (Krotkiewski et al., 1983; Musatov et al., 2007), and promotion and
maintenance of bone mineral density (Delmas et al., 1997; Felson et al., 1993; Sowers et al.,
1998).
Similar contrasting effects of glucocorticoids and estrogens occur in neural tissue and
cognition. For instance, while glucocorticoids are necessary for optimal neuronal functioning
(Doi et al., 1991; Nadeau & Rivest, 2003; Sapolsky et al., 1991), beneficial to memory for
emotional stimuli (Buchanan & Lovallo, 2001; Payne et al., 2007; Smeets et al., 2006), and
possibly necessary for avoidance and aversion learning (Garrido, De Blas, Giné, Santos, & Mora,
2011; Hintiryan et al., 2009; Miele et al., 1988; Nachman & Ashe, 1973; Yerkes & Dodson,
1908), chronic or extreme glucocorticoid exposure results in cell damage or death (Behl et al.,
41
1997; Gerlach & McEwen, 1972; MacPherson et al., 2005; McEwen et al., 1968; Stein-Behrens
et al., 1992; Tombaugh et al., 1992; Tynan et al., 2010; Woolley, Gould, & McEwen, 1990) and
impairments in cognitive performance. Contrary to this pattern, estrogens, particularly estradiol,
can be detrimental to already compromised neural tissue (Chen et al., 2006) and is reported to
interfere with memory for emotional items (Gasbarri et al., 2008; Mather et al., under review;
Sakaki & Mather, 2012), however, most research finds that estradiol promotes neural growth and
protection (Brinton et al., 2000; Chen et al., 2006; Gerstner et al., 2007; Hao et al., 2007; Saravia
et al., 2007; Ter Horst, Wichmann, Gerrits, Westenbroek, & Lin, 2009) and can improve
cognitive function (Baker et al., 2012; Duff & Hampson, 2000; R. Krug, Born, & Rasch, 2006;
Maki et al., 2001; Miller et al., 2002; Smith et al., 2001; Wolf & Kirschbaum, 2002; Wolf et al.,
1999). This pattern of opposing effects on the brain and cognition has important implications for
women throughout the reproductive and post-reproductive lifespan, but particularly for aging
post-menopausal women.
Aging is accompanied by a host of changes, including changes in the ability to respond,
or shut down response, to a stressor (Almela et al., 2011; Dodt et al., 1991; Wilkinson et al.,
2001). Given the deleterious effects of stress exposure on the body (McGuinness et al., 1993;
O'Brien et al., 2004; Pasquali et al., 2006; Rebuffe-Scrive et al., 1992; Rosmond, 2005), brain
(Behl et al., 1997; Gerlach & McEwen, 1972; MacPherson et al., 2005; McEwen et al., 1968;
Stein-Behrens et al., 1992; Tombaugh et al., 1992; Tynan et al., 2010; Woolley, Gould, &
McEwen, 1990), and cognition (Alexander, Hillier, Smith, Tivarus, & Beversdorf, 2007; Duncko
et al., 2009; Elzinga & Roelofs, 2005; Luethi, Meier, & Sandi, 2009; Oei et al., 2006; Schoofs et
al., 2008; Schoofs et al., 2009; Young, Sahakian, Robbins, & Cowen, 1999), and the already
increased incidence of illness in the aged, the inability to effectively shut down the stress
42
response may make aging persons more susceptible to neural and cognitive decline. Aging
women, however, may be able to stave off the decline in effective stress reactivity by
maintaining estradiol levels beyond menopause, as suggested by evidence showing short-term
estradiol interventions, ranging in time from 1 day to 8 weeks, can attenuate the stress response
(Ceresini et al., 2000; Del Rio et al., 1998; Komesaroff et al., 1999; Lindheim et al., 1992; Puder
et al., 2001).
While various benefits of estradiol treatment have been studied, there has not been much
attention paid to the possible benefits of the hormone for attenuating stress reactivity. Protection
of this type may result in protection against declines in neural integrity and therefore cognition,
given that animal studies show that stress hormones target and damage areas important for
cognition, such as the hippocampus and prefrontal cortex (MacPherson et al., 2005; McEwen &
Morrison, 2013; Tynan et al., 2010). Thus, the rapid and dramatic decrease in estradiol
production and circulation during and after menopause may leave middle-aged and older women
more vulnerable to the detrimental effects of stress hormone exposure on neural and cognitive
integrity. However, maintenance of estradiol past the menopause transition may attenuate some
of these age-related changes in the stress response and stress-induced declines in the brain and
cognition.
With this study we aimed to address what effects estradiol might have on the
physiological and behavioral effects of stress. Specifically, we tested three hypotheses regarding
the effects of estradiol treatment after menopause. First we hypothesized that estradiol treatment
after menopause would be associated with attenuated cortisol release in response to a stressor,
the second hypothesis stated women taking estradiol would show less stress-induced interference
43
in working memory performance, and the third hypothesis stated that the women taking estradiol
would display decreased stress-induced enhancement of emotional memory.
2.2 Methods
2.2.1 Participants
Forty-two post-menopausal women (54-87 years) were recruited from the double-blind,
placebo-controlled, Early versus Late Intervention Trail with Estradiol (ELITE) clinical trial.
Average time of enrollment in the ELITE trial, prior to participation in this estradiol-and-stress
study, was 4.9 years. Women began their participation in the ELITE trial either within 6 years of
their last menses (early initiation) or beyond 10 years of their last menses (late initiation).
Women in the early and late initiation groups were randomly assigned to take either 1mg oral
estradiol daily, or a placebo, creating 4 groups: early initiation-estradiol, early initiation-placebo,
late initiation-estradiol, and late initiation-placebo. Participants had normal or corrected vision,
were fluent in English, free of cognitive impairment, and free of conditions or medications that
would enhance risks associated with the stressor or compromise data validity of this study or the
ELITE trial.
2.2.2 Inclusionary and Exclusionary Criteria
Post-menopausal participants were required to be enrolled in the ELITE clinical trial and
be free from the following conditions and medications: heart disease, peripheral vascular disease,
diabetes, Reynaud’s phenomenon, cryoglobulinemia, vasculitis, lupus, tingling or numbness in
the hands and/or feet, or any other serious chronic illness. Subjects also were nonsmokers, were
not taking beta-blocker medications, corticosteroid-based medications, or psychoactive drugs.
Subjects also were free of any cognitive impairment, were fluent in English, and had normal or
corrected vision.
44
2.2.3 Sessions
Participants came for 2 sessions, one stress and one control session, order
counterbalanced. Women were given the option of being tested at the main USC campus (UPC)
or at the USC Health Sciences Campus (HSC). The majority of women elected to be tested at
HSC, where they were seen for their ELITE trial-related medical visits.
Sessions lasted 50-90 minutes, depending of factors such as time needed to produce
sufficient saliva samples, time taken to read and sign informed consent, and time taken to
complete questionnaires at the beginning the session.
2.2.4 Hormone Sampling
Three saliva samples were collected to assess cortisol, estradiol, and progesterone levels.
Samples were collected before the stress manipulation, prior to starting the behavioral tasks (on
average, 15 minutes after stressor onset), and after completion of the behavioral tasks (on
average, 40 minutes after stressor onset). In order to minimize variations in hormone levels,
participants were asked to refrain from exercise and food/drink (except water) within one hour,
sleep within two hours, and caffeine and alcohol within three hours of their session start time.
Participants were then asked to drink 8oz of water upon arrival to the lab in order to ensure
proper hydration for saliva production and collection of a clean saliva sample. The first saliva
sample was not taken until a minimum of 10 minutes had elapsed since the participant finished
the 8oz water bottle.
Salivary samples are a reliable source for determining biologically available, unbound,
levels of hormones (Tunn, Mollmann, Barth, Derendorf, & Krieg, 1992; Vining, McGinley, &
Symons, 1983). Saliva sample 1 was used to assess baseline cortisol and sex hormone levels; to
collect a large enough sample, participants passively drooled 1.25mL of saliva into a collection
45
tube. The remaining 2 samples only assessed cortisol and were collected using two sponge
sorbettes. Samples were labeled with a barcode containing no personal information and then
stored at 0°C until all data collection was completed. Once all samples had been collected they
were packaged and transported frozen to CLIA-certified analytical laboratories for immunoassay
(Salimetrics, LLC, State College, PA).
2.2.5 Stress Manipulation
To induce a stress response, participants completed the Cold Pressor Task (Hines Jr. &
Brown, 1936; Lovallo, 1975), which has been shown to reliably induce cortisol secretion
(Bullinger et al., 1984; Edelson & Robertson, 1986; Lighthall et al., 2009; Lighthall et al., 2011;
Mather et al., 2010). In the Cold Pressor Task (CPT) participants submerge their non-dominant
hand, up to the wrist, in ice water (0-5°C; average temperature in this study was 2.8°C or
37.04°F) for one to three minutes. The control condition uses warm water (37-40°C; average
temperature in this study was 38.3°C or 100.94°F).
2.2.6 Psychological Measures
Participants completed questionnaires for demographics, emotional state, mood, and daily
stress. Some measures were completed upon arrival at the lab and prior to the CPT, while others
were completed after the CPT and before the behavioral tasks.
The measures used were: 1) Health and Demographics form, to determine sleep habits,
stress, education level, and income; 2) the Daily Stress Inventory (Almeida, Wethington, &
Kessler, 2002) to assess the current level of stress, 3) the Positive and Negative Affective Scale
(Watson, Clark, & Tellegen, 1988) to assess emotional state; and 4) the Center for
Epidemiological Studies Depression Scale (Radloff, 1977) to determine mood. Subjects also
were screened for intellectual capabilities using the Wechsler Test of Adult Reading (Wechsler,
46
1981) as a measure of verbal intelligence. Lastly, immediately before and after the stress
manipulation subjects completed a visual analog scale indicating 1) the amount of pain and stress
they felt at the moment before stress and 2) the peak amount of pain and stress they felt during
the stress manipulation. Participants were screened for cognitive impairment using the Telephone
Interview for Cognitive Status – modified (Brandt, Spencer, & Folstein, 1988; Welsh, Breitner,
& Magruder-Habib, 1993) prior to participating in the study.
2.2.7 Behavioral Tasks
Behavioral tasks commenced after collection of Saliva Sample 2, or approximately 18
minutes after stressor onset. The delay between stressor onset and commencement of behavioral
tasks was utilized to ensure participants were experiencing peak cortisol responses during the
tasks. Other studies have shown that peak cortisol responses are observed between 15-45 minutes
after stressor onset (Kern et al., 2008; Kirschbaum et al., 1999; Kirschbaum et al., 1992;
Kudielka, Buske-Kirschbaum, et al., 2004; Kudielka et al., 1998; Kudielka & Kirschbaum, 2005;
Kudielka, Schmidt-Reinwald, Hellhammer, & Kirschbaum, 1999; Kudielka, Schommer,
Hellhammer, & Kirschbaum, 2004). If time remained between completing the second round of
questionnaires (CES-D and WTAR) and the second sample, participants completed a word
search as a filler task until 14 minutes had passed since stressor onset.
2.2.8 Emotional Memory Task: Encoding Phase
We used the same task used in Mather et al. (under review) to assess whether emotional
items were remembered differently depending on estradiol levels and cortisol levels. Participants
viewed pictures of positive, negative, and neutral valence shown in different locations on a
computer screen during an encoding phase. After viewing the pictures participants completed a
working memory task, followed by a free recall test for the pictures they viewed in the encoding
47
phase and were then tested for their memory of for where pictures were presented during the
encoding phase.
Participants viewed pictures of negative, positive, or neutral valence on a computer
screen. They viewed 24 pictures in total, 12 emotional and 12 neutral. Of the 12 emotional
pictures, 6 were negative and 6 were positive. Of the 12 neutral pictures, 6 were neutral versions
of negative photos, and 6 were neutral versions of positive pictures. Each participant only saw
either the negative, positive, or neutral version of each picture. Four versions of the task were
created and each participant viewed two versions. Of the two versions each participant viewed,
none of the negative and neutral or positive and neutral photo pairs overlapped. Photos were
presented on different locations of the screen. Each photo was presented for 2000ms. Between
each individual photograph, subjects were shown a yellow or green dot and indicated the color of
the dot by key press on a keyboard. The dot task helped ensure participants were paying attention
to the screen throughout the duration of the task. Participants were tested for their recall of the
photos as well as for photo-location associations after completing the working memory task. On
average, this portion of the task began approximately 18 minutes after stressor onset and took
approximately 2 minutes to complete. Participants viewed different pictures at each session. The
versions of the task viewed during session 1 and session 2 were counterbalanced.
2.2.9 Working Memory Task
The sentence span task was used to assess working memory. In this task, sentences were
shown one at a time on the computer screen and participants were asked to make their best
judgment as to whether the sentence “makes sense” or was “nonsense” and to remember the last
word of every sentence presented within a given block. “Makes sense” and “Nonsense”
48
judgments were recorded by key press on a computer keyboard, while word recall was recorded
on a paper scoring sheet by the experimenter.
Sentences were collected from various sources and have been used in other reading or
sentence span tasks (Copeland & Radvansky, 2001; Daneman & Carpenter, 1980; N. P.
Friedman & Miyake, 2004). Nonsense sentences were created as done in Turner and Engle
(1989), “[Semantically and Syntactically] ‘Incorrect’ sentences were made nonsense by
reversing the order of the last four…preterminal words e.g. ‘The grades for our finals will be
posted outside the classroom door’ to ‘The grades for our finals will be classroom the outside
posted door’”. Sentences were presented on the center of the screen with the last word in all
capital letters (e.g. The boy said HELLO), for a duration of 5 seconds. At the end of the 5-second
sentence presentation, the screen changed and displayed “Makes Sense” and “Nonsense”, on the
left and right side of screen, respectively. Subjects made a key press indicating whether they
thought the just-viewed sentence made sense or did not. This was followed by a 500ms inter-trial
interval, and then presentation of the next sentence. Participants completed 13 blocks in total.
Blocks 1 and 2 were practice blocks and consisted of a 1-sentence load and 2-sentence load,
respectively. At the end of each load the subject was prompted to tell the experimenter the last
word of the 1 or 2 sentences they just viewed. The main portion of the task included 4 blocks of
2-sentence loads, 3 blocks of 3-sentence loads, and 2 blocks each of 4-, 5-, and 6-sentence loads.
At the end of each block participants were prompted to tell the experimenter the last word of the
2, 3, 4, 5, or 6 sentences they just viewed, in the order they were presented in.
A lenient scoring criterion was used for this task; women were given 1 point for each
word they remembered whether or not they recalled the words in the order presented or
remembered every word in a given block. On average, this task began approximately 21.5
49
minutes after stressor onset and took approximately 11 minutes to complete. Participants saw
different sentences at each session.
2.2.10 Emotional Memory Task: Recall Phase
Immediately following completion of the working memory task, participants were tested
for their free recall of the photos presented during the encoding phase of the emotional memory
task. On average, this portion of the emotional memory task began approximately 34 minutes
after stressor onset and took approximately 4 minutes to complete.
2.2.11 Emotional Memory Task: Association Test Phase
Immediately following the recall portion of the emotional memory task, participants were
tested for their memory of photo-location association of the photos presented during the
encoding phase of the emotional memory task. Subjects viewed the same picture displayed on
two different locations of the computer screen, simultaneously, and indicated the correct location
of the picture by key press. Participants were only shown the same pictures they viewed during
the encoding phase of that same session. On average, this portion of the emotional memory task
began approximately 38 minutes after stressor onset and took approximately 2 minutes to
complete.
2.2.12 Estradiol Condition and Analyses
Drug assignments for the ELITE trial have yet to be unblinded. Due to the delay in
unblinding, we conducted all analyses on estradiol level, rather than condition. Estradiol
influence on cortisol response, working memory, and emotional memory was assessed by
comparing performance of women within the top and bottom quartiles of estradiol levels (n=10).
Average estradiol level for women in the bottom quartile was 1.2935 pg/ml, and was 97.51125
pg/ml for women in the top quartile.
50
2.2.13 Statistics
Pearson’s correlation analyses were conducted on questionnaires completed during
sessions 1 and 2. Positive affect, negative affect, CES-D scores, pre and post water stress ratings,
and pre and post water pain ratings were correlated with estradiol levels. These analyses were
conducted on the entire sample of 42 women. All remaining analyses were conducted using the
quartile-based Low Estradiol (LE) and High Estradiol (HE) groups.
Analyses of variance (ANOVA) were conducted for effects of estradiol level and stress
on cortisol response, working memory performance, recall of emotional and neutral pictures, and
picture-location associations. Significance was set at p≤0.05.
Analyses for working memory performance were conducted for the proportion of words
recalled within each load, as well as the overall proportion of words recalled, collapsed across
loads. Analyses for the emotional memory recall test and emotional memory picture-location
association test also were conducted for the proportion of the pictures and locations remembered
within each valence and collapsed across valences.
2.3 Results
2.3.1 Participants: Demographic Information for Low E2

versus High E2

Women
Low Estradiol (LE) and High Estradiol (HE) women did not differ in subjective ratings of
their stress level at the start of session 1 or session 2, nor did they differ in their ratings for their
stress levels on study days compared to their stress levels on “normal days”. Nor did LE and HE
women differ in their positive affect, negative affect, or depression scores at session 1 or session
2.
LE and HE women did differ in years of education (t=-2.214, df=18, p=.040), and
subjective ratings of overall health at session 1 (t=-2.75, df=18, p=.013) and session 2 (t=-3.349,
51
df=18, p=.004). Overall subjective health ratings were based on a scale from 1 to 9, with 1 being
very poor health and 9 being excellent health, that both groups averaged ratings over 7 suggests
both groups of women felt subjectively healthy. While women in the LE and HE groups differed
in their years of education, they did not differ in their performance on the WTAR (t=-.471,
df=18, p=.643).
Lastly, of course, estradiol levels did differ between the low and high estradiol women
(t=-2.936, df=18, p=.009). See table 2.2.
2.3.2 Correlation Analyses between Estradiol Levels and Questionnaire Responses
There was no relationship between estradiol levels and positive affect, negative affect,
pre-water pain, post water pain, or post water stress during session 1. There was, however, a
significant negative correlation between estradiol levels and pre-water stress ratings (r=-5.30,
p=.016), such that women with higher estradiol levels reported feeling less stressed before
placing their hand in water during session 1. No relationships between estradiol levels and any
questionnaire were found during session 2.
2.3.3 Subjective Ratings of Stress and Pain Immediately Before and After Stress Exposure
Due to the negative correlation between estradiol and pre-water stress ratings, we
conducted a 2 (stress: cold v warm water) x 2 (estradiol) within-by-between MANCOVA,
controlling for session 1 stress condition, on pre-water stress ratings. This analysis revealed that
LE women reported higher levels of subjective stress prior to placing their hands in water than
did HE women (F=6.41, p=.021).
Despite the higher baseline levels of subjective stress immediately prior to water
exposure in LE women, follow-up 2 (stress: cold v warm) x 2 (time: pre v post stress) x 2
(estradiol: low v high) within-by-between MANCOVA, controlling for session 1 condition,
52
found no three way interaction between stress, time, and estradiol (F=1.543, p=.231; see figure
2.1), indicating there was no difference in the magnitude of change in subjective stress ratings
from pre stress to post stress between LE and HE women during either the stress or control
sessions. Follow-up 2 (time: pre v post stress) x 2 (estradiol: low v high) analyses for the stress
session alone, while still controlling for session 1 condition, also found no two way interaction
(F=2.915, p=.106), indicating that there was no difference in the magnitude of change from pre
stress to post stress between LE and HE women during the stress session. Likewise, no
interaction was found when the analysis was conducted for the control session alone (F=.422,
p=.524).
A 2 (stress: cold v warm) x 2 (time: pre v post stress) x 2 (estradiol: low v high) within-
by-between ANOVA revealed no overall stress x time x estradiol interaction, suggesting there
was no difference between LE and HE women in the change of their subjective pain ratings from
immediately before placing their hand in water to the peak amount of pain felt while their hand
was in the water during both the stress and control sessions (F=.351, p=.561; see figure 2.2).
Follow-up 2 (time: pre v post stress) x 2 (estradiol: low v high) analyses for the stress session
only revealed a main effect of time (F=51.676, p<.001), such that subjective pain ratings
increased from pre stress to post stress (means: pre=4.95, post=54.15). However, there was no
estradiol x time interaction (F=1.046, p=.32), suggesting that there was no difference in the
change from pre to post stress between LE and HE women. Similar analyses conducted for the
control session revealed a main effect of time (F=5.766, p=.027), with subjective pain ratings
decreasing from pre stress to post stress (means: pre=4.5, post=0.4). However, there was no
estradiol x time interaction (F=2.885, p=.107), suggesting that there was no difference in the
change from pre to post stress between LE and HE women.
53
2.3.4 Cortisol Response for Low E2

versus High E2

Women During the Stress and Control
Sessions
A 2 (stress: cold v warm) x 2 (estradiol) within-by-between ANOVA comparing baseline
cortisol levels between LE and HE women in the stress and control sessions, revealed that
baseline cortisol levels were comparable during the stress and control sessions (F=.721, p=.407).
The analysis also failed to find a stress x estradiol interaction (F=1.176, p=.292), indicating LE
and HE women had comparable baseline levels of cortisol at each session. In order to account for
the higher pre-water stress levels reported by LE women during session 1, a follow-up
ANCOVA looking at only session 1 baseline cortisol levels while controlling for pre-water stress
levels was conducted. This comparison also revealed that LE and HE women had comparable
baseline cortisol levels (F=.000, p=.989).
A 2 (stress: cold v warm) x 2 (time: baseline v 15 minutes post stress onset) x 2
(estradiol: low v high) within-by-between ANOVA on cortisol levels revealed no main effect of
stress (F=1.313, p=.267), but a main effect of time (F=4.668, p=.044), such that cortisol levels
increased from baseline to 15 minutes post stress onset, and a stress x time interaction (F=8.559,
p=.009), revealing that cortisol levels increased from baseline to 15 minutes post stress onset in
the stress session, but decreased between the two time points during the control session; see
figure 2.3. Follow-up analyses revealed that the increase from baseline to 15 minutes post stress
onset was significant for the stress session (F=9.977, p=.005), but the decrease observed for the
control session was not significant (F=.161, p=.693).
This analysis also uncovered a time x estradiol interaction (F=7.881, p=.012), such that
when collapsed across sessions, LE women displayed increases in cortisol from baseline to 15
minutes post stress onset, while HE women displayed a decrease in cortisol; see figure 2.4.
54
Follow-up analyses conducted on cortisol levels during the stress session only revealed a main
effect of time (F=9.977, p=.005) and a time x estradiol interaction (F=5.58, p=.03) revealing that
LE women showed a much larger increase in cortisol levels from baseline to 15 minutes post
stress onset than the HE women; see figure 2.5. Additional follow-up analyses found that the
increase in cortisol observed in LE women was significant (F=9.047, p=.015), but the increase
observed in HE women was not statistically significant (F=1.005, p=.342).
Further follow-up analyses controlling for pre-water stress levels during the stress session
still found the same pattern of results. A 2 (time: baseline v 15 minutes post stress onset) x 2
(estradiol) within-by-between ANCOVA, using pre-water stress ratings as a covariate, uncovered
a main effect of time (F=10.205, p=.005), such that cortisol levels increased from baseline to 15
minutes post stress onset. The analysis also uncovered a time x estradiol interaction (F=7.30,
p=.015), suggesting that main effect of time was largely attributable to the larger cortisol
increase exhibited by LE women (means: baseline=.148 µg/dl, 15 minutes post onset= .354
µg/dl) versus the lack of increase in HE women (means: baseline=.150 µg/dl, 15 minutes post
onset= .150 µg/dl), despite the two groups having comparable baseline cortisol levels during the
stress session (F=.001, p=.974, means: LE=.148 µg/dl, HE= .150 µg/dl).
2.3.5 Working Memory: Word Recall
A 2 (stress: cold v warm) x 5 (load: 2 sentences through 6 sentences) x 2 (estradiol: low v
high) within-by-between ANOVA revealed a marginal main effect of stress (F=3.843, p=.066),
trending towards worse working memory performance during the stress session than control
session. A main effect of load was uncovered (F=29.870, p<.001), revealing that performance
decreased as load increased. No estradiol x load or stress x load interactions were detected,
suggesting LE and HE women showed the same pattern of performance decrement as loads
55
increased in difficulty, with the same pattern also observed across stress and control conditions.
There was, however, a marginal stress x estradiol interaction (F=4.138, p=.057), where LE
women showed a larger difference in performance across stress and control sessions (means:
control=.598, stress=.511) while HE women performed almost the same in each session (means:
control=.543, stress=.542). The difference in means between LE and HE women for control
session performance did not differ statistically (t=.828, df=18, p=.418).
We next conducted a 2 (stress: cold v warm) x 2 (estradiol: low v high) ANOVA on
overall working memory performance, collapsed across loads. This analysis again revealed a
marginal effect of stress (F=3.815, p=.067) and marginal stress x estradiol interaction (F=3.815,
p=.067). Trends also were for worse performance during the stress session, and greater
impairment after stress in LE women than HE women. Follow-up analyses for the LE women
alone revealed an effect of stress (F=5.099, p=.05), where they exhibited worse performance
during the stress session than during the control session. HE women, however, showed no
decrement in performance after stress compared to their control session (F=.000, p=1; see figure
2.6).
2.3.6 Emotional Memory: Free Recall of Emotional and Neutral Pictures
A 2 (stress: cold v warm) x 2 (valence: emotional pictures v neutral pictures) x 2
(estradiol: low v high) within-by-between ANOVA on picture recall revealed no main effect of
stress (F=.614, p=.444), but did find a main effect of valence; see figure 2.7. This analysis also
failed to find stress x valence, stress x estradiol, or valence x estradiol interactions.
We next conducted follow-up analyses for each valence of picture presented: negative,
positive, and neutral. A 2 (stress: cold v warm) x 2 (valence: negative pictures v neutral pictures)
x 2 (estradiol: low v high) within-by-between ANOVA on picture recall revealed no main effect
56
of stress (F=.297 p=.592), but did find a main effect of valence (F=16.141, p=.001; means:
negative=.304, neutral=.183). This analysis also failed to find stress x valence, stress x estradiol,
or valence x estradiol interactions, although the stress x valence interaction was marginal
(F=3.349, p=.084; see figure 2.8) with a smaller difference between the proportion of negative
pictures and neutral picture remembered during the stress session than the difference observed in
the control session. Further follow-up analyses revealed that the difference between recall of
negative and neutral pictures during the stress session was not statistically significant (F=.673,
p=.423), but the difference between recall of negative and neutral pictures during the control
session was significant (F=9.176, p=.007).
Analyses looking at positive versus neutral pictures found no effects of stress or valence,
or any interactions. However, the main effect of valence was marginal with F=4.315 and p=.052
(means: positive=.275, neutral=.2). Likewise, analyses looking at negative versus positive
pictures found no effects of stress or valence, or any interactions; nor did any of the comparisons
approach significance.
2.3.7 Emotional Memory: Memory for Picture Location
A 2 (stress: cold v warm) x 2 (valence: emotional pictures v neutral pictures) x 2
(estradiol: low v high) within-by-between ANOVA on memory for where on the computer
screen pictures were presented during the encoding phase revealed no main effect of stress
(F=.086, p=.773), valence (F=.208, p=.654) or stress x valence, stress x estradiol, or valence x
estradiol interactions.
We next conducted follow-up analyses for each valence of picture presented: negative,
positive, and neutral. A 2 (stress: cold v warm) x 2 (valence: negative pictures v neutral pictures)
x 2 (estradiol: low v high) within-by-between ANOVA on memory for where pictures were
57
presented on the computer screen during encoding revealed no main effect of stress (F=.497
p=.49), valence (F=2.718, p=.117), or stress x valence, stress x estradiol, or valence x estradiol
interactions. Likewise analyses looking at memory for where pictures were presented on the
computer screen during encoding between positive versus neutral pictures and positive versus
negative pictures found no effects of stress or valence, or any interactions.
2.4 Discussion
Study 1 tested three main hypotheses regarding the effects of estradiol treatment on HPA
reactivity to a stressor. Hypothesis 1 focused on the cortisol response to a stressor and stated that
HE women would display a blunted cortisol response to an ice water stressor compared with LE
women. The results of the study support this hypothesis. The notion that estradiol treatment
would minimize cortisol release created the foundations for the second and third hypotheses,
such that reductions in cortisol release should mitigate the effects of stress on other domains
because less GC would be available to act on and affect other processes. Following this
reasoning, the second hypothesis focused on the ability of estradiol treatment to change the
effects of stress on working memory. We stated that HE women would show less stress-induced
interference of working memory than LE women, and indeed our results support this hypothesis
as well. The third hypothesis focused on how estradiol treatment might alter the effects of stress
on emotional memory. We stated that HE women would show less stress-induced enhancement
of emotional memory than LE women, however this hypothesis was not supported by the current
study.
2.4.1 Differences In Overall Health Ratings And Education Between HE And LE Women
Low E2 and High E2 women differed in their subjective ratings of their overall health
and education, with HE women reporting higher overall health ratings and having a higher
58
education. Overall health ratings were made based a scale of 1 to 9, with 1 being very poor
health and 9 being excellent health. The average rating for both groups of women was above 7,
suggesting that both groups of women felt relatively healthy. The slightly higher rating provided
by HE women, however, is not unexpected. Menopausal women taking estradiol supplements are
often relieved of the discomforts associated with menopause such as hot flashes (Nelson, 2004),
vaginal dryness, other urogenital changes, and decreased libido (Bachmann & Leiblum, 2004).
The HE women, presumably taking estradiol supplements through the ELITE trial, should have
fewer menopausal symptoms than the LE women, and this may contribute to their slightly higher
ratings of overall health.
HE women were found to have more years of education than the LE women, albeit the
difference was relatively small since the average for both groups was over 16 years of education.
In fact, of the 10 women in each group, 9 LE women had 16 or more years of education, versus
all 10 of the HE women. Further, to check that the years of education did not alter vocabulary
capabilities, we compared LE and HE women on their WTAR performance and found no
difference. Given that WTAR performance and years of education were positively correlated
(r=.490, p=.001) and with linear regression analysis revealing years of education to be a
significant predictor of WTAR performance (F=12.654, p=.001) in our sample of women,
finding that LE and HE women did not differ in their WTAR performance despite the difference
in years of education suggests the LE and HE groups were comparable in their overall level of
education.
2.4.2 Subjective Ratings of Pre-water Stress Levels Between LE and HE Women
With regard to the subjective ratings of pre and post water stress and pain levels, we did
find a difference between LE and HE women on their subjective ratings of pre-water stress levels
59
during session 1. This finding suggests that LE women may have felt more anticipatory stress
immediately prior to placing their hand in water than the HE women. Importantly, when first
arriving in the laboratory women completed two scales for the amount of stress they were feeling
that day on a scale from 1 to 9. The first scale asked how stressed they felt that day and was
marked as 1 being very low, 5 being usual, and 9 being very high. They also completed a scale
rating “how does your stress level today compare to your usual stress level?” with 1 being much
lower, 5 being same as usual, and 9 being much higher. LE and HE women did not differ on
either of these scales at the beginning of the session, suggesting the LE women had a larger
subjective increase in stress levels from the beginning to the session to the moments immediately
before placing their hand in water.
This change from having comparable stress levels at the beginning of the session to
higher stress levels immediately before placing their hand in water for the first time does fall
under our general hypothesis that estradiol treatment, or higher estradiol levels, are associated
with a minimized stress response. If HE women are less likely to feel anticipatory stress in the
moments before a stressful event, they may be offered psychological protection from the
anticipation of stressor onset, which may contribute to reducing their stress response.
2.4.3 Baseline Cortisol Levels in LE and HE Women
Despite higher subjective ratings of stress prior to hand immersion during session 1, LE
and HE women had comparable baseline cortisol levels during that session. Unlike session 1,
subjective ratings of pre-water stress levels did not differ during session 2, and again neither did
baseline cortisol levels.
Our hypothesis that HE women experience blunted cortisol responses to a stressor may
partially address why this group had lower subjective stress ratings during session 1, but does not
60
address why LE women did not then have higher baseline cortisol levels during this session. A
potential explanation for this may be a result of the time course of the cortisol response. Recall
that cortisol levels peak between 15-45 minutes after stressor onset (Kern et al., 2008;
Kirschbaum et al., 1999; Kirschbaum et al., 1992; Kudielka, Buske-Kirschbaum, et al., 2004;
Kudielka et al., 1998; Kudielka & Kirschbaum, 2005; Kudielka et al., 1999; Kudielka,
Schommer, et al., 2004), and that the baseline saliva sample was taken immediately before
women rated their pre-water subjective stress levels. It may be the case that baseline cortisol
levels could have differed if stress ratings were collected prior to the saliva sample and if there
had been ample time between the two events. This, however, would not be an ideal protocol for
assessing changes in subjective stress before and after stress onset.
2.4.4 Cortisol Response in LE and HE Women
HE women displayed a blunted cortisol response after ice water exposure compared with
the LE women, suggesting that higher estradiol levels, presumably as a result of taking estradiol
supplements, dampens the cortisol response to a stressor. This pattern was maintained when
controlling for session 1 pre-water stress ratings and stress session pre-water stress ratings,
suggesting this effect is robust.
Based on the literature reviewed here, two possibilities exist, one based on potential
short-term effects of estradiol, and the other based on potential long-term effects of estradiol.
The first, potential short-term, effect is that the HE women had higher levels of corticosteroid
binding globulin (CBG), limiting the amount of unbound, biologically active cortisol available.
Saliva samples assess the amount of unbound cortisol, and so it may be the case that HE and LE
women released the same amount of cortisol but that the HE women were protected from much
of the cortisol released as it bound to CBG. This mechanism would still exert a protective effect
61
on neural tissue and cognition as the proportion of the bound hormone would be unable to act on
tissue.
The second, potential long-term, effect is that the HPA axis of HE women functions more
optimally than the HPA axis of LE women. Women assigned to estradiol in the ELITE trial had
been using estradiol for almost 5 years, on average, which may be enough time to protect the
HPA axis from age-related degradation in the negative feedback loop of the system. Maintenance
of efficient functioning of the axis would allow for better regulation of cortisol release in
response to a stressor, lower peaks in cortisol levels after stress exposure, and faster shutdown of
the HPA axis after stressor removal. The pattern of results for saliva sample 3 suggests that the
HE women do not reach the same peak levels of cortisol as the LE women, at least within the
timeframe of this study (i.e., within 38 minutes of stress onset). Although the difference between
LE and HE groups 38 minutes post stress onset was nonsignificant (t=.923, df=18, p=.368), the
means for the HE and LE groups do trend toward a pattern of HE women remaining much lower
than LE women with regard to cortisol release post stress exposure (means: LE 38 minuets post
onset=.387 µg/dl, HE 38 minutes post onset=.286 µg/dl).
A third possibility based on results stemming from this study, could be that HE women
experience less anticipatory stress prior to potential stress exposure. Whereas LE women may
experience higher anticipatory stress, resulting in an additive effect of anticipatory stress and
actual stress exposure on cortisol response, HE women may only experience cortisol responses to
a direct stressor. An alternative explanation along similar lines is that HE women are more
impervious to the effects of feeling stress. Although there was no interaction between pre-water
and post-water stress ratings between the LE and HE women, and no significant difference when
post-water ratings were compared directly (t=-1.299, df=18, p=.210), HE women did have a
62
numerically higher post-water stress rating than LE women (means: LE=36.2, HE=53.5). Yet,
despite reporting a numerically higher stress level after stressor removal, these women still
showed markedly lower cortisol release than the LE women. This pattern may point to a level of
resistance against stress in women with high estradiol levels, which may aid in maintaining
proper function of the HPA axis and neural integrity of regions most affected by GC exposure.
2.4.5 Effects of Stress on Working Memory and Emotional Memory in LE and HE Women
Based on the cortisol results of this study it would be expected that HE women would
show less decline in working memory performance from the control to stress session than the LE
women, and this seems to be the case. Given that HE women exhibited less cortisol release, less
GC should have been available to interfere with neural regions involved in working memory
performance. Again, the decreased availability of cortisol could result from less cortisol release
in response to the stressor, or from higher CBG levels minimizing the amount of cortisol that can
act on tissue. Unfortunately, this study cannot address which mechanism is at play. In order to
assess total cortisol levels (i.e., bound and unbound) one would need blood samples. Collecting
blood samples for a stress study could have confounding effects in that the insertion of the needle
for blood draw could be considered a stressor and increase cortisol levels in the absence of a
stressor or have an additive effect of stress during the stress session.
Yet, regardless of the means by which cortisol levels were reduced, the effect remains
that HE women seem to have some level of protection against stress when it comes to their
working memory. Protection against stress in this domain of cognition might prove to be
beneficial to postmenopausal women. Recall that perceived high pressures at work and home
were associated with higher cortisol levels in middle-aged women (Eller et al., 2006; Evolahti et
al., 2006). The perceived high pressure was a result of having multiple tasks and timelines to
63
adhere to in both cases, suggesting women experiencing this high pressure are using their
working memory to address their many tasks. Showing that estradiol can minimize the negative
effects of the stress on this cognitive domain could have important implications for how women
cope with, respond to, and perform their daily tasks when feeling this increased pressure and
associated increases in cortisol levels.
Less clear are the lack of effects in the emotional memory tests. Despite the evidence
showing that stress can affect memory for emotional items (Buchanan & Lovallo, 2001; Payne et
al., 2007; Smeets et al., 2006), this study failed to find any direct stress effects for any
combination of valence and stress interactions in picture recall or memory for where pictures
were presented on the computer screen.
With regard to picture recall, valence effects were found for comparisons of emotional
and neutral pictures, and negative and neutral pictures, but no other effects were found.
Meanwhile, no effects at all were uncovered for memory of where pictures were presented
during encoding. While the failure to find stress effects in HE women does make sense given
their minimal cortisol response to the ice water stressor, it is not clear why the LE women also
failed to exhibit an effect of stress of emotional memory given the significant increase in cortisol
elicited by the stressor. In fact, we hypothesized that LE women would show an enhanced effect
of stress on emotional memory as a result of lower estradiol levels being associated with better
emotional memory (Sakaki & Mather, 2012) coupled with exhibiting a higher cortisol response
to the stressor. One possibility is that the effects of stress and low estradiol competed to
eliminate any effects of estradiol, stress or stress-by-estradiol interaction on emotional memory.
Emotional memory, although nonsignificant, was negatively correlated with salivary estradiol
levels (see figure 2.9). However, although also nonsignificant, emotional memory was negatively
64
correlated with cortisol levels in the LE women (see figure 2.10). While the LE women should
have exhibited better emotional memory due to the low estradiol levels, the dramatic increase in
cortisol levels may have dampened this effect, essentially negating any effects, and contributing
to the failure to see effects of stress in this group.
2.4.6 Conclusion
This study suggests that estradiol treatment after menopause may have protective effects
against stress, with regard to cortisol release and working memory. Although this finding may
have important implications for post-menopausal women, the mechanism remains unclear and
demands further attention.
Whether these effects are a result of circulating estradiol levels that can be achieved with
short-term estradiol treatment, or if they are a result of the cumulative five years that women
were using estradiol through the ELITE trial, remains unclear. This is an important issue to parse
out as it is still unclear if extended long-term estradiol treatment is feasible for post-menopausal
women, or if following an intermittent on-off scheduling for estradiol treatment is safer. In the
instance of the short-term mechanism of action for this protective effect, estradiol’s ability to
reduce the stress response would be based solely on circulating estradiol levels. This short-term
mechanism would likely be attributable to upregulation of CBG levels, thereby minimizing the
amount of bioavailable, unbound cortisol. It also is possible that the reintroduction of estradiol
into the system would result in a renewed upregulation of GC receptors leading to more efficient
and optimal HPA axis function. The long-term mechanism would involve the ability of long-
term estradiol treatment to protect the neural regions involved in shutting down the HPA
response. While this mechanism is suggested to be the most beneficial in terms of neural
integrity, cognitive function, and delaying of neurodegenerative disorders of aging, it cannot be
65
implemented until estradiol treatments are found safe for extended long-term use. Once the
mechanism of protection is understood, this line of research has the potential to further inform
the medical community of the effects of estradiol treatment. With the potential application of this
work to the general population, it is important to continue pursuit of the potential additional
benefits of this estradiol-stress interaction, its limitations, and its risks.

66
CHAPTER 3
Study 2: Influence of estradiol fluctuations during the menstrual cycle on the stress
response, and stress effects on working memory and emotional memory.
3.1 Introduction
As briefly discussed in Chapter 1, results are mixed as to whether, or how, stress or
cognition are affected by the natural hormone fluctuations characteristic of the menstrual cycle.
During the course of the 28-day cycle, women experience increases and decreases in a number of
hormones other than estradiol, such as progesterone, follicle stimulating hormone, and
luteinizing hormone. The varying levels of each hormone, and the different combination of
levels of each hormone during different phases of the menstrual cycle, make it difficult to
ascertain which hormones may be involved with any observed changes in behavior. This
increased variability likely contributes to the varying results obtained for effects of the menstrual
cycle on behavior.
For this study we were not interested in the effects of the menstrual cycle, per se, but
rather to determine if the effects of estradiol treatment observed in study 1 also hold true for
young-adult, spontaneously cycling, women. In order to limit our examination to the effects of
estradiol we focused on the portion of the menstrual cycle where only estradiol fluctuates, while
the other HPG hormones remain at relatively low and stable levels, the follicular phase of the
menstrual cycle. The follicular phase of the menstrual cycle, or the first half of the cycle, begins
on the first day of menstruation and lasts until just before ovulation. The follicular phase is
characterized by low levels of progesterone, follicle stimulating hormone, and luteinizing
hormone, but increasing estradiol levels. During the early follicular phase, or when women are
menstruating, estradiol also is at its lowest levels. A few days later, or the days leading up to
67
ovulation, women experience a sharp increase in estradiol levels, until it reaches its highest
levels of the monthly cycle.
In order to pick up on the largest differences between phases, studies often focus on the
follicular phase, where only estradiol levels vary, versus the luteal phase, where both
progesterone and estradiol are higher. Thus limiting our examination to the follicular phase
where estradiol alone varies, may make it difficult to uncover robust differences, however, in
order to examine the effects of estradiol fluctuations in young spontaneously cycling females we
must limit the examination to only this phase. Our hypotheses for this second study were similar
to study 1. Specifically, that higher estradiol levels would be associated with attenuated cortisol
release in response to a stressor, that higher estradiol levels would be associated with less stress-
induced interference in working memory performance, and higher estradiol levels would be
associated with decreased stress-induced enhancement of emotional memory.
3.2 Methods
3.2.1 Participants
Twenty-seven young-adult, spontaneously-cycling, undergraduate females from USC
(18-24 years) were recruited for this study. Each participant attended four sessions. Two sessions
occurred during the low estradiol/low progesterone, Early Follicular phase (EF) and two sessions
occurred during the high estradiol/low progesterone, Late Follicular (LF) phase. Whether
participants were first seen in the early or late follicular phase was counterbalanced.
3.2.2 Inclusionary and Exclusionary Criteria
Young women participating in the study were free from the following conditions and
medications: heart disease, peripheral vascular disease, diabetes, Reynaud’s phenomenon,
cryoglobulinemia, vasculitis, lupus, tingling or numbness in the hands and/or feet, or any other
68
serious chronic illness. Subjects also were nonsmokers, were not taking hormone contraceptives,
beta-blocker medications, corticosteroid-based medications, or psychoactive drugs. In order to
participate subjects further were not allowed to have used hormone birth control in the last 6
months or been pregnant in the last year. Subjects also were fluent in English, and had normal or
corrected vision.
3.2.3 Sessions
Participants came in for 4 sessions, two sessions during the early follicular phase and two
sessions in the late follicular phase, order of first phase seen was counterbalanced. Within each
phase participants underwent one stress and one control session, order also counterbalanced. The
EF phase was defined as days 1-5 of the menstrual cycle, with the first day of menses being day
1. The LF phase was defined as days 8-12 of the menstrual cycle, with the first day of menses
being day 1. Women were screened for cycle regularity before participating. During this
screening, women reported the expected start date of their next menstrual cycle and were not
seen for their sessions until they reported the start of their menses to study personnel. Women
first seen during the EF phase completed all 4 sessions within the same menstrual cycle, whereas
women first seen during the LF phase completed their 4 sessions across two menstrual cycles. In
order to reduce individual variability in stress hormone levels, all sessions were conducted in the
afternoons between 1pm and 7pm, with no session starting later than 5pm.
Sessions lasted approximately 55 minutes, depending on factors such as time needed to
produce sufficient saliva samples, time taken to read and sign informed consent, and time taken
to complete questionnaires at the beginning the sessions.

69
3.2.4 Hormone Sampling
Three saliva samples were taken to assess cortisol, estradiol, and progesterone levels, and
were collected before the stress manipulation, prior to starting the behavioral tasks (on average,
17 minutes after stressor onset), and after completion of the behavioral tasks (on average, 43
minutes after stressor onset). In order to minimize variations in hormone levels, participants were
asked to refrain from exercise and food/drink (except water) within one hour, sleep within two
hours, and caffeine and alcohol within three hours of their session start time. Participants were
then asked to drink an 8oz bottle of water upon arrival to the lab in order to ensure proper
hydration for saliva production and collection of a clean saliva sample. The first saliva sample
was not taken until a minimum of 10 minutes had elapsed since the participant finished the 8oz
water bottle.
Salivary samples are a reliable source for determining biologically available, unbound,
levels of hormones (Tunn et al., 1992; Vining et al., 1983). Saliva sample 1 was used to assess
baseline cortisol and sex hormone levels; to collect a large enough sample, participants passively
drooled 1.25mL of saliva into a collection tube. The remaining 2 samples assessed cortisol levels
after water exposure as well as sex hormones. Samples were labeled with a barcode containing
no personal information and then stored at 0°C until all data collection was completed. Once all
samples had been collected they were analyzed in house. Sex hormone levels have not yet been
processed. Once the samples are analyzed, estradiol and progesterone levels from all three
samples will be averaged to obtain the most accurate estradiol and progesterone levels for that
particular session.

70
3.2.5 Stress Manipulation
To induce a stress response, participants completed the Cold Pressor Task (Hines Jr. &
Brown, 1936; Lovallo, 1975), which has been shown to reliably induce cortisol secretion
(Bullinger et al., 1984; Edelson & Robertson, 1986; Lighthall et al., 2009; Lighthall et al., 2011;
Mather et al., 2010). In the CPT participants submerge their non-dominant hand, up to the wrist,
in ice water (0-5°C) for one to three minutes. The control condition uses warm water (37-40°C).
3.2.6 Psychological Measures
Participants completed questionnaires for demographics, emotional state, mood,
symptoms of PMS, and daily stress. Measures were completed upon arrival at the lab and prior to
the CPT.
The measures used were: 1) Health and Demographics form, to determine sleep habits,
stress, education level, and income; 2) the Daily Stress Inventory (Almeida et al., 2002) to assess
the current level of stress, 3) the Positive and Negative Affective Scale (Watson et al., 1988) to
assess emotional state; 4) the Center for Epidemiological Studies Depression Scale (Radloff,
1977) to determine mood; 5) the Premenstrual Tension Syndrome Visual Analogue Scale to
determine symptoms of PMS (Steiner, Peer, Macdougall, & Haskett, 2011). Subjects also were
screened for intellectual capabilities using the Wechsler Test of Adult Reading (Wechsler, 1981)
as a measure of verbal intelligence. Lastly, immediately before and after the stress manipulation
subjects completed a visual analog scale indicating the 1) the amount of pain and stress they felt
at the moment before stress exposure and 2) the peak amount of pain and stress they felt during
the stress manipulation.

71
3.2.7 Behavioral Tasks
Behavioral tasks began after collection of Saliva Sample 2, or approximately 18 minutes
after stressor onset. The delay between stressor onset and commencement of behavioral tasks
ensured participants were experiencing peak cortisol responses during the tasks. Other studies
have shown that peak cortisol responses are observed between 15-45 minutes after stressor onset
(Kern et al., 2008; Kirschbaum et al., 1999; Kirschbaum et al., 1992; Kudielka, Buske-
Kirschbaum, et al., 2004; Kudielka et al., 1998; Kudielka & Kirschbaum, 2005; Kudielka et al.,
1999; Kudielka, Schommer, et al., 2004). In the time between completing the questionnaires and
collection of the second saliva sample, participants completed a word search as a filler task until
14 minutes had passed since stressor onset.
3.2.8 Emotional Memory Task: Encoding Phase
We used the same task used in Mather et al. (under review) to assess whether emotional
items were remembered differently depending on estradiol levels and cortisol levels. Participants
viewed pictures of positive, negative, and neutral valence shown in different locations on a
computer screen during an encoding phase. After viewing the pictures participants completed a
working memory task, followed by a free recall test for the pictures they viewed during the
encoding phase, and then tested for their memory of picture-location associations.
Participants viewed pictures of negative, positive, or neutral valence on a computer
screen. They viewed 24 pictures in total, 12 emotional and 12 neutral. Of the 12 emotional
pictures, 6 were negative and 6 were positive. Of the 12 neutral pictures, 6 were neutral versions
of negative photos, and 6 were neutral versions of positive photos. Each participant only saw
either the negative, positive, or neutral version of each picture. Eight versions of the task were
created and each participant viewed four versions. Of the four versions each participant viewed,
72
none of the negative and neutral or positive and neutral photo pairs overlapped. Photos were
presented on different locations of the screen. Each photo was presented for 2000ms. Between
each individual photograph, subjects were shown a yellow or green dot and indicated the color of
the dot by key press on a keyboard. The dot task helped ensure participants were paying attention
to the screen throughout the duration of the task. Participants were tested for their recall of the
photos as well as for photo-location associations after completing the working memory task. On
average, this portion of the task began 18 minutes after stressor onset and took approximately 2
minutes to complete. Participants viewed different pictures at each session. The versions of the
task viewed across sessions were counterbalanced.
3.2.9 Working Memory Task
The sentence span task was used to assess working memory. In this task, sentences were
shown one at a time on the computer screen and participants were asked to make their best
judgment as to whether the sentence “makes sense” or was “nonsense” and to remember the last
word of every sentence presented within a given block. “Makes sense” and “Nonsense”
judgments were recorded by key press on a computer keyboard, while word recall was recorded
on a paper scoring sheet by the experimenter.
Sentences were collected from various sources and have been used in other reading or
sentence span tasks (Copeland & Radvansky, 2001; Daneman & Carpenter, 1980; N. P.
Friedman & Miyake, 2004). Nonsense sentences were created as done in Turner and Engle
(1989), “[Semantically and Syntactically] ‘Incorrect’ sentences were made nonsense by
reversing the order of the last four…preterminal words e.g. ‘The grades for our finals will be
posted outside the classroom door’ to ‘The grades for our finals will be classroom the outside
posted door’”. Sentences were presented on the center of the screen with the last word in all
73
capital letters (e.g. The boy said HELLO), for a duration of 4 seconds. At the end of the 4-second
sentence presentation, the screen changed and displayed “Makes Sense” and “Nonsense”, on the
left and right side of screen, respectively. Subjects made a key press indicating whether they
thought the just-viewed sentence made sense or did not. This was followed by a 500ms inter-trial
interval, and then presentation of the next sentence. Participants completed 13 blocks in total.
Blocks 1 and 2 were practice blocks and consisted of a 1-sentence load and 2-sentence load,
respectively. At the end of each load the subject was prompted to tell the experimenter the last
word of the 1 or 2 sentences they just viewed. The main portion of the task included 4 blocks at
the 2-sentence load, 3 blocks of the 3-sentence load, and 2 blocks each of the 4-, 5-, and 6-
sentence loads. At the end of each block participants were prompted to tell the experimenter the
last word of the 2, 3, 4, 5, or 6 sentences they just viewed, in the order they were presented in.
Analyses were conducted using a lenient scoring criterion. The lenient scoring criterion
involved giving women 1 point for each word they remembered whether or not they recalled the
words in the order presented or remembered every word in a given block. On average, this task
began approximately 23 minutes after stressor onset and took approximately 11 minutes to
complete. Participants saw different sentences at each session.
3.2.10 Emotional Memory Task: Recall Phase
At least thirty-five minutes following stress onset, participants were tested for their free
recall of the photos presented during the encoding phase of the emotional memory task. On
average, this portion of the emotional memory task began 36 minutes after stressor onset and
took approximately 3 minutes to complete.

74
3.2.11 Emotional Memory Task: Association Test Phase
Immediately following the recall portion of the emotional memory task, participants were
tested for their memory of where photos were presented during the encoding phase of the
emotional memory task. Subjects viewed the same picture displayed on two different locations of
the computer screen, simultaneously, and indicated the correct location of the picture by key
press. Participants were only shown the same pictures they viewed during the encoding phase of
that same session. On average, this portion of the emotional memory task began approximately
40 minutes after stressor onset and took approximately 1.5 minutes to complete.
3.2.12 Statistics
Pearson’s correlation analyses were conducted on questionnaire responses and cortisol
levels. Specifically, positive affect, negative affect, CES-D scores, PMS symptoms, pre and post
water stress ratings, and pre and post water pain ratings were tested for correlations with baseline
cortisol levels. T-tests were also conducted on these measures to examine potential differences
between the two phases. Since all measures, except the CES-D and WTAR, were completed
before the participant placed their hand in water or knew what water condition they were
receiving, T-tests were conducted on the first session of each phase. CES-D and WTAR scores
were compared across the first session of each phase, as they were considered stable enough to
not be affected by stress exposure.
Analyses of variance (ANOVA) were conducted for effects of menstrual cycle phase and
stress on cortisol response, working memory performance, recall of emotional and neutral
pictures, and picture-location associations. Significance was set at p≤0.05.
Analyses for working memory performance were conducted for the proportion of words
recalled within each load and the overall proportion of words recalled (collapsed across loads).
75
Analyses for the emotional memory recall test and emotional memory picture-location
association test also were conducted for the proportion of the pictures and locations remembered
within each valence and collapsed across valences.
3.3 Results
3.3.1 Participants: Demographic Information
Young women were between the ages of 18 and 24 (average age: 20.77 years) and had
between 12 and 18 years of education (average years of education: 14.79). Ethnic and racial
breakdown was 77.8% non-Hispanic and 22.2% Hispanic, 55.6% Asian, 18.5% Caucasian, 7.4%
biracial, 14.8% other, and 3.7% declined to state. Participants were primarily undergraduate or
graduate students at the University of Southern California, or just recently graduated with
degrees in Occupational Therapy (7.4%).
3.3.2 Correlation Analyses between Cortisol Levels and Questionnaire Responses
We first conducted Pearson’s correlation analyses looking at relationships between
baseline cortisol levels during the first EF and LF phase sessions and questionnaire responses
during the first EF and LF phase sessions. The only relationship found during the first session of
the EF phase was a positive relationship between baseline cortisol levels and the PMS symptom
of experiencing a greater inability to sleep (r=.435, p=.023). The only relationship found during
the first session of the LF phase was a positive relationship between baseline cortisol levels and
scores on the CES-D (r=.405, p=.036).
We next conducted Pearson’s correlation analyses looking at relationships between
baseline cortisol levels during the stress session of the EF and LF phases and questionnaire
responses during the stress session of the EF and LF phase. During the stress session of the EF
phase there was no relationship between baseline cortisol levels and positive affect, negative
76
affect, depression scores, pre and post stress onset pain ratings, or pre and post stress onset stress
ratings. Neither was there a relationship between baseline cortisol levels and the following
symptoms of PMS: 1) depressed mood, 2) overeating or food cravings, 3) changes in sleep
patterns related to sleeping more, 4) feeling overwhelmed or out of control, 5) experiencing more
frequent emotional mood swings, 6) decreased interest in activities, 7) feeling lethargic, easily
fatigued, or having a lack of energy, or 8) breast tenderness, bloating, or water retention.
However, baseline cortisol levels were positively correlated with the following symptoms of
PMS: 1) feeling tense, restless, or anxious (r=.400, p=.039), 2) feeling more irritable or hostile
(r=.434, p=.024), 3) difficulty concentrating (r=.404, p=.037), and 4) changes in sleep patterns
related to being unable to sleep (r=.411, p=.033). There were no relationships between baseline
cortisol and any measures during the stress session of the LF phase.
We next conducted Pearson’s correlation analyses looking at relationships between
baseline cortisol levels during the control sessions of the EF and LF phases and questionnaire
responses during the control sessions of the EF and LF phase sessions. The pattern was similar to
that seen for the stress session. During the control session of the EF phase there was no
relationship between baseline cortisol levels and positive affect, negative affect, depression
scores, or pre- and post-stress onset pain ratings. Neither was there a relationship between
baseline cortisol levels and the following symptoms of PMS: 1) depressed mood, 2) overeating
or food cravings, 3) changes in sleep patterns related to sleeping more, 4) feeling overwhelmed
or out of control, 5) experiencing more frequent emotional mood swings, 6) decreased interest in
activities, 7) feeling lethargic, easily fatigued, or having a lack of energy, 8) feeling tense,
restless, or anxious, 9) difficulty concentrating, or 10) breast tenderness, bloating, or water
retention. However, baseline cortisol levels were positively correlated with the following
77
symptoms of PMS: 1) feeling more irritable or hostile (r=.437, p=.023), 2) changes in sleep
patterns related to being unable to sleep (r=.401, p=.038), as well as pre stress onset stress ratings
(r=.435, p=.023) and post stress onset stress ratings (r=.463, p=.015). There were no
relationships between baseline cortisol and any measures during the control session of the LF
phase.
3.3.3 T-tests between EF and LF phases on Questionnaire Responses
Comparisons between the first session of the EF phase and first session of the LF phase
revealed no differences between phases with exception of 2 PMS symptoms: 1) feeling lethargic,
easily fatigued, and/or having a lack of energy (t=3.045, df=26, p=.005) and 2) breast tenderness,
bloating, and/or water retention (t=3.257, df=26, p=.003). There was marginal difference in
feeling overwhelmed or out of control (t=1.936, df=26, p=.064). In all three cases, ratings were
higher during the EF phase than the LF phase.
We next compared responses on measures during the stress and control sessions with the
EF or LF phases. During stress and control sessions of the EF phase, women only differed on
their post stress onset pain ratings (t=14.496, df=26, p<.001) and post stress onset stress ratings
(t=10.156, df=26, p<.001). Likewise, during the stress and control sessions of the LF phase,
women only differed on their post stress onset pain ratings (t=15.058, df=26, p<.001) and post
stress onset stress ratings (t=10.637, df=26, p<.001). See table 3.2.
3.3.4 Subjective Ratings of Stress and Pain Immediately Before and After Stress Exposure
We conducted a 2 (stress: cold v warm) x 2 (time: pre v post stress) x 2 (phase: EF v LF)
within subjects ANOVA on pre and post stress onset subjective stress ratings. The analysis
revealed no main effect of phase, but did reveal a main effect of stress (F=79.597, p<.001), with
higher subjective stress ratings during the stress sessions than the control sessions, and a main
78
effect of time (F=17.280, p<.001), with lower pre stress onset stress ratings than post-stress onset
stress ratings. There were no phase x stress, phase x time, or three-way phase x stress x time
interactions. There was, however, a stress x time interaction (F=121.972, p<.001; see figure 3.1),
with subjective stress ratings increasing during the stress sessions, and decreasing during the
control sessions. Follow up analyses revealed that the increase in stress ratings during the stress
session was significant in both phases (EF: t=-7.005, df=26, p<.001; LF: t=-6.555, df=26,
p<.001), and the decrease in stress ratings during the control session was significant in both
phases (EF: t=5.122, df=26, p<.001; LF: t=4.058, df=26, p<.001).
We conducted a similar 2 (stress: cold v warm) x 2 (time: pre v post stress) x 2 (phase:
EF v LF) within subjects ANOVA on pre and post stress onset subjective pain ratings. Unlike
the subjective stress ratings, this analysis did reveal a main effect of phase (F=5.284, p=.030),
with women reporting higher pain during the EF phase than during the LF phase, as well as a
main effect of stress (F=224.057, p<.001), with higher subjective pain ratings during the stress
sessions than the control sessions, and a main effect of time (F=361.497, p<.001), with lower pre
stress onset pain ratings than post stress onset pain ratings. However, like the stress ratings there
were no phase x stress, phase x time, or three-way phase x stress x time interactions. There was,
however, a stress x time interaction (F=298.766, p<.001; see figure 3.2), with subjective pain
ratings increasing during the stress session, and slightly decreasing during the control session.
Follow up analyses revealed that the increase in pain ratings during the stress session was
significant in both phases (EF: t=-16.027, df=26, p<.001; LF: t=-15.378, df=26, p<.001), while
the decrease in pain ratings during the control session was nonsignificant in both phases (EF:
t=.851, df=26, p=.402; LF: t=.086, df=26, p=.932).

79
3.3.5 Cortisol Response during the EF and LF phases
A 2 (stress: cold v warm) x 2 (phase: EF v LF) within-subject ANOVA comparing
baseline cortisol levels between the EF and LF phases during the stress and control sessions,
revealed that baseline cortisol levels were comparable during the stress and control sessions
(F=.535, p=.471). The analysis also failed to find a stress x phase interaction (F=.423, p=.521),
indicating women had comparable baseline levels of cortisol at each session during both the EF
and LF phases.
A 2 (stress: cold v warm) x 3 (time: baseline v 15 minutes post stress onset v 40 minutes
post stress onset) x 2 (phase: EF v LF) within subject ANOVA on cortisol levels revealed only a
marginal effect of stress (F=3.326, p=.08; see figure 3.3), but a main effect of time (F=5.757,
p=.006; see figure 3.4), such that cortisol levels increased from baseline to 15 minutes post stress
onset then deceased again from 15 minutes post stress onset to 40 minutes post stress onset.
Further, the significant stress x time interaction (F=10.04, p<.001; see figure 3.5), revealed that
the pattern of cortisol changes differed between stress and control sessions, with cortisol levels
increasing from baseline to 15 minutes post stress onset in the stress session, then decreasing
from 15 minutes post stress onset to 40 minutes post onset, but slightly decreasing during each
time point in the control sessions. However, this analysis failed to find a main effect of phase, or
phase x stress, phase x time, or phase x stress x time interactions.
We then conducted a follow-up 2 (stress: cold v warm) x 2 (time: baseline v 15 minutes
post stress onset) x 2 (phase: EF v LF) within subject ANOVA on cortisol levels. This reduced
design revealed no main effect of stress (F=2.916, p=.1), but a main effect of time (F=8.912,
p=.006), such that cortisol levels increased from baseline to 15 minutes post stress onset. Further,
the significant stress x time interaction (F=20.91, p<.001), revealed that the pattern of cortisol
80
changes differed between stress and control sessions, with cortisol levels increasing from
baseline to 15 minutes post stress onset in the stress session, and slightly decreasing from
baseline to 15 minutes post stress onset in the control sessions. However, this analysis also failed
to find a main effect of phase, or phase x stress, phase x time, or phase x stress x time
interactions.
Lastly, we conducted further analyses on only the stress or control session, and only the
EF or LF phase. A 2 (phase: EF v LF) x 2 (time: baseline v 15 minutes post stress onset) analysis
for the stress session only revealed no effect of phase or a phase x time interaction, but did reveal
an effect of time (F=17.528, p<.001). No effects were found for the same analysis during the
control sessions. We then conducted a 2 (stress: cold v warm) x 2 (time: baseline v 15 minutes
post stress onset) analysis for the EF phase only. This analysis revealed no main effect of stress
(see figure 3.6), but did reveal a main of time (F=4.556, p=.042) and a stress x time interaction
(F=4.431, p=.045; see figure 3.7), with cortisol increasing from baseline to 15 minutes post stress
onset during the stress session, but slightly decreasing across the two time points during the
control session. Follow-up analyses on cortisol levels at baseline and 15 minutes post stress
onset, however, failed to reveal differences (baseline: t=-.637, df=26, p=.530; 15 minutes post
onset: t=1.145, df=26, p=.263).
We next conducted a 2 (stress: cold v warm) x 2 (time: baseline v 15 minutes post stress
onset) analysis for the LF phase only. Unlike the EF comparisons, this analysis did reveal a main
effect of stress (F=7.274,p=.012; see figure 3.8), as well as a main effect of time (F=6.878,
p=.014), and a stress x time interaction (F=11.763, p=.002; see figure 3.9), with cortisol
increasing from baseline to 15 minutes post stress onset during the stress session, but slightly
decreasing across the two time points during the control session. Follow-up analyses on cortisol
81
levels at baseline and 15 minutes post stress onset, did reveal differences at 15 minutes post
stress onset, but not at baseline (baseline: t=-.432, df=26, p=.670; 15 minutes post onset:
t=3.827, df=26, p=.001).
3.3.6 Working Memory: Word Recall
A 2 (phase: EF v LF) x 2 (stress: cold v warm) x 5 (load: 2 sentences through 6
sentences) within subject ANOVA revealed only a main effect of load (F=53.226, p<.001), but
no main effect phase or stress, nor any interactions. Neither were any effects of phase or stress
found when a 2 (phase: EF v LF) x 2 (stress: cold v warm) within subject ANOVA was
conducted on overall word recall (i.e., collapsed across loads).
3.3.7 Emotional Memory: Free Recall of Emotional and Neutral Pictures
Within subject ANOVAs were conducted on phase, stress, and valence. Valence
comparisons were conducted for negative images versus neutral images, positive images versus
neutral images, negative images versus positive images, and emotional (negative and positive)
images versus neutral images.
A 2 (phase: EF v LF) x 2 (stress: cold v warm) x 2 (valence: emotional pictures v neutral
pictures) within subject ANOVA on picture recall revealed main effects of phase (F=4.683,
p=.040; see figure 3.10), with women recalling more pictures during the LF phase than the EF
phase, and valence (F=25.950, p<.001), with more emotional pictures recalled than neutral.
However, the main effect of stress was only marginal (F=3.141, p=.088) with a trend toward
higher recall during the stress session than control session. This analysis also failed to find any
significant interactions.
We next conducted follow-up analyses for each valence of picture presented: negative,
positive, and neutral. A 2 (phase: EF v LF) x 2 (stress: cold v warm) x 2 (valence: negative
82
pictures v neutral pictures) within subject ANOVA on picture recall revealed no main effect of
stress (F=1.856, p=.185), but did find main effects of phase (F=6.166, p=.020; see figure 3.11)
and valence (F=37.357, p<.001; negative > neutral). This analysis also failed to find stress x
valence, stress x phase, valence x phase, or overall phase x stress x valence interactions.
Analyses looking at positive versus neutral pictures found no effects of phase, stress or
valence, nor phase x stress, stress x valence, or phase x stress x valence interactions. However,
the phase x valence interaction was significant (F=4.381 and p=.046; see figure 3.12), with
women showing better recall of positive pictures than neutral pictures during the EF phase, but
better recall of neutral pictures than positive pictures during the LF phase. Contrasts for negative
versus positive pictures revealed no effects of phase or stress, but did reveal a main effect of
valence (F=27.283, p<.001), with negative pictures recalled more than positive pictures. The
only interaction found to be significant was the phase x valence interaction (F=8.074, p=.009;
see figure 3.13), with the difference between the proportion of negative pictures recalled versus
positive pictures being larger during the LF phase than the difference between recall of negative
and positive pictures during the EF phase.
3.3.8 Emotional Memory: Memory for Picture Location
ANOVAs were conducted on phase, stress, and valence. Valence comparisons were
conducted for negative images versus neutral images, positive images versus neutral images,
negative images versus positive images, and emotional (negative and positive) images versus
neutral images.
A 2 (phase: EF v LF) x 2 (stress: cold v warm) x 2 (valence: emotional pictures v neutral
pictures) within subject ANOVA on memory for picture location revealed no main effects of
phase, stress, or valence. Nor did we find any phase x stress, phase x valence, or phase x stress x
83
valence interactions. The stress x valence interaction, however, was significant (F=4.791,
p=.038; see figure 3.14), with better memory for where neutral pictures were presented than
where emotional pictures were presented during the stress session, but better memory for where
emotional pictures were presented than for where neutral pictures were presented during the
control session.
We next conducted follow-up analyses for each valence of picture presented: negative,
positive, and neutral. A 2 (phase: EF v LF) x 2 (stress: cold v warm) x 2 (valence: negative
pictures v neutral pictures) within subject ANOVA on memory for where pictures were
presented during encoding revealed a main effect of phase (F=4.303, p=.048; see figure 3.15),
with better memory for location of presentation during the LF phase than during the EF phase.
There was, however, no main effect of stress or valence. This analysis also failed to find phase x
stress, phase x valence, or phase x stress x valence interactions. However, the analysis did reveal
a marginally significant stress x valence interaction (F=4.166, p=.052; see figure 3.16) with
better memory for where neutral pictures were presented than where negative pictures were
presented during the stress session, but better memory for where negative pictures were
presented than where neutral pictures were presented during the control sessions.
Analyses looking at positive versus neutral pictures found no main effects of phase,
stress, or valence, nor any phase x stress, phase x valence, stress x valence, or phase x stress x
valence interactions. Contrasts for negative versus positive pictures also failed to find any main
effects or interactions.
3.3.9 Cortisol Response between Responders and Nonresponders during the EF and LF phases
Because we failed to observe effects of stress on working memory we ran analyses on
“responders” versus “nonresponders”, as well as responders only and nonresponders only for
84
cortisol levels, working memory, and emotional memory. Responders (n=19 in both phases)
were defined as women who experienced any increase in cortisol from baseline to 15 minutes
post stress onset during the stress session. Nonresponders (n=8 in both phases) were defined as
women who showed no change, or a decrease, in cortisol levels from baseline to 15 minutes post
stress onset during the stress session. Because the responders and nonresponders did not remain
the same across phases analyses were run separately for the EF and LF phases.
A 2 (stress: cold v warm) x 2 (time: baseline v 15 minutes post stress onset) x 2
(response: responders v nonresponders) within-by-between subject ANOVA on cortisol levels
during the EF phase only revealed no main effects of response category, stress, or time. Nor did
it reveal stress x response or stress x time interactions. The response x time interaction, however,
was significant (F=13.38,p=.001), showing that when collapsed across stress conditions
responders experienced an increase between the two time points, while nonresponders
experienced a decrease. The analysis also revealed a marginal overall response x stress x time
interaction (F=3.941, p=.058, see figure 3.17), suggesting the responders and nonresponders
exhibited a different pattern of cortisol responses during stress and control sessions.
A similar 2 (stress: cold v warm) x 2 (time: baseline v 15 minutes post stress onset) x 2
(response: responders v nonresponders) within-by-between subject ANOVA on cortisol levels
during the LF phase only revealed main effects of response (F=4.333, p=.048), with responders
exhibiting higher cortisol levels than nonresponders, and stress (F=6.147, p=.02) with higher
cortisol levels during the stress session than control session, but no main effect of time. Unlike
the EF phase, only the response x stress interaction was found to be nonsignificant, suggesting
that both responders and nonresponders exhibited similar patterns of cortisol levels between
stress and control sessions. The response x time interaction (F=9.259, p=.005) revealed that
85
when collapsed across stress conditions responders experienced an increase between the two
time points, while nonresponders experienced a decrease.
The stress x time interaction (F=5.287, p=.03) revealed that cortisol levels increased from
baseline to 15 minutes post stress onset during the stress session, and very slightly decreased
during the control session. The analysis also revealed an overall response x stress x time
interaction (F=18.301, p<.001; figure 3.18), suggesting the responders and nonresponders
exhibited a different pattern of cortisol responses between time points during stress and control
sessions.
We then looked at responders and nonresponders during the stress session only within
each phase. For the EF phase, a 2 (response: responders v nonresponders) x 2 (time: baseline v
15 minutes post stress onset) between-by-within subject ANOVA revealed a main effect of
response (F=21.936, p<.001), with responders exhibiting higher cortisol levels than
nonresponders, but only a marginal effect of time (F=3.716, p=.065), with a trend toward
increasing from baseline to 15 minutes post stress onset. The response x time interaction,
however, was significant (F=21.936, p<.001; see figure 3.19), with responders exhibiting an
increase between the two time points and nonresponders exhibiting a decrease. The same
analysis for the LF phase revealed a main effect of response (F=17.774, p<.001), with responders
exhibiting higher cortisol levels than nonresponders, and a main effect of time (F=4.769,
p=.039), with cortisol increasing from baseline to 15 minutes post stress onset. The response x
time interaction also was significant (F=17.774, p<.001; see figure 3.20), with responders
exhibiting an increase between the two time points and nonresponders exhibiting a decrease.
These same analyses conducted for the control sessions found no main effects of response, time,
or any response x time interactions in either the EF or LF phases.
86
We next looked at responders during the stress and control sessions within each phase.
For responders in the EF phase, a 2 (stress: cold v warm) x 2 (time: baseline v 15 minutes post
stress onset) within subject ANOVA revealed no main effect of stress (F=.133, p=.719), with
cortisol levels being only slightly higher during the stress session than during the control session,
but there was a main effect of time (F=15.819, p=.001), with cortisol increasing from baseline to
15 minutes post stress onset. The stress x time interaction also was significant (F=6.552, p=.02),
with cortisol significantly increasing between the two time points during the stress session
(F=29.502, p<.001) and slightly but nonsignificantly decreasing during the control session
(F=.03, p=.865; see figure 3.21). The analysis for the LF phase found a marginal effect of stress
(F=3.567, p=.075), with a trend toward higher cortisol levels during the stress session than the
control session, but there was a main effect of time (F=12.942, p=.002), with cortisol increasing
from baseline to 15 minutes post stress onset. The stress x time interaction also was significant
(F=33.122, p<.001), with cortisol significantly increasing between the two time points during the
stress session (F=27.905, p<.001) and numerically but nonsignificantly decreasing during the
control session (F=.352, p=.56; see figure 3.22).
3.3.10 Working Memory: Word Recall in Responders versus Nonresponders during the EF and
LF phases
Because working memory analyses collapsed across responders and nonresponders failed
to reveal any effects, we next split women into groups of responders (n=19 in both phases) and
nonresponders (n=8 in both phases) and conducted between-by-within analyses on working
memory performance during the EF and LF phases separately. We first ran a 2 (response:
responders v nonresponders) x 2 (stress: cold v warm) x 5 (load: 2 sentences through 6
sentences) between-by-within subject ANOVA for the EF phase. This analysis failed to find
87
main effects of response or stress, but did find a main effect of load (F=21.878, p<.001) with the
proportion of words correctly recalled decreasing as loads increased in difficulty. Further,
interactions of response x load and stress x load were found to be nonsignificant. However, the
response x stress interaction was marginal (F=3.923, p=.059; see figure 3.23), with responders
exhibiting worse performance during the stress session than control session and nonresponders
showing better performance during the stress session than control session. Similarly, the overall
response x stress x load interaction was marginal (F=2.142, p=.081), suggesting the different
patterns of performance between the stress and control sessions in responders differed from the
different patterns of performance between the stress and control sessions in nonresponders. This
same analysis for stress and control sessions during the LF phase, failed to find any significant or
marginal main effects or interactions, with the exception of load (F=22.489, p<.001).
We next ran the same contrast but only looking at responders in the EF and LF phase.
Here, a 2 (stress: cold v warm) x 5 (load: 2 sentences through 6 sentences) within subject
ANOVA completed on responders during the EF phase revealed a marginal main effect of stress
(F=7.19, p=.07; see figure 3.24), with worse performance during the stress session than the
control session, and an effect of load (F=19.117, p<.001). The stress x load interaction, however,
was nonsignificant. A similar contrast looking at responders in the LF phase revealed a different
pattern of effects. While there was still a main effect of load (F=22.235, p<.001), there was a
robust nonsignificant effect of stress (F=.002, p=.962; see figure 3.25) and no stress x load
interaction, suggesting that despite showing a larger, albeit nonsignificant, cortisol response to
the stressor during the LF phase, women failed to experience stress-induced interference of
working memory performance. These same analyses examining patterns in nonresponders
revealed only effects of load (EF: F=7.056, p<.001; LF: F=7.05, p<.001). Failure to find effects
88
of stress or stress x load interactions are not surprising given these women did not exhibit a
cortisol response to the stressor.
3.3.11 Emotional Memory: Free Recall of Emotional and Neutral Pictures in Responders versus
Nonresponders during the EF and LF phases
Although we did observe effects in the overall analyses collapsed across responders and
nonresponders, we conducted within-by-between analyses examining picture recall in responders
versus nonresponders to remain consistent with the cortisol and working memory analyses.
Mixed design ANOVAs were conducted on phase, stress, and valence between responders and
nonresponders during the EF and LF phases separately. Valence comparisons were conducted for
negative images versus neutral images, positive images versus neutral images, negative images
versus positive images, and emotional (negative and positive) images versus neutral images.
A 2 (response: responders v nonresponders) x 2 (stress: cold v warm) x 2 (valence:
emotional pictures v neutral pictures) between-by-within subject ANOVA on picture recall
during the EF phase only revealed a main effect of valence (F=11.031, p=.003), with better recall
of emotional pictures than of neutral pictures. We next conducted follow-up analyses for each
valence of picture presented: negative, positive, and neutral. A 2 (response: responders v
nonresponders) x 2 (stress: cold v warm) x 2 (valence: negative pictures v neutral pictures)
within-by-between subject ANOVA on picture recall also failed to find any main effects or
interactions, with the exception of valence (F=13.017, p=.001), with better recall of negative
pictures than neutral pictures. Analyses looking at positive versus neutral pictures, and negative
versus positive pictures, in responders versus nonresponders found no effects of phase, stress,
valence, or any interactions.
89
A 2 (response: responders v nonresponders) x 2 (stress: cold v warm) x 2 (valence:
emotional pictures v neutral pictures) within-by-between subject ANOVA on picture recall
during the LF phase only revealed only a main effect of valence (F=7.442, p=.011), with better
recall of emotional pictures than neutral pictures, and no other effects or interactions. We next
conducted follow-up analyses for each valence of picture presented: negative, positive, and
neutral. A 2 (response: responders v nonresponders) x 2 (stress: cold v warm) x 2 (valence:
negative pictures v neutral pictures) between-by-within subject ANOVA on picture recall also
failed to find any main effects or interactions, with the exception of valence (F=18.215, p<.001),
with better recall of negative pictures than neutral pictures. Analyses looking at positive versus
neutral pictures in responders versus nonresponders found no effects of phase, stress, valence, or
any interactions. However, comparisons between negative and positive pictures found a main
effect of valence (F=21.359, p<.001), with better recall of negative pictures than positive
pictures, but no other main effects or interactions.
3.3.12 Emotional Memory: Memory for Picture Location in Responders versus Nonresponders
during the EF and LF phases
Similar to the cortisol and working memory analyses, we then conducted within-by-
between analyses examining memory for location of picture presentation in responders versus
nonresponders. Mixed ANOVAs were conducted on phase, stress, and valence between
responders and nonresponders in the EF and LF phases separately. Valence comparisons were
conducted for negative images versus neutral images, positive images versus neutral images,
negative images versus positive images, and emotional (negative and positive) versus neutral
images.
90
A 2 (response: responders v nonresponders) x 2 (stress: cold v warm) x 2 (valence:
emotional pictures v neutral pictures) between-by-within subject ANOVA on memory for picture
location during the EF phase revealed no main effects or interactions. We next conducted follow-
up analyses for each valence of picture presented: negative, positive, and neutral. A 2 (response:
responders v nonresponders) x 2 (stress: cold v warm) x 2 (valence: negative pictures v neutral
pictures) between-by-within subject ANOVA on memory for picture location also failed to find
any main effects or interactions. Analyses looking at positive versus neutral pictures and
negative versus positive pictures also failed to reveal any main effects or interactions.
A 2 (response: responders v nonresponders) x 2 (stress: cold v warm) x 2 (valence:
emotional pictures v neutral pictures) between-by-within subject ANOVA on memory for picture
location during the LF phase revealed no main effects or interactions. We next conducted follow-
up analyses for each valence of picture presented: negative, positive, and neutral. A 2 (response:
responders v nonresponders) x 2 (stress: cold v warm) x 2 (valence: negative pictures v neutral
pictures) between-by-within subject ANOVA on memory for picture location also failed to find
any main effects or interactions. However, analyses looking at positive versus neutral pictures in
responders versus nonresponders found a response x stress interaction (F=5.909, p=.023; see
figure 3.26), with responders showing better location memory during the control session than the
stress session and nonresponders showing better location memory during the stress session than
the control session. No other main effects or any interactions were found. Comparisons between
negative pictures and positive pictures revealed no main effects or interactions for memory of
picture locations.

91
3.4 Discussion
Study 2 aimed to determine if estradiol fluctuations during the menstrual cycle mimic
effects observed in post-menopausal women in study 1. Of interest was whether changes in
estradiol levels affected the stress response and/or working memory and emotional memory
performance during stress in the same manner observed in postmenopausal women taking
estradiol or placebo in study 1. We hypothesized that the fluctuations of estradiol during the first
half of the menstrual cycle would result in the same pattern of effects observed in study 1. Our
first hypothesis stated that women would display a blunted cortisol response to an ice water
stressor during the high estradiol LF phase compared to the response exhibited during the low
estradiol EF phase. The results of this study do not strongly support this hypothesis. The notion
that higher estradiol levels would minimize cortisol release created the foundations for the
second and third hypotheses, such that reductions in cortisol release should mitigate the effects
of stress on other domains because less GC would be available to act on and affect other
processes. Following this reasoning, our second hypothesis stated that women would show less
stress-induced interference of working memory during the high estradiol LF phase than the low
estradiol EF phase. Given that the cortisol response results did not support an attenuated cortisol
response to an ice water stressor it was unclear whether or not hypothesis 2 would be supported.
However, our results do support this hypothesis, with responders during the LF phase failing to
show a decrease in performance during the stress session and EF responders exhibiting worse
performance during the stress session. The third hypothesis focused on how estradiol treatment
might enhance the effects of stress on emotional memory. We predicted women would show less
stress-induced enhancement of emotional memory during the high estradiol LF phase than during
the low estradiol EF phase, however while the emotional memory results did reveal an effect of
92
phase, there were no phase-by- stress interactions, even when looking at responders versus
nonresponders – thus not supporting our third hypothesis.
3.4.1 Differences in Subjective Ratings of Pain
Pre and post stress onset stress ratings did not differ between the 2 phases. However,
there was a main effect of phase when looking at pre and post stress onset pain ratings. The
difference in the overall pain ratings was in the direction of greater pain during the EF phase than
the LF phase. Given the within subject nature of our experimental design, this difference seems
to be a result of the phase, rather than a confounding factor like cohort effects. The LF phase has
been associated with higher pain thresholds relative to the EF phase in previous studies
(Hellström & Anderberg, 2003; Riley III, E. Robinson, Wise, & Price, 1999), this may be related
to the higher ratings of PMS symptoms during the EF phase in this study. The general level of
discomfort experienced during menstruation may make women more sensitive to other forms
somatic pain, as would be experienced by holding one’s hand in ice water for an extended period
of time.
3.4.2 Cortisol Response to a Stressful Event during the EF and LF phases
Our stress manipulation failed to result in a main effect of stress, suggesting no difference
in cortisol levels between the stress and control sessions, although we did observe a stress-by-
time interaction with increases during the stress session and decreases during the control session.
Yet, follow up analyses on just the EF and LF phases alone revealed a main effect of stress for
the LF phase, with higher cortisol levels during the stress session than the control session, but
only an effect of time and a stress-by-time interaction during the EF phase. Failure to find
consistent stress effects with the cold pressor is not unheard of, nor the first time to occur in our
lab (Clewett, Schoeke, & Mather, 2013). However, given that the majority of studies conducted
93
in our lab to employ the cold pressor task result in reliable increases in cortisol levels (Lighthall
et al., 2009; Lighthall et al., 2011; Mather et al., 2010), we decided to examine differences in
responders and nonresponders.
These analyses did reveal different patterns between phases, but not in line with our
hypotheses. We hypothesized that cortisol responses would be larger during the EF phase than
the LF phase, however, when analyses focused on responders only there was a robust
nonsignificant effect of stress on overall cortisol levels during the EF phase, but a marginal effect
of stress during the LF phase. This pattern suggests that cortisol levels increased more during the
LF phase than during the EF phase. While our original overall analyses failed to find a
significant difference between cortisol levels at baseline and at 15m post stress onset, a paired
samples t-test comparing the 13 women classified as responders during both phases did show a
significant difference between the change score from baseline to 15m post stress onset (t=-2.889,
df=12, p=.014; means: change for EF=.0635 µg/dl, change for LF=.1492 µg/dl).
Finding that cortisol levels increased more during the LF phase is surprising given the
evidence that women show less HPA activation in response to a stressor than men, presumably
partially due to female sex hormones (Davis & Emory, 1995; Kirschbaum et al., 1999;
Kirschbaum et al., 1992; Kudielka, Buske-Kirschbaum, et al., 2004; Kudielka et al., 1998). This
pattern, however, is in direct opposition to the animal literature which shows that females show
greater HPA activation to a stressor than males, in both rodents (Handa, Burgess, Kerr, &
O'Keefe, 1994; Viau & Meaney, 1991) and non-human primates (Roy, Reid, & Van Vugt, 1999).
This effect in the animal literature may be a result of estradiol’s ability to upregulate CRH
expression (Lalmansingh & Uht, 2008), which should cause a greater release ACTH and
glucocorticoids.
94
This mechanism may have contributed to the higher cortisol release observed during the
LF phase. Unfortunately, we cannot investigate this until the saliva samples collected for this
study have been analyzed for estradiol levels. In this case, estradiol and cortisol should be
positively correlated, as the higher estradiol levels would increase CRH expression leading to
more CRH available for release when faced with a stressor, followed by increased downstream
HPA activity and greater cortisol release.
Another possibility is that our LF women were already experiencing an increase in
progesterone, making their phase profile look more similar to the luteal phase. In this instance,
we would expect a greater cortisol response, as women often exhibit greater cortisol responses to
a stressor during the luteal phase (Andreano et al., 2008; Kirschbaum et al., 1999). Again,
however, we cannot investigate this further until the saliva samples are analyzed for
progesterone.
3.4.3 Relationship between Stress and Working Memory during the EF and LF phases
When analyzed irrespective of whether or not women responded to the ice water stressor,
we found no effects of stress or phase on working memory performance. However, when running
analyses on responders during the EF and LF phases we found different patterns of effects. In
these analyses it was found that EF responders performed worse during the stress session than
during the control session, but that LF responders performed comparably across sessions. This
finding supported our hypothesis that higher estradiol levels would be associated with less stress-
induced interference of working memory. This finding is interesting, and a bit unexpected, since
women experienced greater cortisol responses to the ice water stressor during the LF phase.
Finding that women experienced greater cortisol responses during the LF phase suggests that
they should also have experienced greater interference with working memory performance.
95
That women experienced less stress induced interference during the LF phase when they
were experiencing greater cortisol responses suggests that something about the LF phase is
protective against these higher cortisol levels. One possibility concerns another effect of estradiol
– increasing GC receptor expression in the brain. Thus, it is possible that although higher
estradiol levels during the LF phase may increase CRH expression, the higher estradiol levels
also should increase glucocorticoid receptor expression in the brain (Ferrini & De Nicola, 1991;
Ferrini et al., 1995). The increase in GC receptors should lead to more efficient GC detection and
swifter shut down of the stress response despite releasing more cortisol initially. Thus, while the
brain regions involved in working memory may be initially exposed to more cortisol, the
potential for more rapid reduction of cortisol release due to more efficient detection of higher
levels may limit the amount of time cortisol is available to interfere with working memory
processes.
Another interesting result our analyses uncovered was a response-by-stress interaction for
working memory performance during the EF phase, but not the LF phase. The EF phase
interaction indicated that women classified as nonresponders performed better under stress than
during the control session, versus responders who performed better during the control session
than during stress. One possible explanation refers back to the inverted-U shaped effect of stress
on performance first reported by Yerkes and Dodson (1908). Perhaps during the EF phase, the
cortisol response displayed by nonresponders was enough arousal to place participants in the
range of optimal performance, versus responders, whose cortisol response was too large, causing
too much arousal and thereby interfering with performance.
Although there was no response-by-stress interaction during the LF phase, there was a
trend towards better performance under stress for nonresponders, and no change in performance
96
in responders. Interestingly, the effect of response was also marginal, with nonresponders
trending toward better performance than responders in both the stress and control sessions. In
this instance, the Yerkes-Dodson principle may still apply. In this scenario, the amount of
arousal experienced by the nonresponders during the LF phase may have been near the optimal
level to help facilitate performance, similar to that suggested for the EF nonresponders, despite
the failure to increase cortisol levels. On the other hand, the increase in cortisol levels in the
responders was enough to prevent enhancement of performance, but not enough to hinder
performance as observed during the EF phase. This may be another way in which the LF phase
protects working memory under stress; that the change in the hormonal profile leads to greater
tolerance of stress despite showing larger physiological responses, thereby protecting women
from the maladaptive effects of stress exposure on this form of cognition.
3.4.4 Relationship between Stress and Emotional Memory during the EF and LF phases
Analyses examining memory for where pictures were presented during encoding did not
reveal many effects, with exception of a main effect of phase when comparing memory for
negative versus neutral pictures, and a stress-by-valence interaction for emotional versus neutral
pictures. The failure to find more robust effects of stress on this test of emotional memory may
be due to the relative ease of the task. While on average the encoding phase of this task occurred
21 minutes prior to the location memory test, this interval may be too short to pick up effects of
stress prior to encoding in our sample of women. Recall that each woman only viewed 24
pictures during each session, and that the location memory test consisted of showing only the
pictures they viewed during that session both in the correct location and an incorrect location
simultaneously. It may be the case that the added cue of viewing the pictures again, with one in
97
the correct location, provided enough of a cue to aid in recalling location regardless of whether
stress was applied prior to encoding or not.
With regard to the main effect of phase, women exhibited better memory for picture
presentation location during the LF phase than the EF phase. This effect may reflect overall
better memory during the LF phase relative to the EF phase, although other studies report mixed
results regarding whether various forms of memory are worse during the EF phase (Phillips &
Sherwin, 1992), better during the EF phase (Maki, Rich, & Shayna Rosenbaum, 2002), or
whether no difference exists between menstrual cycle phases (Kuhlmann & Wolf, 2005). The
stress-by-valence interaction suggested memory for the location of emotional pictures was better
than memory for the location of neutral pictures during the control session, and vice versa during
the stress session. This is contradictory to typical reports of stress enhancing memory for
emotional stimuli (Buchanan & Lovallo, 2001; Payne et al., 2007; Smeets et al., 2006). Because
of the failure to find more robust effects of stress, and the contradictory stress-by-valence
interaction, we conducted follow-up analyses looking at responders and nonresponders to see
whether including nonresponders in the overall analysis muddied the results.
Unfortunately, analyses comparing nonresponders and responders during the EF phase
failed to find any effects at all. During the LF phase, the only effect found was a response-by-
stress interaction in the positive versus neutral picture comparison, with responders showing
better location memory during the control session than the stress session and nonresponders
showing better location memory during the stress session than the control session. The sparse
significant results in the location memory analyses, despite looking at responders and
nonresponders alone, again may be due to the participants finding this particular memory test
relatively easy and thus performing well on the test regardless of whether or not they
98
experienced stress. Further, it may be the case that a stronger stressor would have affected
emotional memory. Recall that as tasks become easier greater stress is required to enhance
performance (Dodson, 1915), thus it is possible that the ice water stressor did not induce a
sufficient amount of stress to affect the relatively easy picture-location test.
Analysis of the free recall test did reveal more effects, perhaps because the greater
difficulty of this test was sensitive to the manipulations employed. First, comparing memory for
emotional versus neutral pictures did uncover effects of phase and valence, and even a marginal
effect of stress, but no interactions. Women recalled more total pictures during the LF phase than
during the EF phase and emotionally valenced pictures were better recalled than neutral pictures
regardless of phase. Because the emotional versus neutral pictures comparison includes all 24
pictures viewed during encoding, the failure to find a phase-by-valence interaction may indicate
that women experience better overall memory during the LF phase than they do during the EF
phase, irrespective of content of the pictures. The effect of valence followed the expected pattern
of better recall of emotional pictures than neutral pictures, yet the marginal effect of stress
trended toward better recall during the stress session than the control session. This pattern is not
necessarily what would be expected. Although stress should enhance memory for emotional
items, it should not necessarily cause an overall increase in recall of all items. This trend may
again be a related to arousal induced enhancement of overall performance as discussed for the
stress and working memory interactions. Perhaps the amount of stress experienced was optimal
for performance rather than detrimental, leading to slightly better recall rates after stress
exposure.
Patterns differed for the more specific valence comparisons. When looking at recall of
negative and neutral pictures we only observed effects of phase and valence, with greater recall
99
during the LF phase than during the EF phase, which may have contributed to the effect
observed in the overall analysis. This differed from the analyses comparing recall of positive and
neutral pictures. This analysis revealed no main effects, but did uncover a phase-by-valence
interaction with women showing better recall of positive pictures than neutral pictures during the
EF phase, but showing better recall of neutral pictures than positive pictures during the LF phase.
Interestingly, although not significant in every comparison, women numerically recalled more
pictures during the LF phase then the EF phase. However, the valence most recalled differed
across analyses, with more negative than neutral, more neutral than positive, and more negative
than positive pictures recalled during the LF phase. Thus, while the LF phase may be related to
better memory, the valence may in the end matter, despite the failure to find significant effects in
all comparisons or a phase-by-valence interaction in the overall emotional versus neutral picture
comparison. These same analyses comparing responders versus nonresponders did not change
the pattern of results.
One possible explanation for the better recall of negative pictures during the LF phase
may be related to reports of better recognition of negative facial expressions outside of the high
progesterone luteal phase (Derntl, Kryspin-Exner, Fernbach, Moser, & Habel, 2008). The greater
recognition of the emotionally valenced items may have made them more salient and therefore
more likely to be recalled during the LF phase. However, this assumes that estradiol alone can
modulate the ability to recognize and remember emotionally valenced stimuli, which this study
cannot attest to. Another possibility is that our LF women were already experiencing increases in
progesterone making them exhibit memory performance more commonly reported within the
luteal phase – which, counterintuitive to the above-reported increased recognition of negative
facial expressions outside of the luteal phase – is characterized as having better memory for
100
emotional stimuli, particularly negative stimuli (Ertman, Andreano, & Cahill, 2011; Sakaki &
Mather, 2012). Again, we are unable to examine this possibility until we have processed the
saliva samples for sex hormones levels.
3.4.5 Conclusion
This study suggests that the late follicular phase of the menstrual cycle may exert some
form of protective effects against the effects of stress on working memory and may be related to
better delayed recall over relatively short intervals regardless of stress exposure. Although, firm
conclusions on patterns of emotional memory throughout the menstrual cycle phase currently
cannot be made because sex hormone levels are still unknown.
It is unclear how the LF phase may be exerting its effect against stress given women
displayed higher cortisol responses to the ice water stressor during this phase. Likewise, the
dichotomy of phase and stress interacting for the working memory task, but not the emotional
memory task, despite higher cortisol levels, suggests that the way phase protects against stress
effects on working memory is different from the way phase and stress each affect emotional
memory. This is not entirely surprising since different brain areas govern these two forms of
cognition. Further, since estradiol is not reducing the cortisol levels in the phase that experiences
some buffering from stress, this rules out one potential mechanism of protection – that higher
estradiol levels reduce HPA response to a stressor, but provides possible alternative mechanisms.
First, the estradiol levels of our women during their LF phase may alter the speed with which
HPA activity is shut down after initiation of hormone release due to GC receptor upregulation.
Second, given that 1) different brain regions govern working memory and emotional memory,
and 2) it does not appear that higher estradiol levels are able to attenuate initial cortisol release,
the actual hormone profile of our LF women (versus the anticipated hormone profile based on
101
time since the first day of menses) may alter the way cortisol acts on those specific brain regions.
Third, rather than changing the action of cortisol directly in those regions, it may be that higher
estradiol levels lead to higher binding of estradiol in those same regions and that this higher rate
of estradiol action in the brain somehow counteracts cortisol effects. These last alternative
explanations could also explain why some forms of cognition experience protection from stress
while others seem to be unaffected.

102
CHAPTER 4
Estradiol differentially alters the effects of stress in post-menopausal and young
spontaneously cycling women.
4.1 Introduction
This work set out to investigate the potential protective effects of estradiol against the
stress response and effects of stress on working memory and emotional memory in post-
menopausal and young spontaneously cycling women. We tested 3 general hypotheses in each
study. First, we proposed that higher estradiol levels, either as a result of estradiol
supplementation or menstrual cycle phase, would be associated with reduced cortisol response to
a stressful event. Second, that higher estradiol levels would be associated with less stress-
induced interference of working memory, and third, that higher estradiol levels would be
associated with less stress-induced enhancement of emotional memory. In each study we
uncovered support for some but not all of our a priori hypotheses.
Our first study, examining the effects of estradiol treatment in post-menopausal women,
did reveal that the highest estradiol levels observed in our subsample of ELITE patients was
related to significantly blunted cortisol responses to the ice water stressor – supporting our first
hypothesis in post-menopausal women. We also found that the women with the highest estradiol
levels, who displayed the blunted cortisol response, also were protected from the negative effects
of stress on working memory. So much so, that these women failed to show any difference in
their performance between stress and control sessions, in contrast with the low estradiol women
who displayed significant increases in cortisol, and significantly worse working memory during
the stress session – supporting our second hypothesis in post-menopausal women. However,
103
estradiol treatment did not appear to influence emotional memory in this study, nor did stress –
thus not supporting our third hypothesis in post-menopausal women.
Our second study, examining the effects of low and high estradiol during different
menstrual cycle phases, failed to support all three hypotheses. In this study, although not
significant, cortisol responses to the ice water stressor were greater during the high estradiol LF
phase than during the low estradiol EF phase – thus not concretely negating our first hypothesis,
but at minimum suggesting the higher estradiol phase might be associated with greater cortisol
release in response to a stressor. Surprisingly, despite trending toward greater cortisol release
during the LF phase, women experienced protection from stress-induced interference in working
memory performance during the high estradiol LF phase – supporting our second hypothesis in
young spontaneously cycling women. Lastly, while menstrual cycle phase did seem to influence
emotional memory, stress did not do so reliably, and there were no phase-by-stress interactions –
thus these findings did not support our third hypothesis.
We had proposed that estradiol would have the same effect in both post-menopausal and
younger women. And, in particular, that the variable effects obtained across menstrual cycle
studies was a result of looking at phases with such different hormone profiles, rather than
isolating specific hormones, which is admittedly difficult. However, after the exciting results of
the post-menopause study, we set out to isolate the effects of estradiol as best we could in a
sample of young naturally cycling women, which is why we focused on the low-estradiol/low-
progesterone, early follicular phase and the high-estradiol/low-progesterone, late follicular phase.
Thus, we were surprised that we did not obtain similar effects across the board. Yet, the systems
governing these effects in pre- and post-menopausal women likely differ given the physiological
differences between the two groups.
104
4.2 Differences in cortisol response to an ice water stressor in young spontaneously cycling
and post-menopausal women
The pattern of cortisol responses between low and high estradiol groups differed in the
two studies. In study 1, we observed a estradiol-by-stress interaction where post-menopausal
women with the top quarter of estradiol levels (HE) displayed nonsignificant increases in cortisol
in response to the ice water stressor, while the post-menopausal women with the bottom quarter
of estradiol levels (LE) displayed significant increases in cortisol levels in response to the ice
water stressor. In study 2, we did not detect a phase-by-stress interaction. However, the
numerical patterns trended toward greater increases in cortisol levels in response to the ice water
stressor during the high estradiol phase of the menstrual cycle (LF), while exhibiting smaller
increases in cortisol in response to the ice water stressor during the low estradiol phase of the
menstrual cycle (EF).
The mechanisms by which estradiol is working may differ in the two populations of
women, perhaps as a result of relative HPG axis function. For the postmenopausal women in
study 1, the discontinuation of estradiol and progesterone release from the ovaries results in
dysregulation of the entire HPG system because of the removal of negative feedback. The
absence of ovarian hormones leads to loss of inhibition of GnRH release from the hypothalamus,
and increased synthesis and release of LH and FSH (Atwood et al., 2005). Reintroduction of
estradiol should reactivate some of the negative feedback lost through menopause, but may not
completely reestablish optimal HPG regulation. Although the axis may be partially reestablished,
estradiol treatment may not mimic the cyclicity of activation and inhibition observed in young
naturally cycling women. This difference in HPG axis activity might change the way the HPG
105
axis influences HPA activity and thereby change the effects of estradiol levels on the stress
response.
One way communication between the axes may differ could be the effects of estradiol on
CRH upregulation. The rapid and large increase observed during the LF phase of the menstrual
cycle may lead to upregulation of CRH. The greater expression of this releasing hormone would
be expected to cause greater release of ACTH and cortisol, causing greater stress reactivity
during this phase. In post-menopausal women, estradiol treatment may not elevate E2 levels
enough to induce the upregulation of CRH, thus preventing this heightened stress response. The
failure to induce increased CRH expression, coupled with the possible increase of CBG limiting
the amount of free cortisol available to act on tissue, and the possible increase of GC receptors
leading to more efficient and swift shutdown of the HPA axis, may explain why post-
menopausal women taking estradiol experience smaller cortisol responses to a stressor than
young women in the high estradiol phase of the menstrual cycle.
An additional contributor to these possible differences in HPA-HPG communication is
the role of progesterone. Although women in the clinical trial were also given progesterone,
nurses anecdotally reported that many women did not comply with the treatment that would be
associated with the progesterone (i.e. use of the vaginal cream). On the other hand, the women in
study 2 may have already been experiencing rises in progesterone. If this were the case, then the
women in study 1 would only be experiencing alterations in estradiol, allowing us to observe
only changes in estradiol, whereas the younger women would be experiencing alterations in both
estradiol and progesterone, accounting for the numerically larger increase in cortisol during the
LF phase and potentially for the different patterns observed across the two studies.
106
4.3 How estradiol is preventing stress from interfering with working memory in young
spontaneously cycling and post-menopausal women
Interestingly, despite displaying different cortisol response profiles, both the HE women
from study 1 and women during the LF phase of study 2 experienced protection from stress-
induced interference of working memory. That both groups of women experienced protection,
despite exhibiting different cortisol response profiles, suggests different mechanisms are
responsible for the protection observed in the pre- and post-menopausal women. In the post-
menopausal women from study 1, protection from stress-induced interference in working
memory could be due to estradiol reducing HPA reactivity to a stressor, thereby reducing cortisol
release. Reduced cortisol release means less cortisol would be available to disrupt the brain
regions involved in working memory, thus leading to resilience against stress. A possible
alternative to this is that estradiol treatment upregulated CBG. In this case, the same amount of
cortisol could be released by LE and HE women, but the higher CBG levels in HE women would
result in less biologically active cortisol available to act on tissue. Recall that saliva samples only
measure free, biologically active, levels of cortisol, so our data cannot speak to which of these
mechanisms might contribute to the decreased cortisol levels or protection against stress-induced
interference in working memory.
For the young naturally cycling women in study 2, protection from stress-induced
interference in working memory is likely due to a different mechanism. One possibility is the
potential estradiol-induced upregulation of GC receptors. The greater levels of GC receptors in
the brain would lead to more sensitive detection of increases in cortisol levels, leading to more
efficient and swift reduction of CRH release and HPA reactivity, despite exhibiting larger
cortisol increases initially. Alternatively, the larger cortisol response to the ice water stressor
107
coupled with the protection of working memory processes could involve an inverted-U shaped
function. Resilience against stress in this phase may result in a shift of how much stress one can
experience while still remaining in the optimal range for performance, before reaching the
tipping point that sends one into a detrimental amount of arousal that impairs performance. This
increased resilience could also be a result of the increased GC receptors. In the instance of the
inverted U curve of arousal and performance, basal levels of cortisol bind to the type 1,
mineralocorticoid receptor, and as cortisol levels increase the hormone binds to the type 2,
glucocorticoid receptor. Binding to the type 2 receptor may contribute to stress related
dysfunctions since these are the receptors filled during stress and have been associated with
decreased performance (Seckl & Olsson, 1995). However, since estradiol has been shown to
upregulate both type 1 and type 2 receptors (Ferrini & De Nicola, 1991), women should have
more type 1 receptors available to fill during the LF phase before reaching levels requiring
activation of the type 2 receptor. This type of mechanism would shift the range of GC receptor
occupation related to optimal performance versus detrimental performance.
4.4 Differences in the effects of stress on emotional memory in young spontaneously cycling
and post-menopausal women
Patterns of emotional memory also differed between the two studies. Post-menopausal
women from study 1 showed no effects of stress or estradiol treatment on emotional memory.
Meanwhile, although there were no strong effects of stress on emotional memory in the young
women from study 2, effects of phase and phase-by-valence interactions were detected. We
found that women correctly recalled a greater number of pictures during the LF phase than
during the EF phase and that women recalled more positive pictures than neutral pictures during
108
the EF phase, but more neutral pictures during the LF phase, and more negative pictures than
positive pictures during the LF phase.
For the HE women in study 1, the failure to find effects of stress make sense given their
minimal cortisol response to the ice water stressor. However, the significant increase in cortisol
observed in the LE women suggests strong effects of stress should have been observed in the
emotional memory task. However, while the data did not reveal effects of stress, it is possible
that we are seeing effects of the inverted-U relationship between stress and performance. The
large stress response in the LE women might have induced too much arousal, leading to robust
type 2 GC receptor occupation and activation, leading to detrimental performance. This high-
arousal induced reduction in performance would then look similar to the lower performance
observed during the control session resulting from not enough arousal, making performance in
both sessions look similar.
The failure to observe strong effects of stress in the young women from study 2 may also
be attributed to the inverted U function, just as with the LE women from study 1. However, the
manner in which the mechanism influences performance would be different. The trend was
opposite to what was expected, with recall slightly better during the stress session. In this
instance, the amount of stress experienced would have been just enough to enhance overall
performance. As for the effects of phase and valence, the better memory performance during the
LF phase seemed to be at least partially attributable to greater memory of negative stimuli. This
trend toward better memory of negative stimuli may be a result of better recognition of negative
stimuli outside of the high progesterone luteal phase (Derntl et al., 2008); the better recognition
of the negative stimuli may have made it more salient and thus easier to encode and remember.
However, while recognition of negative stimuli is associated with lower progesterone phases of
109
the menstrual cycle, it is the high progesterone phases that are associated with better memory for
emotional stimuli, particularly negative stimuli (Ertman et al., 2011; Sakaki & Mather, 2012).
Thus, like the cortisol results, it may be that our LF women were already experiencing increases
in progesterone leading to enhanced memory of negative stimuli. Again, unfortunately we cannot
speak to this possibility until saliva samples have been analyzed for sex hormones.
4.5 Closing remarks
The results of these studies do suggest that hormone profiles influence the stress
response, and can change the way stress affects working memory processing, although the
mechanisms by which this is occurring appears to differ between pre- and post-menopausal
women. Further work is required in both age groups to better determine how estradiol is exerting
its effect in women before and after menopause. Nonetheless, the studies do confirm that sex
hormones play an important role in our physiological responses to stressful stimuli and events, as
well as to cognition. Further, this work speaks to the importance of continuing this line to
research to better understand how these systems interact to sometime protect cognition. Although
this work is limited to only adding to the understanding of how changes during the menstrual
cycle affect our responses to stress and cognitive processing, the work with post-menopausal
women can further inform the medical community on potential effects of estradiol after
menopause.

110
References

Alexander, J. K., Hillier, A., Smith, R. M., Tivarus, M. E., & Beversdorf, D. Q. (2007). Beta-
adrenergic modulation of cognitive flexibility during stress. Journal of Cognitive
Neuroscience, 19(3), 468-478.

Almeida, D. M., Wethington, E., & Kessler, R. C. (2002). The Daily Inventory of Stressful
Events: An Interview-Based Approach for Measuring Daily Stressors. Assessment, 9(1),
41-55. doi: 10.1177/1073191102091006

Almela, M., Hidalgo, V., Villada, C., van der Meij, L., Espín, L., Gómez-Amor, J., & Salvador,
A. (2011). Salivary alpha-amylase response to acute psychosocial stress: The impact of
age. Biological Psychology, 87(3), 421-429. doi: 10.1016/j.biopsycho.2011.05.008

Alonso, M., Vianna, M. R. M., Depino, A. M., Mello e Souza, T., Pereira, P., Szapiro, G., . . .
Medina, J. H. (2002). BDNF-triggered events in the rat hippocampus are required for
both short- and long-term memory formation. Hippocampus, 12(4), 551-560. doi:
10.1002/hipo.10035

Altemus, M., Redwine, L., Leong, Y., Yoshikawa, T., Yehuda, R., Detera-Wadleigh, S., &
Murphy, D. L. (1997). Reduced sensitivity to glucocorticoid feedback and reduced
glucocorticoid receptor mRNA expression in th eluteal phase of the menstrual cycle.
Neuropsychopharmacology, 17, 100-109.

Altemus, M., Roca, C., Galliven, E., Romanos, C., & Deuster, P. (2001). Increased Vasopressin
and Adrenocorticotropin Responses to Stress in the Midluteal Phase of the Menstrual
Cycle. Journal of Clinical Endocrinology & Metabolism, 86(6), 2525-2530.

Anderson, K. J. (1994). Impulsitivity, caffeine, and task difficulty: A within-subjects test of the
Yerkes-Dodson law. Personality and Individual Differences, 16(6), 813-829. doi:
http://dx.doi.org/10.1016/0191-8869(94)90226-7

Andreano, J. M., Arjomandi, H., & Cahill, L. (2008). Menstrual cycle modulation of the
relationship between cortisol and long-term memory. Psychoneuroendocrinology, 33(6),
874-882. doi: 10.1016/j.psyneuen.2008.03.009

Arbel, I., Kadar, T., Silberman, M., & Levy, A. (1994). The effects of long-term corticosterone
administration on hippocampal morphology and cognitive performance of middle-aged
rats. Brain Research, 657(227-235).

Atwood, C. S., Meethal, S. V., Liu, T., Wilson, A. C., Gallego, M., Smith, M. A., & Bowen, R.
L. (2005). Dysregulation of the hypothalamic-pituitary-gonadal axis with menopause and
andropause promotes neurodegenerative senescence. Journal of Neuropathology &
Experimental Neurology, 64(2), 93-103.

111
Bachmann, G. A., & Leiblum, S. R. (2004). The impact of hormones on menopausal sexuality: a
literature review. Menopause, 11(1), 120-130.

Baker, L. D., Asthana, S., Cholerton, B. A., Wilkinson, C. W., Plymate, S. R., Green, P. S., . . .
Craft, S. (2012). Cognitive response to estradiol in postmenopausal women is modified
by high cortisol. Neurobiology of Aging, 33(4), 829.e829-829.e820. doi:
10.1016/j.neurobiolaging.2011.07.002

Barker, J. M., & Galea, L. A. M. (2009). Sex and regional differences in estradiol content in the
prefrontal cortex, amygdala and hippocampus of adult male and female rats. General and
Comparative Endocrinology, 164(1), 77-84. doi: 10.1016/j.ygcen.2009.05.008

Bartholomeusz, C., Wesnes, K., Kulkarni, J., Vitetta, L., Croft, R., & Nathan, P. (2008).
Estradiol treatment and its interaction with the cholinergic system: Effects on cognitive
function in healthy young women. Hormones and Behavior, 54(5), 684-693. doi:
10.1016/j.yhbeh.2008.07.007

Baum, A., Gatchel, R. J., & Schaeffer, M. A. (1983). Emotional, Behavioral, and Physiological
Effectsmof Chronic Stress at Three Mile Island. Journal of Consulting and Clinical
Psychology, 51(4), 565-572.

Behl, C. (1998). Effects of glucocorticoids on oxidative stress-iduced hippocampal cell death:
Implications for the pathogenesis of Alzheimer's disease. Experimental Gerontology,
33(7/8), 689-696.

Behl, C., Lezoualch, F., Trapp, T., Widmann, M., Skutella, T., & Holsboer, F. (1997).
Glucocorticoids enhance oxidative stress-induced cell death in hippocampal neurons in
vitro. Endocrinology, 138(1), 101-106.

Bimonte, H. A., & Denenberg, V., H. (1999). Estradiol facilitates performance as working
memory load increases. Psychoneuroendocrinology, 24, 161-173.

Bode, F. J., Stephan, M., Suhling, H., Pabst, R., Straub, R. H., Raber, K. A., . . . von Horsten, S.
(2008). Sex differences in a transgenic rat model of Huntington's disease: decreased 17 -
estradiol levels correlate with reduced numbers of DARPP32+ neurons in males. Human
Molecular Genetics, 17(17), 2595-2609. doi: 10.1093/hmg/ddn159

Brandt, J., Spencer, M., & Folstein, M. (1988). The telephone interview for cognitive status.
Neuropsychiatry, Neuropsychology, and Behavioral Neurology, 1(2), 111-117.

Breen, K. M., & Karsch, F. J. (2006). New insights regarding glucocorticoids, stress and
gonadotropin suppression. Frontiers in Neuroendocrinology, 27(2), 233-245. doi:
http://dx.doi.org/10.1016/j.yfrne.2006.03.335

Brien, T. (1981). Human corticosteroid binding globulin. Clinical Endocrinology, 14(2), 193-
212.
112

Brinton, R. D. (2001). Cellular and Molecular Mechanisms of Estrogen Regulation of Memory
Function and Neuroprotection Against Alzheimer's Disease: Recent Insights and
Remaining Challenges. Learning & Memory, 8(3), 121-133. doi: 10.1101/lm.39601

Brinton, R. D. (2005). Investigative Models for Determining Hormone Therapy-Induced
Outcomes in Brain: Evidence in Support of a Healthy Cell Bias of Estrogen Action.
Annals of the New York Academy of Sciences, 1052(1), 57-74. doi:
10.1196/annals.1347.005

Brinton, R. D. (2008). Estrogen regulation of glucose metabolism and mitochondrial function:
Therapeutic implications for prevention of Alzheimer's disease ☆. Advanced Drug
Delivery Reviews, 60(13-14), 1504-1511. doi: 10.1016/j.addr.2008.06.003

Brinton, R. D., Chen, S., Montoya, M., Hsieh, D., Minaya, J., Kim, J., & Chu, H.-P. (2000). The
women’s health initiative estrogen replacement therapy is neurotrophic and
neuroprotective. Neurobiology of Aging, 21, 475-496.

Brown, S. M., Henning, S., & Wellman, C. L. (2005). Mild, Short-term Stress Alters Dendritic
Morphology in Rat Medial Prefrontal Cortex. Cerebral Cortex, 15(11), 1714-1722. doi:
10.1093/cercor/bhi048

Buchanan, T. W., & Lovallo, W. R. (2001). Enhanced memory for emotional material following
stress-level cortisol treatment in humans. Psychoneuroendocrinology, 26, 307-317.

Bullinger, M., Naber, D., Pickar, D., Cohen, R. M., Kalin, N. H., Pert, A., & Bunney Jr., W. E.
(1984). Endocrine Effects of the Cold Pressor Test: Relationships to Subjective Pain
Appraisal and Coping. Psychiatry Research, 12, 227-233.

Burke, C., & Roulet, F. (1970). Increased exposure of tissues to cortisol in late pregnancy.
British medical journal, 1(5697), 657.

Cabeza, R., Dolcos, F., Graham, R., & Nyberg, L. (2002). Similarities and differences in the
neural correlates of episodic memory retrieval and working memory. NeuroImage, 16(2),
317-330. doi: 10.1006/nimg.2002.1063

Carlson, N. R. (2010). Anxiety Disorders, Autistic Disorder, Attention-Deficit/Hyperactivity
Disorder, and Stress Disorders Physiology of Behavior (10th ed.). Boston: Allyn & Bacon

Ceresini, G., Freddi, M., Marganti, S., Rebecchi, I., Modena, A. B., Rinaldi, M., . . . Valenti, G.
(2000). The effects of transdermal estradiol on the response to mental stress in
postmenopausal women: A randomized trial. American Journal of Medicine, 109, 463-
468.

113
Chen, S., Nilsen, J., & Brinton, R. D. (2006). Dose and temporal pattern of estrogen exposure
determines neuroprotective outcome in hippocampal neurons: therapeutic implications.
Endocrinology, 147(11), 5303-5313. doi: 10.1210/en.2006-0495

Choi, C.-I., Lee, Y.-D., Gwag, B. J., Cho, S. I., Kim, S.-S., & Suh-Kim, H. (2008). Effects of
estrogen on lifespan and motor functions in female hSOD1 G93A transgenic mice.
Journal of the Neurological Sciences, 268(1-2), 40-47. doi: 10.1016/j.jns.2007.10.024

Clewett, D., Schoeke, A., & Mather, M. (2013). Amygdala functional connectivity is reduced
after the cold pressor task. Cognitive, affective & behavioral neuroscience. doi:
10.3758/s13415-013-0162-x

Coburn-Litvak, P. S., Pothakos, K., Tata, D. A., McCloskey, D. P., & Anderson, B. J. (2003).
Chronic administration of corticosterone impairs spatial reference memory before spatial
working memory in rats. Neurobiology of Learning and Memory, 80(1), 11-23. doi:
10.1016/s1074-7427(03)00019-4

Cook, S. C., & Wellman, C. L. (2004). Chronic stress alters dendritic morphology in rat medial
prefrontal cortex. Journal of Neurobiology, 60(2), 236-248. doi: 10.1002/neu.20025

Copeland, D. E., & Radvansky, G. A. (2001). Phonological similarity in working memory.
Memory & Cognition, 29(5), 774-776.

Craig, M., Fletcher, P., Daly, E., Rymer, J., Brammer, M., Giampietro, V., & Murphy, D. (2008).
Physiological variation in estradiol and brain function: A functional magnetic resonance
imaging study of verbal memory across the follicular phase of the menstrual cycle.
Hormones and Behavior, 53(4), 503-508. doi: 10.1016/j.yhbeh.2007.11.005

Curtis, C. E., & D'Esposito, M. (2003). Persistent activity in the prefrontal cortex during working
memory. Trends in Cognitive Sciences, 7(9), 415-423. doi: 10.1016/s1364-
6613(03)00197-9

Daneman, M., & Carpenter, P. A. (1980). Individual differences in working memory and
reading. Journal of verbal learning and verbal behavior, 19, 450-466.

Daniel, J. M., Fader, A. J., Spencer, A. L., & Dohanich, G. P. (1997). Estrogen Enhances
Performance of Female Rats during Acquisition of a Radial Arm Maze. Hormones and
Behavior, 32, 217-225.

Davis, M., & Emory, E. (1995). Sex differences in neonatal stress reactivity. Child Development,
66(1), 14-27.

Del Rio, G., Velardo, A., Menozzi, R., Zizzo, G., Tavernari, V., Venneri, M. G., . . . Petraglia, F.
(1998). Acute estradiol and progesterone administration reduced cardiovascular and
catecholamine responses to mental stress in menopausal women. Neuroendocrinology,
67, 269-274.
114

Del Rio, G., Velardo, A., Zizzo, G., Avogaro, A., Cipolli, C., Della Casa, L., . . . Macdonald, I.
A. (1994). Effect of estradiol on the sympathoadrenal response to mental stress in normal
men. Journal of Clinical Endocrinology & Metabolism, 79(3), 836-840.

Delmas, P. D., Bjarnason, N. H., Mitlak, B. H., Ravoux, A., Shah, A. S., Huster, W. J., . . .
Christiansen, C. (1997). Effects of raloxifene on bone mineral desnity, serum cholesterol,
concentration, and uterine endometrium in portmenopausal women. The new england
journal of medicine, 337(23), 1641-1647.

Derntl, B., Kryspin-Exner, I., Fernbach, E., Moser, E., & Habel, U. (2008). Emotion recognition
accuracy in healthy young females is associated with cycle phase. Hormones and
Behavior, 53(1), 90-95. doi: http://dx.doi.org/10.1016/j.yhbeh.2007.09.006

Diamond, D. M., Bennett, M. C., Fleshner, M., & Rose, G. M. (1992). Inverted-U relationship
between the level of peripheral corticosterone and the magnitude of hippocampal primed
burst potentiation. Hippocampus, 2(4), 421-430. doi: 10.1002/hipo.450020409

Diamond, D. M., Fleshner, M., Ingersoll, N., & Rose, G. M. (1996). Psychological Stress
Impairs Spatial Working Memory: Relevance to Electrophysiological Studies of
Hippocampal Function. Behavioral Neuroscience, 110(4), 661-672.
Diamond, D. M., & Rose, G. M. (1994). Stress Impairs LTP and Hippocampal-Dependent
Memory. Annals New York Academy of Sciences, 30(746), 411-414.

Dodson, J. D. (1915). The relation of strength of stimulus to rapidity of habit-formation in the
kitten. Journal of Animal Behavior, 5(4), 330-336. doi:
http://dx.doi.org/10.1037/h0073415

Dodt, C., Dittman, J., Hruby, J., Spath-Schwalbe, E., Born, J., Schuttler, R., & Fehm, H. L.
(1991). Different regulation of adrenocorticotropin and cortisol in young, mentally
healthy elderly, and patients with senile dementia of Alzheimer's type. Journal of
Clinical Endocrinology & Metabolism, 72, 272-276.

Doi, N., Miyahara, S., & Hori, N. (1991). Glucocorticoids promote localized activity of rat
hippocampal CA1 pyramidal neurons in brain slices. Neuroscience Letters, 123, 99-101.

Donley, M. P., Schulkin, J., & Rosen, J. B. (2005). Glucocorticoid receptor antagonism in the
basolateral amygdala and ventral hippocampus interferes with long-term memory of
contextual fear. Behavioural Brain Research, 164(2), 197-205. doi:
10.1016/j.bbr.2005.06.020

Duff, S. J., & Hampson, E. (2000). A beneficial effect of estrogen on working memory in
postmenopausal women taking hormone replacement therapy. Hormones and Behavior,
38(4), 262-276. doi: 10.1006/hbeh.2000.1625

115
Duncko, R., Johnson, L., Merikangas, K., & Grillon, C. (2009). Working memory performance
after acute exposure to the cold pressor stress in healthy volunteers. Neurobiology of
Learning and Memory, 91(4), 377-381. doi: 10.1016/j.nlm.2009.01.006

Edelson, J. T., & Robertson, G. L. (1986). The effect of the cold pressor test on vasopressin
secretion in man. Psychoneuroendocrinology, 11(3), 307-316.

Eller, N. H., Netterstrøm, B., & Hansen, Å. M. (2006). Psychosocial factors at home and at work
and levels of salivary cortisol. Biological Psychology, 73(3), 280-287. doi:
10.1016/j.biopsycho.2006.05.003

Elzinga, B. M., & Roelofs, K. (2005). Cortisol-induced impairments of working memory require
acute sympathetic activation. Behavioral Neuroscience, 119(1), 98-103. doi:
10.1037/0735-7044.119.1.98

Ertman, N., Andreano, J. M., & Cahill, L. (2011). Progesterone at encoding predicts subsequent
emotional memory. Learn Mem, 18(12), 759-763. doi: 10.1101/lm.023267.111

Espeland, M. A., Rapp, S. R., Shumaker, S. A., Brunner, R., Manson, J. E., Sherwin, B. B., . . .
Hays, J. (2004). Conjugated equine estrogens and global cognitive function in
postmenopausal women. Journal of the American Medical Association, 291(24), 2959-
2968.

Evolahti, A., Hultcrantz, M., & Collins, A. (2006). Women's work stress and cortisol levels: A
longitudinal study of the association between the psychosocial work environment and
serum cortisol. Journal of Psychosomatic Research, 61(5), 645-652. doi:
10.1016/j.jpsychores.2006.07.022

Fader, A. J., Johnson, P. E., & Dohanich, G. P. (1999). Estrogen Improves Working But Not
Reference Memory and Prevents Amnestic Effects of Scopolamine on a Radial-Arm
Maze. Pharmacology Biochemistry and Behavior, 62(4), 711-717.

Felson, D. T., Zhang, Y., Hannan, M. T., Kiel, D. P., Wilson, P. W. F., & Anderson, J. J. (1993).
The effect of postmenopause estrogen therapy on bone density in elderly women. The
new england journal of medicine, 329(16), 1141-1146.

Fernandez, G., Weis, S., Stoffel-Wagner, B., Tendolkar, I., Reuber, M., Beyenburg, S., . . . Elger,
C., E. (2003). Menstrual cycle-dependent neural plasticity in the adult human brain in
hormone, task, and region specific. The Journal of Neuroscience, 23(9), 3790-3795.

Ferrini, M., & De Nicola, A. F. (1991). Estrogens up-regulate type I and type II glucocorticoid
receptors in brain regions from ovariectomized rats. Life Sciences, 48, 2593-2601.

Ferrini, M., Lima, A., & De Nicola, A. F. (1995). Estradiol abolishes autologous down regulation
of glucocorticoid receptors in brain. Life Sciences, 57(26), 2403-2412.

116
Ferrini, M., Piroli, G., Frontera, M., Falbo, A., Lima, A., & De Nicola, A. F. (1999). Estrogens
Normalize the Hypothalamic-Pituitary-Adrenal Axis Response to Stress and Increase
Glucocorticoid Receptor Immunoreactivity in Hippocampus of Aging Male Rats.
Neuroendocrinology, 69, 129-137.

Foy, M. R., Baudry, M., Foy, J. G., & Thompson, R. F. (2008). 17β-estradiol modifies stress-
induced and age-related changes in hippocampal synaptic plasticity. Behavioral
Neuroscience, 122(2), 301-309. doi: 10.1037/0735-7044.122.2.301

Frick, K. M., Fernandez, M., & Bulinski, S. C. (2002). Estrogen replacement improves spatial
reference memory and increases hippocampal synaptophysin in aged female mice.
Neuroscience, 115(2), 547-558.

Friedman, N. P., & Miyake, A. (2004). The reading span test and its predictive power for reading
comprehension ability. Journal of Memory and Language, 51(1), 136-158. doi:
10.1016/j.jml.2004.03.008

Friedman, S. B., Ader, R., Grota, L. J., & Larson, T. (1967). Plasma Corticosterone Response to
Parameters of Electric Shock Stimulation in the Rat. Psychosomatic Medicine, 29(4),
323-328.

Galliven, E. A., Singh, A., Michelson, D., Bina, S., Gold, P. W., & Deuster, P. A. (1997).
Hormonal and metabolic responses to exercise across time of day and menstrual cycle
phase. Journal of Applied Physiology, 83(6), 1822-1831.

Gallo, M., Roldan, G., & Bures, J. (1992). Differential involvement of gustatory insular cortex
and amygdala in the acquisition and retrieval of conditioned taste aversion in rats.
Behavioural Brain Research, 52, 91-97.

Garrett, J. E., & Wellman, C. L. (2009). Chronic stress effects on dendritic morphology in
medial prefrontal cortex: sex differences and estrogen dependence. Neuroscience, 162(1),
195-207. doi: 10.1016/j.neuroscience.2009.04.057

Garrido, P., De Blas, M., Giné, E., Santos, Á., & Mora, F. (2011). Aging impairs the control of
prefrontal cortex on the release of corticosterone in response to stress and on memory
consolidation. Neurobiology of Aging. doi: 10.1016/j.neurobiolaging.2011.06.011

Gasbarri, A., Pompili, A., d’Onofrio, A., Cifariello, A., Tavares, M. C., & Tomaz, C. (2008).
Working memory for emotional facial expressions: Role of the estrogen in young
women. Psychoneuroendocrinology, 33(7), 964-972. doi:
10.1016/j.psyneuen.2008.04.007

Gerlach, J. L., & McEwen, B. S. (1972). Rat brain binds adrenal steroid hormone:
radioautography of hippocampus with corticosterone. Science, 175(4026), 1133-1136.

117
Gerstner, B., Sifringer, M., Dzietko, M., Schüller, A., Lee, J., Simons, S., . . . Felderhoff-Mueser,
U. (2007). Estradiol attenuates hyperoxia-induced cell death in the developing white
matter. Annals of Neurology, 61(6), 562-573. doi: 10.1002/ana.21118

Gibbs, R. B. (1999). Estrogen Replacement Enhances Acquisition of a Spatial Memory Task and
Reduces Deficits Associated with Hippocampal Muscarinic Receptor Inhibition.
Hormones and Behavior, 36, 222-233.

Gibbs, R. B. (2000). Long-term treatment with estrogen and progesterone enhances acquisition
of a spatial memory task by ovariectomized aged rats. Neurobiology of Aging, 21, 107-
116.

Goldstein, J. M., Jerram, M., Abbs, B., Whitfield-Gabrieli, S., & Makris, N. (2010). Sex
Differences in Stress Response Circuitry Activation Dependent on Female Hormonal
Cycle. Journal of Neuroscience, 30(2), 431-438. doi: 10.1523/jneurosci.3021-09.2010

Goldstein, J. M., Jerram, M., Poldrack, R., Ahern, T., Kennedy, D. N., Seidman, L. J., & Makris,
N. (2005). Hormonal Cycle Modulates Arousal Circuitry in Women Using Functional
Magnetic Resonance Imaging. Journal of Neuroscience, 25(40), 9309-9316. doi:
10.1523/jneurosci.2239-05.2005

Goncharova, N. D., & Lapin, B. A. (2002). Effects of aging on hypothalamic–pituitary–adrenal
system function in non-human primates. Mechanisms of ageing and development 123,
1191-1201.

Gore, A. C., Attardi, B., & DeFranco, D. B. (2006). Glucocorticoid repression of the
reproductive axis: Effects on GnRH and gonadotropin subunit mRNA levels. Molecular
and Cellular Endocrinology, 256(1–2), 40-48. doi:
http://dx.doi.org/10.1016/j.mce.2006.06.002

Gould, E., Woolley, C. S., Frankfurt, M., & McEwen, B. S. (1990). Gonadal Steroids Regulate
Dendritic Spine Density in Hippocampal Pyramidal Cells in Adulthood. The Journal of
Neuroscience, 10(4), 1286-1291.

Grasso, G., Lodi, L., Lupo, C., & Muscettola, M. (1997). Glucocorticoid receptors in human
peripheral blood mononuclear cells in relation to age and to sport activity. Life Sciences,
61(3), 301-308.
Green, J. S., Stanforth, P. R., Rankinen, T., Leon, A. S., Rao, D. C., Skinner, J. S., . . . Wilmore,
J. H. (2004). The effects of exercise training on abdominal visceral fat, body
composition, and indicators of the metabolic syndrome in postmenopausal women with
and without estrogen replacement therapy: The HERITAGE family study. Metabolism,
53(9), 1192-1196. doi: 10.1016/j.metabol.2004.04.008

Handa, R. J., Burgess, L. H., Kerr, J. E., & O'Keefe, J. A. (1994). Gonadal Steroid Hormone
Receptors and Sex Differences in the Hypothalamo-Pituitary-Adrenal Axis. Hormones
and Behavior, 28(4), 464-476. doi: http://dx.doi.org/10.1006/hbeh.1994.1044
118

Hao, J., Rapp, P. R., Janssen, W. G. M., Lou, W., Lasley, B. L., Hof, P. R., & Morrison, J. H.
(2007). Interactive effects of age and estrogen on cognition and pyramidal neurons in
monkey prefrontal cortex. Proceedings of the National Academy of Sciences, 104(27),
11465-11470. doi: 10.1073/pnas.0704757104

Haynes, L. E., Lendon, C. L., Barber, D. J., & Mitchell, I. J. (2003). 17β-oestradiol attenuates
dexamethasone-induced lethal and sublethal neuronal damage in the striatum and
hippocampus. Neuroscience, 120(3), 799-806. doi: 10.1016/s0306-4522(03)00167-2

Hellström, B., & Anderberg, U. M. (2003). Pain perception across the menstrual cycle phases in
women with chronic pain. Perceptual and motor skills, 96(1), 201-211.

Herbison, A. E. (1998). Multimodal influence of estrogen upon gonadotropin-releasing hormone
neurons. Endocrine reviews, 19(3), 302-330.

Herman, J. P., Patel, P. D., Akil, H., & Watson, S. J. (1989). Localization and regulation of
glucocorticoid and mineralocorticoid receptor mRNAs in the hippocampal formation of
the rat. Molecular Endocrinology, 3, 1886-1894.

Hines Jr., E. A., & Brown, G. E. (1936). The cold pressor test for measuring the reactibility of
the blood pressure: data concerning 571 normal and hypertensive subjects. The american
heart journal 11(1), 1-9.

Hintiryan, H., Foster, N. N., & Chambers, K. C. (2009). Dissociating the conditioning and the
anorectic effects of estradiol in female rats. Behavioral Neuroscience, 123(6), 1226-1237.
doi: 10.1037/a0017701

Hoffman, G., Moore, N., Fiksum, G., & Murphy, A. Z. (2003). Ovarian steroid modulation of
seizure severity and hippocampal cell death after kainic acid treatment. Experimental
Neurology, 182(1), 124-134. doi: 10.1016/s0014-4886(03)00104-3

Hosoda, T., Nakajima, H., & Honjo, H. (2001). Estrogen protects neuronal cells from amyloid
beta-induced apoptotic cell death. NeuroReport, 12, 1965-1970.

Jia, D., O'Brien, C. A., Stewart, S. A., Manolagas, S. C., & Weinstein, R. S. (2006).
Glucocorticoids Act Directly on Osteoclasts to Increase Their Life Span and Reduce
Bone Density. Endocrinology, 147(12), 5592-5599. doi: 10.1210/en.2006-0459

Jones, E. E. (2009). The female reproductive system. In W. F. Boron & E. L. Boulpaep (Eds.),
Medical Physiology (2 ed.): Elsevier Health Sciences. Retrieved from
http://www.us.elsevierhealth.com.libproxy.usc.edu/product.jsp?isbn=9781437720174&sg
Country=US&isbn=9781437720174

Kern, S., Oakes, T. R., Stone, C. K., McAuliff, E. M., Kirschbaum, C., & Davidson, R. J. (2008).
Glucose metabolic changes in the prefrontal cortex are associated with HPA axis
119
response to a psychosocial stressor. Psychoneuroendocrinology, 33(4), 517-529. doi:
10.1016/j.psyneuen.2008.01.010

Kirschbaum, C., Kudielka, B., Gaab, J., Schommer, N. C., & Hellhammer, D. H. (1999). Impact
of gender, menstrual cycle phase, and oral contraceptives on the activity of the
hypothalamus-pituitary-adrenal axis. Psychosomatic Medicine 61, 154–162.

Kirschbaum, C., Wust, S., & Hellhammer, D. H. (1992). Consistent sex differences in cortisol
responses to psychological stress. Psychosomatic Medicine, 54, 648-657.

Kleen, J. K., Sitomer, M. T., Killeen, P. R., & Conrad, C. D. (2006). Chronic stress impairs
spatial memory and motivation for reward without disrupting motor ability and
motivation to explore. Behavioral Neuroscience, 120(4), 842-851. doi: 10.1037/0735-
7044.120.4.842

Klein, K., & Boals, A. (2001). The relationship of life event stress and working memory
capacity. Applied cognitive psychology, 15, 565-579.

Komesaroff, P. A., Elser, M. D., & Sudhir, K. (1999). Estrogen supplementation attenuates
glucocorticoid and catecholamine responses to mental stress in perimenopausal women.
Journal of Clinical Endocrinology & Metabolism, 84(2), 606-610.

Krotkiewski, M., Bjorntorp, P., Sjostrom, L., & Smith, U. (1983). Impact of obesity on
metabolism in men and women: importance of regional adipose tissue distribution.
Journal of Clinical Investigation, 83(9), 1150-1162.

Krug, R., Born, J., & Rasch, B. (2006). A 3-day estrogen treatment improves prefrontal cortex-
dependent cognitive function in postmenopausal women. Psychoneuroendocrinology,
31(8), 965-975. doi: 10.1016/j.psyneuen.2006.05.007

Krug, R., Stamm, U., Pietrowsky, R., Fehm, H. L., & Born, J. (1994). Effects of menstrual cycle
on creativity. Psychoneuroendocrinology, 19(1), 21-31.

Kudielka, B., Buske-Kirschbaum, A., Hellhammer, D. H., & Kirschbaum, C. (2004). HPA axis
responses to laboratory psychosocial stress in healthy elderly adults, younger adults, and
children: impact of age and gender. Psychoneuroendocrinology, 29(1), 83-98. doi:
10.1016/s0306-4530(02)00146-4

Kudielka, B., Hellhammer, J., Hellhammer, D. H., Wolf, O., Pirke, K., Varadi, E., . . .
Kirschbaum, C. (1998). Sex Differences in Endocrine and Psychological Responses to
Psychosocial Stress in Healthy Elderly Subjects and the Impact of a 2-Week
Dehydroepiandrosterone Treatment. Journal of Clinical Endocrinology & Metabolism,
83, 1756-1761.

Kudielka, B., & Kirschbaum, C. (2005). Sex differences in HPA axis responses to stress: a
review. Biological Psychology, 69(1), 113-132. doi: 10.1016/j.biopsycho.2004.11.009
120

Kudielka, B., Schmidt-Reinwald, A. K., Hellhammer, D. H., & Kirschbaum, C. (1999).
Psychological and endocrine responses to psychosocial stress and
dexamethasone/corticotropin-releasing hormone in healthy postmenopausal women and
young controls: The impact of age and a two-week estradiol treatment.
Neuroendocrinology, 70, 422-430.

Kudielka, B., Schommer, N. C., Hellhammer, D. H., & Kirschbaum, C. (2004). Acute HPA axis
responses, heart rate, and mood changes to psychosocial stress (TSST) in humans at
different times of day. Psychoneuroendocrinology, 29(8), 983-992. doi:
10.1016/j.psyneuen.2003.08.009

Kuhlmann, S., Piel, M., & Wolf, O. T. (2005). Impaired memory retrieval after psychosocial
stress in healthy young men. The Journal of Neuroscience, 25(11), 2977-2982.

Kuhlmann, S., & Wolf, O. T. (2005). Cortisol and memory retrieval in women: influence of
menstrual cycle and oral contraceptives. Psychopharmacology, 183(1), 65-71. doi:
10.1007/s00213-005-0143-z

Lalmansingh, A. S., & Uht, R. M. (2008). Estradiol Regulates Corticotropin-Releasing Hormone
Gene (crh) Expression in a Rapid and Phasic Manner that Parallels Estrogen Receptor-α
and -β Recruitment to a 3′,5′-Cyclic Adenosine 5′-Monophosphate Regulatory Region of
the Proximal crh Promoter. Endocrinology, 149(1), 346-357. doi: doi:10.1210/en.2007-
0372

Lamprecht, R., & Dudai, Y. (1996). Transient expression of c-Fos in rat amygdala during
training is required for encoding conditioned taste aversion memory. Learning &
Memory, 3(1), 31-41. doi: 10.1101/lm.3.1.31

Laroche, S., Davis, S., & Jay, T. M. (2000). Plasticity at hippocampal to prefrontal cortex
synapses: dual roles in working memory and consolidation. Hippocampus, 10, 438-446.

Lighthall, N. R., Mather, M., & Gorlick, M. A. (2009). Acute stress increases sex differences in
risk seeking in the balloon analogue risk task. PLoS ONE, 4(7). doi:
10.1371/journal.pone.0006002.t001

Lighthall, N. R., Sakaki, M., Vasunilashorn, S., Nga, L., Somayajula, S., Chen, E. Y., . . .
Mather, M. (2011). Gender differences in reward-related decision processing under
stress. Social Cognitive and Affective Neuroscience. doi: 10.1093/scan/nsr026

Lindheim, S. R., Legro, R. S., Bernstein, L., Stanczyk, F. Z., Vijod, M. A., Presser, S. C., &
Lobo, R. A. (1992). Behavioral stress responses in premenopausal and postmenopausal
women and the effects of estrogen. American Journal of Obstetrics and Gynecology, 167,
1831-1836.

121
Lovallo, W. (1975). The cold pressor test and autonomic function: a review and integration.
Psychophysiology, 12(3), 268-282.

Lu, K. H., Hopper, B. R., Vargo, T. M., & Yen, S. S. C. (1979). Chronological Changes in Sex
Steroid, Gonadotropin and Prolactin Secretion in Aging Female Rats Displaying
Different Reproductive States. Biology of reproduction, 21, 193-203.
Luethi, M., Meier, B., & Sandi, C. (2009). Stress effects on working memory, explicit memory,
and implicit memory for neutral and emotional stimuli in healthy men. Frontiers in
Behavioral Neuroscience, 2, 1-9. doi: 10.3389/neuro.08.005.2008

Lupien, S. J., McEwen, B. S., Gunnar, M. R., & Heim, C. (2009). Effects of stress throughout the
lifespan on the brain, behaviour and cognition. Nature reviews. Neuroscience, 10(6), 434-
445. doi: 10.1038/nrn2639

MacPherson, A., Dinkel, K., & Sapolsky, R. (2005). Glucocorticoids worsen excitotoxin-induced
expression of pro-inflammatory cytokines in hippocampal cultures. Experimental
Neurology, 194(2), 376-383. doi: 10.1016/j.expneurol.2005.02.021

Maki, P. M., Dennerstein, L., Clark, M., Guthrie, J., LaMontagne, P., Fornelli, D., . . . Resnick,
S. M. (2011). Perimenopausal use of hormone therapy is associated with enhanced
memory and hippocampal function later in life. Brain Research, 1379, 232-243. doi:
10.1016/j.brainres.2010.11.030

Maki, P. M., Rich, J. B., & Shayna Rosenbaum, R. (2002). Implicit memory varies across the
menstrual cycle: estrogen effects in young women. Neuropsychologia, 40(5), 518-529.

Maki, P. M., Zonderman, A. B., & Resnick, S. M. (2001). Enhanced verbal memory in
nondemented elderly woman receiving hormone-replacement therapy. American Journal
of Psychiatry, 158, 227-233.

Malagutti, K. S., da Silva, A. P., Braga, H. C., Mitozo, P. A., Soares dos Santos, A. R., Dafre, A.
L., . . . Farina, M. (2009). 17β-estradiol decreases methylmercury-induced neurotoxicity
in male mice. Environmental Toxicology and Pharmacology, 27(2), 293-297. doi:
10.1016/j.etap.2008.11.005

Marin-Burgin, A., Mongiat, L. A., Pardi, M. B., & Schinder, A. F. (2012). Unique Processing
During a Period of High Excitation/Inhibition Balance in Adult-Born Neurons. Science,
335(6073), 1238-1242. doi: 10.1126/science.1214956

Markowska, A. L., & Savonenko, A. V. (2002). Effectiveness of Estrogen Replacement in
Restoration of Cognitive Function after Long-Term Estrogen Withdrawal in Aging Rats.
The Journal of Neuroscience, 22(24), 10985-10995.

Mateo, J. M. (2008). Inverted-U shape relationship between cortisol and learning in ground
squirrels. Neurobiology of Learning and Memory, 89(4), 582-590. doi:
http://dx.doi.org/10.1016/j.nlm.2007.11.002
122

Mather, M., Cowell, P. E., Whiteside, S. P., Ledger, W. L., & Mangold, S. (under review).
Menstrual cycle and ovarian hormone effects on emotional memory.

Mather, M., Lighthall, N. R., Nga, L., & Gorlick, M. A. (2010). Sex differences in how stress
affects brain activity during face viewing. NeuroReport, 21(14), 933-937. doi:
10.1097/WNR.0b013e32833ddd92

Matt, D. W., Sarver, P. L., & Lu, J. K. H. (1987). Relation of Parity and Estrous Cyclicity to the
Biology of Pregnancy in Aging Female Rats. Biology of reproduction, 37, 421-430.
McEwen, B. S. (2007). Physiology and Neurobiology of Stress and Adaptation: Central Role of
the Brain. Physiological Reviews, 87, 873-904. doi: 10.1152/physrev.00041.2006.-The

McEwen, B. S., & Morrison, John H. (2013). The brain on stress: vulnerability and plasticity of
the prefrontal cortex over the life course. Neuron, 79(1), 16-29. doi:
10.1016/j.neuron.2013.06.028

McEwen, B. S., Weiss, J. M., & Schwartz, L. S. (1968). Selective retention of corticosterone by
limbic structures in the rat brain. Nature, 220, 911-912.

McGuinness, O. P., Fugiwara, T., Murrell, S., Bracy, D., Neal, D., O'Connor, D., & Cherrington,
A. D. (1993). Impact of chronic stress hormone infusion on hepatic carbohydrate
metabolism in the conscious dog. endocrinology & metabolism, 28, E314-E322.

McLaughlin, K. J., Wilson, J. O., Harman, J., Wright, R. L., Wieczorek, L., Gomez, J., . . .
Conrad, C. D. (2010). Chronic 17β-estradiol or cholesterol prevents stress-induced
hippocampal CA3 dendritic retraction in ovariectomized female rats: Possible
correspondence between CA1 spine properties and spatial acquisition. Hippocampus,
NA-NA. doi: 10.1002/hipo.20678

Miele, J., Rosellini, R. A., & Svare, B. (1988). Estradiol Benzoate Can Function as an
Unconditioned Stimulus in a Conditioned Taste Aversion Paradigm. Hormones and
Behavior, 22, 116-130.

Miller, K. J., Conney, J. C., Rasgon, N. L., Fairbanks, L. A., & Small, G. W. (2002). Mood
symptoms and cognitive performance in women estrogen users and nonusers and men.
Journal of the American Geriatrics Society, 50, 1826-1830.

Mizoguchi, K., Ikeda, R., Shoji, H., Tanaka, Y., Maruyama, W., & Tabira, T. (2009). Aging
attenuates glucocorticoid negative feedback in rat brain. Neuroscience, 159(1), 259-270.
doi: 10.1016/j.neuroscience.2008.12.020

Morales-Medina, J. C., Sanchez, F., Flores, G., Dumont, Y., & Quirion, R. (2009).
Morphological reorganization after repeated corticosterone administration in the
hippocampus, nucleus accumbens and amygdala in the rat. Journal of Chemical
Neuroanatomy, 38(4), 266-272. doi: 10.1016/j.jchemneu.2009.05.009
123

Morgan, C. A., 3rd, Doran, A., Steffian, G., Hazlett, G., & Southwick, S. M. (2006). Stress-
Induced Deficits in Working Memory and Visuo-Constructive Abilities in Special
Operations Soldiers. Biological Psychiatry, 60(7), 722-729. doi:
10.1016/j.biopsych.2006.04.021

Morgan, C. A., 3rd, Southwick, S., Steffian, G., Hazlett, G. A., & Loftus, E. F. (2013).
Misinformation can influence memory for recently experienced, highly stressful events.
Int J Law Psychiatry, 36(1), 11-17. doi: 10.1016/j.ijlp.2012.11.002

Morimoto, K., Morikawa, M., Kimura, H., Ishii, N., Takamata, A., Hara, Y., . . . Yoshida, K.
(2008). Mental stress induces sustained elevation of blood pressure and lipid peroxidation
in postmenopausal women. Life Sciences, 82(1-2), 99-107. doi: 10.1016/j.lfs.2007.10.018

Mukai, H., Tsurugizawa, T., Ogiue-Ikeda, M., Murakami, G., Hojo, Y., Ishii, H., . . . Kawato, S.
(2006). Local Neurosteroid Production in the Hippocampus: Influence on Synaptic
Plasticity of Memory. Neuroendocrinology, 84(4), 255-263. doi: 10.1159/000097747

Mulnard, R. A., Cotman, C. W., Kawas, C., van Dyck, C. H., Sano, M., Doody, R., . . . Thal, L.
J. (2000). Estrogen Replacement Therapy for Treatment of Mild to Moderate Alzheimer
Disease: A Randomized Controlled Trial. Journal of the American Medical Association,
283, 1007-1015.

Musatov, S., Chen, W., Pfaff, D. W., Mobbs, C. V., Yang, X. J., Clegg, D. J., . . . Ogawa, S.
(2007). Silencing of estrogen receptor in the ventromedial nucleus of hypothalamus leads
to metabolic syndrome. Proceedings of the National Academy of Sciences, 104(7), 2501-
2506. doi: 10.1073/pnas.0610787104

Nachman, M., & Ashe, J. H. (1973). Learned Taste Aversions in Rats as a Function of Dosage,
Concentration, and Route of Administration of LiC1. Physiology & Behavior, 10, 73-78.

Nadeau, S., & Rivest, S. (2003). Glucocorticoids Play a Fundamental Role in Protecting the
Brain during Innate Immune Response. The Journal of Neuroscience, 23(13), 5536-5544.

Navarro, M., Spray, K. J., Cubero, I., Thiele, T. E., & Bernstein, I. L. (2000). cFos induction
during conditioned taste aversion expression varies with aversion strength. Brain
Research, 887, 450-453.

Nelson, H. D. (2004). Commonly used types of postmenopausal estrogen for treatment of hot
flashes. JAMA: the journal of the American Medical Association, 291(13), 1610-1620.

O'Brien, C. A., Jia, D., Plotkin, L. I., Bellido, T., Powers, C. C., Stewart, S. A., . . . Weinstein, R.
S. (2004). Glucocorticoids act directly on osteoblasts and osteocytes to induce their
apoptosis and reduce bone formation and strength. Endocrinology, 145(4), 1835-1841.
doi: 10.1210/en.2003-0990
124

Oei, N. Y. L., Everaerd, W. T. A. M., Elzinga, B. M., van Well, S., & Bermond, B. (2006).
Psychosocial stress impairs working memory at high loads: An association with cortisol
levels and memory retrieval. Stress: The International Journal on the Biology of Stress,
9(3), 133-141. doi: 10.1080/10253890600965773

Ohl, F., & Fuchs, E. (1999). Differential effects of chronic stress on memory processes in the
tree shrew. Cognitive Brain Research, 7, 379-387.

Park, C. R., Campbell, A. M., & Diamond, D. M. (2001). Chronic Psychosocial Stress Impairs
Learning and Memory and Increases Sensitivity to Yohimbine in Adult Rats. Biological
Psychiatry, 50(994-1004).

Pasquali, R., Vicennati, V., Cacciari, M., & Pagotto, U. (2006). The hypothalamic-pituitary-
adrenal axis activity in obesity and the metabolic syndrome. Annals of the New York
Academy of Sciences, 1083(1), 111-128. doi: 10.1196/annals.1367.009

Pavlides, C., Watanabe, Y., Magariños, A. M., & McEwen, B. S. (1995). Opposing roles of type
I and type II adrenal steroid receptors in hippocampal long-term potentiation.
Neuroscience, 68(2), 387-394. doi: http://dx.doi.org/10.1016/0306-4522(95)00151-8

Pavlides, C., Watanabe, Y., & McEwen, B. S. (1993). Effects of glucocorticoids on hippocampal
long-term potentiation. Hippocampus, 3(2), 183-192. doi: 10.1002/hipo.450030210

Payne, J. D., Jackson, E. D., Hoscheidt, S., Ryan, L., Jacobs, W. J., & Nadel, L. (2007). Stress
administered prior to encoding impairs neutral but enhances emotional long-term
episodic memories. Learning & Memory, 14(12), 861-868. doi: 10.1101/lm.743507

Pefanco, M. A., Kenny, A. M., Kaplan, R. F., Kuchel, G., Walsh, S., Kleppinger, A., &
Prestwood, K. (2007). The Effect of 3-Year Treatment with 0.25â ƒmg/day of
Micronized 17Î²-Estradiol on Cognitive Function in Older Postmenopausal Women.
Journal of the American Geriatrics Society, 55(3), 426-431. doi: 10.1111/j.1532-
5415.2007.01085.x

Phillips, S. M., & Sherwin, B. B. (1992). Variations in memory function and sex steroid
hormones across the menstrual cycle. Psychoneuroendocrinology, 17(5), 497-506. doi:
http://dx.doi.org/10.1016/0306-4530(92)90008-U

Pike, C. J. (1999). Estrogen Modulates Neuronal Bcl-xL Expression and b-Amyloid-Induced
Apoptosis: Relevance to Alzheimer’s Disease. Journal of Neurochemistry, 72, 1552-
1563.

Pike, C. J., Carroll, J. C., Rosario, E. R., & Barron, A. M. (2009). Protective actions of sex
steroid hormones in Alzheimer’s disease. Frontiers in Neuroendocrinology, 30(2), 239-
258. doi: 10.1016/j.yfrne.2009.04.015

125
Puder, J. J., Freda, P. U., Goland, R. S., & Wardlaw, S. L. (2001). Estrogen modulates the
hypothalamic-pituitary-adrenal and inflammatory cytokine responses to endotoxin in
women. Journal of Clinical Endocrinology & Metabolism, 86(6), 2403-2408.

Purves, D., Augustine, G. J., Fitzpatrick, D., Hall, W. C., LaMantia, A.-S., McNamara, J. O., &
Williams, S. M. (Eds.). (2004). Neuroscience (3rd ed.). Sunderland, MA, U.S.A: Sinauer.

Qin, S., Hermans, E. J., van Marle, H. J. F., Luo, J., & Fernández, G. (2009). Acute
psychological stress reduces working memory-related activity in the dorsolateral
prefrontal cortex. Biological Psychiatry, 66(1), 25-32. doi:
10.1016/j.biopsych.2009.03.006

Radloff, L. S. (1977). The CES-D scale: a self-report depression scale for research in the general
population. Applied psychological measurement, 1(3), 385-401.

Rapp, P. R., Morrison, J. H., & Roberts, J. A. (2003). Cyclic Estrogen Replacement Improves
Cognitive Function in Aged Ovariectomized Rhesus Monkeys. The Journal of
Neuroscience, 23(13), 5708-5714.

Rapp, S. R., Espeland, M. A., Shumaker, S. A., Henderson, V. W., Brunner, R. L., Manson, J. E.,
. . . Bowen, D. (2003). Effect of estrogen plus progestin on global cognitive function in
postmenopausal women. The women's health initiative memory study: a randomized
controlled trial. Journal of the American Medical Association, 289(20), 2663-2672.

Reavis, R., & Overman, W. H. (2001). Adult sex differences on a decision-making task
previously shown to depend on the orbital prefrontal cortex. Behavioral Neuroscience,
115(1), 196-206. doi: 10.1037//0735-7044.115.1.196

Rebuffe-Scrive, M., Walsh, U. A., McEwen, B. S., & Rodin, J. (1992). Effect of chronic stress
and exogenous glucocorticoids on regional fat distribution and metabolism. Physiology &
Behavior, 52, 583-590.

Resnick, S. M., Maki, P. M., Rapp, S. R., Espeland, M. A., Brunner, R., Coker, L. H., . . .
Shumaker, S. A. (2006). Effects of Combination Estrogen Plus Progestin Hormone
Treatment on Cognition and Affect. Journal of Clinical Endocrinology & Metabolism,
91(5), 1802-1810. doi: 10.1210/jc.2005-2097

Reul, J. M. H. M., Rothuizen, J., & de Kloet, E. R. (1991). Age-related changes in the dog
hypothalamic-pituitary-adrenocortical system: neuroendocrine activity and corticosteroid
receptors. Journal of Steroid biochemistry and molecular biology, 40(1), 63-69.

Riegle, G. D. (1973). Chronic stress effects on adrenocortical responsiveness in young and aged
rats. Neuroendocrinology, 11, 1-10.

Riegle, G. D., & Hess, G. D. (1972). Chronic and acute dexamethasone supression of stress
activation of the adrenal cortex in young and aged rats. Neuroendocrinology, 9, 175-187.
126
Riley III, J. L., E. Robinson, M., Wise, E. A., & Price, D. (1999). A meta-analytic review of pain
perception across the menstrual cycle. Pain, 81(3), 225-235. doi:
http://dx.doi.org/10.1016/S0304-3959(98)00258-9

Rivier, C., & Rivest, S. (1991). Effect of stress on the activity of the hypothalamic-pituitary-
gonadal axis: peripheral and central mechanisms. Biology of reproduction, 45(4), 523-
532.

Roca, C. A., Schmidt, P. J., Altemus, M., Deuster, P., Danaceau, M. A., Putman, K., & Rubinow,
D. R. (2003). Differential Menstrual Cycle Regulation of Hypothalamic-Pituitary-
Adrenal Axis in Women with Premenstrual Syndrome and Controls. Journal of Clinical
Endocrinology & Metabolism, 88(7), 3057-3063. doi: 10.1210/jc.2002-021570

Roldan, G., & Bures, J. (1994). Tetrodotoxin blockade of amygdala overlapping with poisoning
impairs acquisition of conditioned taste aversion in rats. Behavioural Brain Research, 65,
213-219.

Roozendaal, B., & McGaugh, J. L. (1997). Glucocorticoid Receptor Agonist and Antagonist
Administration into the Basolateral but Not Central Amygdala Modulates Memory
Storage. Neurobiology of Learning and Memory, 67, 176-179.

Roozendaal, B., McReynolds, J. R., Van der Zee, E. A., Lee, S., McGaugh, J. L., & McIntyre, C.
K. (2009). Glucocorticoid Effects on Memory Consolidation Depend on Functional
Interactions between the Medial Prefrontal Cortex and Basolateral Amygdala. Journal of
Neuroscience, 29(45), 14299-14308. doi: 10.1523/jneurosci.3626-09.2009

Rosmond, R. (2005). Role of stress in the pathogenesis of the metabolic syndrome.
Psychoneuroendocrinology, 30(1), 1-10. doi: 10.1016/j.psyneuen.2004.05.007

Roy, B. N., Reid, R. L., & Van Vugt, D. A. (1999). The Effects of Estrogen and Progesterone on
Corticotropin-Releasing Hormone and Arginine Vasopressin Messenger Ribonucleic
Acid Levels in the Paraventricular Nucleus and Supraoptic Nucleus of the Rhesus
Monkey. Endocrinology, 140(5), 2191-2198. doi: doi:10.1210/endo.140.5.6684

Sakaki, M., & Mather, M. (2012). How emotion and reward processing change across the
menstrual cycle. social and personality psychology compass, 6(1), 1-17.

Salehi, B., Cordero, M. I., & Sandi, C. (2010). Learning under stress: the inverted-U-shape
function revisited. Learning & Memory, 17(10), 522-530.

Sandstrom, N. J., & Williams, C. L. (2001). Memory retention is modulated by acute estradiol
and progesterone replacement Behavioral Neuroscience, 115(2), 384-393. doi:
10.1037//0735-7044.115.2.384

Sapolsky, R. M. (1990). Chapter 2 Glucocorticoids, hippocampal damage and the glutamatergic
synapse. 86, 13-23. doi: 10.1016/s0079-6123(08)63163-5
127

Sapolsky, R. M. (1999). Glucocorticoids, stress, and their adverse neurological effects: relevance
to aging. Experimental Gerontology, 34, 721-732.

Sapolsky, R. M. (2000). Glucocorticoids and Hippocampal Atrophy in Neuropsychiatric
Disorders. Archives of general psychiatry, 57, 925-935.

Sapolsky, R. M., Krey, L. C., & McEwen, B. S. (1983). The adrenocortical stress-response in the
aged male rat: impariment of recovery from stress. Experimental Gerontology, 18, 55-64.

Sapolsky, R. M., Krey, L. C., & McEwen, B. S. (1984). Stress down-regulates corticosterone
receptors in a site-specific manner in the brain. Endocrinology, 114(1), 287-292.

Sapolsky, R. M., Krey, L. C., & McEwen, B. S. (1986). The neuroendocrinology of stress and
aging: the glucocorticoid cascade hypothesis. Endocrine Reviews, 7(3), 284-301.

Sapolsky, R. M., Stein-Behrens, B. A., & Armanini, M. P. (1991). Long-Term Adrenalectomy
Causes Loss of Dentate Gyrus and Pyramidal Neurons in the Adult Hippocampus.
Experimental Neurology, 114(246-249).

Saravia, F., Beauquis, J., Pietranera, L., & De Nicola, A. F. (2007). Neuroprotective effects of
estradiol in hippocampal neurons and glia of middle age mice.
Psychoneuroendocrinology, 32(5), 480-492. doi: 10.1016/j.psyneuen.2007.02.012

Schoofs, D., Preuß, D., & Wolf, O. T. (2008). Psychosocial stress induces working memory
impairments in an n-back paradigm. Psychoneuroendocrinology, 33(5), 643-653. doi:
10.1016/j.psyneuen.2008.02.004

Schoofs, D., Wolf, O. T., & Smeets, T. (2009). Cold pressor stress impairs performance on
working memory tasks requiring executive functions in healthy young men. Behavioral
Neuroscience, 123(5), 1066-1075. doi: 10.1037/a0016980

Seckl, J. R., & Olsson, T. (1995). Glucocorticoid hypersecretion and the age-impaired
hippocampus: cause or effect? Journal of endocrinology, 145, 201-211.

Selye, H. (1939). The effect of adaptation to various damaging agents on the female sex organs
in the rat. Endocrinology, 25(4), 615-624.

Sharma, K., & Mehra, R. D. (2008). Long-term administration of estrogen or tamoxifen to
ovariectomized rats affords neuroprotection to hippocampal neurons by modulating the
expression of Bcl-2 and Bax. Brain Research, 1204, 1-15. doi:
10.1016/j.brainres.2008.01.080

Shors, T. J. (2004). Learning During Stressful Times. Learning & Memory, 11(2), 137-144. doi:
10.1101/lm.66604

128
Shumaker, S. A., Legault, C., Rapp, S. R., Thal, L., Wallace, R. B., Ockene, J. K., . . .
Wactawski-Wende, J. (2003). Estrogen plus progestin and the incidence of dementia and
mild cognitive impairment in postmenopausal women. The women's health initiative
memory memory study: A rondomized controlled trial. Journal of the American Medical
Association, 289, 2651-2662.

Smeets, T., Jelicic, M., & Merckelbach, H. (2006). The effect of acute stress on memory depends
on word valence. International Journal of Psychophysiology, 62(1), 30-37. doi:
10.1016/j.ijpsycho.2005.11.007

Smith, Y. R., Giordani, B., Lajiness-O'Neill, R., & Zubieta, J.-K. (2001). Long-term estrogen
replacement is associated with imporved nonverbal memory and attentional measures in
postmenopausal women. Menopause, 76(6), 1101-1107.

Sowers, M., Crutchfield, M., Bandekar, R., Randolph, J. F., Shapiro, B., Schork, M. A., &
Jannausch, M. (1998). Bone mineral density and its change in pre- and perimenopausal
white women: the Michigan bone health study. Journal of Bone and Mineral Research,
13(7), 1134-1140.

Stein-Behrens, B. A., Elliott, E. M., Miller, C. A., Schilling, J. W., Newcombe, R., & Sapolsky,
R. M. (1992). Glucocorticoids exacerbate kainic acid-induced extracellular accumulation
of excitatory amino acids in the rat hippocampus. Journal of Neurochemistry, 58, 1730-
1735.

Steiner, M., Peer, M., Macdougall, M., & Haskett, R. (2011). The premenstrual tension
syndrome rating scales: An updated version. Journal of Affective Disorders, 135(1-3), 82-
88. doi: 10.1016/j.jad.2011.06.058

Ter Horst, G. J., Wichmann, R., Gerrits, M., Westenbroek, C., & Lin, Y. (2009). Sex differences
in stress responses: Focus on ovarian hormones. Physiology & Behavior, 97(2), 239-249.
doi: 10.1016/j.physbeh.2009.02.036

Tetsuka, M. (2007). Actions of glucocorticoid and their regulatory mechanisms in the ovary.
Animal Science Journal, 78(2), 112-120. doi: 10.1111/j.1740-0929.2007.00414.x

Tilbrook, A., Turner, A., & Clarke, I. (2000). Effects of stress on reproduction in non-rodent
mammals: the role of glucocorticoids and sex differences. Reviews of reproduction, 5(2),
105-113.

Tombaugh, G. C., Yang, S. H., Swanson, R. A., & Sapolsky, R. M. (1992). Glucocorticoids
exacerbate hypoxic and hypoglycemic hippocampal injury in vitro: biochemical
correlates and a role for astrocytes. Journal of Neurochemistry, 59, 137-146.

Tunn, S., Mollmann, H., Barth, J., Derendorf, H., & Krieg, M. (1992). Simultaneous
Measurement of Cortisol in Serum and Saliva after Different Forms of Cortisol
Administration. Clinical Chemistry, 38(8), 1491-1494.
129

Turner, M. L., & Engle, R. W. (1989). Is working memory capacity task dependent? Journal of
Memory and Language, 28(2), 127-154. doi: http://dx.doi.org/10.1016/0749-
596X(89)90040-5

Tynan, R. J., Naicker, S., Hinwood, M., Nalivaiko, E., Buller, K. M., Pow, D. V., . . . Walker, F.
R. (2010). Chronic stress alters the density and morphology of microglia in a subset of
stress-responsive brain regions. Brain, Behavior, and Immunity, 24(7), 1058-1068. doi:
10.1016/j.bbi.2010.02.001

Uno, H., Tarara, R., Else, J. G., Suleman, M. A., & Sapolsky, R. M. (1989). Hippocampal
Damage Associated with Prolonged and Fatal Stress in Primates. The Journal of
Neuroscience, 9(5), 1705-1711.

Vaucher, E., Reymond, I., Najaffe, R., Kar, S., Quirion, R., Miller, M. M., & Franklin, K. B. J.
(2002). Estrogen effects on object memory and cholinergic receptors in young anf old
female mice. Neurobiology of Aging, 23, 87-95.

Viau, V., & Meaney, M. J. (1991). Variations in the Hypothalamic-Pituitary-Adrenal Response
to Stress during the Estrous Cycle in the Rat. Endocrinology, 129(5), 2503-2511. doi:
doi:10.1210/endo-129-5-2503

Vining, R. F., McGinley, R. A., & Symons, R. G. (1983). Hormones in Saliva: Mode of Entry
and Consequent Implications for Clinical Interpretation. Clinical Chemistry, 29(10),
1752-1756.

Virgin, J., C.E., Ha, T., Packan, D. R., Tombaugh, G. C., Yang, S. H., Horner, H. C., &
Sapolsky, R. M. (1991). Glucocorticoids inhibit glucose transport and glutamte uptake in
hippocampal astrocytes: implications for glucocorticoid neurotoxicity. Journal of
Neurochemistry, 57, 1422-1428.

Watanabe, Y., Gould, E., & McEwen, B. S. (1992). Stress induces atrophy of apical dendrites of
hippocampal CA3 pyramidal neurons. Brain Research, 588, 341-343.

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures
of positive and negative affect: the PANAS scales. Journal of Personality and Social
Psychology, 54(6), 1063-1070.
Wechsler, D. (1981). Wechsler adult intelligence scale-revised: Psychological Corporation.

Wellman, C. L. (2001). Dendritic Reorganization in Pyramidal Neurons in Medial Prefrontal
Cortex after Chronic Corticosterone Administration. Journal of Neurobiology, 49, 245-
253.

Welsh, K. A., Breitner, J. C. S., & Magruder-Habib, K. M. (1993). Detection of dementia in the
elderly using telephone screening of cognitive status. Neuropsychiatry, Neuropsychology,
and Behavioral Neurology, 6(2), 103-110.
130

Wilkinson, C. W., Petrie, E. C., Murray, S. R., Colasurdo, E. A., Raskind, M. A., & Peskind, E.
R. (2001). Human glucocorticoid feedback inhibition is reduced in older individuals:
evening study. Journal of Clinical Endocrinology & Metabolism, 86(2), 545-550.

Wilson, I. A., Puolivali, J., Heikkinen, T., & Riekkinen Jr., P. (1999). Estrogen and NMDA
receptor antagonism: effects upon reference and working memory. European Journal of
Pharmacology, 381, 93-99.

Wolf, O. T., & Kirschbaum, C. (2002). Endogenous estradiol and testosterone levels are
associated with cognitive performance in older women and men. Hormones and
Behavior, 41(3), 259-266. doi: 10.1006/hbeh.2002.1770

Wolf, O. T., Kudielka, B., Hellhammer, D. H., Torber, S., McEwen, B. S., & Kirschbaum, C.
(1999). Two weeks of transdermal estradiol treatment in postmenopausal elderly women
and its effect on memory and mood: verbal memory changes are associated with the
treatment induced estradiol levels. Psychoneuroendocrinology, 24, 727-741.

Wolf, O. T., Minnebusch, D., & Daum, I. (2009). Stress impairs acquisition of delay eyeblink
conditioning in men and women. Neurobiology of Learning and Memory, 91(4), 431-436.
doi: 10.1016/j.nlm.2008.11.002

Womble, M. D., Andrew, J. A., & Crook, J. J. (2002). 17b-Estradiol reduces excitatory
postsynaptic potential (EPSP) amplitude in rat basolateral amygdala neurons.
Neuroscience Letters, 331, 83-86.

Woolley, C. S., Gould, E., Frankfurt, M., & McEwen, B. S. (1990). Naturally Occurring
Fluctuation in Dendritic Spine Density on Adult Hippocampal Pyramidal Neurons. The
Journal of Neuroscience, 10(12), 4035-4039.

Woolley, C. S., Gould, E., & McEwen, B. S. (1990). Exposure to excess glucocorticoids alters
dendritic morphology of adult hippocampal pyramidal neurons. Brain Research, 531,
225-231.

Woolley, C. S., & McEwen, B. S. (1992). Estradiol Mediates Fluctuation in Hippocampal
Synapse Density during the Estrous Cycle in the Adult Rat. The Journal of Neuroscience,
12(7), 2549-2554.

Woolley, C. S., & McEwen, B. S. (1993). Roles of Estradiol and Progesterone in Regulation of
Hippocampal Dendritic Spine Density During the Estrous Cycle in the Rat. The Journal
of Comparative Neurology, 336, 293-306.
Woolley, C. S., & McEwen, B. S. (1994). Estradiol Regulates Hippocampal Dendritic Spine
Density via an N-Methyl-D-Aspartate Receptor-Dependent Mechanism. The Journal of
Neuroscience, 14(12), 7680-7687.

131
Yaffe, K., Vittinghoff, E., Ensrud, K. E., Johnson, K. C., Diem, S., Hanes, V., & Grady, D.
(2006). Effects of ultra-low-dose transdermal estradiol on cognition and health-related
quality of life. Archives of Neurology, 63, 945-950.

Yerkes, R. M., & Dodson, J. D. (1908). The relation of strength of stimulus to rapidity of habit-
formation. Journal of Comparative Neurology and Psychology, 18(5), 459-482.

Young, A. H., Sahakian, B. J., Robbins, T. W., & Cowen, P. J. (1999). The effects of chronic
administration of hydrocortisone on cognitive function in normal male volunteers.
Psychopharmacology, 145, 260-266.

Zald, D. H. (2003). The human amygdala and the emotional evaluation of sensory stimuli. Brain
Research Reviews, 41, 88-123.

132
Appendix A: Tables
Table 2.1: Average timeline for Study 1 sessions

Time
(min)
Time (min) Since Stressor Onset Task
10 -25 Informed consent/ HIPAA Waiver/Drink
water
10 -10 Questionnaires/Rest
5 -5 Baseline saliva sample/Baseline pain and
stress rating
3 0-3 Stress manipulation
1 3-4 Post stress manipulation pain and stress
rating
10 4-14 Questionnaires and Word Search
3 14-17 Saliva sample 2
2 18-20 Emotional Memory Task, Encoding Phase
11 21-32 Sentence Span Task
4 34-38 Emotional Memory Task, Recall Phase
2 38-40 Emotional Memory Task, Association Test
Phase
3 41-44 Saliva sample 3
5 49-54 Debriefing
Total time: 69 (60-80 to account for individual timing differences)
133
Table 2.2: Sex hormone levels, demographics, emotional state, and mood
LE HE df t p
Estradiol levels 1.29 pg/ml 97.51 pg/ml 18 -2.936 .009
Progesterone levels 80.12 pg/ml 50.72 pg/ml 18 .443 .663
Age at menopause 50.13 51.38 17 -.677 .507
Age at randomization 59.59 62.03 18 -.665 .514
Years on study drugs 4.94 4.88 18 .376 .711
Age at time of participation 64.88 66.91 18 -.644 .527
Years of education 16.2 17.6 18 -2.214 .040
WTAR 44.80 45.80 18 -.471 .643
S1– Overall health rating 7.65 8.70 18 -2.750 .013
S1 – Stress levels on day of session 2.9 1.9 18 1.467 .160
S1 – Stress on day of session
compared to normal non-study days
5.0 5.2 18 -2.169 .791
S1 – Positive affect 34.50 30.30 18 1.143 .268
S1 – Negative affect 11.20 10.50 18 1.505 .094
S1 – Pre stress, pain 9.9 1.0 18 1.619 .123
S1 – Pre stress, stress 9.0 2.5 18 2.652 .016
S1 – CES-D 5.2 2.0 18 1.303 .193
S2 – Overall health rating 7.9 8.8 18 -3.349 .004
S2 – Stress levels on day of session 2.2 1.5 18 1.481 .156
S2 – Stress on day of session
compared to normal non-study days
5.0 4.6 18 .937 .361
134
S2 – Positive affect 36.4 32.1 18 1.233 .234
S2 – Negative affect 10.8 10.3 18 1.197 .247
S2 – Pre stress, pain 6.8 1.2 18 1.496 .152
S2 – Pre stress, stress 11.0 5.1 18 1.079 .295
S2 – CES-D 6.1 4.0 18 .736 .471
*S1=Session 1, S2=Session 2; Significant values are bolded; the df=17 for age at menopause was due to a missing
value from the ELITE data set.

135
Table 3.1: Average timeline for Study 2 sessions

Time
(min)
Time (min) Since Stressor
Onset
Task
10 -10 Informed consent/Drink water/Questionnaire
2 -2 Baseline saliva sample/Baseline pain and stress
rating
3 0-3 Stress manipulation
1 3-4 Post stress manipulation pain and stress rating
10 4-14 Word Search
3 14-17 Saliva sample 2
2 18-20 Emotional Memory Task, Encoding Phase
11 21-32 Sentence Span Task
3 35-38 Emotional Memory Task, Recall Phase
1.5 38-40 Emotional Memory Task, Association Test Phase
3 41-44 Saliva sample 3
5 49-54 Debriefing
Total time: 54.5 minutes (41-62 minutes to account for individual timing differences)
136
Table 3.2: Emotional state, mood, and PMS symptoms during first session of each phase
EF LF df t p
Positive Affect 24.26 25.52 26 -.829 .415
Negative Affect 13.48 13.52 26 -.063 .950
CES-D 18.0 17.52 26 .346 .732
Depressed mood – PMTS-VAS 20.07 22.22 26 -.538 .595
Tense, restless, anxious – PMTS-VAS 29.74 32.19 26 -.807 .427
Emotional, mood swings – PMTS-VAS 26.37 27.52 26 -.254 .802
Irritable, hostile – PMTS-VAS 17.74 16.44 26 .382 .706
Decreased interest in activities – PMTS-VAS 24.11 17.37 26 1.557 .132
Difficulty concentrating – PMTS-VAS 28.59 22.22 26 1.416 .169
Lethargy, easy fatigability, lack of energy –
PMTS-VAS
40.11 23.89 26 3.045 .005
Overeating, food cravings – PMTS-VAS 25.56 22.19 26 .637 .530
Change in sleep patterns: unable to sleep – PMTS-
VAS
16.93 17.59 26 -.144 .886
Change in sleep pattern: sleeping more – PMTS-VAS 26.15 20.70 26 .830 .414
Feeling overwhelmed or out of control – PMTS-
VAS
34.63 26.15 26 1.936 .064
Breast tenderness, bloating, water retention –
PMTS-VAS
33.74 13.74 26 3.257 .003
Pre Stress, Stress 26.37 23.11 26 .756 .459
Pre Stress, Pain 9.74 5.44 26 1.670 .107
*Significant values are bolded

137
Appendix B: Figures

University of New South Wales Embryology:
http://php.med.unsw.edu.au/embryology/index.php?title=File:HPA_
axis.jpg
Hippocampus
Figure 1.1
138

Figure 1.2
University of South Wales Embryology:
http://php.med.unsw.edu.au/embryology/index.php?title=File:HPG_f
emale_axis.jpg
139

-10
0
10
20
30
40
50
60
70
Pre Stress Post Stress
Visual analog acale measurements (mm)
Figure 2.1: Subjective ratings of stress before and after
water exposure
LE Stress
LE
Control
HE Stress
HE
Control
140

-10
0
10
20
30
40
50
60
70
80
Pre Pain Post Pain
Visual Analog Scale Measurements (mm)
Figure 2.2: Subjective ratings of pain before and after
water exposure
LE Stress
LE Control
HE Stress
HE Control
141

0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
baseline 15m post baseline 15m post
cold warm
Salivary Cortisol (ug/dl)
Figure 2.3: Cortisol during the stress and control
sessions collapsed across HE and LE women
p=.005
n.s.
142

0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
baseline 15m post baseline 15m post
Higher E2 Lower E2
Salivary Cortisol (ug/dl)
FIgure 2.4: Cortisol levels in HE and LE women
collapsed across sessions
143

0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
baseline 15m post baseline 15m post
Higher E2 Lower E2
Salivary Cortisol (ug/dl)
Figure 2.5: Cortisol levels in HE and LE women during
the stress session only
n.s.
p=.015
144

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
cold warm cold warm
Higher E2 Lower E2
Proportion of Words Correctly Recalled
Figure 2.6: Working memory performance in HE and LE
women
n.s.

p=.05
145

0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
emotional neutral
Proportion of pictures correctly recalled
Figure 2.7: Picture recall collapsed across session and
estradiol: Emotional versus neutral pictures
146

0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
neg neu neg neu
stress control
Proportion of pictures correctly recalled
Figure 2.8: Emotional memory picture recall during the
stress and control sessions
n.s.
p=.007
147

148

149

0
10
20
30
40
50
60
70
Pre Stress Post Stress
Visual analog scale measurements (mm)
Figure 3.1: Subjective ratings of stress before and after
water exposure
EF Stress
EF Control
LF Stress
LF Control
150

0
10
20
30
40
50
60
70
80
Pre Stress Post Stress
Visual analog scale measurements (mm)
Figure 3.2: Subjective ratings of pain before and after
water exposure
EF Stress
EF Control
LF Stress
LF Control
151

0.05
0.07
0.09
0.11
0.13
0.15
0.17
stress control
Salivary Cortisol (ug/ml)
Figure 3.3: Cortisol levels collapsed across phase and
all 3 time points
152

0.05
0.07
0.09
0.11
0.13
0.15
0.17
Baseline 15m 40m
Salivary Cortisol (ug/ml)
Figure 3.4 Cortisol levels collapsed across session and
phase
153

0.05
0.07
0.09
0.11
0.13
0.15
0.17
0.19
0.21
Baseline 15m 40m
Salivary Cortisol (ug/ml)
Figure 3.5: Cortisol levels collapsed across phase
Stress
Control
154

0.05
0.07
0.09
0.11
0.13
0.15
0.17
Stress Control
Salivary Cortisol (ug/ml)
Figure 3.6: Cortisol levels during the EF phase:
Collapsed across time points
155

0.05
0.07
0.09
0.11
0.13
0.15
0.17
0.19
Baseline 15m
Salivary Cortisol (ug/ml)
Figure 3.7: Cortisol levels during the EF phase:
Baseline versus 15m post onset
Stress
Control
156

0.05
0.07
0.09
0.11
0.13
0.15
0.17
Cold Warm
Salivary Cortisol (ug/ml)
Figure 3.8: Cortisol levels during the LF phase:
Collapsed across time points
157

0
0.05
0.1
0.15
0.2
0.25
Baseline 15m
Salivary Cortisol (ug/ml)
Figure 3.9: Cortisol levels during the LF phase:
Baseline versus 15m post onset
Stress
Control
158

0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
EF LF
Proportion of pictures correctly recalled
Figure 3.10: Picture recall across stress and valence:
Emotional versus neutral pictures
159

0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
EF LF
Proportion of pictures correctly recalled
Figure 3.11: Picture recall across stress and valence:
Negative versus neutral pictures
160

0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
Positive Neutral Positive Neutral
EF LF
Proportion of pictures correctly recalled
Figure 3.12: Picture recall across stress: Positive
versus neutral pictures
161

0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
Negative Positive Negative Positive
EF LF
Proportion of pictures correctly recalled
Figure 3.13: Picture recall across stress: Negative
versus positive pictures
162

0.5
0.55
0.6
0.65
0.7
0.75
0.8
0.85
0.9
Emotional Neutral Emotional Neutral
Stress Control
Proportion of picture locations correctly
remembered
Figure 3.14: Picture location memory across phase:
Emotional versus neutral pictures
163

0.5
0.55
0.6
0.65
0.7
0.75
0.8
0.85
0.9
EF LF
Proportion of picture locations correctly
remembered
Figure 3.15: Picture location memory across stress and
valence: Negative versus neutral pictures
164

0.5
0.55
0.6
0.65
0.7
0.75
0.8
0.85
0.9
Negative Neutral Negative Neutral
Cold Warm
Proportion of picture locations correctly
remembered
Figure 3.16: Picture location memory across phase:
Negative versus neutral pictures
165

0
0.05
0.1
0.15
0.2
0.25
baseline 15m
Salivary Cortisol ug/ml)
Figure 3.17: Cortisol levels between Responders and
Nonresponders during the EF phase: Stress versus
control sessions
Nonresponders
Stress
Nonresponders
Control
Responders Stress
Responders Control
166

0
0.05
0.1
0.15
0.2
0.25
0.3
baseline 15m
Salivary Cortisol ug/ml)
Figure 3.18: Cortisol levels between Responders and
Nonresponders during the LF phase: Stress versus
control sessions
Nonresponders
Stress
Nonresponders
Control
Responders Stress
Responders Control
167

0
0.05
0.1
0.15
0.2
0.25
0.3
baseline 15m
Salivary Cortisol (ug/ml)
Figure 3.19: Cortisol levels in Responders and
Nonresponders during the stress session in the EF
phase
Nonresponders
Responders
168

0
0.05
0.1
0.15
0.2
0.25
0.3
baseline 15m
Salivary Cortisol (ug/ml)
Figure 3.20: Cortisol levels in Responders and
Nonresponders during the stress session in the LF
phase
Nonresponders
Responders
169

0
0.05
0.1
0.15
0.2
0.25
0.3
baseline 15m
Salivary Cortisol (ug/ml)
Figure 3.21: Cortisol levels in Responders in the EF
phase: Stress versus control sessions
Stress
Control
170

0
0.05
0.1
0.15
0.2
0.25
0.3
baseline 15m
Salivary Cortisol (ug/ml)
Figure 3.22: Cortisol levels in Responders in the LF
phase: Stress versus control sessions
Stress
Control
171

0.6
0.65
0.7
0.75
0.8
0.85
Control Stress
Proportion of words correctly recalled
Figure 3.23: Working memory performance in the EF
phase: Stress versus control session
Nonresponders
Responders
172

0.5
0.55
0.6
0.65
0.7
0.75
0.8
0.85
Control Stress
Proportion of words correctly recalled
Figure 3.24: Working memory performance in
responders during the EF phase
173

0.5
0.55
0.6
0.65
0.7
0.75
0.8
0.85
Control Stress
Proportion of words correctly recalled
Figure 3.25: Working memory performance in
responders during the LF phase
174

0.5
0.55
0.6
0.65
0.7
0.75
0.8
0.85
0.9
0.95
Stress Control Stress Control
Nonresponders Responders
Proportion of picture locations correctly
remembered
Figure 3.26: Picture location memory in the LF phase:
Positive versus neutral pictures

Abstract (if available)

Abstract Estradiol and the class of stress hormones called glucocorticoids exert contrasting effects on various systems throughout the body, including neural tissue and cognition. Evidence also exists showing that estradiol can mitigate the damaging effects of excessive glucocorticoid levels on neural tissue. Given the sharp decline in estradiol levels that characterize the menopausal transition in human females, it is important to understand if the loss of estradiol leaves post‐menopausal women at a higher risk of the negative effects of stress on neural tissue, and by extension the negative effects of stress on cognition. ❧ Studies do show that estradiol treatment after menopause can dampen the physiological stress response to a stressful event

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Cognitive functioning following ovarian removal before or after natural menopause

PDF

The neural basis of estradiol conditioned taste aversion learning and estradiol anorexia

PDF

How emotional arousal influences memory and learning in younger and older adults

PDF

Age-related changes in cognitive functioning and their implications for the reacquisition of conditioned taste avoidances

PDF

Emotion, attention and cognitive aging: the effects of emotional arousal on subsequent visual processing

PDF

Mechanisms of stress effects on learning and decision making in younger and older adults

PDF

Optimism when imagining the future in old age influences the positivity effect in memory

PDF

Influence of age and anxiety on recognition of facial expressions of emotion: exploring the role of attentional processes

PDF

Effect of estradiol on circulating levels of inflammatory cytokines in postmenopausal women

PDF

Adolescent life stress and the cortisol awakening response: the moderating roles of emotion regulation, attachment, and gender

PDF

Dedifferentiation of emotion regulation strategies in the aging brain: an MVPA investigation

PDF

The role of inflammation in mediating effects of obesity on Alzheimer's disease

PDF

The effects of curiosity-evoking events on consumption enjoyment

PDF

Age-related androgen depletion and the development of Alzheimer's disease

PDF

Do you see what I see? Personality and cognitive factors affecting theory of mind or perspective taking

PDF

Noradrenergic mechanisms of arousal-enhanced memory selectivity

PDF

Reacting to the negative consequences of your decisions: an ATSS comparison of cognitive and affective reactions in older versus younger adults

PDF

An investigation of factors associated with reliability and validity of family history reports of alcohol‐related problems

PDF

Heart, brain, and breath: studies on the neuromodulation of interoceptive systems

PDF

A functional use of response time data in cognitive assessment

Asset Metadata

Creator Ycaza, Alexandra E. (author)

Core Title Effects of estradiol on cortisol response, working memory, and emotional memory during stress in young and post-menopausal women

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Psychology

Publication Date 04/14/2014

Defense Date 03/04/2014

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

Tag cortisol,estradiol,Estrogen,Menopause,Menstrual Cycle,OAI-PMH Harvest,Stress,working memory

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Mather, Mara (committee chair), Gatz, Margaret (committee member), Pike, Christian J. (committee member), Walsh, David A. (committee member)

Creator Email alexycaza@gmail.com,ycaza@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-378430

Unique identifier UC11297666

Identifier etd-YcazaAlexa-2351.pdf (filename),usctheses-c3-378430 (legacy record id)

Legacy Identifier etd-YcazaAlexa-2351.pdf

Dmrecord 378430

Document Type Dissertation

Format application/pdf (imt)

Rights Ycaza, Alexandra E.

Type texts

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

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

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA