Impairing and Enhancing Effects of Psychosocial Stress on Episodic Memory and Eyewitness Report

Impairing and enhancing effects of psychosocial stress on episodic memory and eyewitness report

Item Type text; Electronic Dissertation

Authors Hoscheidt, Siobhan M.

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 30/09/2021 19:47:55

Link to Item http://hdl.handle.net/10150/202762

IMPAIRING AND ENHANCING EFFECTS OF PSYCHOSOCIAL STRESS ON EPISODIC MEMORY AND EYEWITNESS REPORT

Siobhan Marie Hoscheidt

A Dissertation Submitted to the Faculty of the

Department of Psychology

In Partial Fulfillment of the Requirements For the Degree of

DOCTOR OF PHILOSOPHY

In the Graduate College

THE UNIVERSITY OF ARIZONA

2011 2

THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE

As members of the Dissertation Committee, we certify that we have read the dissertation prepared by Siobhan Marie Hoscheidt entitled IMPAIRING AND ENHANCING EFFECTS OF PSYCHOSOCIAL STRESS ON EPISODIC MEMORY AND EYEWITNESS REPORT and recommend that it be accepted as fulfilling the dissertation requirement for the

Degree of Doctor of Philosophy

______Date: 6/1/2011 Lynn Nadel, Ph.D.

______Date: 6/1/2011 Lee Ryan, Ph.D.

______Date: 6/1/2011 Jake Jacobs, Ph.D.

______Date: 6/1/2011 Katalin Gothard, Ph.D.

Final approval and acceptance of this dissertation is contingent upon the candidate’s submission of the final copies of the dissertation to the Graduate College.

I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement.

______Date: 6/1/2011 Dissertation Director: Lynn Nadel 3

STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgment of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the copyright holder.

SIGNED: Siobhan Marie Hoscheidt 4

ACKNOWLEDGEMENTS

I want to thank my major advisor Dr. Lynn Nadel for supporting my research endeavors financially and intellectually, the opportunities he has bestowed upon me, and his confidence in my abilities during my graduate school career. I want to thank Dr. Lee Ryan for her generous financial support and for fostering my intellectual abilities and research integrity. I want to give a special thanks to Dr. Jake Jacobs for being an inspirational researcher and teacher, and for being a lifelong mentor and dear friend. I want to thank Dr. Katalin Gothard for her intellectual contributions to my research on stress and for being an excellent role model for females in the world of science and academia. Last but not least, I want to thank Dr. Jessica Payne for having provided me with the opportunity to become a significant part of the research community at such as young age and for her guidance as a graduate student and friend.

Also, I would like to acknowledge my brother Ryan Hoscheidt and dear friends Chad Goebel, Bhaktee Dongaonkar and Elsa Baena for their emotional support, unabashed confidence in me, and contagious positive energy during the most challenging times of my graduate school career. 5

DEDICATION

To my parents who fostered in me the ambition to pursue my passions, the creativity to always find a way, the courage to forge ahead, the clarity to know the difference between impossible and improbable and the power of faith and humility. 6

TABLE OF CONTENTS

LIST OF FIGURES……………………………………………………………………….8

ABSTRACT……………………………………………………………………………...10

I. INTRODUCTION……………………………………………………………………..12 1.1 General Introduction………………………………………………………………12 1.2 The Physiological Stress Response………………………………………………..13 1.3 Cortisol Modulation of Amygdala and Hippocampal Plasticity…………………..15 1.4 Stress Modulation of Encoding Processes………………………………………...19 1.5 Stress Modulation of Retrieval Processes…………………………………………24 1.6 Basal Cortisol Levels, Emotional Memory, & Time of Day……………………...26 1.7 The Misinformation Effect………………………………………………………..27

II. RELEVENT RESEARCH……………………………………………………….…...31 2.1 Summary………………………………………………………………………...... 31

III. METHODOLOGICL ASPECTS OF CURRENT RESEARCH…………………….32 3.1 The Trier Social Stress Test……………………………………………………….32 3.1.1 Trier Social Stress Tests (TSST) Methodology………………………………32 3.2 Salivary Cortisol…………………………………………………………………..33 3.2.1 Saliva Sampling Device and Methods………………………………………..35 3.2.1.1 Sampling Device…………………………………………………………35 3.2.1.2 Sampling Methods……………………………………………………….35 3.2.2 Serum and Salivary Cortisol Correlations…………...…………………….....36 3.2.3 Salivary Cortisol Analysis……………………………………………………37 3.2.3.1 Preparation and Quality Control…………………………………………38 3.2.3.2 Materials and Equipment………………………………………………...39 3.2.3.3 Analysis Protocol………………………………………………………...40

IV. EXPERMENT I: MORNING STRESS SIGNIFICANLTY REDUCED THE EMOTIONAL MEMORY ENAHCNEMENT EFFECT WHEN INDUCED AT ENCODING……………………………………………………………………………...42 4.1 Research Aims & Hypotheses…………………………………………………….42 4.1.1 Method……………………………………………………………………..…43 4.1.1.1 Participants…………………………………………………………….…43 4.1.1.2 Stimulus Materials…………………………………………………….…43 4.1.1.3 Procedure………………………………………………………………...44 4.1.2 Results………………………………………………………………………...46 4.1.2.1 Subjective Anxiety Scores……………………………………………….47 4.1.2.2 Salivary Cortisol…………………………………………………………48 4.1.2.3 Recognition Memory…………………………………………………….50

6 7

TABLE OF CONTENTS- Continued

4.1.3 Discussion………………………………………………………………….……53

V. EXPERMENT II: THE SUSCEPTIBIITY OF EPISODIC MEMORY TO MISINFOMRATION WHEN ENCODED UNDER STRESS……………………….…58 5.1 Research Aims & Hypotheses………………………………………………….…58 5.1.1 Method……………………………………………………………………..…59 5.1.1.1 Participants…………………………………………………………….…59 5.1.1.2 Stimulus Materials………………………………………………….……59 5.1.1.3 Procedure……………………………………………………………...…59 5.1.2 Results…………………………………………………………………...……62 5.1.2.1 Subjective Anxiety Scores………………………………………….……62 5.1.2.2 Salivary Cortisol…………………………………………………………64 5.1.2.3 Recognition Memory……………………………………………….……66 5.1.3 Discussion……………………………………………………….……………74

VI. GENRAL DISCUSSION……………………………………………………………77

REFERENCES…………………………………………………………………………..80

LIST OF FIGURES

Figure 1. Mean subjective anxiety ratings for visit one...... 48

Figure 2. Mean subjective anxiety ratings for visit two………………………..………...48

Figure 3. Mean salivary cortisol for visit one……………………………………………49

Figure 4. Mean salivary cortisol for visit two…………………………………...….……49

Figure 5. Mean baseline salivary cortisol for early and late morning groups……………50

Figure 6. Mean corrected recognition scores across groups for neutral and emotional phases of slideshow……………………………..………………………………………..51

Figure 7. Mean corrected recognition scores for emotional and neutral phases of the neutral and emotional slideshows for early morning participants…………………….…52

Figure 8. Mean corrected recognition scores for emotional and neutral phases of the neutral and emotional slideshows for late morning participants………………………...52

Figure 9. Mean subjective anxiety ratings across experimental time points for stress and control groups for session one…………………………………………………………...62

Figure 10. Mean subjective anxiety ratings across experimental time points for stress and control groups for session two…………………………………………………………...63

Figure 11. Mean subjective anxiety ratings across experimental time points for stress and control groups for session three………………………………………………………….63

Figure 12. Mean salivary cortisol for visit one………………………………………..…64

Figure 13. Mean salivary cortisol for visit two……………………………………..……65

Figure 14. Mean salivary cortisol for visit three……...... 65

Figure 15. Negative correlation between subjective anxiety ratings during slideshow presentation (minus baseline) and misinformation hits for the emotional phase of the emotional narration slideshow…………………………………………………………...66

Figure 16. Positive correlation between subjective anxiety ratings during slideshow presentation (minus baseline) and correct hits for misinformed items for the emotional phase of the emotional narration slideshow……………………………………………...67

8 9

LIST OF FIGURES- Continued

Figure 17. Mean hits across the three phases of the slideshow for stress and control, no misinformation, emotional narration groups…………………………………………...... 70

Figure 18. Mean hits across the three phases of the slideshow for stress and control, misinformation, emotional narration groups………………………………………...…...70

Figure 19. Mean hits across the three phases of the slideshow for stress and control, no misinformation, neutral narration groups………………………………………………..71

Figure 20. Mean hits across the three phases of the slideshow for stress and control, misinformation, neutral narration groups………………………………………………..71

Figure 21. Narration type by group interaction………………………………………….72

Figure 22. Main Effect of misinformation……………………………………………….73

ABSTRACT

Research conducted over the past two decades has contributed a wealth of new knowledge to the field’s understanding of stress effects on memory. It has been widely demonstrated that stress can either facilitate or impair memory, depending on 1) the phase of memory processing influenced by stress hormones and 2) the valence or arousing nature of the encoded information. It has also been reported that, when stress levels are significantly elevated at encoding, emotional memory is preserved (or enhanced) while memory for non-emotional information is impaired. These effects have been discussed at the neurobiological level with respect to the stress hormone, cortisol, and the impairing and facilitating modulatory effects it has on regions of the brain involved in emotional learning and memory. Whether diurnal shifts in basal levels of cortisol modulate these effects remains unknown. Additionally, it remains unknown whether enhancing and impairing effects of stress on memory result in memory traces that are more or less open to alteration by subsequent experiences, such as observed in the so-called “misinformation” effect.

The current dissertation aimed to investigate the effects of stress on encoding of thematically negatively arousing and non-emotional events, composed of negatively arousing and neutral stimuli. Our goal in using more complex materials, in lieu of stimuli

(e.g. word lists, images) traditionally used in studies of emotion and memory, was to examine the effects of stress on encoding of information more representative of a real- world event. Within this framework we examined 1) the effects of basal cortisol levels on

10 11

stress modulation of memory encoding, and 2) the integration of subsequent misinformation on memory for negatively arousing versus non-arousing events encoded under stress. The research included in this dissertation aims to further the field’s current understanding of the effects of stress on memory processes. Findings are relevant to the literature on traumatic memory, eyewitness testimony, and the effects of moderate to severe emotion on long-term episodic memory.

I. INTRODUCTION

1.1 General Introduction

When asked to recall past experiences, individuals are more likely to recollect emotionally laden than non-emotional events. Recent and remote emotional memories are typically recalled with relative ease and reported in vivid detail (Denburg et al., 2003;

Kensinger & Schacter, 2006; Mickley & Kensinger, 2009; Sharot, et al., 2007; Talarico,

Berntsen, & Rubin, 2009; Talarico, LaBar, & Rubin, 2004). By comparison, non- emotional memories tend to require effortful recollection and typically lack detailed information when reported. Fast and efficient recollection of emotional information serves an important adaptive function, providing a means by which memory for past emotional events can be used to inform future behaviors (Koster et al., 2007; Nairne,

2007). Over the past two decades, numerous studies have demonstrated memory enhancement for emotional, compared to neutral, information particularly when encoded just immediately prior to or after stress induction (Anderson et al., 2006; Harris &

Pashler, 2005; Kensinger, Garoff-Eaton, & Schacter, 2007b; Payne et al., 2006, 2007;

Segal & Cahill, 2009; Touryan, Marian, & Shimamura, 2007). The effects of stress on memory are not always facilitatory, however. Several studies have demonstrated that while memory for emotional information is enhanced when encoded under stress, memory for neutral information can be impaired (Payne et al., 2006; 2007). Additionally, research examining the effects of stress on memory retrieval has shown that stress induced just prior to memory testing results in impaired memory for emotional compared to neutral information. Differential effects of stress on encoding versus retrieval

12 13

processes and memory for emotional versus neutral information suggests that modulatory effects of stress on memory processes vary as a function of 1) the emotional valence or arousing nature of the relevant stimuli or event and 2) the phase of memory processing

(i.e. attention/encoding, consolidation, or retrieval) affected by stress and stress hormones.

1.2 The Physiological Stress Response

When an organism faces emotional distress or is physically challenged the autonomic nervous system, a subdivision of the sympathetic nervous system, is automatically activated. Once activated, a cascade of physiological changes occurs that better enables an organism to confront (i.e. fight, freeze) or escape (i.e. flee) danger. The term “stress” applies to the condition under which the autonomic nervous system is activated and stress hormones are released. Autonomic arousal is partially mediated through activation of the central medial nucleus of the amygdala (CeM), which provides outputs to brainstem and hypothalamic nuclei. Activation of these nuclei elicits changes in the central nervous system that affect the brain, such as release of norepinephrine from the locus coeruleus (a nucleus located in the rostral pons that provides the primary source of noradrenergic innervation to the cerebrum) and release of acetylcholine, from the nucleus basalis. While norepinephrine and acetylcholine promote vasoconstriction in the peripheral vasculature, their effects on neurons in the brain are quite different. Norepinephrine modifies the resting membrane potential of neurons throughout the cortex, causing a positive shift in these neurons. When the resting potential is more positive, neurons are more easily

depolarized and activated when excited by inputs. Acetylcholine release also modifies neuronal functioning in cortical regions, eliciting increased coherence amongst regions.

Such coherence amongst cortical regions is believed to play an important role in the consolidation of information. For the purpose of this discussion, the most important system that is activated under stress is the hypothalamic-pituitary-adrenal (HPA) axis.

This axis is composed of a collection of nuclei and organs that release stress hormones.

The end result of HPA activation is release of the stress hormone cortisol by the adrenal gland.

Once released from the adrenal gland cortisol circulates through the bloodstream and crosses the blood-brain barrier. In the brain cortisol binds to two types of corticosteroid receptors, mineralocorticoid receptors (MRs) and glucocorticoid receptors (GRs). Two medial temporal lobe structures that play integral roles in memory and emotional learning, the hippocampus and amygdala, possess an abundance of both of these stress receptors. Animal and neuroimaging studies have provided evidence that activation of

MRs and GRs by high levels of cortisol modulates normal learning within the amygdala and hippocampus quite differently, facilitating neural plasticity in the former and impairing neural plasticity in the latter (Joëls & de Kloet, 1990; 1991; Joëls & de Kloet,

1996; Kim & Diamond, 2002; Vyas et al., 2002; Pruessner et al., 2008; Henckens et al.,

2009).

14 15

1.3 Cortisol Modulation of Amygdala and Hippocampal Plasticity

Animal and neuroimaging studies have shed light on a number of mechanisms by which cortisol modules neuronal plasticity and functional activation within the hippocampus and amygdala. By and large, these studies have provided evidence that glucocorticoids impair plasticity and decrease functional activation in the hippocampus and facilitate plasticity and increase functional activation in the amygdala.

Cortisol modulates both intracellular and extracellular neural processes in the amygdala, making neurons in these regions more excitable. Research has demonstrated that high levels of cortisol facilitate neuronal excitability and plasticity in the amygdala by shifting the resting membrane potential of neurons in this region positively, enhancing output resistance and decreasing the amplitude of inhibitory post-synaptic potentials of GABA-

A receptors (Duvarci & Paré, 2007). Research has shown that the magnitude of these changes is directly proportional to cortisol dose. This was demonstrated in a study conducted by Duvarci and Paré (2007) who showed that the magnitude of facilitatory effects of cortisol on basolateral amygdala (BLA) neurons was dependent on whether a

1nM or 100nM cortisol dose was administered, in vitro, to cells. Cells treated with

100nM of cortisol showed more robust neural excitability. These results demonstrate that plasticity in BLA principle cells is significantly enhanced by cortisol, larger doses of cortisol resulting in greater facilitation of synaptic plasticity. The functional implication of these effects seems clear: learning and memory can be facilitated by the amygdala when stress levels are high.

While in vitro studies have provided evidence that cortisol facilitates excitability of BLA neurons, these studies do not speak to how cortisol-driven neuronal changes affect emotional learning and memory. Behavioral studies, in animals, have provided evidence that the influence of cortisol alone does not lead to enhanced emotional learning and that the presence of noradrenergic activation is necessary. Studies have shown that facilitating effects of cortisol on BLA plasticity are modulated by, and perhaps even dependent upon, arousal-induced noradrenergic activation within the BLA. Blockade of beta- adrenoceptors in the BLA prevents the memory enhancing effects of cortisol on amygdala function (McGaugh & Roozendaal, 2002; Roozendaal et al., 2006b). The implications of these results seem clear: emotional arousal and stress must co-occur for subsequent memory for emotional materials to be enhanced.

Neuroimaging studies have demonstrated that amygdala activation is modulated by both stress and arousal in humans, providing evidence that arousing materials as well as increases in cortisol elicit increases in amygdala activation and emotional learning. A neuroimaging study conducted by Canli et al., (2000) demonstrated that amygdala activation is modulated by subjective arousal and that increased activation predicted subsequent memory performance. Results showed that negative emotional scenes, subjectively rated as highly arousing, elicited greater increases in amygdala activation and the amount of activation correlated with better subsequent memory of negative images (Canli et al., 2000). While this study examined the effects of arousal specifically,

16 17

other studies have demonstrated that cortisol and noradrenergic activation interact synergistically to increase functional activation observed in the amygdala and enhance subsequent memory for emotional materials. Van Stegeren et al. (2007) demonstrated this in a study that compared amygdala activation and subsequent memory for emotional information across high and low cortisol responders treated with a placebo or a betablocker (i.e. propranolol). Results showed a significant interaction effect of endogenous cortisol level with increasing activation in the amygdala in the placebo, but not the betablocker, group. Overall, these data suggest that noradrenergic activation and cortisol levels modulate amygdala activation and subsequent memory for negative emotional materials.

While noradrenergic activation and cortisol appear to predominantly facilitate neural plasticity and learning in the amygdala, cortisol modulates hippocampal plasticity differentially depending on the amount of cortisol-bound stress receptors. Cortisol binds first to mineralocorticoid receptors (MRs), which have a high affinity for cortisol.

Conditions under which MRs become largely saturated results in unbound cortisol binding to the low affinity glucocorticoid receptors (GRs). In vitro studies, examining cortisol modulation of CA1 pyramidal cells of the hippocampus, have demonstrated that the binding of cortisol to these two receptors results in opposing effects. In conditions where MRs are predominantly cortisol bound, hippocampal neurons show increased excitability and long-term potentiation (LTP; the cellular mechanism shown to underlie learning). By contrast, when MRs are saturated and GRs become bound by cortisol,

neuronal activity significantly declines, LTP is impaired and long-term depression (LTD) is enhanced. These studies suggest that activation of MRs mediates facilitation of neural plasticity in the hippocampus, while activation of GRs has the opposite effect. The relationship between cortisol levels and hippocampal-dependent learning has been traditionally described as an inverted-U function (Yerkes & Dodson, 1908) where low and high cortisol levels result in poor hippocampal-dependent learning (lack of MR activation; activation of GRs) and moderate cortisol levels result in optimal hippocampal- dependent learning (predominant MR activation). Neuroimaging studies have provided evidence for modulated hippocampal activation in humans under acute conditions of stress. A neuroimaging study conducted by Pruessner et al., (2008) revealed that acute stress resulted in significant hippocampal deactivation and that the degree of hippocampal deactivation significantly correlated with the cortisol stress response.

Overall, results from in vitro and neuroimaging studies demonstrate that emotional arousal and cortisol modulate activation in the amygdala and the hippocampus, high cortisol levels facilitating plasticity and functional activation in the amygdala while impairing plasticity and functional activation in the hippocampus. Dissociative effects of cortisol on these medial temporal lobe regions is believed to be the primary neurobiologically-defined basis by which memory for emotional information is enhanced and non-emotional memory is impaired when encoded under conditions of acute stress. It is also hypothesized that dissociative effects of cortisol on these regions mediates differential effects observed when stress is induced just prior to encoding compared to at

18 19

the time of retrieval.

1.4 Stress Modulation of Encoding Processes

Negative emotional material is typically remembered better than positive and non- emotional information. Enhanced memory for emotional compared to non-emotional material is particularly robust when stress is induced just prior to encoding of a negatively arousing event or stimulus (de Quervain et al., 2009, review; Payne et al.,

2006; 2007). Activation of the amygala and arousal-induced release of noradrenergic and stress hormones produce ideal neurophysiologic conditions for optimal encoding of aversive stimuli (Brosch, Pourtois, & Sander, 2010; Compton, 2003; Kensinger &

Corkin, 2004; Ohman, 2005; Pessoa & Ungerleider, 2004; Talmi et al., 2008) that typically results in emotional information having privileged access, over positive and non-emotional information, to working memory and subsequent memory systems (see review by Mather, 2007).

Encoding of negative information begins with spontaneous biasing of perception and attention towards negative stimuli, often referred to as “attention-narrowing” (Alpers,

2008; Calvo & Lang, 2005; Laney, Heuer, & Reisberg, 2003; Nobata, Hakoda, & Ninose,

2010; Nummenmaa, Hyönä, & Calvo, 2006). Attention-narrowing is mediated in part by amygdala innervation of low-level visual areas (Adolphs, 2004) which occurs well before conscious awareness of stimulus perception (i.e. 100ms to 300ms after stimulus exposure) (Bradley et al., 2007; Huang & Luo, 2006; Schaefer, Pottage, & Rickart,

2011). Studies conducted on amygdala patients provide evidence for the essential role of the amygdala in attention-narrowing and encoding of aversive stimuli. Amygdala patients show minimal physiological arousal (Bernston et al., 2007) and unbiased visual attention when encoding negative emotional materials (Adolphs et al., 2005; Anderson & Phelps,

2001). These deficits may explain, in part, why memory for emotional information is typically not enhanced in amygdala patients as they are in intact individuals.

Pharmacological studies have demonstrated similar results in healthy individuals with administration of the betablockeer, propranolol, just prior to acquisition. Participants in the betablocker group showed decreased functional activation in the amygdala during encoding of aversive stimuli and subsequent memory impairment for emotional information (Strange & Dolan, 2004; van Stegeren et al., 2005).

Enhancement of memory for emotional information, at the expense of non-emotional and positive materials, is often referred to as emotion-induced memory trade-off (see

Buchanan & Adolphs, 2002; Kensinger, 2009; Reisberg & Heuer, 2004 for reviews).

Memory trade-offs are particularly robust when negative information is presented within or in close proximity to non-emotional or positive information (Kensinger, Garoff-Eaton,

& Schacter, 2007b; Touryan, Marian, & Shimamura, 2007). More recent research has demonstrated, however, that emotion-induced memory trade-offs can be reduced, and in some cases eliminated, when non-emotional and emotional materials are spatially (e.g. the font color or spatial location of an emotional word; Kensinger, Garoff-Eaton, &

Schacter, 2007a), temporally or conceptually (Adolphs, Denburg, & Tranel, 2001;

20 21

Schmidt, 2002) integrated or are equally as important to obtain a goal (Adolphs,

Denburg, & Tranel, 2001; Compton, 2003; Gable & Harmon-Jones, 2008; Levine &

Edelstein, 2009, review; Levine & Pizarro, 2004). Explicit encoding instructions have also been shown to modulate memory tradeoff effects. For example, Kensinger et al.

(2005) showed that participants given instructions to remember all components of an emotional scene were significantly more likely to later recall both emotional and non- emotional components of stimuli compared with participants who were not provided with instructions. Similar results have been observed in studies were participants were told that the goal of the encoding task was to later describe the scene to an individual in detail such that it could be accurately recreated. These findings suggest that, although attention and perception may automatically shift towards emotional components of an event, that these effects may be overridden by conscious control of attention processes or automatic integration of emotional and non-emotional stimuli.

The nature of the arousing component of an experience has been argued to play an integral role in automatic allocation of attention and subsequent memory performance.

Research has shown that when non-emotional aspects of an event or stimulus are encoded in a context that is thematically arousing, or under conditions of empathy, memory is enhanced overall with no memory narrowing effects (Laney, Heuer, & Reisberg, 2003;

Laney et al., 2004). In a meta-analytic review Deffenbacher et al. (2004) illustrated that attention is modulated differentially by the nature of the arousing experience. In his meta- analytic review, Deffenbacher (2004) argues that arousal can evoke one of two responses,

an orienting response in which attention is restricted to a particular stimulus (i.e. arousal mode of attention) or an orienting response in which attention is allocated to a motivational component of a stimulus (i.e. activation mode of attention control).

Deffenbacher (2004) defines the first as an arousal state in which it is more likely that an individual will allocate attention across various aspects of a stimulus, compared to the latter case, in which attention is likely restricted to a specific feature of a stimulus. These results suggest, that although emotion-induced memory trade-offs are typically robust, particularly in incidental encoding tasks, explicit encoding strategies and the nature of the arousal induction mediate these effects (Deffenbacher et al., 2004). These effects have also been shown to vary depending on individual differences. For example, levels of subjective anxiety, working memory capacity and executive function abilities correlate with more prominent emotion-induced memory trade-offs compared to individuals without these traits (Waring et al., 2010).

Research suggests that both arousal and physiological stress are necessary components of the emotional memory enhanced effect, the occurrence of one or the other in isolation not leading to memory enhancing effects (Roozendaal et al., 2006a; 2006b; Segal & Cahill,

2009). Behavioral studies have provided evidence for the critical interaction between stress and arousal during encoding of emotional stimuli by examining memory for arousing versus non-arousing episodes, using identical stimuli across arousal and non- arousal groups. These studies have demonstrated that non-arousal stress groups showed impaired subsequent memory while, by contrast, the arousal stress group showed

22 23

enhanced memory, particularly for the emotional components of the slideshow (Payne et al., 2007). Pharmacological studies that have examined independent effects of cortisol and norepinephrine on memory have provided further evidence for the notion that both hormones are critical components of the emotional memory enhancement effect. The majority of these studies have demonstrated that administration of a noradrenergic antagonist just prior to encoding emotionally arousing stimuli impairs long-term memory for emotional materials (Maheu et al., 2004; van Steregen et al., 2005; 2007).

Neuroimaging studies provide evidence that blockade of norepinephrine release correlates with significant decreased activation in the amygdala, even under conditions of stress (van Steregen et al., 2005; 2007). Clearly, cortisol and norepinephrine interact in important ways to produce memory enhancement effects in the amygdala during encoding.

The hippocampus also plays an integral role in the encoding of emotional events and stimuli. Although the binding of the stress hormone cortisol decreases hippocampal function overall, under conditions of stress and arousal, hippocampal activation may be facilitated via interactions with the amygdala (Dolcos, LaBar, & Cabeza, 2004;

Richardson, Strange, & Dolan, 2004). Some studies suggest that the amygdala and other regions involved in affective attentional processes (i.e. orbitofrontal cortex, ventral striatum, and anterior cingulated gyrus) modulate encoding processes in the hippocampus to ensure that salient information is attended to, encoded and stored in a declarative memory system for future recollection (Kensinger, Garoff-Eaton, & Schacter, 2007a).

Additionally, similar to the amygdala, norepinephrine has been shown to modulate activation in the hippocampus under stress. It is postulated that norepinephrine-driven modulation of hippocampal function allows for preferential encoding of emotionally relevant information (van Stegeren et al., 2010). A small subset of animal studies have demonstrated that an intact hippocampus is necessary for modulation of learning by stress, showing that hippocampal lesions prevent stress-induced enhancements, as well as impairments, of learning after stress (Bangasser & Shors, 2007). Research conducted with MTL patients has demonstrated that the extent of impaired memory for unpleasant events corresponds to whether damage is confined to the hippocampus or extents into surrounding cortical areas. The latter case has been shown to yield more severe impairment of negative autobiographical memories (Adolphs et al., 2005). Additionally, right- compared to left-brain MTL damaged patients show differential performance on episodic memory tasks. Left-sided anteriormedial temporal lobe damage patients demonstrate severely impaired recollection of unpleasant memories compared to right temporal lobectomy patients who perform similarly to intact individuals (Buchanan,

Tranel & Adolphs, 2006). More research is needed to determine factors that may modulate the effects of MTL or amygdala damage on memory for negative emotional episodic events such as whether or not memories that can be retrieved from patients were acquired pre or post MTL or amygdala injury.

1.5 Stress Modulation of Retrieval Processes

A number of studies have demonstrated deleterious effects of exogenous and endogenous

24 25

increases in cortisol levels just prior to memory testing (de Quervain et al., 2009;

Kuhlmann, Piel, & Wolf, 2005; Merz, Wolf, & Hennig, 2010; Tollenaar et al., 2008b).

Impairing effects of stress at retrieval appear to be robust, generalizing across various experimental materials with varying levels of valence and arousal (e.g. words (Domes et al., 2004; Kuhlmann, Piel, & Wolf, 2005), word pair associations (Tollenaar et al., 2008a;

2008b), slideshows (Buchanan & Tranel, 2008), and socially relevant information (Merz,

Wolf, & Hennig, 2010)) and memories for past experienced events (Tollenaar et al.,

2009a; 2009b). Decreased hippocampal function, caused by the binding of cortisol to

GRs, is believed to be the underlying mechanism by which retrieval processes are hindered by stress. Neuroimaging studies have provided evidence for this notion, demonstrating that hippocampal activation is significantly decreased under stress compared to non-stressed controls (Dedovic et al., 2005; Wang et al., 2005). A number of these studies have provided evidence that subsequently impaired memory retrieval significantly correlates with decreased hippocampal activation and the cortisol stress response (Dedovic et al., 2009; Pruessner et al., 2008).

While the effects of cortisol on hippocampal function are largely deleterious, whether or not noradrenergic hormones modulate cortisol effects on hippocampal function, as in the amygdala, remains unknown. Some research suggests that effects of cortisol on retrieval processes is mediated by noradrenergic hormones by demonstrating that exogenous cortisol administration prior to testing does not affect retrieval of non-arousing materials

(Kuhlmann, Kirschbaum, & Wolf, 2005) or materials remembered in a non-arousing test

situation (Kuhlmann & Wolf, 2006b). Impaired retrieval during stress and arousal may serve adaptive purposes allowing for salient information to be encoded efficiently without congruent information being simultaneously retrieved. Additionally, impaired retrieval may allow stressful events to remain salient and distinct by restricting newly encoded emotional information from being integrated with previously experienced events

(Roozendaal, Barsegyan, & Lee, 2008). Retrieval impairments are usually temporary with acute stress, detrimental effects of cortisol on memory retrieval processes wearing off once high cortisol levels have abated (de Quervain, Aerni, & Roozendaal, 2007).

1.6 Basal Cortisol Levels, Emotional Memory, & Time of Day

Aside from the issue of memory phase, time of day has also been shown to play a modulatory role in the effects of stress on memory performance. Basal levels of glucocorticoids follow a diurnal cycle, being highest in the morning (30-60mins post wake) and lowest in the early evening (Maheu et al., 2005). As a consequence, the impact of stress, and the release of cortisol triggered by stress, depends upon levels of endogenous hormone, which in turn depend upon time of day. Stress induced in the morning, when basal glucocorticoid levels are high, may push glucocorticoid levels past an upper threshold, causing disrupted plasticity within regions such as the hippocampus.

By contrast, stress induced in the afternoon, when basal glucocorticoid levels are low, may increase glucocorticoid levels moderately, facilitating plasticity across regions (i.e. hippocampus and amygdala). To our knowledge, only one study has examined the behavioral effects of stress induced in the morning versus the afternoon in a single

26 27

experimental paradigm (Mahua et al., 2005). Results showed that stress induced in the morning just prior to encoding led to subsequent memory impairments for emotional information, specifically. By contrast, the expected emotional memory enhancement effect was observed in the afternoon stress group. A meta-analysis conducted by Het,

Ramlow, & Wolf (2005) support these findings, suggesting that mixed results from stress and memory studies may be attributable, in part, to stress induction occurring at different times of day across experiments. Overall, stress studies conducted in the morning hours showed memory impairments (Fehm-Wolfsdorf et al., 1993; Hsu et al., 2003 Kirshbaum et al., 1996; Lupien et al., 1999) while studies conducted in the afternoon showed emotional memory enhancements (Abercrombie et al., 2003; Buchanan & Lovallo, 2001; de Quervain et al., 2000). These data suggest that differences in baseline cortisol levels, from morning to afternoon, significantly alter the effects of stress on memory performance. To date, no one has examined the effects of stress induced in the morning and afternoon on different memory phases for emotional compared to neutral stimuli.

1.7 The Misinformation Effect

The misinformation effect, first coined by Elizabeth Loftus in the early 1970’s, refers to the phenomenon that episodic memory is susceptible to the integration of subsequently provided false information (Loftus & Palmer, 1974; Loftus, 1975; Loftus & Hoffman,

1989). The classic method used to examine the misinformation effect entails individuals viewing a slideshow of a car running a stop sign and hitting a pedestrian. Sometime after slideshow acquisition participants are questioned about the event. A subset of participants, randomly assigned to the misinformation group, is exposed to false

information during the questioning session (e.g. “how fast was the car going when it ran the yield sign?”). Results from such studies have shown that participants who receive false information during questioning are significantly more likely to incorporate this false information into their memory recalls for the original episode during subsequent testing

(i.e. misinformed participants were significantly more likely to report that the car ran a yield sign instead of a stop sign).

Over the past two decades, hundreds of studies have been conducted on the misinformation effect (Cabeza et al., 2001; Cahill, McGaugh, & Weinberger, 2001;

Loftus, 1975; Loftus & Hoffman, 1989; Loftus & Palmer, 1974; Okado & Stark, 2005).

Many of these studies have examined factors that may modulate the incorporation of misinformation into memory by making memory more or less resistant to the integration of false information. This line of research has predominantly demonstrated that the integration of misinformation into episodic memory can be modulated by various factors such as the passage of time and the source of misinformation. For example, a study conducted by Loftus et al., 1978 showed that false information is less likely to be accepted into episodic memory storage when presented directly after encoding of the initial event. Results of this study were interpreted to mean that the passage of time results in the weakening of an event memory and this weakening increases susceptibility of a memory to the integration of false information.

A number of studies have demonstrated that the source of misinformation is also

28 29

important in determining whether misinformation will be remembered as a true aspect of that event by the rememberer (Loftus, 1993; Loftus and Pickrell, 1995). For example,

Loftus and Pickrell (1995) showed that false information was more likely to be integrated into memory for the original event if presented by an individual in a respected social position (e.g. doctor, policeman, fireman) or an individual who was considered credible to the person being misinformed (e.g. family member or close friend). Under some circumstances, where family was used to deliver false information, whole false events could be implanted, what has been called “rich false memories”.

Since its first description in the late 1970’s, the misinformation effect has been accepted into mainstream psychology as a robust and reliable effect that has received fairly minimal scrutiny. The general principle of this effect, that memory is susceptible to the integration of false information without conscious awareness, has been raised in U.S. courtrooms to challenge and contest the credibility of eyewitness testimony. To date, the misinformation effect has been examined in the context of non-arousing physiological conditions and non-arousing stimuli almost exclusively. Overall, results from this line of research suggest that misinformation is integrated into memory for events. This research, however, has failed to examine the effects of stress on memory, when induced at encoding, and the susceptibility of subsequent memory to integration of false information. Provided that the misinformation effect is modulated by the “strength” of the original memory trace, it is plausible that aspects of a memory enhanced by stress may be less susceptible to the integration of false information. Such a theory has

important implications for the application of the misinformation effect in court cases where eyewitnesses encoded an emotionally arousing event under acute stress.

30 31

II. RELEVANT RESERCH 2.1 Summary

Research conducted in the current dissertation was primarily based on the methodology and results of two studies previously conducted by our research group (Payne et al., 2006;

Payne et al., 2007). These studies were amongst the first in the area of stress and memory to examine the effects of psychosocial stress on encoding of complex negative arousing and non-arousing events composed of neutral and emotional visual information. Overall, these studies provided evidence that emotional and neutral information, and negative arousing and non-arousing events, are affected differentially when encoded under stress.

Subsequent memory for negative arousing events and information was enhanced when encoded under stress while memory for non-arousing events and neutral information was impaired. Enhancing and impairing effects of stress on emotional compared to neutral information are critical factors in examining the susceptibility of particular memories to the incorporation of false information. Currently, the predominant view is that episodic memory is susceptible to the integration of misinformation, overall, regardless of encoding conditions.

III. METHODOLGICAL ASPECTS OF CURRENT RESESARCH

3.1 The Trier Social Stress Test

The Trier Social Stress Test (TSST) is a well-established method developed in Trier,

Germany, that reliably induces a significant physiological stress response in laboratory settings (Kirschbaum et al., 1993, 1996; Kuhlmann, Piel, & Wolf, 2005). The TSST combines social evaluative threat with stressor uncontrollability, consisting of a 5 minute speech preparation period, followed by a 5-min speech given without notes (notes are abruptly taken away from participants just prior to speech) and a 5 minute surprise subtraction task. The TSST was used in the present dissertation studies as a stress induction manipulation.

3.1.1 Trier Social Stress Test (TSST) Methodology

On the initial visit, participants randomly assigned to the stress group were guided to a room in which they were told that they would soon give a speech in front of a panel of judges specially trained to analyze behavior. Video cameras, a microphone, judge’s chairs and a two-way mirror were explicitly pointed out to the participant while they were informed that a video and audio recording would be collected during their speech and judges would be present in front of and behind the two-way mirror to evaluate their verbal and non-verbal behavior. After a brief introduction to the task and room, participants were given 5 minutes to prepare a written speech. During this time interval the experimenter left the room. After 5 minutes, the experimenter returned with two

32 33

judges (one male, one female), took away the participant’s notes and asked them to proceed to the microphone to give their speech. Written notes were handed to one of the judges for evaluation during the speech. Two side stage lights, a video camera with direct feed into a television (on which the participant could see themselves being recorded) and microphone were turned on by the experimenter in preparation for the speech. After a brief microphone check, the participant was cued to begin their speech. If at any time the participant stopped talking for more than 30 seconds during the speech period they were told that they still had time left and were asked to continue. After the 5-minute speech task, participants were provided with instructions for a surprise subtraction task.

Participants were asked to subtract 17 from 1873 and to continue to subtract 17 from every subsequent answer, as fast and as accurately as they could, for 5-minutes.

Participants were instructed to verbalize their answers to the judges. If the participant provided an incorrect answer, they were asked to begin the math task again from 1873.

Completion of the math task marked the end of the stress induction. Participants were not given feedback regarding their performance during any of the experimental sessions.

3.2 Salivary Cortisol

In studies of stress and memory salivary cortisol is often collected as an objective measure of stress hormone changes (Abercrombie et al., 2003; Buchanan & Lovallo,

2001; Buchanan & Tranel, 2008; Cahill, Gorski, & Le, 2003; Dedovic et al., 2005;

Domes et al., 2004; Henckens et al., 2009; Kuhlmann & Wolf, 2006a; Kuhlmann,

Kirschbaum, & Wolf, 2005; Kuhlmann, Piel, & Wolf, 2005; Maheu et al., 2004; Marin,

Pilgrim, & Lupien, 2010; Payne et al., 2006; 2007; Tollenaar et al., 2008a). Cortisol is

released into the bloodstream, via the adrenal gland, where 85 to 95% binds to cortisol- binding globulin (CBG) and albumin (Galbois et al., 2010). The remaining 1 to 15% remains unbound (Robin, Predine, and Milgrom, 1978) and exits the circulatory system via intracellular mechanisms that allow it to be secreted through salivary glands in saliva.

Cortisol levels may be tested through collection of protein bound cortisol in the blood or may be tested through collection of unbound cortisol in saliva. Measuring salivary cortisol is typically preferred because it is less invasive, poses few health risks during sampling and testing, and doesn’t require phlebotomy training. Saliva samples are collected using various saliva collection devices, depending on the sample population

(i.e. adults versus young children) and are analyzed using enzyme-linked immunosorbent assaying (i.e. ELISA or EIA), a biochemical technique used to detect the presence of an antibody or antigen in a sample.

Salivary cortisol measures are unaffected by salivary flow rate (provided there is enough sample to be tested in duplicate) or salivary enzymes (Vining & McGinley, 1987).

Cortisol measures are affected, however, by various medications, particularly those that contain corticosteroids such as inhalers. Additionally, diseases or disorders of the hypothalamic pituitary adrenal (HPA) axis, such as Cushing’s disease, significantly affect cortisol levels. For this reason, participants in the current studies were screened for past and current medication use and diseases or disorders that would exclude them from a representative sample. In healthy individuals, salivary cortisol levels fluctuate as a function of the circadian rhythm (Dorn et al., 2007), reaching peak levels approximately

34 35

30 minutes after waking and steadily dropping throughout the day. Cortisol levels have been shown to rise independent of circadian rhythm in response to stress (Krieger, 1975) however, it is important to note that circadian rhythm determines basal cortisol levels, increases being relative to circadian rhythm modulated baselines. In the following dissertation studies, salivary cortisol was collected as an objective measure of physiological stress.

3.2.1 Saliva Sampling Device and Methods

3.2.1.1 Sampling Device

Saliva samples were collected using Salimetrics Oral Swabs (SOS) manufactured by

Salimetrics Inc. SOSs are made from inert polymer, shaped into 30 x 10 mm cylinders, and are stored in a capped, conical centrifuge tube with a separate tube insert that has a hole in its bottom to allow passage of saliva sample into the conical centrifuge tube during centrifuging. Polymer is preferable over cotton swabs, traditionally used for saliva collection, because it is highly absorbent but allows for maximal amount of sample to be extracted and yields greater cortisol recoveries. Additionally, polymer has been reliably shown to not interfere with salivary immunoassay results, a problem that has been demonstrated with the use of cotton-based sample collection methods (Granger et al.,

2007; Groschl & Rauh, 2006; Shirtcliff, Granger, Schwartz, & Curran, 2001).

3.2.1.2 Sampling Methods

In the following dissertation studies, cortisol was accessed through collection of whole saliva, which provides an index of bioavailable free cortisol levels. All samples were

collected during experimental sessions. At the beginning of each session, participants were required to rinse their mouth with water ten minutes prior to the first saliva collection. This was done to ensure that oral contaminants (e.g. food ruminants) were cleared out of the mouth before sampling. Participants were also instructed to remove any chapstick, lip-gloss or lipstick from their lips before the experiment. After rinsing, participants were asked not to eat or drink anything for the remainder of the session. At the time of sampling, participants were instructed to tip the conical centrifuge tube up to their mouth and, without touching it to their lips, let the swab fall into their mouth.

Participants were told to lightly chew on the swab for one minute, saturating it with as much saliva as possible. After one-minute had passed the experimenter instructed participants to spit the swab back into the tube, without touching the tube to their mouth.

Saliva samples were placed in a freezer for storage within ten minutes of sample collection to prevent bacterial growth, which can compromise assay validity (Whembolua et al., 2006). Samples remained frozen at --20 o C until thawed for analysis.

3.2.2 Serum and Salivary Cortisol Correlations

Numerous studies that have examined the relationship between salivary and serum cortisol measures have demonstrated that these two methods of cortisol collection yield significantly correlated cortisol concentrations (Dorn et al., 2007; Duplessis et al., 2010;

Galbois et al., 2010; Perogamvros et al., 2010). It has been consistently demonstrated across studies that salivary and free serum cortisol measures yield a significant correlation ranging from 0.82 to 0.914 (p<0.001) (Duplessis et al., 2010; Galbois et al.,

2010). These data support earlier research conducted on salivary and serum cortisol

36 37

correlations (Bober et al., 1988; Kahn et al., 1988; Obminski et al., 1997; Vining et al.,

1983) and suggest that although only 1-15% of unbound cortisol passes through saliva, salivary cortisol nonetheless estimates free serum cortisol reliably. Salivary cortisol is thus a useful surrogate of circulating free cortisol that has been shown to closely reflect free serum cortisol after adrenal stimulation and hydrocortisone administration

(Perogamvros et al., 2010). Due to the high correlation between salivary and serum cortisol measures, it is argued that collection methods (saliva versus serum) may be more dependent on the population being sampled, particularly clinical populations. For example, studies have demonstrated that salivary cortisol measures are preferred for assessment of adrenal function in patients with cirrhosis, as a peripheral symptom of this disease is reduced synthesis of cortisol-binding globulin and albumin. If serum cortisol is measured in these patients it could potentially lead to an overestimation of adrenal insufficiency (Galbois et al., 2010). Furthermore, serum cortisol measurement is complex, expensive, labor intensive and generally is performed by specific laboratories.

By contrast, salivary cortisol measures are not advised if sampling from populations in an

ICU due to inadequate sample volumes and the possibility of contamination. For the purpose of sampling healthy young adult populations, salivary cortisol is a non-invasive, convenient, expedient, and acceptable measure.

3.2.3 Salivary Cortisol Analysis

Saliva samples were collected in all dissertation studies for the purpose of obtaining salivary cortisol measures. In the first study, saliva samples were sent to the University of

Trier, to Dr. Kirschbaum’s lab, for salivary cortisol analysis. Saliva samples for the latter two studies were analyzed in our lab at University Medical Center. Methods and procedures discussed in the following sections are specific to testing done at the

University of Arizona. Analysis protocols used at the University of Trier are not included in this discussion.

3.2.3.1 Preparation and Quality Control

Twenty-four hours prior to testing, samples were left at room temperature to thaw. For the purpose of quality control, two experimenters were present at all times for analysis of each plate. One experimenter conducted the analysis while the other supervised. Single and multichannel pipettors were checked prior to each analysis session to ensure that pipetting instruments were accurate. Pipetting techniques were standardized across experimenters (e.g. plate wells were always analyzed from left to right and pipette tips were changed at the same time points across experimenters and plates). Each testing plate was actively handled by one experimenter only, with a maximum of two experimenters conducting analyses on all plates. Each sample was tested in duplicate and standard deviations of optical density readings and cortisol concentrations were compared for intra-and inter-assays. Outliers for unknown samples were retested or discarded if the second testing sample was insufficient for analysis. Saliva samples with insufficient sample for duplicate testing or retesting were excluded from the analysis. Standards and control well data were compared to known values to ensure accurate calculation of the standard curve for each plate.

38 39

3.2.3.2 Materials and Equipment

Samples were analyzed using solid phase enzyme-linked immunosorbent assay (ELISA) kits manufactured by IBL International (Hamburg, Germany). ELISA analysis is based on the competition principle, where a fixed amount of enzyme labeled antigen and an unknown amount of antigen (i.e. experimental samples) are combined in plate wells where they compete for binding sites of antibodies coated onto the bottom of the well.

After an incubation period, the wells are washed to stop the competition reaction.

Intensity of the developed color, present due to the substrate reaction, is inversely proportional to the amount of the antigen in the sample. A measure of optical density allows one to determine cortisol concentrations of unknown samples from a standard curve calculated using an automated 4 Parametric Logistics nonlinear regression model

(Y=(A-D)(1+(X/C) ^ B + D); where X is the concentration). The standard curve represents optical densities of the standards (y-axis, linear) plotted against corresponding concentrations (x-axis, logarithmic). Due to possible variance across plates (e.g. manufacturing or testing protocol variance), a standard curve is calculated for each plate and cortisol concentrations are determined using each plate’s specific curve. Optical densities and cortisol concentrations of standards and controls were compared across plates to ensure that standard curves were nearly identical.

Each ELISA kit included a 96-well microtiter plate coated with anti-cortisol antibodies

(rabbit), enzyme conjugate composed of cortisol conjugated to HRP stabilizers, standards

(0 !g/dL to 4.00 !g.dL), high and low cortisol controls, TMB substrate and stop solution, wash buffer and adhesive foil. All standards, controls and unknown samples were tested in duplicate to identify potential pipetting errors. Average optical density reading was used to determine cortisol concentrations for all duplicate wells. An adjustable single channel pipette was used to pipette standards, controls and unknown samples into respective wells of the microtiter plate. An electronic 8-channel Biohit micropipettor fitted with disposable, plastic, filtered, Biohit Optifit Tips was used to pipette enzyme conjugate, wash buffer, and TMB substrate and stop solutions. Plates were placed on an orbital shaker during incubation. Optical density readings were done with a Bio Tek

Absorbance Microplate Reader ELx800. Gen5 Data Analysis Software was used to calculate standard curves and determine cortisol concentrations for each plate.

3.2.3.3 Analysis Protocol

All reagents and specimens were warmed to room temperature (18-25o C) one hour prior to analysis. Once plate testing started, all steps were completed without interruption. To avoid contamination of reagents, pipettes, or wells, new disposable plastic pipette tips were used for each component and specimen. Wells and reagents were never reused across plates or kits. All wells were handled in the same order and time sequence. To minimize variance introduced by washing techniques, an 8-channel micropipettor was used and all wells were checked to confirm precise filling with wash buffer, and that there were no residues present, during rinsing. Additionally, wells were never allowed to dry between incubations or washes and protocol precautions prevented coated wells from

40 41

being scratched during rinsing and aspiration.

Samples were centrifuged in a Fisher HealthCare Model 614B for 15 minutes at 3000 x g to extract sample saliva and remove particulate materials. An adjustable single channel pipette, set to 50!L, was used to pipette standards, controls and samples into respective wells of the microtiter plate. Enzyme conjugate was pipetted into each well using an 8- channel micropipettor set to 100 !L. The plate was covered with adhesive foil and incubated on an orbital shaker (450rpm) for two hours at room temperature. After the incubation period, adhesive foil was removed, incubation solution was discarded and the plate was washed with 250uL of diluted wash buffer (10mL wash buffer to 100mL biodistilled water) four times. Excess solution present after washing was removed by tapping the plate inverted on a paper towel. One-hundred !L of TMB substrate solution was then applied to each well. The plate underwent a second incubation period on an orbital shaker (450rpm) uncovered at room temperature for 30minutes. After the incubation period 100 !L of TMB stop solution was applied to each well to stop the substrate reaction. The plate was shaken briefly by hand. Within one minute of application of TMB stop solution the plate was placed in the optical density reader with a photometer set at 450nm. Both experimenters recorded optical density and cortisol concentration data output to ensure that readings for each sample were correctly documented. Cortisol concentrations were entered into SPSS for statistical analysis.

IV. EXPERIMENT I: MORNING STRESS SIGNIFICANLTY REDUCES THE EMOTIONAL MEMORY ENHANCEMENT EFFECT WHEN INDUCED AT ENCODING

4.1 Research Aims & Hypotheses

Research in the area of emotional memory and stress has shown that stress enhances or preserves memory for emotional information but impairs memory for neutral information. To our knowledge, the majority of these studies have been conducted in the afternoon hours, when basal cortisol levels are typically low. The effects of stress on memory for emotional and neutral information under physiological conditions in which basal cortisol levels are high (in the morning) have not yet been fully examined. This is particularly true for stress induced just prior to retrieval.

The aim of the current experiment was to examine whether elevated basal cortisol levels modulate traditionally reported emotional memory enhancing and neutral memory impairing effects of stress on episodic memory. In addition we examined the effects of morning stress on both encoding and retrieval processes.

The major working hypotheses were the following: 1) salivary cortisol will be significantly greater in the early compared to late morning groups. This hypothesis is based on the fact that cortisol levels are highest 30mins to 1hr post wake, 2) stress induction at encoding in the early morning group will impair memory overall. This hypothesis is based on the notion that hippocampal MRs and GRs are near saturation due to high basal levels of cortisol and that further increases in cortisol levels, elicited by

42 43

stress, will significantly impairing hippocampal plasticity and function. As a consequence, emotional information will not be efficiently encoded, 3) stress induction at encoding in the late morning, when basal cortisol levels are relatively lower, will result in comparatively enhanced memory for emotional information. This hypothesis is based on the notion that as basal cortisol levels drop, MRs and GRs become less saturated and hippocampal encoding of emotional information can be effectively modulated by stress induced release of cortisol, 4) stress will predominantly impair encoding but not affect retrieval processes, particularly of emotional information. This notion is based on research showing that memory retrieval is typically not impaired for thematically arousing events.

4.1.1 Method

4.1.1.1 Participants

Ninety University of Arizona undergraduate students (50 females, 40 males; mean age

=19) were tested in two experimental sessions, held exactly 1 week apart. Participants were tested in the morning, between the hours of 7am and 11am.

4.1.1.2 Stimulus Materials

Participants viewed a slideshow accompanied by either a non-emotional or negative narration. The slideshow was composed of 12 slides in total and had three phases. The first phase, referred to as the neutral phase, was composed of the first four slides, accompanied by a non-emotional narration and composed exclusively of neutral visual stimuli. The second phase, referred to as the emotional phase, was composed of the

subsequent three slides that were accompanied by a negative narration and composed of negative visual stimuli. The final phase, referred to as the post emotional phase, was composed of the remaining five slides that were accompanied by a moderately negative emotional narration and composed of neutral visual stimuli. Overall, the slideshow depicted a mother taking her young son to see his father who works as a surgeon in a nearby hospital (i.e. neutral phase). Earlier in the day there was a horrible car accident, the victim of the accident sustained an injury that severed both legs. The car crash victim is brought to the hospital the father works at and the son and mother watch as the father fights to save the victim and reattach the severed limbs (i.e. emotional phase). The mother leaves after the surgery, emotionally disturbed by what she saw. She makes a phone call to get out of work and hails a taxi home (i.e. post emotional phase).

4.1.1.3 Procedure

This study was a 3 (stress at encoding vs. stress at retrieval vs. control) x 2 (emotional vs. neutral narration) x 2 (time of day (i.e. early vs. late morning)) mixed factorial design, yielding twelve experimental conditions; (1) stress at encoding & emotional narration early morning, (2) stress at encoding & emotional narration late morning, (3) stress at encoding & neutral narration early morning, (4) stress at encoding & neutral narration late morning (5) stress at retrieval & emotional narration early morning, (6) stress at retrieval & emotional narration late morning (7) stress at retrieval & neutral narration early morning, (8) stress at retrieval & neutral narration late morning, (9) control (no stress) & emotional narration early morning, (10) control (no stress) & emotional narration early late morning, (11) control (no stress) & neutral narration early morning,

44 45

and (12) control (no stress) & neural narration late morning.

In the first experimental session, participants arrived in the lab, gave informed consent, and were asked to perform the Trier Social Stress Test (TSST) (Kirschbaum et al., 1993), or a matched control task, both of which lasted ~20mins. The TSST, as described earlier reliably induces a significant physiological stress response in laboratory settings

(Kirschbaum et al., 1993, 1996; Kuhlmann et al., 2005). The matched control consisted of a set of cognitive tasks designed to be similar to those of the stress task, without stress induction; (1) a written sentence completion task, (2) reading of sentences aloud, and (3) a pencil and paper math task (addition & subtraction only). Immediately following stress induction, or the matched control task, participants watched the slide show and were instructed to carefully attend to each of the 12 color slides while listening to an emotional narration (Cahill et al. 1994; Cahill & McGaugh, 1995). This procedure was developed by Heuer and Reisberg (1990) and Cahill et al. (1994) to investigate memory for closely matched neutral and emotional episodes depicted in a slide show (Cahill & McGaugh,

1995).

One week later participants returned to the lab to complete the second session.

Participants assigned to the stress at encoding and control groups performed the matched control task and participants assigned to the stress at retrieval group performed the TSST.

Immediately following the stress (or control) task participants were asked to perform an unexpected memory test for the slide show they viewed the previous week. The main test

was a four-alternative forced choice recognition test, administered over a computer, which included eight questions per slide and assessed memory for both the general theme and details of the slide show.

Subjective anxiety reports were collected, using the Speilberger State Anxiety Inventory

(Speilberger et al. 2004), at eight time points (i.e. four x per visit) during the course of the experiment. Ratings were collected during both visits (1) upon arrival, (2) during preparation of the TSST (or control task), and (3) after completing the TSST (or control) procedure. Additional measures were collected after participants viewed the slide show on the first visit and again after memory testing on the second visit. Saliva samples were collected, using Salivette collection devices (Sarstedt, Inc.), as a physiological measure of stress levels, at six time points throughout the experiment. These samples were collected at the same time subjective anxiety measures were taken (1) baseline and (2) after stress induction (or control task) on each visit and at memory testing.

4.1.2 Results

Upon completion of data collection of recognition accuracy scores, subjective anxiety reports and cortisol levels were compared across the twelve groups. A repeated-measures

ANOVA was used to determine whether significant main effects or interactions existed across slide show conditions and/or experimental groups. Additional analyses were performed to determine correlations amongst memory scores, subjective ratings, and cortisol measures.

46 47

4.1.2.1 Subjective Anxiety Scores

A repeated-measures ANOVA, with group and story as between-subject factors and assessment time as the within-subject factor, for visit one, revealed a significant anxiety x group interaction (F[6,73] = 20.59, P <.0001) (Fig.1 ). Stress at encoding participants rated themselves as significantly more stressed both during (M=60, SD=11.31) and after

(M=53; SD=9.65) the stress manipulation, than those in the control (during M=43;

SD=12.91; after M=37.5; SD=10.22) and stress at retrieval (during M=44.7; SD=10.21; after M=37.5; SD=10.6) conditions. Independent samples T-tests revealed that subjective anxiety ratings for visit one were significantly greater in the stress at encoding group during and after the stress manipulation compared to the control (during t(64) = 5.54, p=.0001; after t(64)=-6.23, p=.0001) and stress at retrieval (during t(53)=5.22, p=.0001; after t(53)=4.15, p=.0001) groups. Similarly, stress at retrieval participants rated themselves as significantly more stressed both during (M=55.8 SD=11.27) and after

(M=51;SD=12.73) the stress manipulation, then those in the stress at encoding (during

M=37.31,SD=8.97; after 34.76; SD=9.32) and control (during M=40.5,SD=9.88; after

M=36.8, SD=9.18) conditions (Fig. 2). Independent samples T-tests revealed that subjective anxiety ratings for visit two were significantly greater in the stress at retrieval group during and after the stress manipulation compared to the control (during t(59)=4.69, p=.0001; after t(59)=4.47, p=.0001) and stress at encoding (during t(53)=5.46, p=.0001; after t(53)=4.78, p=.0001) groups.

Figures 1 & 2. Mean subjective anxiety ratings across experimental time points, for visits 1 and 2, compared across 1) stress at encoding, 2) stress at retrieval and 3) control groups. For visit 1, the stress at encoding group reported significantly higher anxiety ratings, compared to the stress at retrieval and control groups, during and directly after the stress manipulation. For visit 2, the stress at retrieval group reported significantly higher anxiety ratings, compared to the stress at encoding and control groups, during and directly after the stress manipulation.

4.1.2.2 Salivary Cortisol

A repeated measures ANOVA comparing group, time of day, and salivary cortisol sampled at various time points throughout the experiment revealed a significant cortisol x group interaction (F[5,88] = 6.23, P <.0001) (Fig. 3 & 4) and a significant between groups main effect of time of day (early versus late morning) (F[1,88] = 21.40, P <.0001) (Fig. 5).

Salivary cortisol levels were significantly higher for the stress at encoding group immediately after (M=22.76; SD=9.55) and 20minutes post (M=23.57; SD=10.92) the stress manipulation compared to the control (after M=13.14; SD=6.49; 20mins post

M=13.04; SD=7.09) and stress at retrieval (after M=13.43; SD=8.66; 20mins post

M=16.59; SD=10.65) groups.

48 49

Follow-up independent samples T-tests revealed that during visit one the stress at encoding group showed significantly increased cortisol levels immediately after and 20- mins post the stress manipulation compared to the control (post t(61)=4.57; p=.0001;

20mins post t(61)=4.65; p=.0001) and stress at retrieval (post t(52)=3.75; p=.0001;

20mins post t(52)=2.37; p=.021) groups. The stress at retrieval group showed a similar pattern of results during visit two, significantly higher cortisol levels 30 minutes after the stress manipulation (M=24; SD=12.15), compared to the control (M=13.32; SD=7.12) and stress and encoding (M=16.74; SD=9.72). Independent samples T-tests revealed that the stress at retrieval group had significantly higher cortisol levels 30-minutes post the stress manipulation compared to the control (t(59)=4.26, p=.001) and stress at encoding

(t(52)=2.39, p=.020) groups.

Figures 3 & 4. Mean salivary cortisol across experimental time points 1) pre-task, 2) post-task and 3) 30mins post-task for visits 1 and 2, compared across 1) stress at encoding, 2) stress at retrieval and 3) control groups. For visit 1, the stress at encoding group showed significantly elevated salivary cortisol levels immediately after and 30mins post stress manipulation, compared to the stress at retrieval and control groups. For visit 2, the stress at retrieval group showed significantly elevated salivary cortisol levels 30mins post stress manipulation compared to the stress and encoding and control groups.

A main effect of time of day on cortisol levels was also found, significantly higher cortisol levels observed for early (M=20.36; SD=7.75) compared to the late (M=13.52;

SD=5.71) morning group. An independent samples T-test revealed that salivary cortisol levels were significantly higher in the early morning than late morning group (t(87)=4.67, p=.0001).

Figure 5. Mean baseline salivary cortisol levels for participants tested in the early (7am- 9am) compared to late (9am-11am) morning. Mean salivary cortisol baseline was significantly higher in the early compared to late morning group.

4.1.2.3 Recognition Memory

A repeated measures ANOVA comparing group, slideshow (emotional and neutral narration), time of day, and slide show phase (emotional and neutral stimuli) revealed a significant phase x group interaction (F[2,88] = 3.02, P <.020) (Fig. 6). Between group

50 51

comparisons revealed that the control (M=52.38; SD=12.14) and stress at retrieval

(M=56.18; SD=8.75) groups performed significantly better on the emotional phase of both slide shows than the stress at encoding group (M=42.9; SD=14.08) (vs. control t(62)=2.88, p=.005; vs. stress at retrieval t(53)=4.13, p=.0001). Follow-up paired T-tests revealed that control and stress at retrieval groups performed significantly better on the emotional (control M=52.38, SD=12.14; stress at retrieval M=56.18; SD=8.75), compared to the neutral phase (control M=38.85; SD=12.44; stress at retrieval M=43.06;

SD=8.19) of both slide shows (control t(34)=5.89, p=.0001; stress at retrieval t(25)=6.07, p=.0001), while the stress at encoding group did not show this difference.

Figure 6. Mean corrected recognition (i.e. hits-FA) for stress at encoding, stress at retrieval, and control groups for neutral and emotional phases, collapsed across slideshow narration (i.e. emotional versus neutral). Control and stress at retrieval groups showed significantly better memory for the emotional, compared to neutral, phase of the slideshows. The stress at encoding group showed significantly poorer memory for the emotional phase of the slideshow, compared to control and stress and encoding, groups.

A repeated measures ANOVA comparing slideshow phase, story, and time of day

revealed a significant phase x story x time of day interaction (F[2,88] = 4.27, P <.016; Fig.

7 & 8). Paired samples T-tests revealed that the emotional phase of both slideshows was remembered significantly better than the neutral phase, in both the early morning (neutral phase M=40.61; SD=10.44; emotional phase M=50.6; SD=11.84; t(48)=4.69, p=.0001) and late morning (neutral phase M=38.44; SD=13.17; emotional phase M=50.26;

SD=14.42; t(40)=4.95, p=.0001) participants. In the late morning group, participants who viewed the emotional slideshow showed better memory for the emotional phase

(M=55.61; SD=10.60) than participants who viewed the neutral slideshow (M=45.17;

SD=15.13). An independent samples T-test revealed that this difference was significant

(t(39)=2.45,p=.019).

Figures 7 & 8. Mean corrected recognition score (i.e. hits-FA) across all groups for emotional and neutral phases of neutral and emotional slideshows for early (Fig. A) compared to late (Fig. B) morning participants. Recognition memory for the emotional compared to neutral phases is significantly better than recognition memory for the neutral phase for both early and late morning participants for both the emotional and neutral slideshow. The late morning participants show a significantly better recognition memory for the emotional phase of the emotional slideshow.

52 53

4.1.3 Discussion

The current study supports findings that basal cortisol levels are significantly higher in the early morning, compared to late morning, hours and provides evidence that these differences modulate stress effects on memory for emotional information and events.

Results are congruent with the findings of Maheu et al., (2005) and support results reported in the meta-analysis conducted by Het, Ramlow, & Wolf (2005), demonstrating that stress, induced in the morning impairs memory for emotional materials. Maheu et al.,

(2005) and the current dissertation study are amongst the first to demonstrate, within a single paradigm, basal cortisol level modulation of stress effects on memory for emotional materials. These findings are congruent with animal and human research suggesting that the relationship between hippocampal-dependent memory (i.e. episodic memory) and cortisol is represented by an inverted-U function (Lupien et al., 2002a,b;

Abercrombie et al., 2003; de Kloet et al., 1999; 2002; Roozendaal, 2002), the effects of cortisol on hippocampal neural plasticity dependent on the ratio of MR to GR activation, which varies as a function of the diurnal cycle. The probability of activating a significant number of GRs is higher if stress is experienced in the morning, as approximately 80% of

MRs are already occupied at morning cortisol basal levels (Joëls, 1999). Stress induction in the morning, when basal levels of cortisol are highest, likely elevates cortisol levels beyond a critical threshold. At this level, LTP processes in the hippocampus are impaired, nullifying potential facilitatory interactions with the amygdala. By contrast, in the afternoon when basal stress levels are declining to reach a nocturnal trough, a significant increase in cortisol levels likely results in significant MR activation but only

modest GR activation, conditions under which hippocampal-LTP can be facilitated through interactions with the amygdala, permitting the successful encoding of emotional information under stress (de Quervain et al., 2009; Lupien et al., 2007; McGaugh &

Roozendaal, 2002; Nathan et al., 2004; Roozendaal, 2002; Strange & Dolan, 2004).

Furthermore, in this study, morning stress impaired emotional memory in the stress at encoding group but did not affect memory for neutral information in either of the two stress conditions. The absence of stress effects on memory for neutral information could be attributable to mean recognition scores for the neutral phase of the slideshow being low, even under control conditions (Fig. 6). Control means for the neutral phase of the current study are similar to those obtained in studies that have reported impaired neutral memory under conditions of stress, with the same experimental materials (Payne et al.,

2006; 2007). Low means for neutral memory in the control group may create conditions in which an effect of stress on memory for neutral information is difficult, if not impossible, to detect. Low means obtained for controls in the current study, may suggest that high basal cortisol levels (compared to low afternoon levels) may impair encoding processing for neutral information, overall. This explanation, however, seems to conflict with the findings from Maheu et al. (2005), who reported low means for controls on memory for neutral information in both morning and afternoon control participants. They argue that results may be due to the stimuli used, a narrated slideshow containing both neutral and emotional aspects. However, research from our group has shown relatively high means for controls, tested in the afternoon, on memory for neutral phases of a

54 55

narrated slideshow with emotional and neutral features. One critical difference between these studies is the form of memory testing. Maheu et al. (2005) tested recall memory while Payne et al. (2006; 2007) tested recognition memory, a form of memory testing shown to elicit higher memory scores than recall tasks. Therefore, differences in control group performance across experimental designs may be a function of the type of memory being tested. Further research is needed to determine how basal cortisol levels affect memory encoding in general and encoding of information under conditions of stress.

As discussed earlier, stress induced just prior to memory testing has been shown to disrupt memory retrieval. Impaired memory retrieval, under conditions of stress, has been reported across a number of studies and across various types of stimuli with varying levels of valence and arousal (e.g. words (Domes et al., 2004; Kulmann, Piel, & Wolf,

2005), word pair associations (Tollenaar et al., 2008a; 2008b) and slideshows (Buchanan

& Tranel, 2008). The vast majority of these studies, however, used recall to assess memory performance. Very few studies have examined the effects of stress on recognition. The current study used a recognition task and revealed that recognition memory performance for the stress at retrieval group, for neutral and emotional phases of the slideshow, did not significantly differ from controls. In conjunction with previously reported stress at retrieval findings for recall tasks, these data suggest that recognition memory is unaffected by stress when induced at the time of retrieval. Cues provided at time of testing facilitate memory retrieval, which may enable the retrieval of episodic memory even with stress levels that impair functional activation in the hippocampus.

This may be particularly true in cases where recognition cues are thematically related, providing more information about the original event as the test proceeds. More studies are needed to determine if stress effects on retrieval processes are modulated by the amount of information provided by cues during retrieval. Neuroimaging studies would be particularly useful in this investigation to examine brain activation during recall compared to recognition tasks, under conditions of stress, and subsequent memory performance.

Lastly, the emotional memory enhancement effect was observed for the emotional phase of both slideshows in the early and late morning groups; however, in the late morning group recognition memory for the emotional phase of the emotional slideshow was significantly better compared to memory for the emotional phase of the neutral slideshow. Comparison of means across early and late morning groups revealed that memory performance for the emotional phase of the neutral slideshow was lower than that of the early morning group and memory performance for the emotional phase of the emotional slideshow was higher than that of the early morning group. These findings suggest that when cortisol levels are high (i.e. early morning) emotional information is remembered better than neutral information, overall. This may be due to an interaction amongst high cortisol levels, facilitated amygdala learning, and arousing visual stimuli.

As cortisol levels begin to drop (i.e. late morning) the emotionally arousing nature of the event theme becomes more important than the visual information, alone. As seen in our results (Fig. 8) emotional memory performance was significantly better in the emotional

56 57

narration, than neutral narration, condition in the late morning group.

Overall, results of the current study provide evidence that basal cortisol levels modulate the effects of stress on memory for emotional information. Emotional events encoded under stress, when basal cortisol levels are high (i.e. early morning) are subsequently remembered quite poorly. This is in contrast to what has been shown when stress is induced at encoding in the afternoon, where enhanced memory for emotional information was observed. Fluctuating basal cortisol levels also modulate memory for emotional events and material, regardless of stress induction. In the morning, memory for emotional information was enhanced overall, regardless of the emotional context of the event; a result most likely driven by facilitation of emotional memory encoding by high cortisol levels. By contrast, although the late morning group showed the same pattern of results

(enhanced memory for emotional information) cortisol-driven facilitation of emotional information was less prevalent, particularly in the group that viewed the neutral narration.

Memory for the emotional phase was significantly better for the group that viewed the emotional narration slideshow suggesting that as basal cortisol levels begin to decline; the emotional content of the event has more of an effect on memory processes. Future research is needed to compare the effects of stress directly across different times, particularly times of day or night when negative emotional events, such as rapes or assaults, typically occur. These data may speak to why in some instances individuals do not develop traumatic memories.

V. EXPERIMENT II: THE SUSCEPTIBILITY OF EPISODIC MEMORY TO MISINFORMATION WHEN ENCODED UNDER STRESS

5.1 Research Aims & Hypotheses

Research in the area of stress and memory shows that, when encoded under conditions of moderate to severe arousal (i.e.), memory for emotional information is enhanced while memory for non-emotional information is often impaired. The dissociative effect of stress on memory for emotional versus non-emotional information may create conditions in which non-emotional information is more susceptible to integration of misinformation, than emotional information, when encoded under stress.

In the current study, we examined the integration of misinformation into memory for a slideshow, composed of negative emotional and neutral valenced stimuli. The slideshow was either accompanied by a negative or neutral narration for the purpose of examining memory and the integration of misinformation for negative emotional versus neutral thematically arousing events, when encoded under stress. Hypotheses for the current study were the following: 1) participants that receive misinformation will retrieve significantly more misinformation than participants that did not receive misinformation,

2) participants in the stress group will show impaired memory for neutral compared to emotional components of the slideshow, 3) participants in the misinformation/stress group will show the integration of significantly more misinformation into the neutral, compared to emotional, phases of the slideshow.

58 59

5.1.1 Method

5.1.1.1 Participants

One hundred and three University of Arizona undergraduate students (59 females, 44 males; mean age =19) were tested in three experimental sessions, held exactly 48 hours apart. All sessions were conducted in the afternoon, between the hours of 1pm and 5pm.

5.1.1.2 Stimulus Materials

Stimulus materials were identical to those described in the previous study (see on page 46 and 47).

5.1.1.3 Procedure

This study was a 2 (stress vs. control) x 2 (emotional slide show vs. neutral slide show) x

2 (misinformation versus no misinformation) mixed factorial design, yielding eight experimental conditions; (1) stress, emotional narration, and misinformation, (2) stress, emotional narration, and no misinformation, (3) stress, neutral narration, and misinformation, (4) stress, neutral narration, and no misinformation, (5) no stress, emotional narration, and misinformation, (6) no stress, emotional narration, and no misinformation, (7) no stress, neutral narration, and misinformation, and (8) no stress, neutral narration, and no misinformation.

In the first experimental session, participants arrived in the lab, gave informed consent, and were asked to perform the Trier Social Stress Test (TSST) (Kirschbaum et al., 1993),

or a matched control task, both of which lasted ~20mins. The matched control consisted of a set of cognitive tasks designed to be similar to those of the stress task, without stress induction; (1) a written sentence completion task, (2) reading of sentences aloud to themselves, and (3) a pencil and paper math task (addition & subtraction only).

Immediately following stress induction, or the matched control task, participants watched a slide show and were instructed to carefully attend to each of 12 color slides while listening to either a neutral or emotional narrative (Cahill et al. 1994; Cahill & McGaugh,

1995). This procedure was developed by Heuer and Reisberg (1990) and Cahill et al.

(1994) to investigate memory for closely matched neutral and emotional episodes depicted in a slideshow (Cahill & McGaugh, 1995).

Forty-eight hours after the initial visit, participants returned to the lab to complete the second session. During this session participants were asked sixty questions; five questions for each of the twelve slides that composed the slideshow viewed forty-eight hours earlier. Participants randomly assigned to the misinformation group were unknowingly exposed to false information for each slide, a total of five false details per slide. Participants randomly assigned to the no misinformation group were not exposed to false information. Questions across these two groups were identical with the exception of one false detail provided in each question in the misinformation group.

Forty-eight hours after the second visit, participants returned to the lab to complete the last visit of the experiment. During this visit, participants were asked to complete an

60 61

unexpected memory test for the slideshow they viewed earlier in the week. Memory tests involved a four-alternative forced choice recognition test, administered over a computer, with eight questions relating to each slide that assessed memory for both the general theme and details of the slideshow.

Subjective anxiety reports were collected, using the Speilberger State Anxiety Inventory

(Speilberger et al. 2004), at eight time points during the course of the experiment as a measure of subjective anxiety. These eight time points were the following: First visit (1) upon arrival, (2 & 3) directly after completing the TSST (or control) procedure, (4) directly after viewing the slideshow. Second visit (1) upon arrival, (2) in the middle of the question and answer session, (3) upon completion of the question and answer session.

Third visit (1) upon arrival, (2) directly after memory testing.

Saliva samples were collected, using Salivette collection devices (Sarstedt, Inc.), as a physiological measure of stress levels, at seven time points throughout the experiment.

These samples were collected at the same time subjective anxiety measures were taken:

First visit (1) upon arrival, (2) 20mins after initial onset of stress induction (or control task), (3) after viewing the slideshow. Second visit (1) upon arrival, (2) during questioning and answer session, and (3) upon departure. Third visit (1) upon arrival and

(2) directly after memory testing.

5.1.2 Results

5.1.2.1 Subjective Anxiety Scores

A repeated-measures ANOVA, with group and story as between-subject factors and assessment time as the within-subject factor, revealed a significant anxiety x group interaction (F[9,102] = 6.15, P <.0001) (Fig.9). Follow-up paired T-tests revealed that the stress at encoding group showed significant increases in subject anxiety ratings for visit one during (M=53.44; SD=14.91) and immediately after (M=45.15; SD=13.79) the stress manipulation, compared to the control group (during M=43.22; SD=12.05; after

(M=37.64; SD=10.58; during t(101)=3.85, p=.0001; after t(101)=3.12, p=.002).

Subjective anxiety scores did not significantly differ for any time points for visits two

(Fig. 10) or three (Fig. 11).

65 TSST 60 55 Control

50 45 40

35 30 Mean subject anxiety rating 25

Figure 9. Mean subjective anxiety ratings across experimental time points for Visit 1 for stress at encoding and control groups. The stress at encoding group reported significantly higher anxiety ratings, compared to the control group, during and directly after the stress manipulation. Anxiety ratings did not significantly differ between groups at baseline (pre(V1)) or during or after the slideshow presentation.

62 63

65 TSST 60 Control 55 50

40 35 30

Mean subject anxiety rating 25

pre(V2) Q&A post Q&A

Figure 10. Mean subjective anxiety ratings across experimental time points for Visit 2 for stress at encoding and control groups. Anxiety ratings did not significantly differ between groups at baseline (pre(V2)) or during or after the questioning session.

65 TSST 60 Control 55 50 45

40 35 30 Mean subject anxiety rating 25 pre(V3) memory test

Figure 11. Mean subjective anxiety ratings across experimental time points for Visit 3 for stress at encoding and control groups. Anxiety ratings did not significantly differ between groups at baseline (pre(V3)) or at memory test.

5.1.2.2 Salivary Cortisol

A repeated measures ANOVA comparing salivary cortisol measures for stress, misinformation and narration type revealed a significant cortisol x group interaction

(F[4,72] = 8.53, P <.0001) (Fig. 12). Follow-up independent samples T-tests revealed that the stress at encoding group showed significantly increased cortisol levels (M=13.9;

SD=8.63), compared to the control group (M=7.23; SD=4.51) 20-minutes after stress induction on visit one (t(71)=7.33, p=.0001). Salivary cortisol measures did not significantly differ for sessions two or three (Figs. 13 & 14). Note: An average baseline was calculated for visit three from visit one and two baselines.

16.0 TSST

14.0 Control 12.0

10.0

8.0

6.0 Mean salivary cortisol nmol/L Mean salivary cortisol nmol/L pre(V1) post

Figure 12. Mean salivary cortisol across experimental time points for Visit 1 for stress at encoding and control groups. Salivary cortisol significantly differed between groups post manipulation (i.e. TSST or control task).

64 65

16.0 TSST

Control 14.0

12.0

10.0

8.0

Mean salivary cortisol nmol/L Mean salivary cortisol nmol/L 6.0

pre(V2) post Q&A

Figure 13. Mean salivary cortisol across experimental time points for Visit 2 for stress at encoding and control groups. Salivary cortisol did not significantly differ between group for baseline (pre(V2)) or post questioning session.

16.0 TSST

14.0 Control

12.0

10.0

8.0 Mean salivary cortisol nmol/L Mean salivary cortisol nmol/L

6.0 Ave(V1&V2) memory test

Figure 14. Mean salivary cortisol across experimental time points for Visit 3 for stress at encoding and control groups. Salivary cortisol did not significantly differ between groups at time of memory test. The baseline time point for this visit is the average baseline for visits 1 and 2.

5.1.2.3 Recognition Memory

A Pearson-r correlation examining the relationship between subjective anxiety during encoding of the slideshow and memory performance across the three slideshow phases revealed a significant positive correlation between anxiety score and correct hits for misinformed items, for the emotional phase of the emotional slideshow in the stress group [r(14)=.807, p < .01] (Fig.15). This correlation revealed that participants in the stress group that rated themselves as having high anxiety during encoding of the emotional slideshow were more likely than those who rated themselves as having low anxiety levels to choose correct answers for misinformed items in the emotional phase of the slideshow during the recognition task.

40 !"#$#%&'()# 30 20

-10

-20 Anxiety Slideshow - base1 0 0.2 0.4 0.6 0.8

Figure 15. Significant negative correlation between subjective anxiety during slideshow presentation (anxiety slideshow – baseline V1) and misinformation effect for the emotional phase of the slideshow with the emotional narration.

66 67

A Pearson-r correlation examining the relationship between subjective anxiety during encoding of the slideshow and memory performance across the three slideshow phases revealed a significant negative correlation between anxiety score and mean percentage of misinformed items for the emotional phase of the emotional slideshow, within the stress group [r(14)= -.598, p < .05] (Fig.16). This correlation revealed that participants in the stress group that rated themselves as having high anxiety during encoding of the emotional slideshow were less likely than those who rated themselves as having low anxiety to choose misinformed items as correct answers for the emotional phase of the slideshow during the recognition task.

40 !"##$#%&*("# 30 20

-10 -20

Anxiety Slideshow - base1 0.2 0.3 0.4 0.5 0.6 0.7

Figure 16. Significant positive correlation between subjective anxiety during slideshow presentation (anxiety slideshow – baseline V1) and correct misinformation hits for the emotional phase of the slideshow with the emotional narration.

A repeated measures ANOVA comparing within subject measures for hits for three distinct phases of the slideshow (1) neutral, (2) emotional and (3) post emotional and between subject factors (1) stress, (2) misinformation and (3) narration type revealed a significant slideshow phase x group x misinformation x narration interaction (F[2,102] =

3.91, P < .022) (Figs. 17, 18, 19, & 20) and a group x narration type interaction (F[1,102] =

7.86, P <.006) (Fig. 21).

Follow-up paired and independent T-tests for the hits phase x group x misinformation x narration type interaction revealed the following: 1) stress and control groups that did not receive misinformation and viewed the slideshow with the emotional narration showed significantly better memory for the emotional phase (stress M=.63; SD=.124; control

M=.62; SD=.114), compared to the neutral (stress M=.56; SD=.094, t(11)=2.54, p=.05; control M=.56; SD=.089, t(20)=2.56, p=.019) and post emotional (stress M=.53;

SD=.097, t(11)=3.06, p=.011; control M=.51; SD=.093, t(20)=4.57, p=.0001) phases of the slideshow, 2) stress and control groups that were exposed to misinformation and viewed the slideshow with the emotional narration showed differential memory performance, particularly with regard to the emotional phase. Independent samples T- tests revealed that the stress group showed significantly better memory (M=.65;

SD=.063) for the emotional phase of the slideshow than the control group (M=.52;

SD=.136, t(33)=3.16, p=.003). Additionally, the stress group showed significantly better memory for the emotional (M=.65; SD=.063), than neutral (M=.60; SD=.044, t(14)=2.30, p=.038) and post emotional (M=.51; SD=.104, t(14)=4.34, p=.001), phases of the

68 69

slideshow, 3) Stress and control groups that viewed the slideshow with the neutral narration and did not receive misinformation showed differential performance on neutral and post emotional phases of the slideshow. The stress group showed significantly impaired memory for the neutral (M=.53; SD=.059) and post emotional (M=.40;

SD=.073) phases of the slideshow, compared to the control group (neutral phase M=.61;

SD=.071, t(15)=2.31, p=.035; post emotional M=.49; SD=.079, t(15)=2.21, p=.043) but did not significantly differ from controls on memory performance for the emotional phase

(t(15)=.631, p=.538; NS). The stress group showed significantly better recognition memory for the emotional phase (M=.62; SD=.112), compared to the neutral (M=.53;

SD=.059, t(10)=.2.94, p=.015) and post emotional (M=.40; SD=.074, t(10)=7.31, p=.0001), phases. 4) Both stress and control groups that viewed the slideshow with the neutral narration and received misinformation showed significantly poorer recognition memory from the neutral (control M=.58; SD=.125; stress M=.55; SD=.110) to the post emotional (control M=.46; SD=.102; stress M=.39; SD=.114) phase. Independent samples T-tests revealed that both groups showed significantly poorer memory for the post emotional phase of the slideshow compared to the neutral phase (control t(14)=2.67,p=.032; stress t(14)=3.43,p=.008).

0.7 TSST

0.65 + ++# Control 0.6 0.55

0.5

0.45

Mean phase hits 0.4

0.35 Neut Emo Post Emo

Figure 17. Mean hits across the three phases of the slideshow for stress and control, no misinformation, emotional narration groups. Both groups showed significantly better recognition memory for the emotional, compared to neutral and post emotional, phases.

0.7 TSST + Control 0.65 !" +# 0.6 0.55

0.5

0.45 Mean phase hits 0.4

0.35 Neut Emo Post Emo

Figure 18. Mean hits across the three phases of the slideshow for stress and control, misinformation, emotional narration groups. The stress at encoding group showed significantly better recognition for the emotional compared to neutral and post emotional phases of the slideshow. The stress group also showed significantly better memory for the emotional phase of the slideshow compared to the control group.

70 71

0.7 TSST + ++# Control 0.65 +# 0.6

0.55 +#

0.5

0.45 Mean phase hits 0.4 0.35 Neut Emo Post Emo

Figure 19. Mean hits across the three phases of the slideshow for stress and control, no misinformation, neutral narration groups. Both stress and control groups showed significantly better recognition for the emotional phase, compared to the neutral and post emotional phases, of the slideshow. The stress group showed significantly impaired memory for the neutral and post emotional phases, compared to the control group.

0.7 TSST 0.65 +# Control

0.6

0.55

0.5

0.45 Mean phase hits 0.4

0.35

Neut Emo Post Emo

Figure 20. Mean hits across the three phases of the slideshow for stress and control, misinformation, neutral narration groups. Both stress and control groups showed significantly better recognition for neutral than post-emotional phases of the slideshow.

Independent samples T-tests on a group x slideshow narration interaction revealed that the stress group showed significantly impaired memory for the neutral narration slideshow (M=.50; SD=.076; t(33)=2.20,p=.031) and significantly better memory for emotional narration slideshow (M=.59; SD=.065; t(66)=2.09,p=.040), compared to the control group (neutral narration M=.55; SD=.075; emotional narration M=.54; SD=.083).

0.7 TSST 0.65 Control

0.6 +# !" 0.55

0.5

0.45 Mean phase hits 0.4

0.35 Neut Emo

Figure 21. Significant interaction of narration type (i.e. emotional versus neutral) and group (i.e. stress vesus control). Overall, the stress group showed significantly impaired memory for the neutral narration slideshow and enhanced memory for the emotional narration slideshow, compared to the control group.

A repeated measures ANOVA of misinformation, slideshow phase, narration type and group revealed a significant main effect of misinformation condition (F[1,102) = 84.55, P

<.0001) (Fig. 22). A follow up T-test revealed that the misinformation group chose significantly more misinformed items (M=.39; SD=.089) as the correct responses overall,

72 73

than the group that was not misinformed (M=.25; SD=.054); t(101)=9.80,p=.0001).

0.6 0.55 0.5

0.45 +# 0.4 0.35

Mean MIQ hits 0.3 0.25 0.2 No M.I. M.I.

Figure 22. Significant main effect of misinformation. All groups that received misinformation showed significantly greater misinformation hits compared to groups that were not exposed to misinformation.

5.1.3 Discussion

The current dissertation study is the first to examine the misinformation effect in the context of stress. Results provide evidence that the misinformation effect is modulated by stress at encoding, arousal, and the valence of relevant stimuli. The stress-misinformation group showed significantly better memory for non-misinformed items for the emotional phase of the slideshow, compared to controls. In addition, this group showed significantly better memory for the emotional phase, compared to the neutral and post emotional phases, of the slideshow. By contrast, the stress and control, no misinformation groups, that viewed the emotional slideshow, did not significantly differ from one another with respect to recognition performance for neutral, emotional and post emotional phases.

Both groups showed significantly better memory for the emotional phase, compared to the neutral and post emotional phases, of the slideshow for items that were never misinformed. These results suggest that emotional information encoded under stress is less susceptible to alteration by misinformation, specifically those details for which false information was never provided.

Groups that viewed the neutral slideshow showed different effects. Stress and control groups that did not receive misinformation showed differential performance on neutral and post emotional phases of the slideshow. The stress group showed significantly impaired memory for these phases of the slideshow, compared to the control group. The stress group showed significantly better memory for the emotional phase, compared to

74 75

the neutral and post emotional phases of the slideshow, while the control group showed better memory for the emotional phase compared to the post emotional phase. These results support findings from our previous study (Payne et al., 2007) that demonstrated that stress at encoding significantly impaired subsequent memory for neutral, but not emotional, information. The stress and control groups that received misinformation showed a very different pattern for results. Stress and control groups did not significantly differ in memory performance for phases. Both groups showed significantly impaired memory for the post emotional phase of the slideshow, compared to the neutral phase.

Data trends in a linear decline, with the neutral phase of the slideshow being remembered the best, specifically in the stress group. Comparison of results across misinformed and non-misinformed groups suggests that misinformation disrupts memory for the emotional phase of the neutral slideshow in the stress group.

Narration type and group interaction revealed that, regardless of misinformation, the stress group showed significantly impaired memory on the neutral slideshow overall, compared to controls but significantly better memory for the emotional slideshow, overall, compared to controls. These data replicate findings of our previous study (Payne et al. 2007) that showed that stress at encoding impaired memory for neutral information but enhanced memory for emotional information. Data also showed a significant main effect of misinformation. These results support findings that when provided with misinformation, individuals are more likely to view that information as true during memory testing. A significant slideshow phase by narration type interaction revealed that,

regardless of stress condition, misinformation was more likely chosen as the correct response for the neutral portion of the neutral narrated slideshow.

Results additionally suggest that anxiety and stress interact in a way that makes memory for emotionally arousing materials less susceptible to the integration of false information.

Stressed individuals who rated themselves as high on a measure of subjective anxiety while viewing the slideshow were less likely to choose misinformation as the correct answer for the emotional phase of the slideshow. Additionally, these individuals were also significantly more likely to choose the correct answer. These results suggest that the interaction of stress and anxiety at encoding significantly reduces incorporation of misinformation into memory for an original event. These data are congruent with behavioral and pharmacological studies showing that arousal, and the release of noradrenergic hormones, significantly modulated the effects of stress and cortisol on encoding by further facilitating memory for emotional information (Maheu et al., 2004; van Steregen et al., et al., 2005; 2007).

76 77

VI. GENERAL DISCUSSION

The current dissertation research is amongst the first, to our knowledge, to examine the effects of basal cortisol levels on stress modulation of emotional and neutral information, when induced at encoding and the susceptibility of memories encoded under stress to the integration of misinformation. The data suggest that basal cortisol levels modulate encoding of emotional information, impairing these processes under conditions when basal cortisol levels are at their highest (e.g. morning). The arousing nature of the event interacted with basal cortisol levels, showing an enhancing effect of memory for visually emotionally arousing materials, but only when cortisol levels began to decrease. Current results suggest that the time of day when a traumatic event occurs may be one of many critical factors in determining the development of traumatic memory.

Additionally results from the current research provide evidence that episodic memory is not always susceptible to the integration of misinformation, as currently assumed.

Emotional information encoded under conditions of high arousal and stress appears to be comparatively inoculated against the integration of misinformation. Furthermore, these memories appear to be less susceptible to false information overall, even for details of the event that were never misinformed. The extent to which these memories remain uninfluenced by memory updating, and factors that may modulate this effect, remain unknown. One possibility is that emotional memories encoded under stress, and arousal, are represented by strong neutral networks that are less permeable to updating. Although

a system that ensures that emotional information is well remembered and largely accurate seems reasonable in terms of evolution and survival mechanisms, such impermeability of a memory trace to changes seems unlikely given our knowledge of the general malleability of memory and memory traces. Perhaps it is the case that reactivation of the memory must occur under the same or similar physiological conditions (i.e. high levels of stress and arousal) for the memory trace to be significantly reactivated for updating. If this were the case, integration of misinformation would be observed under conditions in which participants are stressed and aroused at encoding as well as during the time they are misinformed. Future research is needed to examine the extent to which emotional memories encoded under stress, and arousal, are unaffected by the introduction of false information. The research included in this dissertation furthers the field’s current understanding of the effects of stress on memory processes. Findings are relevant to the literature on traumatic memory, eyewitness testimony, and the effects of moderate to severe emotion on long-term episodic memory.

78 79

REFERENCES

Abercrombie, H. C., Kalin, N. H., Thurow, M. E., Rosenkranz, M. A., & Davidson, R. J. (2003). Cortisol variation in humans affects memory for emotionally laden and neutral information. Behavioral Neuroscience. 117(3), 505-516.

Adolphs, R. (2004). Emotional vision. Nature Neuroscience, 7(11), 1167-1168.

Adolphs, R., Denburg, N. L., & Tranel, D. (2001). The amygdala's role in long-term declarative memory for gist and detail. Behavioral Neuroscience, 115(5), 983-992.

Adolphs R., Gosselin F., Buchanan T.W., Tranel D., Schyns P., & Damasio, A.R. (2005). A mechanism for impaired fear recognition after amygdala damage. Nature, 433(7021), 68-72.

Alpers, G.W. (2008). Eye-catching: Right hemisphere attentional bias for emotional pictures. Laterality, 13(2), 158-178.

Anderson, A.K., & Phelps, E.A. (2001). Lesions of the human amygdala impair enhanced perception of emotionally salient events. Nature, 411(6835), 305-309.

Anderson, A.K., Yamaguchi, Y., Grabski, W., & Lacka, D. (2006). Emotional memories are not all created equal: Evidence for selective memory enhancement. Learning & Memory, 13(6), 711-718.

Bangasser, D. A., & Shors, T. J. (2007). The hippocampus is necessary for enhancements and impairments of learning following stress. Nature Neuroscience, 10(11), 1401- 1403.

Bober, J., Weller, E., Weller, R., Tait, A., Fristad, M., & Preskorn, S. (1988). Correlation of serum and salivary cortisol levels in prepubertal school-aged children. Jounral of American Academy of Child and Adolescent Psychiatry, 27(6), 748-750.

Bradley, M. M., Hamby, S., Löw, A., & Lang, P. J. (2007). Brain potentials in perception: Picture complexity and emotional arousal. Psychophysiology, 44(3), 364-373..

Brosch, T., Pourtois, G., & Sander, D. (2010). The perception and categorisation of emotional stimuli: A review. Cognition and Emotion, 24(3), 377-400.

Buchanan, T. W., & Adolphs, R. (2002). The role of the human amygdala in emotional modulation of long-term declarative memory. In S. Moore & M. Oaksford (Eds.),

Emotional cognition: From brain to behavior (pp. 9-34). Amsterdam: John Benjamins Publishing.

Buchanan, T. W., & Lovallo, W. R. (2001). Enhanced memory for emotional material following stress-level cortisol treatment in humans. Psychoneuroendocrinology, 26(3), 307-317.

Buchanan, T.W., & Tranel, D. (2008). Stress and emotional memory retrieval: Effects of sex and cortisol response. Neurobiology of Learning and Memory, 89(2), 134-141.

Buchanan, T. W., Tranel, D., & Adolphs, R. (2006). Memories for emotional autobiographical events following unilateral damage to medial temporal lobe. Brain, 129(1), 115-127.

Cabeza, R., Rao, S., Wagner, A., Mayer, A., & Schacter, D. (2001). Can medial temporal lobe regions distinguish true from false? An event related functional MRI study of veridical and illusory recognition memory. Proc. National Academy of Sciences, 98(8), 4804-4810.

Cahill, L., Gorski, L., & Le, K. (2003). Enhanced human memory consolidation with post-learning stress: Interaction with the degree of arousal at encoding. Learning & Memory, 10(4), 270-274.

Cahill, L., & McGaugh, J. L. (1995). A novel demonstration of enhanced memory associated with emotional arousal. Consciousness and Cognition, 4(4), 410-421.

Cahill, L., & McGaugh, J., & Weinberger, N. (2001). The neurobiology of learning and memory: some reminders to remember. Trends in Neuroscience, 24(10), 578-581.

Cahill, L., Prins, B., Weber, M., & McGaugh, J. (1994). Beta-adrenergic activation and memory for emotional events. Nature, 371(6499), 702-704.

Calvo, M. G., & Lang, P. J. (2005). Parafoveal semantic processing of emotional visual scenes. Journal of Experimental Psychology: Human Perception and Performance, 31(3), 502-519.

Canli, T., Zhao, Z., Brewer, J., Gabrieli, J. D. E., & Cahill, L. (2000). Event-related activation in the human amygdala associates with later memory for individual emotional experience. Journal of Neuroscience, 20(RC99), 1-5

Compton, R.J. (2003). The interface between emotion and attention: A review of evidence from psychology and neuroscience. Behavioral and Cognitive Neuroscience Reviews, 2(2), 115-129. de Kloet, E. R. (2002). Stress in the brain: implications for treatment of depression. Acta

80 81

Neuropsychiatrica, 14(4), 155-166.

de Kloet, E. R., Oitzl, M. S., & Joëls, M. (1999). Stress and cognition: are corticosteroids good or bad guys. Trends in Neuroscience, 22(10), 422-426. de Kloet, E. R., Vreugdenhil, E., Oitzl, M. S., & Joëls, M. (1998). Brain corticosteroid receptor balance in health and disease. Endocrine Reviews, 19(3), 269-301. de Quervain, D.J., Aerni, A., & Roozendaal, B. (2007). Preventive effect of beta- adrenoceptor blockade on glucocorticoid-induced memory retrieval deficits. The American Journal of Psychiatry, 164(6), 967-969. de Quervain, D.J., Aerni, A., Shelling, G., & Roozendaal, B. (2009). Glucocorticoids and the regulation of memory in health and disease. Frontiers in Neuroendocrinology, 30(3), 358-370. de Quervain, D.J., Roozendaal, B., Nitsch, R.M., McGaugh, J.L., Hock, C. (2000). Acute cortisone administration impairs retrieval of long-term declarative memory in humans. Nature Neuroscience, 3, 313-314.

Dedovic, K., D’Aguiar, C., & Pruessner, J.C. (2005). What stress does to your brain: A review of neuroimaging studies. Canadian Journal of Psychiatry, 54(1), 6-15.

Dedovic, K., Renwick, R., Mahani, N.K., Engert, V., Lupiën, S.J., & Pruessner, J.C. (2005). The Montreal imaging stress task: Using functional imaging to investigate the effects of perceiving and processing psychosocial stress in the human brain. Journal of Psychiatry & Neuroscience, 30(5), 319-25.

Dedovic, K., Rexroth, M., Wolff, E., Duchesne, A., Scherling, C., Beaudry, T., Lue, S.D., Lord, C., Engert, V., & Pruessner, J.C. (2009). Neural correlates of processing stressful information: An event-related fMRI study. Brain Research, 1293, 49-60.

Deffenbacher, K. A., Bornstein, B. H., Penrod, S. D., & McGorty, E. K. (2004). A meta- analytic review of the effects of high stress on eyewitness memory. Law and Human Behavior, 28(6), 687-706.

Denburg, N.L., Buchanan, T.W., Tranel, D., & Adolphs, R. (2003). Evidence for preserved emotional memory in normal older persons. Emotion, 3(3), 239-253.

Dolcos, F., LaBar, K.S., & Cabeza, R. (2004). Interaction between the amygdala and the medial temporal lobe memory system predicts better memory for emotional events. Neuron, 42(5), 855-863.

Domes, G., Heinrichs, M., Rimmele, U., Reichwald, U., & Hautzinger, M. (2004). Acute

stress impairs recognition for positive words-association with stress-induced cortisol secretion. Stress, 7(3), 173-181.

Dorn, L., Lucke, J., Loucks, T., & Berga, S. (2007). Salivary cortisol reflects serum cortisol: analysis of circadian profiles. Annals Clinical Biochemistry, 44(3), 281- 284.

Duplessis, C., Rascona, D., Yeung, E., & Cullum, M. (2010). Salivary and free serum cortisol evaluation. Military Medicine, 175(5), 340-346.

Duvarci, S., & Paré, D. (2007). Glucocorticoids enhance the excitability of principal basolateral amygdala neurons. Journal of Neuroscience, 27(16), 4482-4491.

Fehm-Wolfsdorf, G., Reutter, K., Zenz, H., Born, J., Lorenz, H. (1993). Are circadian variations in taste thresholds cortisol-dependent. Journal of Psychophysiology, 7, 65-72.

Gable, P.A., & Harmon-Jones, E. (2008). Approach-motivated positive affect reduces breadth of attention. Psychological Science, 19(5), 476-482.

Galbois, A., Rudler, M., Massard, J., Fulla, Y., Bennani, A., et al. (2010). Assessment of adrenal function in cirrhotic patients: salivary coritsol should be preferred. Journal of Hepatology, 52, 839-845.

Granger, D., Kivlighan, K., Furtunato, C., Harmon, A., Hibel, L., Schwartz, E., & Whembolua, G. (2007). Integration of salivary biomarkers into developmental and behaviorally-oriented research: problems and solutions for collecting specimens. Physiology & Behavior, 92, 583-590.

Groschl, M., & Rauh, M. (2006). Influence of commercial collection devices for saliva on the reliability of salivary steroids analysis. Steroids, 71(13-14), 1097-1100.

Harris, C.R., & Pashler, H. (2005). Enhanced memory for negatively emotionally charged pictures without selective rumination. Emotion, 5(2), 191-199.

Henckens, M.J., Hermans, E.J., Pu, Z., Joëls, M., & Fernändez, G. (2009). Stressed memories: How acute stress affects memory formation in humans. Journal of Neuroscience, 29(32), 10111-10119.

Het, S., Ramlow, G., & Wolf, O.T. (2005). A meta-analytic review of the effects of acute cortisol administration on human memory. Psychoneuroendocrinology, 30, 771- 784.

Heuer, F., & Reisberg, D. (1990). Vivid memories of emotional events: the accuracy of remembered minutiae. Memory & Cognition, 18(5), 496-506.

82 83

Hsu, F.C., Garside, M.J., Massey, A.E., McAllister-Williams, R.H. (2003). Effects of a single dose of cortisol on the neural correlates of episodic memory and error processing in healthy volunteers. Psychopharmacology, 167, 431-442.

Huang, Y., & Luo, Y. (2006). Temporal course of emotional negativity bias: An ERP study. Neuroscience Letters, 398(1), 91-96.

Joëls, M. (1999). Effects of corticosteroid hormones in the hippocampus. Acta Physiologica Scandinavica, 167(2), A3.

Joëls, M., & de Kloet, E. R. (1990). Mineralocorticoid receptor-mediated changes in membrane properties of rat CA1 pyramidal neurons in vitro. Proceedings of the National Academy of Sciences, 87(12), 4495-4498.

Joëls, M., & de Kloet, E.R. (1991). Effect of corticosteroid hormones on electrical activity in rat hippocampus. Journal of Steroid Biochemistry and Molecular Biology, 40(1-3), 83-86.

Joëls, M., & de Kloet, E.R. (1996). Corticosteriod hormones in neuroprotection and brain damage. Current Opinion in Endocrinology & Diabetes, 3(2), 184-832.

Kahn, J, Rubinow, D., Davis, C., Kling, M., & Post, R. (1988). Salivary cortisol: a practical method of evaluation of adrenal function. Biological psychiatry, 23(4), 335-349.

Kensinger, E. A. (2009). Remembering the details: Effects of emotion. Emotion Review, 1(2), 99-113.

Kensinger, E. A., & Corkin, S. (2004). Two routes to emotional memory: Distinct neural processes for valence and arousal. Proceedings of the National Academy of Sciences, 101(9), 3310-3315.

Kensinger, E. A., Garoff-Eaton, R. J., & Schacter, D. L. (2007a). How negative emotion enhances the visual specificity of a memory. Journal of Cognitive Neuroscience, 19(11), 1872-1887.

Kensinger, E. A., Garoff-Eaton, R. J., & Schacter, D. L. (2007b). Effects of emotion on memory specificity: Memory trade-offs elicited by negative visually arousing stimuli. Journal of Memory and Language, 56(4), 575-591.

Kensinger, E. A., Piguet, O., Krendl, A. C., & Corkin, S. (2005). Memory for contextual details: Effects of emotion and aging. Psychology and Aging, 20(2), 241-250.

Kensinger, E. A., & Schacter, D. L. (2006). When the Red Sox shocked the Yankees: Comparing negative and positive memories. Psychonomic Bulletin & Review,

13(5), 757-763.

Kim, J.J., & Diamond, D.M. (2002). The stressed hippocampus, synaptic plasticity and lost memories. Nature Reviews Neuroscience, 3(6), 453-462.

Kirschbaum, C., Wolf, O.T., May, M., Wippich, W., Hellhammer, D.H. (1996). Stress- and treatment-induced elevations of cortisol levels associated with impaired declarative memory in healthy adults. Life Sci., 58, 1475-1483.

Kirschbaum, C., Pirke, D., Hellhammer, D.H. (1993). The ‘Trier Social Stress Test’—a tool for investigating psychobiological stress responses in the laboratory setting. Neuropsychoniology, 28(1-2), 76-81.

Koster, E.H., Crombez, G., Verschuere, B., Vanvolsem, P., & De Houwer, J. (2007). A time-course analysis of attentional cueing by threatening scenes. Experimental Psychology, 54(2), 161-171.

Krieger, D. (1975). Circadian pituitary adrenal rhythms. Advances in Experimental Medicine & Biology, 54, 169-189.

Kroes, M.C., Strange, B.A., & Dolan, R.J. (2010). "-Adrenergic blockade during memory retrieval in humans evokes a sustained reduction of declarative emotional memory enhancement. Journal of Neuroscience, 30(11), 3959-3963.

Kuhlmann, S., Kirschbaum, C., & Wolf, O.T. (2005). Effects of oral cortisol treatment in healthy young women on memory retrieval of negative and neutral words. Neurobiology of Learning and Memory, 83(2), 158-162.

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

Kuhlmann, S., & Wolf, O. T. (2006a). Arousal and cortisol interact in modulating memory consolidation in healthy young men. Behavioral Neuroscience, 120(1), 217-223.

Kuhlmann, S., & Wolf, O. T. (2006b). A non-arousing test situation abolishes the impairing effects of cortisol on delayed memory retrieval in healthy women. Neuroscience Letters, 399(3), 268-272.

Laney, C., Campbell, H. V., Heuer, F., & Reisberg, D. (2004). Memory for thematically arousing events. Memory & Cognition, 32(7), 1149-1159.

Laney, C., Heuer, F., & Reisberg, D. (2003). Thematically-induced arousal in naturally-

84 85

occurring emotional memories. Applied Cognitive Psychology, 17, 995-1004.

Levine, L.J., & Edelstein, R.S. (2009). Emotion and memory narrowing: A review and goal-relevance approach. Cognition and Emotion, 23(5), 833-875.

Levine, L. J., & Pizarro, D. A. (2004). Emotion and memory research: A grumpy overview. Social Cognition, 22(5), 530-554.

Loftus, E. (1975). Leading questions and the eyewitness report. Cognitive Psychology, 7, 550-572.

Loftus, E., & Hoffman, H. (1989). Misinformation and memory: the creation of new memories. Journal of Experimental Psychology, 118(1), 100-104.

Loftus, E., & Palmer, J. (1974). Reconstruction of automobile destruction: an example of the interaction between language and memory. Journal of Verbal Learning & Verbal Behavior, 13(5), 585-589.

Loftus, E., & Pickrell, J. (1995). The formation of false memories. Psychiatric Annals, 25(12), 720-725.

Lupien, S.J., Gillin, C.J., Hauger, R.L. (1999). Working memory is more sensitive than declarative memory to the acute effects of corticosteroids: a dose-response study in humans. Behavioral Neuroscience, 113, 420-430.

Lupien, S.J., Maheu, F., Tu, M., Fiocco, A., & Schramek, T.E. (2007). The effects of stress and stress hormones on human cognition: Implications for the field of brain and cognition. Brain Cognition, 65, 209-237.

Lupien, S.J., Wilkinson, C., Briere, S., Menard, C., Ying Kin, N., & Nai, N. (2002a). The modulatory effects of corticosteroids on cognition: studies in young human populations. Psychoneuroendocrinology, 27(3), 401-416.

Lupien, S.J., Wilkinson, C., Briere, S., Menard, C., Ying Kin, N., & Nai, N. (2002b). Acute modulation of aged human memory by pharmacological manipulation of glucocorticoids. Journal of Clinical Endocrinology & Metabolism, 87(8), 3798- 3807.

Maheu, F. S., Collicutt, P., Komik, R., Moszkowski, R., & Lupien, S. (2005). The perfect time to be stressed: a differential modulation of human memory by stress applied in the morning or in the afternoon. Progress in Neuro-Psychopharmachology & Biological Psychiatry , 29, 1281-1288.

Maheu, F. S., Joober, R., Beaulieu, S., & Lupien, S. J. (2004). Differential effects of adrenergic and corticosteroid hormonal systems on human short- and long-term

declarative memory for emotionally arousing material. Behavioral Neuroscience, 118(2), 420-428.

Mather, M. (2007). Emotional arousal and memory binding: An object-based framework. Perspectives on Psychological Science, 2(1), 33-52.

Marin, M., Pilgrim, K., & Lupien, S. (2010). Modulatory effects of stress on reactivated emotional memories. Psychoneuroendocrinology, 35(9), 1388-1396.

McGaugh, J. L., & Roozendaal, B. (2002). Role of adrenal stress hormones in forming lasting memories in the brain. Current Opinion in Neurobiology, 12(2), 205-210.

Merz, C.J., Wolf, O.T., & Hennig, J. (2010). Stress impairs retrieval of socially relevant information. Behavioral Neuroscience, 124(2), 288-293.

Mickley, K.R., & Kensinger E.A. (2009). Phenomenological characteristics of emotional memories in younger and older adults. Memory Memory, 17(5), 528-543.

Murchison, C.F., Zhang, X.Y., Zhang, W.P., Ouyang, M., Lee, A., & Thomas, S.A. (2004). A distinct role for norepinephrine in memory retrieval. Cell, 117(1), 131- 143.

Nairne, J. S. (2007). The foundations of remembering : Essays in honor of Henry L. Roediger III. New York: Psychology Press.

Nathan, S. V., Griffith, Q. K., McReynolds, J. R., Hahn, E. L., & Roozendaal, B. (2004). Basolateral amygdala interacts with other brain regions in regulating glucocorticoid effects on different memory functions. Annals of the New York Academy of Sciences, 1032, 179-182.

Nobata T., Hakoda Y., & Ninose Y. (2010). The functional field of view becomes narrower while viewing negative emotional stimuli. Cognition and Emotion, 24(5), 886-891.

Nummenmaa, L., Hyönä, J., & Calvo, M.G. (2006). Eye movement assessment of selective attentional capture by emotional pictures. Emotion, 6(2), 257-268.

Obminski, Z., Wojtkowiak, M., Stupnick, R., Golec, L., & Hackney, A. (2005).Effect of acceleration strss on salivary coritsol and plasma cortisol and testosterone levels in cadet pilots. Journal of Physiology and Pharmacology, 48(2), 193-200.

Ohman, A. (2005). The role of the amygdala in human fear: Automatic detection of threat. Psychoneuroendocrinology, 30(10), 953-958.

Okado, Y., & Stark, C. (2005). Neural activity during encoding predicts false memories created by misinformation. Learning and Memory, 12(1), 3-11.

86 87

Payne, J., Jackson, E., 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.

Payne, J., Jackson, E., Ryan, L., Hoscheidt, S., Jacobs, J., & Nadel, L. (2006). The impact of stress on neutral and emotional aspects of episodic memory. Memory, 14(1), 1- 16.

Perogamvros, B.G., Keevil, D., Ray, W., & Trainer, P. J. (2010). Salivary cortisone is a potential biomarker for serum free cortisol. Endocrine Care, 95(11), 4951-4958.

Pessoa, L., & Ungerleider, L. G. (2004). Section III - perception and attention - 12 neuroimaging studies of attention and the processing of emotion-laden stimuli. Progress in Brain Research, 144, 171-182.

Pruessner J.C., Dedovic, K., Khalili-Mahani, N., Engert V., Pruessner M., Buss C., Renwick, R., Dagher, A., Meaney, M., & Lupien, S. (2008). Deactivation of the limbic system during acute psychosocial stress: Evidence from positron emission tomography and functional magnetic resonance imaging studies. Biological Psychiatry, 63(2), 234-240.

Reisberg, D., & Hertel, P. (2004). Memory and emotion. Oxford; New York: Oxford University Press.

Richardson, M.P., Strange, B.A., & Dolan, R.J. (2004). Encoding of emotional memories depends on amygdala and hippocampus and their interactions. Nature Neuroscience, 7, 278-285.

Richter-Levin G. (2004). The amygdala, the hippocampus, and emotional modulation of memory. Neuroscientist, 10(1), 31-39.

Robin, P., Predine, J., & Milgrom, E. (1978). Assay of unbound cortisol in plasma. Journal of clinical endocrinology and metabolism, 46(2), 277-283.

Roozendaal, B. (2002). Stress and memory: Opposing effects of glucocorticoids on memory consolidation and memory retrieval. Neurobiology of Learning and Memory, 78(3), 578-595.

Roozendaal, B., Barsegyan, A., & Lee, S. (2008). Adrenal stress hormones, amygdala activation, and memory for emotionally arousing experiences. Progress in Brain Research. 167, 79-95.

Roozendaal, B., Okuda, S., de Quervain, D. J. F., & McGaugh, J. L. (2006a). Glucocorticoids interact with emotion-induced noradrenergic activation in influencing different memory functions. Neuroscience, 138(3), 901-910.

Roozendaal, B., Okuda, S., Van der Zee, E. A., & McGaugh, J. L. (2006b). Glucocorticoid enhancement of memory requires arousal-induced noradrenergic activation in the basolateral amygdala. Proceedings of the National Academy of Sciences, 103(17), 6741-6746.

Schaefer, A., Pottage, C.L., & Rickart, A.J. (2011). Electrophysiological correlates of remembering emotional pictures. NeuroImage, 54(1), 714-724.

Schmidt, S. R. (2002). Outstanding memories: The positive and negative effects of nudes on memory. Journal of Experimental Psychology: Learning, Memory, and Cognition, 28(2), 353-361.

Segal, S.K., & Cahill, L. (2009). Endogenous noradrenergic activation and memory for emotional material in men and women. Psychoneuroendocrinology, 34(9), 1263- 1271.

Sharot, T., Martorella, E. A., Delgado, M. R., & Phelps, E. A. (2007). How personal experience modulates the neural circuitry of memories of September 11. Proceedings of the National Academy of Sciences, 104(1), 389-394.

Shirtcliff, E., Granger, D., & Schwartz, E., & Curran (2001). Use of salivary biomarkers in biobehavioral research: cotton-based sample collection methods can interfere with salivary immunoassay results. Psychoneuroendocrinology, 26(2), 165-173.

Strange, B. A., & Dolan, R. J. (2004). Beta-adrenergic modulation of emotional memory- evoked human amygdala and hippocampal responses. Proceedings- National Academy of Sciences, 101(31), 11454-11458.

Talarico, J.M., Berntsen, D., & Rubin, D.C. (2009). Positive emotions enhance recall of peripheral details. Cognition and Emotion, 23(2), 380-398.

Talarico, J. M., LaBar, K. S., & Rubin, D. C. (2004). Emotional intensity predicts autobiographical memory experience. Memory & Cognition, 32(7), 1118-1132.

Talmi, D., Anderson, A.K., Riggs, L., Moscovitch, M., & Caplan, J.B. (2008). Immediate memory consequences of the effect of emotion on attention to pictures. Learning and Memory, 15(3), 172-182.

Tollenaar, M.S., Elzinga, B.M., Spinhoven, P., & Everaerd, W. (2008a). Long-term outcomes of memory retrieval under stress. Behavioral Neuroscience, 122(3), 697- 703.

Tollenaar, M.S., Elzinga, B.M., Spinhoven, P., & Everaerd, W. (2008b). The effects of cortisol increase on long-term memory retrieval during and after acute psychosocial stress. Acta Psychologica, 127(3), 542-552.

88 89

Tollenaar, M.S., Elzinga, B.M., Spinhoven, P., & Everaerd, W. (2009a). Autobiographical memory after acute stress in healthy young men. Memory, 17(3), 301-310.

Tollenaar, M.S., Elzinga, B.M., Spinhoven, P., & Everaerd, W. (2009b). Immediate and prolonged effects of cortisol, but not propranolol, on memory retrieval in healthy young men. Neurobiology of Learning and Memory, 91(1), 23-31.

Touryan, S. R., Marian, D. E., & Shimamura, A. P. (2007). Effect of negative emotional pictures on associative memory for peripheral information. Memory, 15(2), 154- 166. van Stegeren, A. H., Goekoop, R., Everaerd, W., Scheltens, P., Barkhof, F., Kuijer, J. P., & Rombouts, S. (2005). Noradrenaline mediates amygdala activation in men and women during encoding of emotional material. NeuroImage, 24(3), 898-909. van Stegeren, A.H., Kindt, M., Roozendaal, B., Wolf, O.T., & Joëls, M. (2010). Interacting noradrenergic and corticosteroid systems shift human brain activation patterns during encoding. Neurobiology of Learning and Memory, 93(1), 56-65. van Stegeren, A. H., Wolf, O. T., Everaerd, W., Scheltens, P., Barkhof, F., & Rombouts, S. (2007). Endogenous cortisol level interacts with noradrenergic activation in the human amygdala. Neurobiology of Learning and Memory, 87(1), 57-66.

Vining, R., & McGinley, R. (1987). The measurement of hormones in saliva: possibilities and pitfalls. Journal of Steroid Biochemistry, 27(1-3), 81-94.

Vining, R., McGinley, R., Maksvytis, J., & Ho, K. (1983). Salivary cortisol: a better measure of adrenal cortisol function than serum cortisol. Annals of Clinical Biochemistry, 20(Pt 6), 329-335.

Vyas, A., Mitra, R., Rao, B.S., & Chattarji, S. (2002). Chronic stress induced contrasting patterns of dendritic remodeling in hippocampal and amygdaloid neurons. Journal of Neuroscience, 22(15), 6810-6818.

Wang, J., Rao, H., Wetmore, G. S., Furlan, P. M., Korczykowski, M., Dinges, D. F., & Detre, J. (2005). Perfusion functional MRI reveals cerebral blood flow pattern under psychological stress. Proceedings of the National Academy of Sciences, 102(49), 17804-17809.

Waring, J.D., Kensinger, E.A., Payne, J.D., & Schacter, D.L. (2010). Impact of individual differences upon emotion-induced memory trade-offs. Cognition and Emotion, 24(1), 150-167.

Whembolua, G., Granger, D., Singer, S., Kivlighan, K., & Marguin, J. (2006). Bacteria in the oral mucosa and its effects on the measurement of cortisol, dehydroepiandrosterone, and testosterone in saliva. Hormones & Behavior, 49(4), 478-483.

Yerkes, R. M., & Dodson, J. D. (1908). The relation of strength of stimulus to rapidity of habit formation. Journ.Comp.Neruol.& Psychol., , 459-482.