Mara Mather - US grants

The proposed research focuses on people's memory for past choices. Memory for choices has the potential
to affect people's self-concept, well-being, attitudes, and future decisions. Regret for options not taken can cast a shadow, whereas satisfaction at having made the right choice can make a good outcome seem even better. Our memories of past options provide essential information about our tastes and experience in the context of current and future decisions. These memories have implications for other people as well (e.g., as we give advice about others' choices or participate in collective decisions). Understanding the causes and consequences of distortion in memory for choices, and possible ways to avoid such distortion, can have far- reaching benefits in domains ranging from therapy and consumer education to medical decision making and the law.

Theoretical explanations of memory distortion share many common themes and emphasize the constructive
nature of memory (e.g., Bransford & Johnson, 1973; Loftus, 1979; Reyna & Lloyd, 1997; Roediger &
McDermott, 2000; Schacter, Norman, & Koutstaal, 1998). The current proposal uses the source monitoring
framework (e.g., Johnson, Hashtroudi, & Lindsay, 1993), one of the most well-developed theoretical
approaches to memory and its distortions, to help predict how factors that have been shown to be associated
with memory distortion will affect memory for choices. Preliminary work (e.g., Mather, Shafir, & Johnson,
2000) indicates that people have choice-supportive biases in memory, favoring the option selected over the
one rejected. This choice-supportive tendency goes beyond simply selectively remembering the strengths of
chosen options and the weak attributes of rejected alternatives. People also tend to make errors that favor
their chosen option, misattributing positive features to their chosen option and negative features to the
rejected option. The studies in the first section of the proposal investigate the mechanisms underlying such
choice-supportive memory distortion. The second section investigates some potential effects of choice-
supportive memory on well-being and on future decisions, and the final section investigates how choice-
supportive biases may be avoided or attenuated.

Remembering choices plays a central role in making future decisions. A better understanding of memory
distortion and its influences may help improve decision making in important domains. Better insight into the
role of memory also promises to add to our understanding of intuitive decision making practices and
emergent biases. Finally, an understanding of how memory of past decisions affects our satisfaction, goals,
perceived responsibility, and remembered experience may contribute to well-being in numerous ways. The
proposed studies are intended to shed light on how past choices may be remembered or misremembered, on
how these tendencies may be altered, and on the hedonic consequences that such alterations might have.

DESCRIPTION (provided by applicant): Memory loss is one of older adults' greatest concerns about the aging process. Yet it is possible that some of what passes for simple forgetfulness may result from a complex set of mental trade-offs, a balancing act that shifts priorities from knowledge acquisition in youth to maintenance of a positive emotional state in older years. Across the adult life span, emotion-related goals gain in importance and this shift in focus may affect what older adults are likely to notice and remember. Indeed, the preliminary studies outlined in this application reveal emotionally-oriented biases in older adults' attention and memory, including diminished memory for negative information relative to positive information. We hypothesize that, motivated by emotional goals, older adults recruit executive processes to maintain pleasing and inhibit unpleasant information when storing, rehearsing, or retrieving memories. In this application, we outline a series of studies to test both this hypothesis and other competing accounts of older adults' diminished memory for negative information. Expt. 1 examines the time course of attention to see whether older adults are impaired at detecting threatening stimuli or actively avoid paying attention to anything negative. Expt. 2 tests whether age differences in physiological arousal can account for the changes in emotional memory. Expt. 3 tests emotional memory while participants also complete another task to examine whether cognitive resources are required to diminish memory for negative information. Expt. 4 examines whether older adults who successfully regulate emotion in their everyday lives show more forgetting of negative information and better executive functioning than older adults who are poor regulators. Expt. 5 examines whether older adults are better at suppressing negative than neutral or positive information. Expt. 6 tests whether older adults use memory as a mood repair tool when in a bad mood. In addition, throughout these studies, we examine the role of circadian arousal rhythms (which differ for older and younger adults) in emotional memory. This series of experiments should help us understand whether the diminished potency of negative information as we age is a fortuitous by-product of age-related decline or whether it is the result of increased emotion regulation. In addition, it should help us understand how memory processes can contribute to emotional well-being.

[unreadable] DESCRIPTION (provided by applicant): Stress affects brain systems involved in risk and reward processing and influences learning and retention of emotionally arousing information. Yet, despite the fact that important decisions and stress often go hand in hand, little is known about the influence of stress on decision processes. This proposal outlines studies to investigate the effect of acute stress on learning and performance in investment tasks for younger and older adults. Stress is likely to have a significant impact on both younger and older adults' decision processes, but the effects may differ for the two groups. Previous research indicates that as people age, they are increasingly likely to prioritize emotion regulation and that this leads to age differences in attention and memory, with older adults showing a positivity (or anti-negativity) effect. Because of this increased focus on affect regulation, older adults are more likely than younger adults to be disproportionately influenced by reward outcomes (rather than loss outcomes) in their decisions. Stress should to amplify (Experiment 1) or diminish (Experiment 2) these age differences, depending on the timing of the stress. In addition, both age and stress should influence decision makers' ability to abandon a no-longer successful strategy and learn a new one (Experiment 3). Understanding the ways in which stress can influence decision processes differently for older and younger adults should promote development of ways to avoid decision biases and errors. [unreadable] [unreadable] Important decisions, such as those regarding personal finances and health care, often involve substantial stress, especially among older adults. The studies in this proposal will provide important information about how stress affects learning and decisions and how those effects may vary with age. [unreadable] [unreadable] [unreadable]

DESCRIPTION (provided by applicant): Our attention, memory and the decisions we make are affected by the stresses of the moment and our emotional goals. How emotion and stress affect these cognitive efforts changes with age, yet the underlying mechanisms of these age differences are not clear. This proposal for a Career Development Award outlines two studies that combine structural and functional magnetic resonance imaging (MRI) to examine how age affects the brain mechanisms that coordinate emotion and cognition. The first study tests the role of cognitive control in older adults'positivity effect by scanning younger and older participants while they watch negative, positive, or neutral pictures either while distracted or not. Previously, we have found that older adults ignore negative stimuli and attend to positive stimuli when not distracted, but that distraction eliminates this positivity effect. Study 1 tests the hypothesis that older adults engage prefrontal-based cognitive control mechanisms to modulate the amygdala's response to emotional stimuli. Study 2 examines how stress affects decision making differently for younger and older adults. Previous results indicate that acute stress reduces risk taking in older adults but not younger adults. In this study, brain activity during rest and a risky decision task will be compared in a stress and a control condition. The role of chronic stress in modulating brain activity will be examined in both studies. Substantial training in neuroanatomy and analyzing structural and functional MRI data will occur via tutorials, workshops and graduate-level courses and seminars. Understanding the brain mechanisms of emotion-cognition interactions is a key goal of the PI;this Career Development Award will give her expertise in MRI and neuroanatomy that will help her develop and test theories about age differences in the mechanisms of emotion, stress and cognition. RELEVANCE (See instructions): Understanding age differences in the brain mechanisms of how emotion and stress affect cognitive processing is key to understanding how people can maintain mental sharpness and emotional well-being throughout life.

DESCRIPTION (provided by applicant): Stress is part of daily life; in a national study 40% of respondents reported experiencing stress that day. We know from many studies that stress can affect health, memory, and well-being. But does it affect how we make decisions? And if so, what are the implications-do we become more or less risk averse? Are we more influenced by the potential downsides and less by potential upsides to an option than we would be otherwise? Making decisions involves a complex yet flexible set of brain mechanisms that compute the potential value of options and adjust that value based on the level of uncertainty about the outcome. In this proposal, we examine the effects of acute stress on the basic elements of decision making: the neural signals that predict and represent value, integrate the positive and negative aspects of options, and assess risk based on predicted volatility and uncertainty of options. We test two main hypotheses. The first is that stress makes rewarding stimuli even more attractive and salient, via dopaminergic effects in the ventral striatum. The second is that stress affects risk processing via the insula, but that it has different effects depending on one's sex and age. Some of our most important decisions must be made under stressful circumstances; thus it is paramount that we understand the basic mechanisms of how stress changes the way we evaluate decision options.

DESCRIPTION (provided by applicant): At the core of our sense of self and personal history are emotional memories. Although emotional or stressful experiences tend to be memorable, emotional arousal can also impair various aspects of memory. In recent years, research into arousal and memory has focused on the key role of the amygdala in enhancing perception and memory of emotionally arousing stimuli. But enhanced memory for arousing information is only part of the story-there is also abundant evidence that arousal enhances some aspects of memory while impairing other aspects. The novel arousal-biased competition (ABC) theory tested in this application states that arousal enhances high-priority neural representations but suppresses low-priority neural representations of stimuli. This can account for a broad range of arousal-induced selectivity effects in memory, including findings that initially appear contradictory such as retrograde amnesia vs. enhancement for neutral stimuli preceding arousing stimuli. The proposed studies will test the effect of arousal on low- and high-priority representations and examine how the effects of arousal differ for younger and older adults. Priority will be systematically manipulated both by bottom-up perceptual salience and by top-down goal relevance; studies will examine the nature of age differences in the effects of arousal on short-term memory, long-term memory and memory binding. One key question is whether arousal's enhancement of high priority and suppression of low priority information are two outcomes from the same competitive process, or whether they arise from independent mechanisms. Because older adults show a deficit in suppression, their performance will help address this question. Although the focus of the application is emotional memory, we are likely to learn about the basic mechanisms of enhancement vs. suppression in aging. Despite a long history of findings that older adults show inhibitory deficits, there have been only a few studies examining the neural mechanisms of older adults' impaired suppression. Our studies examine whether their suppression deficits are greater when stimuli priority is determined by top-down goals or by bottom-up salience and how suppression deficits affect long-term memory.

DESCRIPTION (provided by applicant): Almost everyone has at least one social identity that is associated with a negative stereotype. For some people it is their age, for others it is their race, weight, religion, socioeconomic status or political affiliation. Problematically, according t stereotype threat theory, when people encounter these negative stereotypes they often underperform compared to their potential, and in doing so inadvertently confirm the stereotype. For example, older adults are stereotyped as having poor memory abilities. When this stereotype becomes salient to older adults their memory performance tends to decrease. This has serious clinical implications. In one study, stereotype threat increased the proportion of older adults scoring below the clinical cut-off for dementia from 14 to 70%. Given that 20% of Americans will be over the age of 65 by 2030, and that annual dementia screenings are now covered by Medicare, it is important to understand why stereotype threat impairs older adults' performance and how stereotype-threat-related performance deficits can be ameliorated. In Aim 1, we test the role of regulatory focus in contributing to older adults' stereotype threat effects. Regulatory focus theory leads to the counterintuitive hypothesis that although stereotype threat impairs performance when the task emphasizes gains, it should improve performance when the task emphasizes losses. Our preliminary data supports this; stereotype threat impaired older adults' memory when remembering led to gains, but improved memory when forgetting led to losses. In this application we test the generalizability of these results by examining how stereotype threat affects older adults' performance in other cognitive (e.g., verbal fluency, visuospatial skills, decision quality) and non-cognitive (i.e., screened vision and hearing) domains. Many of these domains have not previously been examined. If our regulatory focus predictions are consistently supported, it would suggest that emphasizing the importance of avoiding errors during assessments could be a simple, no-cost clinical intervention to eliminate stereotype threat effects. In Aim 2, we examine the contribution of physiological arousal to older adults' stereotype threat effects. Recently we proposed an arousal-biased competition (ABC) theory. According to our ABC theory, arousal increases the processing of high priority stimuli and decreases processing of low priority stimuli. Preliminary studies support this hypothesis. Of relevance to this application, based upon the regulatory focus theory we predict that under stereotype threat loss-related information is high priority and gain-related information is low priority. Thus, according to ABC theory, arousal should enhance the regulatory focus effects predicted above. Finally, Aim 3 is to examine the functional implications of stereotype threat for older adults. Thus, the domains in which we test our Aim 1 and 2 hypotheses each have relevance for older adults' health, practical, and financial well-being. We also test the hypothesi that genetic testing for Alzheimer's disease carries the risk of inducing stereotype threat and reducing memory performance in asymptomatic, healthy older adults.

? DESCRIPTION (provided by applicant): The attribution of value to prospects is a fundamental element of decision-making, as most day- to-day decisions involve comparing items. Studies on aging document behaviors reflecting difficulties in making comparisons between options, in particular when those options are complex. Given the growing complexity of economic products (insurance, savings, mortgages or even telephone plans), older adults may have difficulty making decisions that accurately reflect their underlying preferences. It is plausible that this difficulty is linked to age-related brain function decline within sectors of the lateral prefrontal cortex implicated in working memory and cognitive control. This study assesses age-related changes in how the brain computes value and makes value comparisons using a well-established economic paradigm, the generalized axiom of revealed preference (GARP) Task, that tests the internal consistency of a subject's preferences by offering repeated choices between bundles of goods. Our preliminary study suggests that aging is associated with greater GARP-Task inconsistency. Although the neural correlates of GARP inconsistency have not been directly established, indirect evidence suggests that the medial orbitofrontal cortex is important in all value-based decision-making, and that areas in the lateral prefrontal and parietal cortices (fronto-parietal network) are important for maintaining consistency in complex decisions (e.g., multi-attribute decisions). Aging is associated with structural and functional deficits withn the fronto-parietal network. Therefore, we believe that studying the neural correlates of the GARP-Task is a promising approach to investigate decision-making deficits in aging. We will recruit 45 young adults, and 45 old adults. Participants will complete the GARP-Task and an fMRI variant designed to isolate neural correlates of valuation of single items, of multiple instances of the same item (scaling) and of sets of distinct items (bundles). Brain activity wil be related to diagnosis and to variance in GARP-Task inconsistency. Given the prominence of age-related decline in working memory, we hypothesize that age will be associated with higher GARP-Task inconsistency and to deficits in conditions that require manipulation of value signals (scaling and bundles). We anticipate that these deficits will be associated with low recruitment within the fronto-parietal network and with reduced functional connectivity between this network and the medial OFC.

DESCRIPTION (provided by applicant): Stress often has detrimental effects on memory and cognition, effects that are especially challenging for older adults to sustain on top of existing age-related declines. In this project, we investigate whether estradiol (E2) protects against such effects. In humans E2 reduces the magnitude of the stress response and aids cognition. Further, in vitro and in vivo animal research shows that E2 protects the brain from the negative effects of stress hormones. Together, this pattern of results suggests the E2 may protect aging women's neural and cognitive integrity during times of stress. The current application tests this hypothesis by examining the effects of short-term E2 treatment on various systems affected by stress as well as testing a mechanism of action for E2 protection. First, we will examine the ability of a short-term E2 intervention to reduce physiological effects of stress, stress-induced impairments in working memory performance, and associated changes in brain activation. Work in our laboratory reveals that post-menopausal women with high salivary E2 levels as a result of taking E2 supplements release less of the stress hormone, cortisol, in response to a stressor than women with low salivary E2 levels. We also found that working memory performance in the high-E2 women was unhindered by stress, whereas low-E2 women performed significantly worse under stress than under control conditions. As a result of the E2-related reduction in stress hormone release we expect to find that E2 will be associated with 1) dampening the stress-induced changes in hippocampal cerebral blood flow and bilateral connectivity at rest during stress and control conditions, 2) smaller stress-induced decreases in heart-rate variability, and 3) hippocampal and prefrontal cortex activity while playing a working memory game. Second, we will test a potential mechanism of action for estradiol protection against stress. We hypothesize E2 limits vulnerability of the hippocampus to the effects of stress, allowing the hippocampus to effectively shut down the stress response, which curtails the levels of cortisol available and the amount of time cortisol is available to act on neural tissue. To test this we will compare estradiol and placebo groups on hippocampal cerebral blood flow and bilateral functional connectivity at rest, under stress and control conditions. These measures will be correlated with all estradiol levels, cortisol response, HRV, and working- memory-related brain activity and performance. The proposed research aims to uncover whether E2 can in fact reduce the negative effect of stress on memory in post-menopausal human females, as well as the brain mechanisms involved in this protection against stress. This research will further inform the medical field on the effects of E2 on stress and memory, which could lead to better guidance and advice for patients seeking information on E2 treatment during or after menopause.

ABSTRACT Success requires focus and the ability to screen out distractions. This is especially important in arousing situations: When avoiding imminent danger, tackling a cognitive challenge or trying to achieve a desired goal, focusing on what has highest priority or is most salient is often helpful. We plan to test the hypothesis that when arousing or challenging situations activate the locus coeruleus (LC) in healthy younger adults, it enhances processing high priority or highly salient items but impairs processing of less active competing representations. The framework behind this hypothesis is the first to explain how arousal both enhances and impairs attention and memory and this project will be the first to systematically examine the relationship between LC structural decline and cognitive function in normal aging and in Alzheimer's disease. We predict that, in older adults, the LC has a less targeted effect than it does in younger adults, meaning that arousal still has a significant impact on older adults' cognitive processing but that it is less likely to increase the selectivity of their attention or the specificity of their memories. Furthermore, we predict that in Alzheimer's disease, declines in structural connectivity will be extensive enough to impair both the excitatory and inhibitory effects of LC activity. This work to understand the role of the LC in cognitive function becomes particularly urgent in light of recent striking findings indicating that the LC is the first place in the brain that sporadic (or late-onset) Alzheimer's related tau pathology emerges, and that by young adulthood, most people have at least some tau pathology in the LC. Here we use neuromelanin-weighted structural MRI images and diffusion tensor imaging structural connectivity measures to examine how LC integrity relates to function on a variety of tasks that assess cognitive selectivity. We will test healthy older adults, older adults with late-onset Alzheimer's and younger and middle-aged adults with genetic subtypes of Alzheimer's disease (due to mutations in the PSEN1 and APP genes) that lead to early onset of the disease.

ABSTRACT Having a heart with a steady beat is not optimal. The brain sends signals about various body states including blood pressure and breathing that contribute oscillatory rhythms to the speed of heartbeats and increase heart rate variability (HRV). For reasons that are not yet clear, having greater HRV at rest is associated with better emotion regulation and well-being. People with higher HRV tend to be less anxious, less depressed, less hostile and produce more context-appropriate emotional responses. One potential explanation for the relationship between HRV and emotion regulation is that the same set of brain regions regulates autonomic states and emotions, and so both HRV and emotion regulation reflect the general health and efficacy of this central autonomic network in the brain. However, it appears that the influences do not just flow from the brain to the heart. Paced breathing at the resonance frequency of the heart rate-baroreceptor feedback loop (around 6 breaths/min, a pace often attained during meditative practice) stimulates resonance characteristics of the cardiovascular system and so increases total HRV amplitude dramatically. Recent studies using this approach have shown that increasing HRV during short daily sessions can improve longer-term emotional outcomes. But it is unclear why episodes of high HRV have a positive impact. This project would be the first to examine the brain mechanisms of these effects, testing the hypothesis that episodes of high HRV induced by resonance frequency breathing lead to positive outcomes because they induce dynamic blood flow oscillations in brain regions that monitor and regulate physiological body states. Our experimental manipulation will be a 5-week protocol with random assignment to either a daily session with paced breathing at resonance frequency or one of two control conditions. We test the hypothesis that resonance frequency breathing will enhance measures of emotional well-being, resting state functional connectivity among brain regions involved in emotion regulation, and flexible up- and down-regulation of the amygdala during emotional experience. Furthermore, we test the hypothesis that these outcomes will be mediated by blood flow variability during paced breathing rather than by alternative mechanisms. The expected findings would indicate that HRV is more than just an indicator of health, with an active role in stimulating brain regions in the central autonomic network to improve their coordination and function.

Abstract This USC multidisciplinary research training program in Gerontology provides predoctoral and postdoctoral training within key disciplines in aging. Trainees become expert in one aspect of aging while also learning about the theory and methodological approaches of related disciplines so they can incorporate multidisciplinary thinking and models in their own research. Predoctoral training can be in Biology, Gerontology, Psychology, Neuroscience, Policy, and Medicine. Postdoctoral students can be appointed in these fields as well as Economics, Neurology, and several multidisciplinary research institutes. The program focuses on physical and cognitive health and the lifecycle circumstances influencing health. Current emphases of the program reflect the research strengths of current faculty: cognitive decline in aging and Alzheimer disease and related disorders; emotional change associated with the aging process; physical and functional change across the lifespan; the role of social, contextual, environmental and institutional factors in promoting or delaying physical and cognitive changes; the biological pathways, including the role of genetic and epigenetic factors, promoting or delaying changes in physical and cognitive health. The training program is housed in the Davis School of Gerontology which is devoted to the study of aging; training is directed by a multidisciplinary faculty across the University who are linked in many ways because of their focus on aging research. The environment is rich with resources and research opportunities. Training is accomplished through completion of (1) disciplinary degree requirements for the Ph.D.; (2) participation in multidisciplinary courses on health and aging and research on Alzheimer disease; (3) training in appropriate methods and analytic techniques; (4) research experience within and across disciplines; (5) close mentoring relationships between trainees and one or more preceptors; (6) participation of trainees in training experiences such as courses, workshops and professional meetings; and (7) support of trainees to develop individual research profiles, publish papers, and make presentations at professional meetings. The program goal is to develop scholars with independent research careers who become leading scholars in in the science of aging. The program draws from a large pool of talented students. Almost all students who enter the program complete it successfully and the program has produced a number of leaders and emerging leaders in aging- related research, a number of whom are diverse scholars. Support is requested for 5 predoctoral trainees, who will generally be supported for 3 years, and 5 postdoctoral trainees, who will generally be supported for 2 years.