<%BANNER%>

Memory-Related Possible Selves: Exploring Age-Related Differences


PAGE 1

MEMORY-RELATED POSSIBLE SELVES: EXPLORING AGE-RELATED DIFFERENCES By ALISSA DARK-FREUDEMAN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Alissa Dark-Freudeman

PAGE 3

iii ACKNOWLEDGMENTS I would like to thank my advisor and mentor, Robin Lea West, for her guidance and encouragement. I would also like to thank my friends and family for their support. I thank my parents for their kind words and continuing interest in my education. And last, but certainly not least, I would like to thank my husband and friend, Jason Thomas Fleming, for his patience and assistance throughout this journey. As I am always fond of saying, we have only just begun.

PAGE 4

iv TABLE OF CONTENTS Page ACKNOWLEDGMENTS..................................................................................... iii LIST OF TABLES............................................................................................... vi ABSTRA CT ....................................................................................................... vii CHAPTER 1 IN TRODUCTION ..........................................................................................1 The Nature and Relevance of Possible Selv es..............................................2 Current Research on Possible Selves...........................................................3 Information Processi ng and Behav ior .....................................................4 Contextual Effects on Possible Selves....................................................9 Differences and Changes in Possible Selves with Age ................................12 Changes in Memo ry Belie fs ........................................................................ 13 General Beliefs About Me mory .............................................................14 Personal Beliefs About Me mory............................................................19 Memory-related Poss ible Selves.................................................................25 Methods Used to Measur e Possible Selves ................................................ 26 2 STUDY 1 INTRODUCTION .........................................................................29 3 STUDY 1 METHODS..................................................................................32 Partici pants .................................................................................................32 Measures....................................................................................................33 Possible Selves ....................................................................................33 Memory Beliefs.....................................................................................34 Procedur e................................................................................................... 39 4 STUDY 1 RESULTS ...................................................................................42 Spontaneous Se lves ...................................................................................42 Selves Endorsed on Cl osed-ended Ques tionnaire ......................................44 Additional A nalyses .....................................................................................48

PAGE 5

v 5 STUDY 2 INTRODUCTION .........................................................................64 6 STUDY 2 METHODS..................................................................................67 Partici pants .................................................................................................67 Procedur e................................................................................................... 67 7 STUDY 2 Results........................................................................................69 8 DICUS SION................................................................................................70 APPENDIX A CLOSED-E NDED POSS IB LE SE LV ES MEASURES..................................81 B ME MORY VA LUE QUESTIONNA IRE.........................................................89 LIST OF RE FE RE NCES ...................................................................................90 BIOGRAPHICAL SKETCH................................................................................94

PAGE 6

vi LIST OF TABLES Table page 3-1. Mean Education and Health Ratings .........................................................40 3-2. Percentage of Agreem ent Between Independent Rate rs ...........................41 4-1. Percentage of S pontaneous Memory Selves R eported by A ge Group .......51 4-2. Type of Spontaneous Memo ry Selves by Age Gr oup .................................52 4-3. Spontaneous C ognitive Selves by Age Group ...........................................53 4-4. Mean Number of Memory Selv es Endorsed on Clos ed Questi onnaire .......54 4-5. Intercorrelations Between Dependent Subsca les .......................................55 4-6. Mean Memory Belief Ratings by Age Gr oup ..............................................56 4-7. Analysis of Variance for Dependent Su bsca les by Type of Me mo ry Se lf Spontaneously R eported ...........................................................................57 4-8. Type of Memory Selves Reported By Age Group ......................................59 4-9. Mean Number of Selves by Age Gr oup .....................................................60 4-10. Mean Capability and Li kelihood Ratings : Fear ed..................................... 61 4-11. Mean Capability and Like lihood Ratings: Hoped-fo r.................................62 4-12. Mean Number of Goal Acti vities ..............................................................63 6-1. Mean Demographic Information for Pr imed and Unprimed Young Adults...68

PAGE 7

vii Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science MEMORY-RELATED POSSIBLE SELVES: EXPLORING AGE-RELATED DIFFERENCES By Alissa Dark-Freudeman August 2004 Chair: Robin L. West Major Department: Psychology Thoughts about the self in the future are an important component of the current identity of any individual. Although we all have future-oriented thoughts, we each have our own unique map of avenues we wish to travel and dead ends we plan to avoid. Markus and Nurius refer to such thoughts about our selves in the future as possible selves. Age-related changes and differences in possible selves have been documented, as have age-related differences and changes in beliefs about memory. The current study extended research in both of these areas. This investigation sought to determine if memory-related possible selves exist, to examine the nature of these selves as hoped-for or feared, and to understand how the existence of these memory-related selves may differ with age. The study also examined the relationship between memory-related possible selves and a wide range of memory beliefs that have been shown to be affected by the aging process. This study also examined whether younger adults who

PAGE 8

viii were primed to think about memory and aging might be more likely to spontaneously generate memory-related selves. In general, the results of the current study show that older adults do have memory-related possible selves, whereas younger adults do not. Further, when these selves appear in an older individual’s repertoire, they are primarily feared selves like becoming senile or losing one’s mind. Primed younger adults were no more likely to report a memory-related self than unprimed younger adults. The relationship between the memory-related selves and memory beliefs reported by older adults approached significance and warrants additional investigation in the future.

PAGE 9

1 CHAPTER 1 INTRODUCTION We all think about the future. We have dreams and we have nightmares about the myriad of possibilities that await us somewhere off in the distance. Can we achieve our goals? Will we fail to make our dreams reality? What will happen to us? What will we be like in ten years, twenty years, or fifty years? Thoughts about the self in the future have become an interesting topic in recent years and are considered to be an important component of the current identity of any individual (Markus & Nurius, 1986). Although we all share these types of future-oriented thoughts, we each have our own unique map of avenues we wish to travel and of dead ends we plan to avoid. Markus and Nurius (1986) have called these thoughts about our selves in the future, possible selves. These include the hoped-for selves we are striving toward and the feared selves we wish to escape. Age-related changes and differences in possible selves have been documented (Cross & Markus, 1991; Hooker, 1992; Hooker, Fiese, Jenkins, Morfei, & Schwagler, 1996; Hooker & Kaus, 1992; Hooker & Kaus, 1994; Ryff, 1991), as have age-related differences and changes in memory performance (Dixon & Hultsch, 1999; Schaie, 1994) and in beliefs about memory (Berry, West, & Dennehey, 1989; Hultsch, Hertzog, & Dixon, 1987; Lachman, Bandura, Weaver, & Elloitt, 1995). The current study extends research in these areas. First, this research will focus on memory-related possible selves to determine whether or not older and younger adults have possible selves in the

PAGE 10

2 memory domain. Second, this study will determine whether or not these selves are primarily hoped-for or feared. Third, this study will investigate the relationship between aging, beliefs about memory, and memory-related possible selves. Fourth, the impact of priming will be examined. The following review discusses what possible selves are and why they are of interest, age-related differences and changes in possible selves and memory beliefs, and how the study of possible selves may contribute to our understanding of memory and aging. The Nature and Relevance of Possible Selves Markus and Nurius defined (1986) possible selves as elements of the selfconcept that represent what an individual could become, would like to become, or is afraid of becoming. A possible self is an idea we have created that represents what we might be like in the future. These ideas are part of the selfconcept. These ideas or selves can be positive, negative, or neutral. They can be concrete and realistic and they can be vague and improbable. They can be relatively close in the future or quite distant, and they can contrast and complement other possible selves. Although each individual has certain selves that are central to his or her identity, the meanings of these selves may change over time (Cross & Markus, 1991). Possible selves are not set in stone. Possible selves change as each individual changes: as some selves are achieved, new selves will take their place, whereas other selves will be revised or dropped altogether. Which selves will be dominant depends on the individual and the context he or she experiences at any given point in time (Markus & Wurf, 1987). This online

PAGE 11

3 shifting set of selves is what Markus and Nurius (1986) call the working selfconcept. On-line selves will be selected from an individual’s entire repertoire of possible selves, according to environmental demands. Thus, possible selves are sensitive to the environment and to the individual. These personally tailored, environmentally responsive selves motivate and guide actions and emotions in personally meaningful ways (Markus & Nurius, 1986). Possible selves guide behavior in several ways. Shifts in possible selves create shifts in information processing (Kato & Markus, 1993), allowing information that is related or relevant to a particular self to be processed more quickly and efficiently (Cross & Markus, 1994; Ruvolo & Markus, 1992). Changes in possible selves result in changes in levels of motivation and feelings of self-efficacy (Cameron, 1999; Cross & Markus, 1991; Hooker, 1992). Possible selves have also been related to psychological well-being and life satisfaction (Cross & Markus, 1991; Ryff, 1991; Smith & Freund, 2002). Further, changes in possible selves have been shown to result in changes in length of persistence and amount of effort expended on specific tasks (Ruvolo & Markus, 1992). Current Research on Possible Selves This section briefly reviews current research on possible selves. First, the selected studies emphasize the influence of possible selves on information processing, including the effects possible selves exert on levels of persistence and effort. Lastly, the effects of context on possible selves are highlighted, in particular, how life course and age may affect our repertoire of possible selves.

PAGE 12

4 Information Processing and Behavior Ruvolo and Markus (1992) completed a series of three studies to determine if possible self manipulations affected persistence and effort on tasks, and accessibility of possible selves. In the first study, participants were exposed to one of three imagery manipulations: success, failure, or positive affect. Participants in the first two conditions were asked to imagine and write about themselves in the future, experiencing either success or failure as a direct consequence of their own actions, whereas participants in the positive affect condition read a neutral passage and received candy. Persistence was measured as the amount of time a participant spent copying numbers with his or her non-dominant hand. Effort was measured as a participant’s accuracy on a circling e’s task. The success group persisted most, followed by the positive affect and failure groups respectively. The success group also put forth more effort than the positive affect and failure groups. In the second study, participants were exposed to one of four imagery manipulations: success, failure, positive affect and other’s success. Three of the four manipulations were the same as in study one. The new group was asked to imagine another person being successful, to determine if thoughts of success in general might be responsible for the results of the prior study. Immediately after the imagery manipulation, participants completed a computerized self-description task in which they answered possible for me or not possible for me for a selection of words and phrases, and response latencies were measured. Although the groups exhibited no differences in the types of words they selected as possible or not possible, their response latencies did differ indicating

PAGE 13

5 that “categories of possible selves were differentially accessible” (Ruvolo & Markus, 1992, p. 108) for all four groups. Participants in the success condition exhibited faster endorsements of positive items and faster rejections of negative items. Those in the other’s success group were quick to endorse positive items; however they did not differ from the failure group in response latencies for negative items. Those in the positive affect group revealed response latencies similar to the failure response latencies for both positive and negative items. No differences were found between the four groups on response latencies for control items. In the third and final study, participants were exposed to one of four imagery manipulations: success work, success luck, failure work, and failure luck. Again, participants completed a computerized self-description task and in addition to this, they completed an achievement value questionnaire. Participants also completed performance tasks measuring effort and persistence. Effort was measured by performance on a mental arithmetic test and persistence was measured as time spent copying numbers with a non-dominant hand. No differences were found on the achievement scale, suggesting that all participants valued achievement to the same degree. On the effort task, the success work group answered a higher percentage of problems correctly compared to the other three groups, but this difference was not significant. The success work group persisted longest on the task, followed by the two luck conditions and the failurework condition, respectively. Again, no differences were found in the types of words endorsed as possible or not possible among the four groups. Participants

PAGE 14

6 in the success work group did endorse positive and reject negative words more quickly than the other three groups, whereas the failure work group was the slowest to respond to both positive and negative items. The two luck conditions fell between the success work and failure work conditions. These results indicate that our online set of possible selves can be manipulated. Further, changes in an individual’s online set of selves affect behavior in meaningful ways. Specifically, changes in information processing occur in response to changes in an individual’s online set of possible selves. These changes can also affect the level of persistence and effort an individual is willing to put forth. Kato and Markus (1993) presented college students with negative and positive adjectives and phrases that were encoded on one of four conditions: semantic, ability, actual self or possible self. In the semantic condition, participants determined whether adjacent items were similar in meaning. In the ability condition, participants provided a synonym for each target item. In the actual-self condition, participants answered whether each item was descriptive of their current self. In the possible-self condition, participants answered whether each item would be descriptive of a future self. Possible-self orientation was established by asking each participant to think about an important hoped-for self that they often thought about attaining. After the encoding activity, participants completed a surprise recall test of the items. Those in the actual and possible selves conditions recalled more items than those in the semantic and ability

PAGE 15

7 conditions. Participants also recalled more positive items than negative items across all four conditions. These results suggest that items related to an individual’s self-concept are more likely to be remembered by that individual. Thus, items relevant to a possible self, especially a hoped-for self, are more likely to be attended to and remembered. Cross and Markus (1994) investigated the effects of an individual’s schematicity on performance and feedback. Self-schemas are identities that are central to an individual. These schemas are core self-representations that define who we are. Such schemas would be an important influence on the possible selves that an individual might endorse, especially those that are spontaneously mentioned. For both studies, a pre-selection questionnaire identified individuals that were either aschematic or schematic for logical problem solving abilities. A schematic individual has an organized and potentially automated way of perceiving and responding to a particular stimulus or situation, in this case, logical problem solving. In this study, schematic individuals indicated that they perceived themselves as competent problem solvers and that this ability was important to them. Aschematic individuals did not indicate that they had any exceptional problem solving abilities, nor did they report that this ability was particularly important to them. Participants also completed a logic activity, after which they answered questions about the test and indicated their perceptions about the difficulty of the logic test.

PAGE 16

8 In the first study, both schematic and aschematic participants reported similar levels of self-esteem and optimism, and both performed equally well on the logic test. Schematic adults reported enjoying the logic test more than aschematic adults. They also indicated that the activity was easier and that they felt more control over their results on the activity. Further, schematic individuals endorsed more positive logic-related items as possible for me than did aschematic individuals. Aschematic individuals were also significantly faster to respond possible for me for the negative logic-related items. No other significant differences in response latencies were discovered. The second study investigated the effect of feedback on subsequent performance for schematic and aschematic individuals. Again, no differences were found in initial problem solving abilities. This study consisted of the same measures in the first study; however, feedback was given after the first logic test. Participants either received failure feedback or no feedback at all. Schematic individuals in the feedback condition responded possible for me more quickly for the positive logic-related terms than any other group. Aschematic individuals in the feedback condition responded possible for me significantly faster for the negative logic-related terms. No other significant differences in response latencies were discovered. Following the first test, a second logic test was given. Aschematic individuals in the no feedback condition performed significantly worse on the second logic test than did those individuals in the other three groups. Aschematic individuals in the failure feedback condition performed similarly to the

PAGE 17

9 schematic individuals in both conditions. Cross and Markus (1994) suggested that feedback acted as an external incentive for aschematic individuals to redeem and prove themselves capable. Without feedback, aschematic individuals lost interest. Therefore, this external push was necessary to compensate for the lack of positive logic-related self-conceptions and consequent lack of motivation these individuals face within the domain of logic. Conversely, schematic individuals relied on positive logic-related selfconceptions to cope with or discount negative feedback and maintain performance. In this situation, schematic individuals can be thought of as having a positive logic-related possible self. The existence of this positive logic-related self affected information processing, and behavior. This self allowed schematic individuals confronted with failure to shift their attention away from negative, toward more positive self-descriptors. Further, the existence of this possible self acted as a goal, motivating schematic individuals to persist during difficulty and maintain a higher level of effort. Although possible selves can influence information processing and behavior, possible selves can also be influenced by external circumstances or conditions. External influences include specific situations and broader contextual factors, such as life stage or age. These contextual factors call certain selves to action and fine tune existing repertoires of selves. Contextual Effects on Possible Selves Hooker et al. (1996) measured parenting possible selves between two different groups of parents in different stages of parenthood. The two groups of

PAGE 18

10 parents each completed an open-ended possible selves questionnaire. The parent-infant group consisted of individuals between the ages of 18 and 43 years old, whose first child was 12 months old or younger. The parent-preschool group consisted of individuals between the ages of 24 and 45 years old, whose oldest child was between 2 and 5 years of age. Both groups of parents listed occupational and parenting selves most frequently, in fact, two-thirds of all participants listed a parenting self as one of their most important selves. The parent-infant group was more likely to have hoped-for parenting selves than the parent-preschool group, while the parentpreschool group was more likely to have feared parenting selves than the parentinfant group, suggesting that the different life stages of the children created a context in which parents saw their future selves. A study by Hooker and Kaus (1994) investigated the possible selves of young and middle-aged adults to determine if health-related selves existed for these age groups. Young adults were between the ages of 24 and 39 years old, while middle-aged adults were between 40 and 59 years old. Both young and middle-aged adults completed an open-ended possible selves questionnaire followed by efficacy and goal activity questions. After the possible selves questionnaire, each group completed a health behavior measure (e.g., yearly physical examinations, exercise) and a health value questionnaire (e.g., If you don’t have your health you don’t have anything). Middle-aged adults reported significantly more health-related selves than young adults. Both groups were more likely to have a feared health-related self

PAGE 19

11 than a hoped-for one. For participants with a health-related self selected as most important, health value, perceived self-efficacy, and number of goal oriented activities (taken to avoid a feared health-related self) were all significant predictors of health behaviors. Perceived self-efficacy and number of goaloriented activities were the strongest predictors of health behavior scores. Hooker (1992) also investigated the possible selves of younger and older adults to determine if health-related selves existed for these age groups. Young adults were between the ages of 18 and 23 years old, while older adults were between 55 and 89 years old. Both younger and older adults completed an open-ended possible selves questionnaire, followed by efficacy and goal activity questions. After the possible selves questionnaire, each group completed a perceived health and health value questionnaire. Initially, Hooker reported that 71% of younger adults listed a health-related self, and 17% selected a health-related self as most important. However, these percentages included weight-related selves. For younger adults, weight-related selves are probably more associated with concerns about appearance than with concerns for health. When weight-related selves were removed, 64% of younger adults reported no health-related self at all. On the other hand, 86% of older adults reported a health-related self. Of these, 73% selected a health-related self as one of their most important selves. In general, participants with health-related selves scored higher on the health value measure; consequently, older adults scored higher on health value than young adults. Lastly, younger adults generated more hoped-for and feared selves than older adults.

PAGE 20

12 Again, life stage plays an important role in the differences and changes that occur in any persons’ set of possible selves. These shifting sets of selves will ultimately affect behavior in meaningful ways, prioritizing selves, ultimately motivating the current self into action. Differences and Changes in Possible Selves with Age Given the fluid nature of possible selves, recent studies have started to examine age-related differences and changes in possible selves, and several age-related differences have been found. Differences exist not only in the types of selves reported by age, but in the number of selves reported. Older adults consistently report fewer possible selves than younger adults (Cross & Markus, 1991; Hooker, 1992). This was true for the number of hoped-for and the number of feared selves. Older adults also report more specific, more realistic possible selves than younger adults (Cross & Markus, 1991; Hooker, 1992). In general, the possible selves older adults report are more in line with their current selves. They typically involve the continuation or maintenance of current activities and relationships, but also include room for continued learning and growth (Cross & Markus, 1991; Frazier, Hooker, Johnson, & Kaus, 2000; Hooker, 1992). Age differences in specific domains have also been found. Hooker and Kaus (1994) found that health-related selves begin to increase in frequency in middle age. Older adults spontaneously generated more health-related selves than middle-aged adults, whereas younger adults reported fewer health-related selves than both middle-aged and older adults (Hooker, 1992; Hooker & Kaus, 1992; Hooker & Kaus, 1994). Thus, the frequency of health-related selves may increase with age as health becomes more salient during later life.

PAGE 21

13 Later, in a longitudinal study by Frazier et al. (2000) patterns of change and stability in the possible selves of men and women between the ages of 55 and 89 were identified. The participants initially reported possible selves in a study by Hooker (1992). Then the same participants were contacted again five years later to determine whether their possible selves would remain stable over time. Although continuity was found in most domains, physical and health-related selves were mentioned more frequently over time. Age-related changes in possible selves do occur, and these shifts in possible selves may be important components of successful aging (Markus & Herzog, 1992), providing “blueprints for personal change and growth across the life span” (Cross & Markus, 1991, p. 232). Changes in Memory Beliefs Just as the aging process influences views of the self in relation to health, it also influences views of the self in relation to memory. Age-related changes and differences in memory abilities have been widely established. As a consequence, we all learn to expect memory decline as a part of life, specifically as a part of growing old. Our expectations and beliefs about memory are composed of general beliefs and personal beliefs (Hertzog & Dixon, 1994). General beliefs about memory include expectations we have about how memory will change over time for general population. On the other hand, personal beliefs are opinions we have about our own memory abilities. Personal beliefs are affected by general beliefs but they need not mirror one another. This review discusses general and personal memory beliefs, why they are important, and some of the current research findings regarding these beliefs.

PAGE 22

14 General Beliefs About Memory General beliefs about memory are common beliefs about the process of aging that relate to memory (Hertzog, Lineweaver, & McGuire, 1999). These beliefs broadly apply to all people. General beliefs about memory include beliefs about the expected developmental trajectory of memory performance, stereotypes about memory and aging, and perspectives on memory failure. Heckhausen, Dixon, and Baltes (1989) investigated age differences in the perceived trajectory of development throughout adulthood. Young, middle-aged and older adults, between the ages of 20 to 85, rated adjectives in terms of expected developmental increase, desirability, onset age, and closing age. All participants were instructed to think about people in general, rather than a specific target, during the entire exercise. Results indicated that all age groups expected a similar developmental path and this path was characterized by both gains and losses. However, the percentage of perceived losses increased across the lifespan. Gains actually outnumbered losses, with the oldest group endorsing more possible gains than the middle-aged and young adults, respectively. Older adults also used a greater range of onset and closing ages than younger adults, indicating a more flexible and gradual view of decline. Lineweaver and Hertzog (1998) examined general beliefs about memory in both younger and older adults. Participants were between 18 and 93-years-old. Each participant completed a questionnaire designed to measure general beliefs about memory. In general, all participants believed that memory self-efficacy and memory control declined with age, with the greatest decline occurring after 50 years of age. Further, all participants believed memory ability declined with age;

PAGE 23

15 however, older adults differed from both young and middle-aged adults regarding the onset and magnitude of memory decline. In comparison to the two younger groups, older adults believed memory declined later in life and more gradually (Lineweaver & Hertzog, 1998). In addition, Hertzog et al. (1999) reviewed several studies investigating age-related changes in opinions and feelings about memory in older and younger adults. Again, older and younger adults believed that memory ability, memory self-efficacy, and control over memory all decline with age; however, the patterns of decline differ by age group. Younger adults reported earlier, steeper declines in memory control and memory self-efficacy, with memory decline beginning after age 30. Older adults reported peaks in performance around age 40, and more gradual declines in memory efficacy, control, and abilities. Ryan (1992) also examined beliefs about the developmental trajectory of memory, by investigating age differences in memory beliefs. In her first study, adults between 18 and 74 years of age completed a questionnaire based on the Short Inventory of Memory Experiences. Each participant completed three versions of this questionnaire. They rated each question in terms of beliefs about themselves, beliefs about a typical 25-year-old, and beliefs about a typical 70year-old. In general, memory ratings were more positive for the typical 25-yearold than for the typical 70-year-old, whereas self-ratings fell between the two. In the second study, participants completed the same questionnaire, rating each item for themselves and for a typical person of 25, 45, 65, or 85 years of age. The typical 25-year-old was rated more positively than all other age groups. The

PAGE 24

16 typical 85-year-old was rated as having the poorest memory of all age groups. The typical 45-year-old and the typical 65-year-old fell between the youngest and oldest groups. Thus, memory decline was perceived to start during middle age and continue on until later life. In comparison to participants with positive selfratings, participants with the poorest self-ratings expected all age groups to have poorer memories. These results confirm that a socially accepted pattern of memory development exists. Again, memory abilities are expected to start declining in middle age and to continue declining through old age. The majority of individuals are believed to conform to this fixed pattern of development. Although these standardized expectations apply to all individuals, they are particularly relevant to older adults, as they often lead to social stereotypes about aging and memory. Levy (1996) has examined the effects of age-related stereotypes. Her first study investigated the effect of age-related stereotypes on the memory performance of older adults between the ages of 60 and 90-years-old. Participants initially completed a series of memory tests. After completing the memory tests, participants completed a computerized priming task. The priming task exposed each participant to either positive or negative words, related to either wisdom or memory decline. The words flashed quickly on a computer screen, so the participants were not aware of having seen them. This task was designed to activate participants’ internal stereotypes about aging and memory. Levy (1996) found that the priming activity was successful and influenced memory performance, attitudes toward aging, and beliefs about memory in older

PAGE 25

17 adults. The older adults primed in the negative condition performed more poorly and reported more negative attitudes about aging and memory. The older adults primed in the wisdom condition, performed better and reported more positive attitudes toward aging and memory. The results indicate that older adults respond to stereotypes directly and without knowing that they are doing so. This has important implications regarding the effects cultural stereotypes about aging and memory have on older adults and the aging process they will expect and experience. Expectations of memory decline and negative age-related social stereotypes reinforce one another. These beliefs affect the judgments we make about others in everyday life, and bias us to look for instances that verify these negative beliefs. Consequently, memory failures will become particularly salient, especially when experienced by an older individual. Such general beliefs and biases become particularly relevant as we interact with older adults, and as we, ourselves age. Erber (1989) explored age differences in the assessment of memory failures. In the first study, participants between the ages of 18 and 29 were given a booklet containing 30 memory failure episodes. Half of the episodes were designed to show serious memory failures (e.g., forgetting the name of someone the protagonist has known for years) and half were designed as nonserious errors (e.g., forgetting the name of someone the protagonist just met). The subject in each episode was always a 30 year-old woman or a 70 year-old woman. Participants rated the seriousness of each episode on a 7-point Likert

PAGE 26

18 scale, and consistently rated the memory failures of the 70 year-old woman as more serious than those same failures made by the 30 year-old woman (Erber, 1989). In a second study, older adults between 62 and 82 years of age completed the same exercise. The older adults did not rate the seriousness of the memory failures any differently for the 30 year-old and the 70 year-old. In addition, older adults rated the serious failures less seriously than the younger adults had in the previous study. Thus, younger adults rate the memory failures of older adults more harshly than failures experienced by younger people. Further, older adults are more moderate in their opinions regarding the seriousness of a memory failure, regardless of age. This is most likely due to a greater acceptance of memory decline as normal and inevitable among older adults (Erber, 1989). Thus, expectations of memory deterioration and negative age-related stereotypes do contribute to different perceptions of memory ability. Compared to older adults, younger adults believe that memory failure is more serious. Younger adults are also more concerned when memory failures occur among older individuals. This bias reflects a belief that memory failures may indicate early stages of serious cognitive impairment among older individuals. However, with increasing age, adults consider memory failure a less serious offense, because memory decline is an expected part of life. These general expectations and beliefs about memory affect each of us in important ways. Not only do they affect our perceptions of others, they influence our personal beliefs about our own memory.

PAGE 27

19 Personal Beliefs About Memory Personal beliefs about memory are self-referent beliefs or beliefs that refer or relate directly to the self. Self-referent beliefs about memory include memory self-efficacy beliefs, memory control beliefs, and perceived changes in individual memory abilities. Self-efficacy is an individual’s personal judgment of his or her capability to perform a specific task or reach a specific goal. Self-efficacy beliefs are important for several reasons. First, self-efficacy beliefs affect performance. These beliefs guide which tasks and goals will be approached and which will be avoided (Berry & West, 1993). Second, self-efficacy beliefs influence the amount of effort and persistence an individual sustains once engaged in a task (Berry & West, 1993). Thus individuals with high levels of self-efficacy will be more motivated to continue working toward a related goal. Self-efficacy beliefs are also domain specific (Berry & West, 1993). It is therefore reasonable to expect each individual to have different self-efficacy beliefs over multiple domains. Memory self-efficacy beliefs are specific judgments related to an individual’s capability to engage in and successfully complete memory tasks. Given the general beliefs we hold regarding memory decline, it is also reasonable to expect memory-related self-efficacy beliefs to decline with age. Berry et al. (1989) developed a measure to assess memory self-efficacy beliefs in older and younger adults. Participants were between 18 and 80 years of age. Each participant completed the Memory Self-efficacy Questionnaire and eight memory tasks. Memory tasks were divided among familiar, everyday

PAGE 28

20 memory activities like remembering phone numbers and less familiar, laboratory tasks such as recalling word lists. These tasks were used to estimate the accuracy of memory self-efficacy judgments. In general, younger adults had higher self-efficacy scores on all memory tasks. Further, older adults efficacy ratings were more accurate for everyday tasks than laboratory tasks. Thus, the authors concluded that task familiarity is an important factor in accurate estimates of self-efficacy (Berry et al., 1989). In addition to memory self-efficacy beliefs, memory control beliefs are also important, as both types of beliefs affect behavior. Memory control beliefs are internal or external attributions regarding memory loss. If a person believes memory loss is inevitable, it is viewed as completely out of ones control. If, on the other hand, one believes memory loss is preventable or controllable, loss can be minimized through individual efforts to maintain and improve upon ones abilities. Control beliefs and self-efficacy beliefs are related, however they are not the same. Self-efficacy beliefs are personal judgments about ones ability to complete a task or goal. An individual may believe he or she is able to complete a memory task, but may still feel that memory loss is an inevitable part of aging (Hertzog et al., 1999). Lachman et al. (1995) developed a measure to assess beliefs about memory control and memory ability. They also examined how these beliefs relate to actual memory performance in older and younger adults. Participants were between 20 and 90 years of age. Each participant completed the Memory Controllability Inventory and several different memory tasks. Participants who

PAGE 29

21 felt secure about their current memory ability also felt they had more control over their memory ability. These participants indicated that they believed making an effort could improve their memory, and reported that they could find strategies to do so. Participants who agreed with the inevitability of memory loss with age felt less secure about their current memory ability and less able to make improvements. With increasing age, people became more likely to report that memory deterioration is inevitable and also became more likely to doubt their own memory ability (Lachman et al., 1995). Participants who predicted they could handle their daily memory tasks independently, without depending on others, also rated their current memory ability more positively. These individuals believed in their ability to improve their memory, endorsed the use of effort, and did not believe that memory loss was inevitable. In contrast, participants who were worried about developing Alzheimer’s disease rated their current memory ability less positively. These individuals also believed in the inevitability of memory loss with age (Lachman et al., 1995). Interestingly, control beliefs were related to actual memory performance. Beliefs in inevitable memory loss were associated with poorer performance on the memory tasks. Positive beliefs about current memory ability and the ability to improve memory were associated with better performance on the memory tasks. Clearly, believing one has control over the maintenance of memory ability affects behavior.

PAGE 30

22 In addition to differences in memory self-efficacy and memory control beliefs, older and younger adults have different perceptions regarding when and how they will experience changes in t heir memory abilities. The personal changes that older and younger adults report, generally coincide with the generic expectations we have for the population, however, individual variation does occur. McDonald-Miszczak, Hertzog, and Hultsch (1995), conducted two longitudinal studies in an attempt to uncover age-related changes in memory beliefs and abilities. The first study looked at 231 adults between 22 and 78 years of age, over a period of two years. Participants completed several metamemory questionnaires (Metamemory in Adulthood, Memory Functioning Questionnaire, Memory Self-efficacy Questionnaire). In addition, participants completed questions regarding perceived changes in memory ability and several recall tasks. This study found considerable stability over the two-year period; however, memory was reported as getting worse over time (McDonald-Miszczak et al., 1995). The second study looked at adults between 55 and 86 years of age, at three points in time over a period of six years. At the end of the 6-year period, 234 of the initial 477 participants remained. Participants completed similar measures as in the first study. Memory ability and self-efficacy both declined over time, whereas anxiety about memory increased, as did reliance on external aids and memory strategies (McDonald-Miszczak et al., 1995).

PAGE 31

23 Hultsch et al. (1987) also explored age differences in beliefs about memory change in two different communities. Participants were between 20 and 78 years old, and each completed the Metamemory in Adulthood Scale and the Memory Functioning Questionnaire. In general, when compared to younger adults, older adults perceived themselves as having less memory capacity. Older adults also reported that their memory had declined over time; further, they believed they could do very little to improve their current memory abilities or to prevent memory decline in the future. Lastly, Lineweaver and Hertzog (1998) examined general beliefs about memory and personal beliefs about memory in both younger and older adults, to determine if personal beliefs and general beliefs differ, and if adults of different ages have different beliefs. Participants were between 18 and 93-years-old. Each participant completed two questionnaires; each designed to measure either general or personal beliefs about memory. Personally, older adults reported lower memory self-efficacy, and less control over memory than both middle-aged and younger adults. Participants also reported on perceived memory change. When comparing current memory ability to past memory ability, younger adults reported improvement, middle-aged adults reported little change, and older adults reported significant memory loss. Participants also reported anticipated change between current and future memory ability. In the future, younger adults anticipated improvement for themselves, but not for the general population. Middle-aged adults anticipated

PAGE 32

24 small declines, and older adults expected substantial decline (Lineweaver & Hertzog, 1998). Thus, general beliefs and personal beliefs are related, but not identical. Personal beliefs appear responsive to life experience. Although older adults report declines in memory ability, memory control, and memory self-efficacy, they also report a later and more gradual decline than younger adults. Presumably this is due to the effects of personal experience on general beliefs. In conclusion, the results reviewed here confirm that memory decline is seen as an inevitable and normative developmental process by adults of all ages (Heckhausen et al., 1989; Lineweaver & Hertzog, 1998; Ryan, 1992). This general belief results in age-related stereotypes and biases regarding memory and memory failure. Older adults internalize these general beliefs and stereotypes. Consequently, they learn to anticipate memory deterioration as a normal part of the aging process (Levy, 1996). As a result, older adults tend to downplay instances of memory failure as a normal part of life, whereas younger adults tend to magnify instances of memory failure among older adults (Erber, 1989). Further, younger adults may use memory failure as a signal of potential cognitive impairment among older adults. Given the general beliefs we hold, it is not surprising that older adults also report lower levels of self-efficacy than younger adults (Berry et al., 1989; Hultsch et al., 1987), and less control over memory than younger adults (Hultsch et al., 1987; Lachman et al., 1995).

PAGE 33

25 Older adults also report experiencing declines in memory ability (Erber, 1989; Hultsch et al., 1987; Lachman et al., 1995; Ryan, 1992). However, the decline for older adults occurs later in life and is more gradual than the decline expected by younger adults. This indicates that personal experience can influence personal beliefs, resulting in slight deviations from the generic path of memory loss. Thus, for older adults, memory beliefs become of notable significance, or more salient, as they experience normative age-related declines and respond to social stereotypes (Hultsch et al., 1987; Cavanaugh, Feldman, & Hertzog, 1998). As a consequence, these age-related opinions and beliefs about memory should be reflected in the possible selves of older adults. Memory-related Possible Selves Possible selves and memory beliefs fit together well. Sehulster (1981ab) proposed a self-theory of memory, in which beliefs about personal memory ability, and memory in general, are considered part of the beliefs we have about our selves. Our memory-related beliefs and experiences create a memory selfschema or identity. Further, these beliefs and experiences play a role not only in our current identity, but also in how we view the trajectory of our future memory abilities. More specifically, our current memory self-schema outlines a future memory self. Like health, memory is believed to become more salient with age (Berry & West, 1993; Cavanaugh et al., 1998; Hultsch, et al., 1987); consequently, memory-related possible selves should become more apparent in an older adults repertoire of possible selves. These memory-related selves and

PAGE 34

26 beliefs should also affect behavior, such as persistence, effort, goal setting, and strategy use on memory tasks (West & Berry, 1994). Methods Used to Measure Possible Selves Several methods have been used to measure possible selves. Initially possible selves were measured by using lists of nouns and adjectives. Participants indicated whether each item from the list of possibilities was selfdescriptive. Markus and Nurius (1986) gave participants a list of 150 possibilities. For each item, participants indicated if the item described them currently, or had described them in the past. Participants were also asked if they had ever considered the item as a possible self, and if so, how probable the self was and how much they would like the self to be true for them. This approach has several advantages and disadvantages. It may be possible to determine if particular selves in specific domains are consistently endorsed or not; however, individuals may have important selves that are not included, and consequently not discovered. Possible selves have also been measured by open-ended questionnaires (Cameron, 1999; Cross & Markus, 1991; Frazier et al., 2000; Hooker, 1992; Hooker & Kaus, 1992; Hooker & Kaus, 1994; Ruvolo & Markus, 1992). In these studies, participants were asked to spontaneously list all of the possible selves they would like to achieve, and all those that they would like to avoid. Participants selected the two or three most important feared and most important hoped-for selves and then explained why each was important. They also rated the likelihood of attaining or avoiding these selves, and listed any steps actively taken to achieve or avoid these selves. Although this method allows each

PAGE 35

27 individual to share their unique set of selves, it may be more time consuming and difficult for participants to understand (Hooker, 1992; Hooker, 1999). Ryff (1991) asked participants to answer several questions from various psychological well-being measures. The questions were answered multiple times, in terms of each participant’s current self, ideal self, past self, and future self. Possible selves were not directly assessed in this study, but were used as a manipulation “to determine perceptions of improved or worsened functioning over time” (Ryff, 1991, p. 287). Possible selves have also been assessed using latency response measures (Cross & Markus 1994, Ruvolo & Markus, 1992). These measures indicate which selves are most active in an individual’s working self-concept after some experimental manipulation. Ruvolo and Markus (1992) asked participants to imagine themselves in the future. Participants were then instructed to imagine their own success or failure as a direct outcome of their own actions and write about what it would be like. After this imagery manipulation, participants completed a self-description task, rating positive, negative, and control words as either possible for me or not possible for me Response latencies for each word indicated whether or not a corresponding self was active or accessible in an individual’s working self-concept. The current study plans to investigate a specific domain of possible selves. An open-ended format can provide a glimpse into the most salient selves of each individual (Hooker et al., 1996); however, a closed format allows participants to endorse memory-related responses that are not prominent enough to lead to

PAGE 36

28 spontaneous mention. The combination of both methods in this investigation allowed us to fully tap the possibilities existing within each individual.

PAGE 37

29 CHAPTER 2 STUDY 1 INTRODUCTION Memory-related possible selves were examined in the current study by administering both open-ended and closed-ended possible selves questionnaires to younger and older adults. Within the older adult group, two age groups were created: a young-old group with participants between 50 and 69 years of age, and an old-old group with participants 70 years of age and older. This was done to examine possible differences and similarities between the types of selves generated by older adults at different points in life. In addition to the possible selves measures, all participants completed several questionnaires assessing their beliefs about memory to examine the relationship between age, possible selves, and a number of memory beliefs, including locus of control, anxiety, and beliefs about recent memory performance. Lastly, each participant completed a health survey and a participant information form to provide basic demographic information. Pilot data had indicated the younger adults were unlikely to spontaneously mention memory-related selves. Therefore, the current study also examined whether younger adults might be more likely to spontaneously generate memoryrelated selves when exposed to specific memory activities prior to completing the possible selves questionnaires. Research by Levy (1996) suggested that exposure to social stereotypes about aging and memory can affect memory beliefs and memory performance of older adults on subsequent memory tasks.

PAGE 38

30 Further, Rahhal, Hasher, and Colcombe (2001) have demonstrated that instructions can activate the different self-schemas older adults have related to aging and memory. The current study planned to examine if the working selfconcepts of younger adults could be influenced in a similar manner, by exposing primed individuals to a memory task and several memory beliefs measures before the possible selves questionnaires. Hypotheses As health becomes more important later in life, this is reflected in the selfconceptions of older adults (Hooker, 1992). Older adults reliably report healthrelated possible selves, whereas younger adults rarely report selves in this domain (Frazier et al., 2000; Hooker, 1992; Hooker & Kaus, 1992; Hooker & Kaus, 1994). Memory is also believed to increase in importance with age, as age-related declines in performance must be managed, and social stereotypes come to affect self-conceptions. As a consequence, we expected that older adults would generate memory-related possible selves, especially feared memory selves, and that the number of reported memory-related selves would increase with age. Further, we expected these selves to be interrelated with a constellation of memory beliefs. 1. Older adults would report a greater number of memory-related selves than younger adults. 2. In particular, the old-old would report a greater number of memoryrelated selves than the young-old. 3. Given the pattern of age-related changes in memory beliefs, older adults are expected to primarily generate feared memory-related selves.

PAGE 39

31 4. In keeping with the experimental evidence on possible selves, we expected to successfully manipulate the relevance of memory selves through priming. Younger adults were expected to be more likely to report memory-related possible selves when primed to do so. 5. Lastly, beliefs about memory (control, anxiety, achievement, memory value, fear of Alzheimer’s disease and memory selfevaluation) would be related to whether a person reported a memory-related possible self, and the content of that self as hoped-for or feared.

PAGE 40

32 CHAPTER 3 STUDY 1 METHODS Participants Participants included both younger adults and older adults. Forty-five younger adults, ages 18 to 22 (M= 18.84, SD= .852) were recruited from the University of Florida’s research participant pool. Eighty-five older adults between the ages of 52 and 88 (M= 71.69, SD= 7.52) were recruited from local communities. Older adults were split into two groups, a young-old and an old-old group. The young-old group included 36 individuals between the ages of 50 and 70 years old (M= 64.94, SD= 5.06). The old-old group included 49 individuals between the ages of 70 and 88 years of age (M= 76.65, SD= 4.59). Two older adults were excluded from the study. One older adult was excluded for depression, and another was excluded for failing to answer the questionnaire in a consistent manner. Education and health were examined as a function of age. The young-old group had significantly more years of education than both the oldold group and the younger adults, F (2, 129) = 8.43, p < .000, eta 2 = .117. No significant differences were found in self-rated health between any of the three age groups, F (2, 123) < 1, p > .25. See Table 3.1.

PAGE 41

33 Measures Possible Selves To thoroughly examine possible selves, both closed and open-ended measures were used. Initially, participants completed the open-ended questionnaire developed by Cross and Markus (1991). This questionnaire was designed to spontaneously elicit both hoped-for and feared possible selves. Participants were first asked to list all of their hoped-for selves. Once they completed this list, they were then asked to list their most important, hoped-for self. Next participants answered two questions to assess self-efficacy and perceived control over attaining their most important hoped-for self, and listed any goal-oriented activities they had undertaken to accomplish this hoped-for self. Next, participants were asked to list all of their feared selves. After completing this list, they were asked to list their most dreaded, feared self. Participants then answered two questions to assess self-efficacy and perceived control over avoiding their most dreaded self, and listed any goal-oriented activities they had undertaken to avoid this feared self. The open-ended questionnaire was followed by an adaptation of the closedended questionnaire developed by Markus and Nurius (1986). The closed-ended questionnaire consisted of 73 specific descriptors, considered to be negative (unpopular, paralyzed), or positive (long-lived, admired). The majority of the items used in this study were taken directly from the original questionnaire; however, 14 memory-specific descriptors were added. The memory-specific descriptors included negative items such as “Alzheimer’s patient” and “unable to

PAGE 42

34 remember,” and positive items such as “mentally alert all my life” and “sound mind.” The memory-related items were selected as either positive or negative based on pilot data. Each participant was asked to complete four versions of this questionnaire, endorsing each item as either descriptive or not descriptive of their current self, their past self, their future hoped-for self, and their future feared self. The current study focused solely on the future hoped-for and feared selves. The current and past selves were included to help participants think about themselves in relation to a personal timeline. Memory Beliefs The Metamemory in Adulthood scale developed by Dixon, Hultsch, and Hertzog (1988), was used to assess beliefs about memory ability. This measure was selected because it is a standard measure of memory beliefs that has shown consistent validity and reliability. The following three subscales were used: achievement, anxiety, and locus of control. These scales were selected because of their expected relationship to hoped or feared memory selves. For instance, a participant listing a feared memory-related self may also report greater anxiety over memory-related tasks and less control over their memory ability. These subscales included items like the following: “It is important to me to have a good memory,” “I get upset when I cannot remember something,” and “I have little control over my memory ability.” Each item was rated on a 5-point Likert scale (1 = agree strongly, 5 = disagree strongly). Achievement, anxiety and locus were calculated by averaging an individual’s responses within each subscale. Internal consistency reliability was good for the three scales: achievement (alpha = .77), anxiety (alpha = .86), and locus (alpha = .71).

PAGE 43

35 Questions specifically related to Alzheimer’s disease were included from the Memory Controllability Inventory (MCI) developed by Lachman et al. (1995). The current study expected a greater number of feared memory-related selves among older adults, therefore, it was important to also assess the degree to which individuals feared the most common type of memory loss in late life. Typical items on this scale included: “When I forget something I am apt to think I have Alzheimer’s disease,” and “I think there’s a good chance I will get Alzheimer’s disease.” Each item was also rated on a 5-point Likert scale (1 = agree strongly, 5 = disagree strongly). The four responses were averaged to create a scale measuring fear of Alzheimer’s disease. This MCI scale had good internal consistency reliability (alpha = .81). Four additional items assessed how each individual values memory. The memory value items were adapted from the Health Value questionnaire used by Hooker (1992), and developed by Lau, Hartman, and Ware (1986). These items were included to determine whether individuals reporting memory-related selves valued memory more than those who did not report future selves related to memory. Individuals responded on a 5-point Likert scale (1 = strongly agree, 5 = strongly disagree) to the following items: (1) “If you don’t have your memory you don’t have anything;” (2) “There are many things I care about more than my memory;” (3) “A good memory is of only minor importance in a happy life;” (4) “There is nothing more important than a good memory.” A memory value score was calculated by averaging an individual’s answers across these four questions. The memory value scale had good internal consistency reliability (alpha = .75).

PAGE 44

36 Opinions about recent memory performance were assessed with four questions developed by West, Thorn, and Bagwell (2003). These questions were included to examine whether participants reporting memory-related selves rated their recent memory performance differently than those who did not. Participants were first primed to think about their everyday memory experience by answering this item: ”How important has it been to you to perform well on memory activities in your everyday life?” (1 = not important at all, 7 = very important). This was followed by the three critical scale items on a 7-point Likert scale: “How have you performed on most memory tasks you have done recently?” (1 = very poor, 7 = very good); “How do you think your memory compares with most other people your age?” (1 = much worse, 7 = much better); “How satisfied are you with your recent memory performance?“ (1 = very satisfied, 7 = very unsatisfied). The responses to these three questions were summed to create one memory evaluation scale. The memory self-evaluation scale had good internal consistency reliability (alpha = .88). Lastly, the SF-36 short-form health survey developed by Ware and Sherbourne (1992) assessed the degree to which health and depression affected an individual’s participation in daily activities. This survey was used because it provides information on eight comprehensive subscales in a relatively short and simple format. The SF-36 is composed of these subscales: General Health, Physical Limitations, Physical Functioning, Bodily Pain, Vitality, Emotional Limitations, Social Functioning, and Mental Health. General Health included four questions rated on a 5-point Likert scale, such as “In general, would you say your

PAGE 45

37 health is…” (1 = excellent, 5 = poor) and “I seem to get sick a little easier than other people” (1 = definitely true, 5 = definitely false). Physical limitations were examined by four yes-no questions, such as, “During the past 4 weeks, have you had any of the following problems with your work or other regular daily activities as a result of your physical health: Accomplished less than you would like” (yes/no). Physical functioning was measured using ten questions rated on a 3point Likert scale, such as “Does your health now limit you in climbing several flights of stairs?” (1 = yes, limited a lot, 3 = no, not limited at all). Bodily Pain was assessed by two questions rated on a 6-point Likert scale, such as “How much bodily pain have you had during the past 4 weeks?” (1 = none, 6 = very severe). Vitality was measured by four questions rated on a 6-point Likert scale, such as, “How much of the time during the past 4 weeks did you feel full of pep?” (1 = all of the time, 6 = none of the time). Emotional limitations were assessed using three yes/no questions, such as “During the past 4 weeks, have you had any of the following problems with your work or other regular daily activities as a result of any emotional problems (such as feeling depressed or anxious): Cut down the amount of time you spent on work or other activities?” (yes/no). Social functioning was measured by two questions rated on a 5-point Likert scale, such as “During the past 4 weeks, to what extent has your physical health or emotional problems interfered with your normal social activities with family, friends, neighbors, or groups?” (1 = not at all, 5 = extremely). Last, mental health was assessed with five questions rated on a 6-point Likert scale, such as “How much of the time during the past 4 weeks have you been a nervous person?” (1 = all of

PAGE 46

38 the time, 5 = none of the time). The responses within each scale were summed for each individual (Ware, 1993). Internal consistency reliability was good for each of the eight scales: general health (alpha = .83), physical limitations (alpha = .81), physical functioning (alpha = .91), bodily pain (alpha = .79), vitality (alpha = .83), emotional limitations (alpha = .85), social functioning (alpha = .74), and mental health (alpha = .82). To examine the spontaneously-generated possible selves, four categories of selves were created: health, dependency, memory, and cognitive selves. Health-related selves included direct statements such as “maintaining or losing my health,” “avoiding illness,” and “remaining physically active.” Health-related selves also included more specific statements such as “becoming a cancer patient,” “having a stroke or heart attack,” and “losing the ability to see.” Dependency-related selves included general statements like “maintaining my independence” or “becoming dependent on another,” but also included such statements as becoming a “burden on my family” and “moving into a nursing home.” Memory-related selves included “becoming an Alzheimer’s patient,” “senile,” or “demented,” and also included positive items such as “improving my memory ability” and remembering specific items, like vocabulary and dance steps. Lastly, cognitive selves included statements such as “staying mentally alert or active,” or “losing my mind” and also included cognitive activities such as “learn to use a computer” or “learn to play an instrument.” Although many of these cognitive selves refer to the mind and to memory, selves that fit this

PAGE 47

39 category were counted separately to ensure that memory-related selves would be defined as conservatively as possible. The spontaneously-generated selves were scored into the four categories by two independent raters. The overall agreement between the two raters was 93%. Table 3.2 shows the percentage of agreement, by category of self, for hoped-for and feared selves. Procedure Interviews were conducted in groups of 3 to 20 individuals. Participants were instructed to complete the questionnaire packet, in order, without looking back at any of their previous answers. Experimenters were present at all times to answer any questions. Each packet began with the open-ended portion of the possible selves questionnaire, followed by the closed portion of the possible selves questionnaire. Participants then completed the remaining items concerning memory beliefs. Last, participants completed a Participant Information form, in order to provide us with basic demographic information and the SF-36 health survey. Due to experimenter error, the majority of the older adults completed the SF-36 one week after completing the questionnaire packet.

PAGE 48

40 Table 3-1. Mean Education and Health Ratings Standard Standard Age group Mean education deviation Mean health deviation Young 13.22 .765 15.37 2.85 Young-old 14.75 2.99 16.06 3.14 Old-old 13.08 2.11 15.11 3.58

PAGE 49

41 Table 3-2. Percentage of Agreement Between Independent Raters Type of selves Percentage agreement Hoped-for Health 85.9% Dependency 94.5% Memory 100% Cognitive 88.3% Feared Health 89% Dependency 93.7% Memory 99.2% Cognitive 93.8%

PAGE 50

42 CHAPTER 4 STUDY 1 RESULTS Analyses of variance were conducted to examine spontaneous memoryrelated selves as a function of age group. In addition to the analyses of variance, cross tabs were also conducted because the actual number of individuals spontaneously reporting memory-related selves was low. These additional analyses examined whether the number of individuals listing these selves differed significantly by age group. Significant group differences were identified in each case by using Tukey’s post hoc tests at the .05 level. Spontaneous Selves An analysis of variance was conducted to examine whether the three age groups (young, young-old, and old-old) differed by the total number of memoryrelated selves each group spontaneously listed. Memory-related selves included items such as: “improving or losing my memory,” and “becoming an Alzheimer’s patient.” In general, older adults spontaneously reported significantly more memory-related selves than younger adults, F (2, 127) = 7.82, p < .000, eta 2 = .110, in the open-ended questionnaire, in support of Hypothesis 1. In fact, younger adults did not spontaneously report any memory-related selves. Contrary to the prediction in Hypothesis 2, no significant differences were found in the number of memory-related selves reported by the old-old and young-old. To further explore the spontaneous reports of the older adult groups, older participants were divided into those who spontaneously mentioned a memory self

PAGE 51

43 and those who did not. A cross-tabulation of the two older groups (young-old and old-old) listing spontaneous memory-related selves was not significant, Pearson X 2 (N = 85, df = 1) = .17, p > .60, further supporting the notion that these two older groups did not differ in their spontaneous report of memory selves. See Table 4.1. Further, a closer look at the types of memory-related selves (none, hopedfor, or feared) reported by the young-old and the old-old groups revealed a significant difference in the types of memory-related selves reported by each age group, F (2, 129) = 7.468, p < .001, eta 2 = .105, such that the old-old and youngold groups both reported a greater number of feared memory-related selves than hoped-for memory-related selves as predicted in Hypothesis 3. To further examine the types of memory-related selves reported by the older adults, participants were categorized into: those who primarily reported hoped-for memory-related selves, those who primarily reported feared memory-related selves, and those who did not report any memory-related selves. Because the majority of participants’ spontaneous memory selves represented fears, any participant who mentioned a hoped-for memory self was categorized as hopedfor. Those with feared selves and no hoped-for selves mentioned spontaneously were classified as feared. A cross-tabulation of the two older adult groups (young-old, old-old) by spontaneous self-categorization (none, hoped-for, feared) was not significant: Pearson X 2 (N = 85, df = 2) = -.04, p > .70. In the two older groups, there were about the same number of individuals reporting hoped-for and feared memory selves. See Table 4.2.

PAGE 52

44 In addition to memory-related selves, participants also included items such as remaining mentally alert, improving mental capacity and losing mental faculties on the open-ended portion of the questionnaire. To be conservative about the memory category, these items were not initially coded as memoryrelated selves. These cognitive items were later combined with the memoryrelated selves above to form a general cognitive category. An analysis of variance was conducted examining the number of cognitive selves reported by the three age groups (young, young-old, and old-old). The younger adults reported fewer cognitive selves than both the young-old and the old-old, F (2, 129) = 9.02, p < .000, eta 2 = .045. No significant differences were found between the number of cognitive selves reported by the young-old and old-old. Individuals were divided into two categories, those who spontaneously mentioned a cognitive and/or memory-related self and those who did not. Again, a cross-tabulation of the three age groups (young, young-old, old-old) showed significant differences when including the spontaneously reported cognitive selves with the memory-related selves: Pearson X 2 (N =130, df = 2) = 16.025, p < .000. The younger adults were less likely to report cognitive selves than the two older adult groups. See Table 4.3. Selves Endorsed on Closed-ended Questionnaire A mixed analysis of variance was conducted to examine the total number and type of memory-related selves (hoped-for or feared) endorsed on the closedended questionnaire, by age group (young, young-old, and old-old). Age group was the between-subjects factor and type of memory self was the within-subjects factor. Overall, no significant age differences were revealed in the number of

PAGE 53

45 selves endorsed on the closed-ended questionnaire by the three age groups, F (1, 127) = 1.58, p > .20. However, all three groups endorsed more feared than hoped-for memory-related selves on the closed-ended questionnaire, F (1, 127) = 82.68, p < .000, eta 2= .394. See Table 4.4. An additional mixed analysis of variance was conducted to examine the total number and type of memory-related selves (hoped-for or feared) endorsed on the closed-ended questionnaire by age group (young-old and old-old) and by type of memory self spontaneously reported (none, hoped-for, or feared) on the open-ended portion of the questionnaire. The analysis confirmed a main effect for closed selves, F (1, 79) = 30.19, p < .000, eta 2 = .277. Again, the young-old and old-old endorsed more feared selves than hoped-for selves on the closed portion of the questionnaire. No other significant differences were found. Next both closed and open memory-related selves were examined together. Individuals were categorized into two groups: those with predominately hoped-for selves and those with predominately feared selves. This categorization utilized both the open and closed item responses. This involved a two-step process. Most individuals on the open and closed portions of the possible selves questionnaires tended to mention feared memory-related selves. Therefore, anyone reporting a spontaneous feared memory-related self, but no spontaneous hoped-for self, was categorized as “feared.” Next, individuals were categorized as “hoped-for” if they reported a spontaneous hoped-for memory-related self and endorsed an equal or greater number of hopes than fears about memory on the closed portion of the questionnaire. A cross-tabulation of the three age groups

PAGE 54

46 (young, young-old, old-old) by self-categorization (hoped-for or feared) did not reveal any significant differences: Pearson’s X2 (N = 130, df = 1) = .13, p = .937 (See Table 4.8). The memory beliefs measures were highly related to each other, both theoretically and empirically as shown by the age-partialled correlations in Table 4.5. Previous research has shown that older and younger adults answer memory beliefs questions differently. Therefore, age was partialled out to ensure that the correlations reflected the relationships between the variables of interest independent of age. A multivariate analysis of variance was conducted to compare the three age groups (young, young-old, old-old) across the following dependent variables: achievement, locus, anxiety, memory value, MCI, and memory self-evaluation. Age differences were significant, F (12, 230) = 4.77, p < .000, eta 2 = .199, and this significant result was further explored in univariate tests Significant age differences were found for memory value, F (2, 119) = 24.59, p < .000, eta 2 = .292, due to the fact that older adults in both the old-old and young-old groups placed a greater value on memory than younger adults. The young-old and old-old did not significantly differ from one another, as shown in Table 8. Significant age differences were also found for the MCI scale which measured fear of Alzheimer’s disease, F (2, 121) = 5.99, p < .003, eta 2 = .091, with older adults in both groups reporting greater fear of Alzheimer’s disease than younger adults. The old-old also reported greater levels of anxiety in response to memory-related tasks than both the young-old and young adults, F

PAGE 55

47 (2, 121) = 4.731, p < .01, eta 2 = .074. Next, significant age differences were found for achievement, F (2, 121) = 3.57, p = .03, eta 2 = .057. For achievement, the old-old placed significantly greater importance on performing well on memory tasks than did younger adults. The young-old did not significantly differ from the old-old or the younger adults. No significant age differences were found for Locus, F (2, 121) = 1.56, p > .20, suggesting that participants in each age group believed they had similar amounts of control over their memory ability. Further, no significant age differences were found on the Memory Self-Evaluation scale, F (2, 121) < 1, p > .50. Thus young adults, the young-old and the old-old all reported similar levels of ability and satisfaction regarding their recent memory performance. See Table 4.6. An additional multivariate analysis of variance examined the memory beliefs measures as a function of age and the type of memory-related selves primarily reported in the open-ended questionnaire, using the categorization described previously (none, hoped-for, feared), and including only the young-old and old-old (because the younger adults reported no memory selves on the open-ended questionnaire). The dependent variables were achievement, locus, anxiety, memory value, MCI, and memory self-evaluation. No significant age differences in memory beliefs were found between the young-old and old-old groups, F (6, 66) = .77, p > .50. Type of memory self spontaneously reported was close to significance, F (12, 134) = 1.73, p > .06, so univariate analyses were conducted. Significant differences were found for anxiety, F (2, 76) = 5.54, p = .006, eta 2 = .135, due to the fact that participants who spontaneously listed

PAGE 56

48 hoped-for memory-related selves also reported greater levels of anxiety when faced with memory-related activities than participants who did not list any memory-related selves. Individuals reporting feared memory selves did not significantly differ from those with hoped-for selves or with no memory selves at all. A significant difference was also found for the MCI scale, F (2, 76) = 4.92, p = .01, eta 2 = .122. Participants spontaneously reporting hoped-for memoryrelated selves also reported a greater fear of Alzheimer’s disease than participants who reported no memory selves. Again, individuals reporting feared memory selves did not differ from those with hoped-for selves or with no memory selves. Last, a significant difference was observed for the memory selfevaluation scale, F (2, 76) = 5.83, p = .003, eta 2 = .141. Participants reporting hoped-for memory-related selves evaluated their recent memory performance less favorably than participants with feared memory-related selves and those reporting none at all. All other memory beliefs measures showed no variation as a function of type of spontaneous memory self (See Table 4.7). Additional Analyses Several additional analyses were conducted to compare the current results with prior research on possible selves. Although this study focused on memoryrelated selves, participants were asked to generate as many future selves as they could think of, both hoped-for and feared. We compared the mean number of hoped-for and feared selves spontaneously generated by age group. Overall, the old-old spontaneously reported significantly fewer hoped-for selves than the younger adults, F (2, 129) = 6.83, p < .002, eta 2 = .097. The young-old did not significantly differ from either the old-old or the younger adults in the number of

PAGE 57

49 hoped-for selves they generated. The old-old also spontaneously reported significantly fewer feared selves than both the young-old and younger adults, F (2, 129) = 10.15, p < .000, eta 2 = .14. See Table 4.9. Participants were also asked to identify their most important hoped-for self from the list of possible selves they generated. Next they were asked to indicate how capable they felt of accomplishing this most important self, and how likely it is that this self would come true. We have compared these mean capability and likelihood ratings for the most important hoped-for self by age group. The old-old and young-old rated themselves as significantly less capable of accomplishing their most important hoped-for self than the younger adults, F (2, 125) = 11.85, p < .000, eta 2 = .16. Further, the old-old and young-old rated their most important hoped-for self as significantly less likely to come true than the younger adults, F (2, 125) = 15.84, p < .000, eta 2 = .21. See Table 4.10. Participants were also asked to identify their most dreaded feared self from the list they generated. Again, they were asked to indicate how capable they felt of preventing this self and how likely it is that this self would come true. We have compared these mean capability and likelihood ratings for the most dreaded feared self by age group. The old-old and young-old rated themselves as significantly less capable of preventing their most dreaded feared self than the younger adults, F (2, 123) = 8.00, p < .001, eta 2 = .12. The old-old and youngold also reported that their most dreaded feared self was more likely to come true than did the younger adults, F (2, 123) = 5.89, p < .004, eta 2 = .09. See Table 4.11.

PAGE 58

50 Last, participants were asked to list the number of goal-oriented activities recently undertaken to attain their most important hoped-for self and to avoid their most dreaded feared self. The old-old reported significantly fewer goaloriented activities recently undertaken to attain their most important hoped-for self than both the young-old and younger adults, F (2, 129) = 9.38, p < .000, eta 2 = .13. No significant differences were found between the three age groups for the number of goal-oriented activities undertaken to avoid their most dreaded feared self, F (2, 129) = 2.96, p > .05 (Table 4.12). These results are generally consistent with previous research on possible selves. In general, older adults report fewer possible selves than younger adults (Cross & Markus, 1991). Further, previous research has also shown that older adults feel less capable of attaining or avoiding their most important or most dreaded selves than younger adults (Cross & Markus, 1991). The current results are not consistent with previous research showing that older adults report undertaking a greater number of goal-oriented activities to attain or avoid their most important or most dreaded selves than younger adults (Cross & Markus, 1991).

PAGE 59

51 Table 4-1. Percentage of Spontaneous Memory Selves Reported by Age Group Age Group No memory self Yes memory self Young 100% 0% Young-old 69.4% 30.6% Old-old 73.5% 26.5%

PAGE 60

52 Table 4-2. Type of Spontaneous Memory Selves by Age Group Age group Hoped-for Feared None Young-old 11.1% 19.5% 69.4% Old-old 10.2% 16.3% 73.5%

PAGE 61

53 Table 4-3. Spontaneous Cognitive Selves by Age Group Age group No cognitive self Yes cognitive self Young 71.1% 28.9% Young-old 27.8% 72.2% Old-old 42.9% 57.1%

PAGE 62

54 Table 4-4. Mean Number of Memory Selves Endorsed on Closed Questionnaire Standard Standard Age group Mean feared deviation Mean hoped-for deviation Young 6.71 1.14 5.04 1.69 Young-old 6.97 .91 5.69 1.14 Old-old 6.67 1.94 5.24 1.70

PAGE 63

55 Table 4-5. Intercorrelations Between Dependent Subscales ANXIETY ACHIEVE LOCUS MCI VALUE EVAL ANXIETY --ACHIEVE .3664 --(176) P= .00 LOCUS -.1006 .2415 --(176) (177) P= .18 P= .001 MCI .4386 .1757 -.2875 --(176) (177) (177) P= .00 P= .02 P= .00 VALUE .1106 .3542 .2058 .0491 --(176) (177) (177) (177) P= .14 P= .00 P= .006 P= .51 EVAL -.4695 -.1459 .0724 -.3915 -.0077 --(175) (176) (176) (176) (176) P= .00 P= .05 P= .34 P= .00 P= .92 Note: These partial correlations have age variance removed.

PAGE 64

56 Table 4-6. Mean Memory Belief Ratings by Age Group Standard Standard Standard Age Group Anxiety deviation Locus deviation Achievement deviation Young 2.92 .58 3.53 .53 3.80 .41 Young-old 2.90 .69 3.69 .51 3.92 .46 Old-old 3.30 .73 3.70 .50 4.03 .35 Mean Memory Belief Ratings by Age Group (continued) Fear of Standard Memory Standard Self Standard Age Group Alzheimer’s deviation value deviation evaluation deviation Young 1.87 .46 2.42 .52 14.40 3.27 Young-old 2.22 .84 3.33 .91 14.57 4.06 Old-old 2.45 1.01 3.48 .82 14.31 2.70

PAGE 65

57 Table 4-7. Analysis of Variance for Dependent Subscales by Type of Memory Self Spontaneously Reported Source df F p eta2 Age Group (Young-old, Old-old) Anxiety 1 0.38 0.54 .005 Achievement 1 0.81 0.37 .011 Locus 1 1.80 0.19 .025 MCI 1 0.68 0.41 .010 Value 1 0.98 0.33 .014 Eval 1 0.09 0.76 .001

PAGE 66

58 Table 4-7. Continued Source df F p eta2 Type of Memory Self (None, Hoped-for, Feared) Anxiety 2 5.54 .006 0.14 Achievement 2 0.58 0.56 0.02 Locus 2 0.94 0.40 0.03 MCI 2 4.92 0.01 0.12 Value 2 0.46 0.64 0.01 Eval 2 5.83 .005 0.14

PAGE 67

59 Table 4-8. Type of Memory Selves Reported By Age Group Age Group Feared selves Hoped-for selves Young 35.7% 32.6% Young-old 27.4% 28.3% Old-old 36.9% 39.1%

PAGE 68

60 Table 4-9. Mean Number of Selves by Age Group Mean Standard Age group selves deviation Hoped-for Young 6.13 2.75 Young-old 5.47 2.77 Old-old 4.29 1.85 Feared Young 4.29 1.66 Young-old 3.92 2.10 Old-old 2.76 1.44

PAGE 69

61 Table 4-10. Mean Capability and Likelihood Ratings: Feared Mean Standard Mean Standard Age group capability deviation likelihood deviation Young 5.73 1.19 2.93 1.51 Young-old 4.54 1.80 4.09 1.56 Old-old 4.61 1.63 3.91 1.87

PAGE 70

62 Table 4-11. Mean Capability and Likelihood Ratings: Hoped-for Mean Standard Mean Standard Age group capability deviation likelihood deviation Young 6.22 .82 6.31 .76 Young-old 5.60 .98 5.00 1.35 Old-old 5.04 1.50 4.93 1.61

PAGE 71

63 Table 4-12. Mean Number of Goal Activities Mean Standard Age group activities deviation Hoped-for Young 2.09 1.38 Young-old 1.06 .90 Old-old 2.03 1.56 Feared Young 1.98 .21 Young-old 2.31 .24 Old-old 1.55 .21

PAGE 72

64 CHAPTER 5 STUDY 2 INTRODUCTION The current study also sought to examine whether younger adults could be primed to spontaneously list memory-related possible selves. In Study 1, younger adults did not spontaneously report any memory-related possible selves, either hoped-for or feared. These initial results suggested that memory, and memory loss specifically, was not salient to younger adults. This study examined whether or not memory could be brought into focus for younger adults, thus prompting them to spontaneously report memory-related selves. Prior research suggests that specific testing conditions can serve as primes to activate belief schemas during memory testing. Levy (1996) examined the effects of agerelated stereotypes in a series of two studies. In the first study a priming task exposed each participant to either positive or negative words, related to either wisdom or memory decline, after a series of memory tests. The words flashed quickly on a computer screen, so the participants were not aware of having seen them. This task was designed to activate participants’ internal stereotypes about aging and memory. After the priming task, a third of the participants immediately completed a second series of memory tests and questionnaires. Levy (1996) found that the priming activity was successful and influenced memory performance, attitudes toward aging, and beliefs about memory in older adults. The older adults primed in the negative condition performed more poorly and reported more negative attitudes about aging and memory. In contrast, the older

PAGE 73

65 adults primed in the wisdom condition, performed better and reported more positive attitudes toward aging and memory. The results indicated that priming was effective in accessing the different beliefs older adults hold toward memory and aging. Rahhal et al. (2001) also manipulated beliefs in relation to memory testing. They examined the effects of instructional manipulations (memory-emphasis and memory-neutral instructions). In the memory-emphasis condition, participants were told that the activity was a memory test, and the need to remember information was stressed. The memory-neutral condition focused on the participant’s ability to learn new facts and did not mention memory or remembering. In each condition, participants were presented with 60 trivia questions labeled with no answer or labeled as true or false. During the recall phase, 12 new items were mixed in with the original 60 questions. Participants identified each statement as new or old and as true, false, or blank. For both studies, the authors found that when instructions emphasized memory and remembering facts, older adults performed worse than younger adults on the recall task. However, when the instructions did not emphasize memory, and instead focused on learning new information, no significant age differences were obtained in recall. These studies suggest that the type of instruction presented to older adults can influence their performance by reinforcing age-related beliefs and stereotypes about memory and memory performance (Rahhal et al., 2001). The previous studies suggest that memory testing under particular instructional conditions can prime beliefs and stereotypes in older adults. There

PAGE 74

66 is also considerable literature showing that beliefs and feelings can be altered in memory studies with younger adults as well. Bower and colleagues (e.g., Bower, 1994) have repeatedly shown that mood can be manipulated in college students, with a subsequent effect of memory. Further, such manipulations affect the types of past memories recalled, judgments of personal competence, and judgments about the likelihood of future events (Bower, 1994). Taken together, the studies above suggest that memory testing under certain circumstances can affect thoughts about the self in relation to memory. That is the effect we expect here. The current study planned to use memory tasks and questionnaires in a similar fashion as the studies described above. Exposing younger adults to a challenging memory task should access their memory-related beliefs, making it more likely that younger adults might spontaneously mention a memory-related possible self. In the current study, priming was accomplished in two ways. First, participants in the primed condition completed a challenging memory activity. Second, they completed several memory beliefs questionnaires. Both of those activities preceded their completion of the open-ended and closed possible selves questionnaires. By having the primed group complete the memory activity and memory questionnaires before the possible selves questionnaires, memory should be more salient to these individuals. Consequently, these younger adults should be more likely to generate memory-related possible selves on the openended questionnaire than the younger adults in the unprimed condition.

PAGE 75

67 CHAPTER 6 STUDY 2 METHODS Participants Of 97 younger adults recruited from the University of Florida’s research participant pool, 52 were primed and 45 were unprimed. Age, health and education were examined as a function of priming condition. No significant differences in age, F (1, 96) = 2.019, p > .10; health, F (1, 94) < 1, p > .25; or education, F (1, 94) < 1, p > .50 were found between primed and unprimed young adults. See Table 6.1. Procedure Young adults in the primed condition received the same questionnaire packet used in Study 1, but in an altered order. First each participant studied a 15-item shopping list created by West, Welch, & Thorn (2001). Participants were given 1 minute to study the list and a maximum of 4 minutes to recall the list. Next, primed participants completed the memory belief measures, followed by the open-ended and closed possible selves questionnaires, respectively. Lastly, primed participants completed the participant information form and SF-36 health survey.

PAGE 76

68 Table 6-1. Mean Demographic Information for Primed and Unprimed Young Adults Mean Standard Mean Standard Mean Standard Condition age deviation health deviation education deviation Primed 19.15 1.23 2.13 .742 13.37 1.27 Unprimed 18.84 .852 2.28 .766 13.22 .765

PAGE 77

69 CHAPTER 7 STUDY 2 RESULTS The younger adults in the unprimed condition did not spontaneously report any memory-related possible selves as expected. Further, the majority of primed younger adults did not report memory-related selves (only two younger adults in the primed condition listed a memory-related self), therefore comparisons between the number of spontaneous memory-related selves generated by the primed and unprimed groups were not conducted. An analysis of variance was conducted to examine the total number of memory-related selves (spontaneously generated plus those endorsed on the closed-ended questionnaire) by priming condition (primed and unprimed). No significant difference was observed in the number of memory-related selves endorsed by primed and unprimed younger adults, F (1, 96) < 1, p > .50, (primed, M = 11.52, SD= 2.07; unprimed, M = 11.76, SD= 2.24).

PAGE 78

70 CHAPTER 8 DICUSSION Memory-related possible selves are important for two main reasons. First, these selves identify what a person hopes to become or fears becoming. Cross and Markus (1994) stated that possible selves can help describe the direction of development an individual is anticipating. Thus the presence or absence of memory selves will provide a more comple te picture of how individuals view their own cognitive development toward the latter half of the lifespan. Next, the presence or absence of these selves may ultimately affect behavior. Hooker has shown that the presence of health-related possible selves are related to health behaviors such as seeking preventative medical treatment and exercising. Specifically, individuals with a health-related self were more likely to report participating in health behaviors than were individuals that placed a high value on health alone (Hooker & Kaus, 1992). From this perspective, possible selves act as goals, and may help motivate individuals into action. These future selves provide a measure to which current selves can be compared and evaluated (Cross & Markus, 1991). When the discrepancy between the current self and a future hoped-for self becomes too great (or the distance between the current self and a future feared self becomes too close), something must be done. From a developmental perspective, Cross and Markus (1991) have suggested that possible selves may actually become more motivating with age. Older adults experience less conflict among their competing future selves.

PAGE 79

71 Further, their future selves are more closely tied to their current selves, and are therefore more specific, well defined, and personally meaningful. As such, the steps required to avoid or achieve a certain self may be more clearly defined and motivating, as one gets older. This study was the first to look specifically at memory-related possible selves. The results confirmed that the types of possible selves spontaneously generated, that is, those selves that are most important or central to an individual’s self-concept, differ with age. Younger adults did not spontaneously report any memory-related possible selves. Even after cognitive and memoryrelated selves were combined, younger adults spontaneously reported fewer cognitive selves than older adults. From this we can conclude that younger adults are not concerned about their memory, nor are they particularly focused on their future cognitive abilities. The future hopes and fears of younger adults are focused elsewhere. In contrast, older adults did report memory-related selves. Further, the memory selves reported by older adults were primarily feared selves, like dementia. Both older and younger adults expect memory decline to occur with age, and the greatest decline is expected to occur in the later years of life (Hertzog et al., 1999; Lachman et al., 1995; Lineweaver & Hertzog, 1998; Ryan, 1992). Older adults also face social expectations of memory decline. Episodes of forgetting among the elderly are rated more seriously than when the same episodes occur for a younger person (Erber, 1989). Further, when forgetting occurs among older adults, it is more likely to be viewed as indicative of a need

PAGE 80

72 for medical attention or cognitive evaluation (Erber, 1989). These social expectations and stereotypes contribute to the views older adults have about themselves and their memory ability. On a personal level, older adults also report lower levels of control over improving or maintaining their memory abilities (Hultsch et al., 1987; Lachman et al., 1995) and feel less capable of completing memory tasks (Berry et al., 1989; Hultsch et al., 1987). With personal and social expectations of memory loss looming on the horizon, it is not surprising to find that older adults have incorporated these beliefs and expectations into their selfconcepts. Cavanaugh et al. (1998) have hypothesized that memory is schematic for older adults. Individuals are said to be schematic for a particular domain or trait when they consider it both highly important to and highly descriptive of themselves (Cavanaugh et al., 1998; Cross & Markus, 1994). Thus, for older adults, memory and memory loss have more personal importance (Cavanaugh et al., 1998). The current results support this idea. If an individual spontaneously generates a possible self in a specific domain, like health or memory, that domain is of central importance to their self-concept. For these individuals, memory or health are important in general but also define who they are in some way. The older adult data in this research may reflect this kind of schematicity. When memory and cognitive selves were combined, 63.5% of older adults spontaneously listed at least one memory or cognitive-related self. This suggests the prominence of cognition as a key element in the self-concept of the older group.

PAGE 81

73 As part of the self-concept, these memory-related selves may play an important role, guiding behavior in several ways. These selves may create shifts in information processing (Kato & Markus, 1993), allowing information related to a particular self to be processed more quickly and efficiently (Cross & Markus, 1994; Ruvolo & Markus, 1992). For inst ance, an episode of forgetting will be differentially attended to and may take on a different meaning for an individual with a memory-related possible self. Next, the existence of a memory self within the self-concept can also result in changes in levels of motivation and selfefficacy (Cameron, 1999; Cross & Markus, 1991; Hooker, 1992). For example, Hooker has shown that individuals with health-related selves were more likely to report engaging in health-related behaviors and felt more capable of attaining health-related goals than individuals who reported valuing health, but listed no health-related selves (Hooker & Kaus, 1992). Last, these selves may affect the level of effort and persistence an individual is willing to put forth on activities related to a specific self (Ruvolo & Markus, 1992). For instance, individuals with a memory-related self may be more likely to seek out challenging cognitive activities, further once engaged in a task, they may be more likely to stick with it. Interestingly, the three age groups did not significantly differ in the number of memory-related selves endorsed on the closed portion of the questionnaire. Thus, when presented with positive and negative future memory outcomes, both groups identified and endorsed similar items. However, all three groups endorsed a greater number of feared items than hoped-for items on the closed portion of the questionnaire. When asked about future selves they would like to

PAGE 82

74 avoid, both groups were likely to circle memory-related items like “Alzheimer’s patient” or “senile.” However, when asked about future selves they would like to attain, they were less likely to circle the positive memory-related items such as “mentally alert all my life” or “sound mind.” There may be several reasons for these differences. First, negative memory outcomes may be more salient to both groups. Previous research has shown that younger and older adults expect memory decline to occur as they age (Hertzog et al., 1999; Lachman et al., 1995; Lineweaver & Hertzog, 1998; Ryan, 1992). The current results appear to confirm a shared stereotype among older and younger adults regarding aging and memory. Second, we know what it means to have a poor memory, to become senile or demented. We have terms to describe and diagnose these problems in a very concrete way. It may be that these terms are simply easier to identify than the more general positive memory-related items. The three age groups differed across several of the memory beliefs measures, verifying results from earlier research examining age differences in memory beliefs. In general, the results indicate that the old-old may be more concerned about their memory. The old-old reported a greater fear of Alzheimer’s disease than younger adults, however the young-old did not significantly differ from the old-old or the younger adults. Further, the old-old reported greater levels of anxiety in response to memory tasks than both the young-old and younger adults. Consequently the old-old may fear that participating in a memory activity might confirm expected memory loss. Again, these results are consistent with previous studies showing that older adults report

PAGE 83

75 greater levels of anxiety and a greater fear of Alzheimer’s disease than younger adults (Lachman et al., 1995) and suggest that memory and memory loss are more salient to older adults than to younger adults. Along these lines, the old-old and the young-old both valued memory more than the younger adults. Thus both the old-old and the young-old placed more importance on performing well on memory tasks than did younger adults. For older adults, performing well on a memory task might act as evidence that their memory is intact. It may also be that the older adults place a greater value on performing well in general, regardless of the task, because of heightened concerns about perceived age-related losses. The younger adults, in contrast, may accept their cognitive abilities as strong and stable. For them, memory performance may not be as important because it has less diagnostic value (Hess, Auman, Colcombe, & Rahhal, 2003). Contrary to what may have been expected, individuals who reported only feared memory-related selves did not appear to differ from those with only a hoped-for memory self or those with no memory selves at all, on the memory self-evaluation scale. These individuals listed a memory-related fear, but still reported feeling confident about their recent memory ability. Further, individuals with a feared memory self or no memory self at all were less anxious and reported less fear of Alzheimer’s disease than individuals with a hoped-for memory self. Cross and Markus (1991) have discussed this phenomena in terms of defensive elaboration. Individuals spontaneously listing a memory-related fear may be experiencing declines in memory ability or increased episodes of

PAGE 84

76 forgetting, resulting in the presence of a memory fear. However, these individuals may be able to feel content with their current memory ability by comparing current ability to a future that appears far worse (Cross & Markus, 1991). As a result they are less anxious and have less immediate fear because their current memory abilities appear intact when compared to the severe impairment expected in the future. This view has not been extensively examined in the literature, and may be a useful concept for future investigation. Last, in Study 2, priming did not elicit any memory-related selves from younger adults. We expected that younger adults in the primed condition would be more likely than unprimed younger adults to generate memory-related possible selves. The memory task included in the priming activities may not have been sufficiently challenging to activate memory-related possible selves in younger adults. The primed younger adults correctly recalled approximately 11 out of 15 shopping list items (M = 11.17, SD = 1.72). A more difficult memory task might be more successful in eliciting these types of selves from younger adults in the future. On the other hand, it may be the case that memory is so low in salience for younger adults that they will not spontaneously generate memoryrelated selves, even after a taxing memory activity. A primed older adult group would be interesting to examine to determine whether more memory selves would be generated when older adults are confronted with a challenging memory task. Stereotype threat research suggests that older adults, for whom memory is self-defining, may react differently when faced with a memory task (Hess et al., 2003). Specifically, their performance may suffer as they anxiously strive to

PAGE 85

77 debunk the negative stereotypes surrounding aging and memory (Hess et al., 2003). More importantly, after experiencing anxiety over a challenging memory task, thoughts and stereotypes about the self, aging, and memory become more accessible. If this is the case it should be apparent in the possible selves generated by these older adults after priming. We must be cautious when interpreting these results due to the fact that only 28 percent of the older participants spontaneously reported memory-related hopes and fears. In particular, we must point out that the group of individuals who reported a hoped-for memory self was small. Further, half of these individuals endorsed both a hoped-for and a feared memory self. Although memory may be particularly salient for these individuals with both a memoryrelated hope and fear, further investigation is warranted before any conclusion can be made. The cross-sectional nature of the current study also limits the conclusions we can currently draw. From the current data, we cannot conclude that memoryrelated selves increase with age, however the current results warrant further exploration. A longitudi nal study of memory-related possible selves will be a necessary and exciting endeavor, and will help us explore when memory selves begin to appear, and if the nature of these selves as hoped-for or feared change over time. Along the same lines, the current study did not recruit a middle-aged sample. It will be important to examine this age group in the future, to understand when memory-related possible selves begin to appear in the selves spontaneously generated by older adults. Greater cultural diversity would also

PAGE 86

78 extend this research, and help determine whether memory selves are equally apparent across different racial and cultural backgrounds. A baseline test of memory ability was not included in the current study. Although the memory evaluation scale showed that younger and older adults in the current sample reported similar levels of performance and satisfaction regarding their recent memory performance, a direct measure of actual ability may be useful. A direct measure of memory performance would allow us to examine how individuals with hoped-for or feared memory-related selves actually performed in comparison to individuals not reporting a memory-related self. It might be the case, for instance, that feared selves are more strongly related to actual performance than to other beliefs about memory. Last, group testing may have limited the amount of data collected on the open-ended portion of the possible selves questionnaire. Most of the previous research on possible selves has been conducted using individual interview sessions. This was done because the questionnaire could potentially be confusing to older adults (Hooker, 1992; Hooker & Kaus, 1992; Hooker, 1999). However, our results are consistent with those obtained from individual interviews, suggesting that this methodological change was not problematic. On average, 86% of the older participants in these prior studies reported healthrelated selves, which included both physical and cognitive-related selves such as having a heart attack or becoming an Alzheimer’s patient. Consistent with these previous results, in the current sample, 83.5% of our participants reported a health-related self when health and memory-related selves were combined.

PAGE 87

79 Although it is true that we may have elicited richer descriptions of possible selves in a one-on-one setting, participants in the current study reported a diverse and rich array of future selves in a group setting. The groups were small and an interviewer was always present to answer any questions. The relationship between memory-related possible selves and subsequent behavior is not clear. Stereotypes about aging and memory create a culture in which memory decline is an expected part of growing old. In such an environment, instances of forgetting may become more salient to older adults, acting as proof of inevitable decline. As a consequence, older adults may feel less capable of and less control over maintaining or improving their memory abilities. These expectations and experiences may be of even greater importance for older adults who place a high value on memory ability (Hess et al., 2003). Individuals who place a high value on memory ability may be more likely to have a memory-related possible self. Further, an individual who has a memory self as a central feature of their self-concept may be more likely to actively pursue or avoid this self than someone for whom memory is less schematic. When stereotypes, decreased feelings of efficacy and control, and actual age-related declines in memory performance are combined with the knowledge that the causes of Alzheimer’s disease are largely unknown, attempting to take the necessary actions to avoid a feared memory-related possible self may be seen as futile for aschematic individuals. This has been the first study to examine memory-related possible selves. Although memory beliefs have been examined extensively, the relationship

PAGE 88

80 between these beliefs and an individual’s self-concept, especially in terms of possible selves, is not well understood. Older adults are more likely to incorporate a memory-related self into their self-concept. Further these memoryrelated selves are more often feared selves, not hoped-for. The ultimate goal of future investigations is to determine how these selves affect older adults, by exploring how the presence of these memory-related hopes and fears motivate or debilitate behavior.

PAGE 89

81 APPENDIX A CLOSED-ENDED POSSIBLE SELVES MEASURES Circle all the items that describe you now : Happy Senile Confident Depressed Intelligent Good-looking Sound mind Blind Rich Not able to fit in Successful Important Alzheimer’s patient Competent In good shape Have lots of friends Owner of a business Paralyzed Make own decisions Manipulate people Trusted Unimportant Remember what I need to Offensive Welfare recipient Sexy Brain dead Spouse or child abuser Athletic Wrinkled Active social life Being health conscious Remember every detail Good friend Not in control Likely to die young Admired Mentally alert all my life Unwanted A good parent Alone Interesting Adjusted Underachiever Incurable memory loss Optimistic

PAGE 90

82 Circle all the items that describe you now : Alcohol dependent Able to fix things Secure Reasonably alert Able to influence people Unable to remember Knowledgeable about art Failing memory Being appreciated Loved Unpopular Able to cook well Respected A cancer victim Long-lived Nonaggressive A failure Have a heart attack Not mentally impaired Win high honors Artistic Stupid Attractive Keen memory Independent Have a nervous breakdown Fuzzy memory

PAGE 91

83 Circle all the items that described you in the past : Happy Senile Confident Depressed Intelligent Good-looking Sound mind Blind Rich Not able to fit in Successful Important Alzheimer’s patient Competent In good shape Have lots of friends Owner of a business Paralyzed Make own decisions Manipulate people Trusted Unimportant Remember what I need to Offensive Welfare recipient Sexy Brain dead Spouse or child abuser Athletic Wrinkled Active social life Being health conscious Remember every detail Good friend Not in control Likely to die young Admired Mentally alert all my life Unwanted A good parent Alone Interesting Adjusted Underachiever Incurable memory loss Optimistic

PAGE 92

84 Circle all the items that described you in the past : Alcohol dependent Able to fix things Secure Reasonably alert Able to influence people Unable to remember Knowledgeable about art Failing memory Being appreciated Loved Unpopular Able to cook well Respected A cancer victim Long-lived Nonaggressive A failure Have a heart attack Not mentally impaired Win high honors Artistic Stupid Attractive Keen memory Independent Have a nervous breakdown Fuzzy memory

PAGE 93

85 Circle all the items that describe a future self that you would like to be: Happy Senile Confident Depressed Intelligent Good-looking Sound mind Blind Rich Not able to fit in Successful Important Alzheimer’s patient Competent In good shape Have lots of friends Owner of a business Paralyzed Make own decisions Manipulate people Trusted Unimportant Remember what I need to Offensive Welfare recipient Sexy Brain dead Spouse or child abuser Athletic Wrinkled Active social life Being health conscious Remember every detail Good friend Not in control Likely to die young Admired Mentally alert all my life Unwanted A good parent Alone Interesting Adjusted Underachiever Incurable memory loss Optimistic

PAGE 94

86 Circle all the items that describe a future self that you would like to be: Alcohol dependent Able to fix things Secure Reasonably alert Able to influence people Unable to remember Knowledgeable about art Failing memory Being appreciated Loved Unpopular Able to cook well Respected A cancer victim Long-lived Nonaggressive A failure Have a heart attack Not mentally impaired Win high honors Artistic Stupid Attractive Keen memory Independent Have a nervous breakdown Fuzzy memory

PAGE 95

87 Circle all the items that describe a future self that you would NOT like to be: Happy Senile Confident Depressed Intelligent Good-looking Sound mind Blind Rich Not able to fit in Successful Important Alzheimer’s patient Competent In good shape Have lots of friends Owner of a business Paralyzed Make own decisions Manipulate people Trusted Unimportant Remember what I need to Offensive Welfare recipient Sexy Brain dead Spouse or child abuser Athletic Wrinkled Active social life Being health conscious Remember every detail Good friend Not in control Likely to die young Admired Mentally alert all my life Unwanted A good parent Alone Interesting Adjusted Underachiever Incurable memory loss Optimistic

PAGE 96

88 Circle all the items that describe a future self that you would NOT like to be: Alcohol dependent Able to fix things Secure Reasonably alert Able to influence people Unable to remember Knowledgeable about art Failing memory Being appreciated Loved Unpopular Able to cook well Respected A cancer victim Long-lived Nonaggressive A failure Have a heart attack Not mentally impaired Win high honors Artistic Stupid Attractive Keen memory Independent Have a nervous breakdown Fuzzy memory

PAGE 97

89 APPENDIX B MEMORY VALUE QUESTIONNAIRE MEMORY VALUE On this page, there are some questions asking for your opinions. To answer each question, you should circle the number that best indicates your opinion Please read each question carefully before you decide how to answer. There are no right or wrong answers on these questions. Please circle the number that best represents your opinion. 1. If you don’t have your memo ry you don’t have anything. 1 2 3 4 5 6 7 Strongly Strongly disagree agree 2. There are many things I care about more than my memory. 1 2 3 4 5 6 7 Strongly Strongly disagree agree 3. A good memory is of only minor importance in a happy life. 1 2 3 4 5 6 7 Strongly Strongly disagree agree 4. There is nothing more important than a good memory. 1 2 3 4 5 6 7 Strongly Strongly disagree agree

PAGE 98

90 LIST OF REFERENCES Berry, J.M. & West, R.L. (1993). Cognitive self-efficacy in relation to personal mastery and goal setting across the life span. International Journal of Behavioral Development, 16 351-379. Berry J.M., West, R.L. and Dennehey, D.M. (1989). Reliability and validity of the memory self-efficacy questionnaire. Developmental Psychology, 25 701713. Bower, G. (1994). Temporary emotional states act like multiple personalities. In R. M. Klein & B. K. Doane (Eds.), Psychological concepts and dissociative disorders (pp. 207-234). Hillsdale, NJ: Lawrence Erlbaum Associates. Cameron, J.E. (1999). Social identity and the pursuit of possible selves: Implications for the psychological well-being of university students. Group Dynamics: Theory, Research, and Practice, 3 179-189. Cavanaugh, J. C., Feldman, J. M., and Hertzog, C. (1998). Memory beliefs as social cognition: A reconceptualization of what memory questionnaires assess. Review of General Psychology, 2 48-65. Cross, S., & Markus, H. (1991). Possible selves across the life span. Human Development, 34 230-255. Cross, S.E. & Markus, H.R. (1994). Self-schemas, possible selves, and competent performance. Journal of Educational Psychology, 86 423-438. Dixon, R. A. & Hultsch, D. F. (1999). Intelligence and cognitive potential in late life. In J. C. Cavanaugh & S. K. Whitbourne (Eds.), Gerontology: An interdisciplinary perspective (pp. 213-237). London: Oxford University Press. Dixon, R. A., Hultsch, D. F., & Hertzog, C. (1988). The Metamemory in Adulthood (MIA) Questionnaire. Psychopharmacology Bulletin, 24 671-688. Erber, J. T. (1989). Young and older adults’ appraisal of memory failures in young and older adult target persons. Journals of Gerontology, 44, P170P175.

PAGE 99

91 Frazier, L.D., Hooker, K., Johnson, P.M., & Kaus, C.R. (2000). Continuity and change in possible selves in later life: A 5-year longitudinal study. Basic and Applied Social Psychology, 2 237-243. Heckhausen, J., Dixon, R. A., & Baltes, P. B. (1989). Gains and losses in development throughout adulthood as perceived by different adult age groups. Developmental Psychology, 25, 109-121. Hertzog, C., & Dixon, R. A. (1994). Metacognitive development in adulthood and old age. In J. Metcalfe & A. P. Shimamura (Eds.), Metacognition: Knowing about knowing Cambridge, MA: The MIT Press, pp. 227-251. Hertzog, C., Lineweaver, T. T., & McGuire, C. L. (1999). Beliefs about memory and aging. In T. M. Hess & F. Blanchard-Fields, (Eds.) Social cognition and aging San Diego, CA: Academic Press, Inc. pp. 43-68. Hess, T. M., Auman, C., Colcombe, S. J., & Rahhal, T. A. (2003). The impact of stereotype threat on age differences in memory performance. Journal of Gerontology, 58 P3-11. Hooker, K. (1992). Possible selves and perceived health in older adults and college students. Journal of Gerontology, 47 P85-95. Hooker, K. (1999). Possible selves in adulthood: Incorporating teleonomic relevance into studies of the self. In T. M. Hess & F. Blanchard-Fields (Eds.) Social cognition and aging (pp. 97-122). San Diego, CA: Academic Press. Hooker K., Fiese, B.H., Jenkins, L., Morfei, M.Z., & Schwagler, J. (1996). Possible selves among parents of infants and preschoolers. Developmental Psychology, 32 542-550. Hooker, K. & Kaus, C.R. (1992). Possible selves and health behaviors in later life. Journal of Aging and Health, 4 390-411. Hooker, K. & Kaus, C.R. (1994). Health-related possible selves in young and middle adulthood. Psychology and Aging, 9, 126-133. Hultsch, D. F., Hertzog, C., & Dixon, R. A. (1987). Age differences in metamemory: Resolving the inconsistencies. Canadian Journal of Psychology, 41, 193-208. Kato, K. & Markus, H. R. (1993). The role of possible selves in memory. Psychologia, 36 73-83. Lachman, M. E., Bandura, M., Weaver, S. L., & Elliott, E. (1995) Assessing memory control beliefs: The Memory Controllability Inventory. Aging & Cognition, 2, 67-84.

PAGE 100

92 Levy, B. (1996). Improving memory in old age through implicit self-stereotyping. Journal of Personality and Social Psychology, 71, 1092-1107. Lineweaver, T. T. & Hertzog, C. (1998) Adults’ efficacy and control beliefs regarding memory and again: Separating general from personal beliefs. Aging, Neuropsychology & Cognition, 5, 264-296. Markus, H.R. & Herzog, A.R. (1992). The role of self-concept in aging. In K. W. Schaie& M. P. Lawton (Eds.), Annual Review of Gerontology and Geriatrics (Vol 11, pp. 110-143). New York: Springer. Markus, H. & Nurius P. (1986). Possible selves. American Psychologist, 41 954969. Markus, H. & Wurf, E. (1987). The dynamic self-concept: A social psychological perspective. Annual Review of Psychology, 38 299-337. McDonald-Miszczak, L., Hertzog, C., & Hultsch, D. F. (1995). Stability and accuracy of metamemory in adulthood and aging: A longitudinal analysis. Psychology & Aging, 10, 553-564. Rahhal, T. A., Hasher, L., & Colcombe, S. J. (2001). Instructional manipulation and age differences in memory: Now you see them, now you don’t. Psychology and Aging, 16 697-706. Ruvolo, A.P., & Markus, H.R. (1992). Possible selves and performance: The power of self-relevant imagery. Social Cognition, 10 95-124. Ryan, E. B. (1992). Beliefs about memory changes across the adult lifespan. Journals of Gerontology, 47, P41-P46. Ryff, C.D. ( 1991). Possible selves in adulthood and old age: A tale of shifting horizons. Psychology and Aging, 6 286-295. Selhulster, J. R. (1981a). Phenomenological correlates of a self theory of memory. American Journal of Psychology, 94 527-537. Sehulster, J. B. (1981b). Structure and pragmatics of a self-theory of memory. Memory & Cognition 9, 263-276. Schaie, K. W. (1994). The course of adult intellectual development. American Psychologist, 49 304-313. Smith, J. & Freund, A. M. (2002). The dynamics of possible selves in old age. Journal of Gerontology, 57B P492-P500.

PAGE 101

93 Ware, J. E. (1993). SF-36 Health Survey: Manual and Interpretation Guide. Nimrod Press: Boston, MA. Ware, J. E. & Sherbourne, C. D. (1992). The MOS 36-item short-form health survey (SF-36). Medical Care, 30(6), 473-483. West, R. L. & Berry, J. M. (1994). Age declines in memory self-efficacy: general or limited to particular tasks and measures? In J. D. Sinnott (Ed.) Interdisciplinary handbook of adult lifespan learning (pp. 426-445). Westport, CT: Greenport Publishing Group. West, R. L., Thorn, R. M., & Bagwell, D. K. (2003). Memory performance and beliefs as a function of goal setting and aging. Psychology and Aging, 18 111-125. West, R. L., Welch, D. C., & Thorn, R. M. (2001). Effects of goal-setting and feedback on memory performance and beliefs among older and younger adults Psychology and Aging, 16 240-250.

PAGE 102

94 BIOGRAPHICAL SKETCH Alissa Dark-Freudeman was born in Akron, Ohio, on December 13th, 1975. She attended Saint Thomas Aquinas High School in Ft. Lauderdale, Florida. She went on to attend the University of Florida and received a Bachelor of Arts in linguistics with a minor in business administration in 1998. She graduated with highest honors after completing a senior honors thesis on sexism and violence in children’s literature. After taking some time off to work and experience the “real” world, she returned to the University of Florida as a graduate student in developmental psychology in 2001.


Permanent Link: http://ufdc.ufl.edu/UFE0005642/00001

Material Information

Title: Memory-Related Possible Selves: Exploring Age-Related Differences
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0005642:00001

Permanent Link: http://ufdc.ufl.edu/UFE0005642/00001

Material Information

Title: Memory-Related Possible Selves: Exploring Age-Related Differences
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0005642:00001


This item has the following downloads:


Full Text











MEMORY-RELATED POSSIBLE SELVES: EXPLORING AGE-RELATED
DIFFERENCES














By

ALISSA DARK-FREUDEMAN


A THESIS PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
MASTER OF SCIENCE

UNIVERSITY OF FLORIDA


2004
































Copyright 2004

by

Alissa Dark-Freudeman















ACKNOWLEDGMENTS

I would like to thank my advisor and mentor, Robin Lea West, for her

guidance and encouragement. I would also like to thank my friends and family

for their support. I thank my parents for their kind words and continuing interest

in my education. And last, but certainly not least, I would like to thank my

husband and friend, Jason Thomas Fleming, for his patience and assistance

throughout this journey. As I am always fond of saying, we have only just begun.















TABLE OF CONTENTS

Page

A C K N O W LE D G M E N T S ............ ............................. .................. .............. iii

LIST O F TA BLES ................................................................... ............ vi

ABSTRACT ................................. ............................... ......... vii

CHAPTER

1 IN T R O D U C T IO N ......................................................................................... 1

The Nature and Relevance of Possible Selves.................................... 2
Current Research on Possible Selves .................. .. ............ ...............3
Information Processing and Behavior .................. ......... ............... 4
Contextual Effects on Possible Selves ............... ................... ..............9
Differences and Changes in Possible Selves with Age.............................. 12
Changes in Memory Beliefs .............. ..................... .................. 13
General Beliefs About Memory ............... ....... ................... 14
Personal Beliefs About Memory ..................................... ............. 19
Memory-related Possible Selves ......... ............ ....... .... ............. 25
Methods Used to Measure Possible Selves ............... ....... ............... 26

2 STUDY 1 INTRODUCTION........... .... ..... .. ....................... ...............29

3 STUDY 1 M ETHODS ................. ........ ................ .. ....... ....... ....32

Participants .. .................... .. ......... .................................. 32
M measures .............. ................... ............ .. ..................... 33
Possible Selves .................................... ............... 33
M em ory B eliefs ............. ........................... ............... .... ........ 34
Procedure ......... ....... ............ .. ............... ...................... 39

4 STUDY 1 RESULTS ........................... .. ........................42

Spontaneous Selves ............... ............ .. ......... ...... ........... 42
Selves Endorsed on Closed-ended Questionnaire ............... ...... ....... 44
A additional A nalyses ........................................ ............... ............... 48









5 STUDY 2 INTRODUCTION........... ......... ........ .................. 64

6 STUDY 2 METHODS .......... .............. .... ............. ....... 67

Participants ............................. ...................... 67
Procedure ............................ ....... ..... ............... 67

7 STUDY 2 Results ................................................. 69

8 DICUSSION ..................................... ....... ...............70

APPENDIX

A CLOSED-ENDED POSSIBLE SELVES MEASURES .........................81

B MEMORY VALUE QUESTIONNAIRE ......... ........ ........ .................... 89

LIST OF REFERENCES ............... ..... ......... .............. 90

BIOGRAPHICAL SKETCH .. ... ................................ .......94


































V















LIST OF TABLES


Table page

3-1. Mean Education and Health Ratings .............. ...................... ................. 40

3-2. Percentage of Agreement Between Independent Raters...........................41

4-1. Percentage of Spontaneous Memory Selves Reported by Age Group.......51

4-2. Type of Spontaneous Memory Selves by Age Group .............................. 52

4-3. Spontaneous Cognitive Selves by Age Group............................ ...... 53

4-4. Mean Number of Memory Selves Endorsed on Closed Questionnaire .......54

4-5. Intercorrelations Between Dependent Subscales........... ... ............... 55

4-6. Mean Memory Belief Ratings by Age Group .................. .................. 56

4-7. Analysis of Variance for Dependent Subscales by Type of Memory Self
Spontaneously Reported............ ...................... ............. 57

4-8. Type of Memory Selves Reported By Age Group ............... ............... 59

4-9. Mean Number of Selves by Age Group ............ .. .... .................. 60

4-10. Mean Capability and Likelihood Ratings: Feared......................... 61

4-11. Mean Capability and Likelihood Ratings: Hoped-for..............................62

4-12. Mean Number of Goal Activities ..... ................. ................. 63

6-1. Mean Demographic Information for Primed and Unprimed Young Adults...68















Abstract of Thesis Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Master of Science

MEMORY-RELATED POSSIBLE SELVES: EXPLORING AGE-RELATED
DIFFERENCES

By

Alissa Dark-Freudeman

August 2004

Chair: Robin L. West
Major Department: Psychology

Thoughts about the self in the future are an important component of the

current identity of any individual. Although we all have future-oriented thoughts,

we each have our own unique map of avenues we wish to travel and dead ends

we plan to avoid. Markus and Nurius refer to such thoughts about our selves in

the future as possible selves. Age-related changes and differences in possible

selves have been documented, as have age-related differences and changes in

beliefs about memory. The current study extended research in both of these

areas. This investigation sought to determine if memory-related possible selves

exist, to examine the nature of these selves as hoped-for or feared, and to

understand how the existence of these memory-related selves may differ with

age. The study also examined the relationship between memory-related possible

selves and a wide range of memory beliefs that have been shown to be affected

by the aging process. This study also examined whether younger adults who









were primed to think about memory and aging might be more likely to

spontaneously generate memory-related selves. In general, the results of the

current study show that older adults do have memory-related possible selves,

whereas younger adults do not. Further, when these selves appear in an older

individual's repertoire, they are primarily feared selves like becoming senile or

losing one's mind. Primed younger adults were no more likely to report a

memory-related self than unprimed younger adults. The relationship between

the memory-related selves and memory beliefs reported by older adults

approached significance and warrants additional investigation in the future.














CHAPTER 1
INTRODUCTION

We all think about the future. We have dreams and we have nightmares

about the myriad of possibilities that await us somewhere off in the distance.

Can we achieve our goals? Will we fail to make our dreams reality? What will

happen to us? What will we be like in ten years, twenty years, or fifty years?

Thoughts about the self in the future have become an interesting topic in recent

years and are considered to be an important component of the current identity of

any individual (Markus & Nurius, 1986). Although we all share these types of

future-oriented thoughts, we each have our own unique map of avenues we wish

to travel and of dead ends we plan to avoid. Markus and Nurius (1986) have

called these thoughts about our selves in the future, possible selves. These

include the hoped-for selves we are striving toward and the feared selves we

wish to escape. Age-related changes and differences in possible selves have

been documented (Cross & Markus, 1991; Hooker, 1992; Hooker, Fiese,

Jenkins, Morfei, & Schwagler, 1996; Hooker & Kaus, 1992; Hooker & Kaus,

1994; Ryff, 1991), as have age-related differences and changes in memory

performance (Dixon & Hultsch, 1999; Schaie, 1994) and in beliefs about memory

(Berry, West, & Dennehey, 1989; Hultsch, Hertzog, & Dixon, 1987; Lachman,

Bandura, Weaver, & Elloitt, 1995). The current study extends research in these

areas. First, this research will focus on memory-related possible selves to

determine whether or not older and younger adults have possible selves in the









memory domain. Second, this study will determine whether or not these selves

are primarily hoped-for or feared. Third, this study will investigate the

relationship between aging, beliefs about memory, and memory-related possible

selves. Fourth, the impact of priming will be examined. The following review

discusses what possible selves are and why they are of interest, age-related

differences and changes in possible selves and memory beliefs, and how the

study of possible selves may contribute to our understanding of memory and

aging.

The Nature and Relevance of Possible Selves

Markus and Nurius defined (1986) possible selves as elements of the self-

concept that represent what an individual could become, would like to become,

or is afraid of becoming. A possible self is an idea we have created that

represents what we might be like in the future. These ideas are part of the self-

concept. These ideas or selves can be positive, negative, or neutral. They can

be concrete and realistic and they can be vague and improbable. They can be

relatively close in the future or quite distant, and they can contrast and

complement other possible selves. Although each individual has certain selves

that are central to his or her identity, the meanings of these selves may change

over time (Cross & Markus, 1991).

Possible selves are not set in stone. Possible selves change as each

individual changes: as some selves are achieved, new selves will take their

place, whereas other selves will be revised or dropped altogether. Which selves

will be dominant depends on the individual and the context he or she

experiences at any given point in time (Markus & Wurf, 1987). This online









shifting set of selves is what Markus and Nurius (1986) call the working self-

concept. On-line selves will be selected from an individual's entire repertoire of

possible selves, according to environmental demands. Thus, possible selves are

sensitive to the environment and to the individual. These personally tailored,

environmentally responsive selves motivate and guide actions and emotions in

personally meaningful ways (Markus & Nurius, 1986).

Possible selves guide behavior in several ways. Shifts in possible selves

create shifts in information processing (Kato & Markus, 1993), allowing

information that is related or relevant to a particular self to be processed more

quickly and efficiently (Cross & Markus, 1994; Ruvolo & Markus, 1992).

Changes in possible selves result in changes in levels of motivation and feelings

of self-efficacy (Cameron, 1999; Cross & Markus, 1991; Hooker, 1992). Possible

selves have also been related to psychological well-being and life satisfaction

(Cross & Markus, 1991; Ryff, 1991; Smith & Freund, 2002). Further, changes in

possible selves have been shown to result in changes in length of persistence

and amount of effort expended on specific tasks (Ruvolo & Markus, 1992).

Current Research on Possible Selves

This section briefly reviews current research on possible selves. First, the

selected studies emphasize the influence of possible selves on information

processing, including the effects possible selves exert on levels of persistence

and effort. Lastly, the effects of context on possible selves are highlighted, in

particular, how life course and age may affect our repertoire of possible selves.









Information Processing and Behavior

Ruvolo and Markus (1992) completed a series of three studies to determine

if possible self manipulations affected persistence and effort on tasks, and

accessibility of possible selves. In the first study, participants were exposed to

one of three imagery manipulations: success, failure, or positive affect.

Participants in the first two conditions were asked to imagine and write about

themselves in the future, experiencing either success or failure as a direct

consequence of their own actions, whereas participants in the positive affect

condition read a neutral passage and received candy. Persistence was

measured as the amount of time a participant spent copying numbers with his or

her non-dominant hand. Effort was measured as a participant's accuracy on a

circling e's task. The success group persisted most, followed by the positive

affect and failure groups respectively. The success group also put forth more

effort than the positive affect and failure groups.

In the second study, participants were exposed to one of four imagery

manipulations: success, failure, positive affect and other's success. Three of the

four manipulations were the same as in study one. The new group was asked to

imagine another person being successful, to determine if thoughts of success in

general might be responsible for the results of the prior study. Immediately after

the imagery manipulation, participants completed a computerized self-description

task in which they answered possible for me or not possible for me for a selection

of words and phrases, and response latencies were measured.

Although the groups exhibited no differences in the types of words they

selected as possible or not possible, their response latencies did differ indicating









that "categories of possible selves were differentially accessible" (Ruvolo &

Markus, 1992, p. 108) for all four groups. Participants in the success condition

exhibited faster endorsements of positive items and faster rejections of negative

items. Those in the other's success group were quick to endorse positive items;

however they did not differ from the failure group in response latencies for

negative items. Those in the positive affect group revealed response latencies

similar to the failure response latencies for both positive and negative items. No

differences were found between the four groups on response latencies for control

items.

In the third and final study, participants were exposed to one of four

imagery manipulations: success work, success luck, failure work, and failure

luck. Again, participants completed a computerized self-description task and in

addition to this, they completed an achievement value questionnaire.

Participants also completed performance tasks measuring effort and persistence.

Effort was measured by performance on a mental arithmetic test and persistence

was measured as time spent copying numbers with a non-dominant hand. No

differences were found on the achievement scale, suggesting that all participants

valued achievement to the same degree. On the effort task, the success work

group answered a higher percentage of problems correctly compared to the other

three groups, but this difference was not significant. The success work group

persisted longest on the task, followed by the two luck conditions and the failure-

work condition, respectively. Again, no differences were found in the types of

words endorsed as possible or not possible among the four groups. Participants









in the success work group did endorse positive and reject negative words more

quickly than the other three groups, whereas the failure work group was the

slowest to respond to both positive and negative items. The two luck conditions

fell between the success work and failure work conditions.

These results indicate that our online set of possible selves can be

manipulated. Further, changes in an individual's online set of selves affect

behavior in meaningful ways. Specifically, changes in information processing

occur in response to changes in an individual's online set of possible selves.

These changes can also affect the level of persistence and effort an individual is

willing to put forth.

Kato and Markus (1993) presented college students with negative and

positive adjectives and phrases that were encoded on one of four conditions:

semantic, ability, actual self or possible self. In the semantic condition,

participants determined whether adjacent items were similar in meaning. In the

ability condition, participants provided a synonym for each target item. In the

actual-self condition, participants answered whether each item was descriptive of

their current self. In the possible-self condition, participants answered whether

each item would be descriptive of a future self. Possible-self orientation was

established by asking each participant to think about an important hoped-for self

that they often thought about attaining. After the encoding activity, participants

completed a surprise recall test of the items. Those in the actual and possible

selves conditions recalled more items than those in the semantic and ability









conditions. Participants also recalled more positive items than negative items

across all four conditions.

These results suggest that items related to an individual's self-concept are

more likely to be remembered by that individual. Thus, items relevant to a

possible self, especially a hoped-for self, are more likely to be attended to and

remembered.

Cross and Markus (1994) investigated the effects of an individual's

schematicity on performance and feedback. Self-schemas are identities that are

central to an individual. These schemas are core self-representations that define

who we are. Such schemas would be an important influence on the possible

selves that an individual might endorse, especially those that are spontaneously

mentioned. For both studies, a pre-selection questionnaire identified individuals

that were either aschematic or schematic for logical problem solving abilities. A

schematic individual has an organized and potentially automated way of

perceiving and responding to a particular stimulus or situation, in this case,

logical problem solving. In this study, schematic individuals indicated that they

perceived themselves as competent problem solvers and that this ability was

important to them. Aschematic individuals did not indicate that they had any

exceptional problem solving abilities, nor did they report that this ability was

particularly important to them. Participants also completed a logic activity, after

which they answered questions about the test and indicated their perceptions

about the difficulty of the logic test.









In the first study, both schematic and aschematic participants reported

similar levels of self-esteem and optimism, and both performed equally well on

the logic test. Schematic adults reported enjoying the logic test more than

aschematic adults. They also indicated that the activity was easier and that they

felt more control over their results on the activity. Further, schematic individuals

endorsed more positive logic-related items as possible for me than did

aschematic individuals. Aschematic individuals were also significantly faster to

respond possible for me for the negative logic-related items. No other significant

differences in response latencies were discovered.

The second study investigated the effect of feedback on subsequent

performance for schematic and aschematic individuals. Again, no differences

were found in initial problem solving abilities. This study consisted of the same

measures in the first study; however, feedback was given after the first logic test.

Participants either received failure feedback or no feedback at all. Schematic

individuals in the feedback condition responded possible for me more quickly for

the positive logic-related terms than any other group. Aschematic individuals in

the feedback condition responded possible for me significantly faster for the

negative logic-related terms. No other significant differences in response

latencies were discovered.

Following the first test, a second logic test was given. Aschematic

individuals in the no feedback condition performed significantly worse on the

second logic test than did those individuals in the other three groups.

Aschematic individuals in the failure feedback condition performed similarly to the









schematic individuals in both conditions. Cross and Markus (1994) suggested

that feedback acted as an external incentive for aschematic individuals to

redeem and prove themselves capable. Without feedback, aschematic

individuals lost interest. Therefore, this external push was necessary to

compensate for the lack of positive logic-related self-conceptions and

consequent lack of motivation these individuals face within the domain of logic.

Conversely, schematic individuals relied on positive logic-related self-

conceptions to cope with or discount negative feedback and maintain

performance.

In this situation, schematic individuals can be thought of as having a

positive logic-related possible self. The existence of this positive logic-related

self affected information processing, and behavior. This self allowed schematic

individuals confronted with failure to shift their attention away from negative,

toward more positive self-descriptors. Further, the existence of this possible self

acted as a goal, motivating schematic individuals to persist during difficulty and

maintain a higher level of effort.

Although possible selves can influence information processing and

behavior, possible selves can also be influenced by external circumstances or

conditions. External influences include specific situations and broader contextual

factors, such as life stage or age. These contextual factors call certain selves to

action and fine tune existing repertoires of selves.

Contextual Effects on Possible Selves

Hooker et al. (1996) measured parenting possible selves between two

different groups of parents in different stages of parenthood. The two groups of









parents each completed an open-ended possible selves questionnaire. The

parent-infant group consisted of individuals between the ages of 18 and 43 years

old, whose first child was 12 months old or younger. The parent-preschool group

consisted of individuals between the ages of 24 and 45 years old, whose oldest

child was between 2 and 5 years of age.

Both groups of parents listed occupational and parenting selves most

frequently, in fact, two-thirds of all participants listed a parenting self as one of

their most important selves. The parent-infant group was more likely to have

hoped-for parenting selves than the parent-preschool group, while the parent-

preschool group was more likely to have feared parenting selves than the parent-

infant group, suggesting that the different life stages of the children created a

context in which parents saw their future selves.

A study by Hooker and Kaus (1994) investigated the possible selves of

young and middle-aged adults to determine if health-related selves existed for

these age groups. Young adults were between the ages of 24 and 39 years old,

while middle-aged adults were between 40 and 59 years old. Both young and

middle-aged adults completed an open-ended possible selves questionnaire

followed by efficacy and goal activity questions. After the possible selves

questionnaire, each group completed a health behavior measure (e.g., yearly

physical examinations, exercise) and a health value questionnaire (e.g., If you

don't have your health you don't have anything).

Middle-aged adults reported significantly more health-related selves than

young adults. Both groups were more likely to have a feared health-related self









than a hoped-for one. For participants with a health-related self selected as most

important, health value, perceived self-efficacy, and number of goal oriented

activities (taken to avoid a feared health-related self) were all significant

predictors of health behaviors. Perceived self-efficacy and number of goal-

oriented activities were the strongest predictors of health behavior scores.

Hooker (1992) also investigated the possible selves of younger and older

adults to determine if health-related selves existed for these age groups. Young

adults were between the ages of 18 and 23 years old, while older adults were

between 55 and 89 years old. Both younger and older adults completed an

open-ended possible selves questionnaire, followed by efficacy and goal activity

questions. After the possible selves questionnaire, each group completed a

perceived health and health value questionnaire.

Initially, Hooker reported that 71% of younger adults listed a health-related

self, and 17% selected a health-related self as most important. However, these

percentages included weight-related selves. For younger adults, weight-related

selves are probably more associated with concerns about appearance than with

concerns for health. When weight-related selves were removed, 64% of younger

adults reported no health-related self at all. On the other hand, 86% of older

adults reported a health-related self. Of these, 73% selected a health-related self

as one of their most important selves. In general, participants with health-related

selves scored higher on the health value measure; consequently, older adults

scored higher on health value than young adults. Lastly, younger adults

generated more hoped-for and feared selves than older adults.









Again, life stage plays an important role in the differences and changes that

occur in any persons' set of possible selves. These shifting sets of selves will

ultimately affect behavior in meaningful ways, prioritizing selves, ultimately

motivating the current self into action.

Differences and Changes in Possible Selves with Age

Given the fluid nature of possible selves, recent studies have started to

examine age-related differences and changes in possible selves, and several

age-related differences have been found. Differences exist not only in the types

of selves reported by age, but in the number of selves reported. Older adults

consistently report fewer possible selves than younger adults (Cross & Markus,

1991; Hooker, 1992). This was true for the number of hoped-for and the number

of feared selves. Older adults also report more specific, more realistic possible

selves than younger adults (Cross & Markus, 1991; Hooker, 1992). In general,

the possible selves older adults report are more in line with their current selves.

They typically involve the continuation or maintenance of current activities and

relationships, but also include room for continued learning and growth (Cross &

Markus, 1991; Frazier, Hooker, Johnson, & Kaus, 2000; Hooker, 1992).

Age differences in specific domains have also been found. Hooker and

Kaus (1994) found that health-related selves begin to increase in frequency in

middle age. Older adults spontaneously generated more health-related selves

than middle-aged adults, whereas younger adults reported fewer health-related

selves than both middle-aged and older adults (Hooker, 1992; Hooker & Kaus,

1992; Hooker & Kaus, 1994). Thus, the frequency of health-related selves may

increase with age as health becomes more salient during later life.









Later, in a longitudinal study by Frazier et al. (2000) patterns of change and

stability in the possible selves of men and women between the ages of 55 and 89

were identified. The participants initially reported possible selves in a study by

Hooker (1992). Then the same participants were contacted again five years later

to determine whether their possible selves would remain stable over time.

Although continuity was found in most domains, physical and health-related

selves were mentioned more frequently over time. Age-related changes in

possible selves do occur, and these shifts in possible selves may be important

components of successful aging (Markus & Herzog, 1992), providing "blueprints

for personal change and growth across the life span" (Cross & Markus, 1991, p.

232).

Changes in Memory Beliefs

Just as the aging process influences views of the self in relation to health, it

also influences views of the self in relation to memory. Age-related changes and

differences in memory abilities have been widely established. As a

consequence, we all learn to expect memory decline as a part of life, specifically

as a part of growing old. Our expectations and beliefs about memory are

composed of general beliefs and personal beliefs (Hertzog & Dixon, 1994).

General beliefs about memory include expectations we have about how memory

will change over time for general population. On the other hand, personal beliefs

are opinions we have about our own memory abilities. Personal beliefs are

affected by general beliefs but they need not mirror one another. This review

discusses general and personal memory beliefs, why they are important, and

some of the current research findings regarding these beliefs.









General Beliefs About Memory

General beliefs about memory are common beliefs about the process of

aging that relate to memory (Hertzog, Lineweaver, & McGuire, 1999). These

beliefs broadly apply to all people. General beliefs about memory include beliefs

about the expected developmental trajectory of memory performance,

stereotypes about memory and aging, and perspectives on memory failure.

Heckhausen, Dixon, and Baltes (1989) investigated age differences in the

perceived trajectory of development throughout adulthood. Young, middle-aged

and older adults, between the ages of 20 to 85, rated adjectives in terms of

expected developmental increase, desirability, onset age, and closing age. All

participants were instructed to think about people in general, rather than a

specific target, during the entire exercise. Results indicated that all age groups

expected a similar developmental path and this path was characterized by both

gains and losses. However, the percentage of perceived losses increased

across the lifespan. Gains actually outnumbered losses, with the oldest group

endorsing more possible gains than the middle-aged and young adults,

respectively. Older adults also used a greater range of onset and closing ages

than younger adults, indicating a more flexible and gradual view of decline.

Lineweaver and Hertzog (1998) examined general beliefs about memory in

both younger and older adults. Participants were between 18 and 93-years-old.

Each participant completed a questionnaire designed to measure general beliefs

about memory. In general, all participants believed that memory self-efficacy and

memory control declined with age, with the greatest decline occurring after 50

years of age. Further, all participants believed memory ability declined with age;









however, older adults differed from both young and middle-aged adults regarding

the onset and magnitude of memory decline. In comparison to the two younger

groups, older adults believed memory declined later in life and more gradually

(Lineweaver & Hertzog, 1998).

In addition, Hertzog et al. (1999) reviewed several studies investigating

age-related changes in opinions and feelings about memory in older and younger

adults. Again, older and younger adults believed that memory ability, memory

self-efficacy, and control over memory all decline with age; however, the patterns

of decline differ by age group. Younger adults reported earlier, steeper declines

in memory control and memory self-efficacy, with memory decline beginning after

age 30. Older adults reported peaks in performance around age 40, and more

gradual declines in memory efficacy, control, and abilities.

Ryan (1992) also examined beliefs about the developmental trajectory of

memory, by investigating age differences in memory beliefs. In her first study,

adults between 18 and 74 years of age completed a questionnaire based on the

Short Inventory of Memory Experiences. Each participant completed three

versions of this questionnaire. They rated each question in terms of beliefs about

themselves, beliefs about a typical 25-year-old, and beliefs about a typical 70-

year-old. In general, memory ratings were more positive for the typical 25-year-

old than for the typical 70-year-old, whereas self-ratings fell between the two. In

the second study, participants completed the same questionnaire, rating each

item for themselves and for a typical person of 25, 45, 65, or 85 years of age.

The typical 25-year-old was rated more positively than all other age groups. The









typical 85-year-old was rated as having the poorest memory of all age groups.

The typical 45-year-old and the typical 65-year-old fell between the youngest and

oldest groups. Thus, memory decline was perceived to start during middle age

and continue on until later life. In comparison to participants with positive self-

ratings, participants with the poorest self-ratings expected all age groups to have

poorer memories.

These results confirm that a socially accepted pattern of memory

development exists. Again, memory abilities are expected to start declining in

middle age and to continue declining through old age. The majority of individuals

are believed to conform to this fixed pattern of development. Although these

standardized expectations apply to all individuals, they are particularly relevant to

older adults, as they often lead to social stereotypes about aging and memory.

Levy (1996) has examined the effects of age-related stereotypes. Her first

study investigated the effect of age-related stereotypes on the memory

performance of older adults between the ages of 60 and 90-years-old.

Participants initially completed a series of memory tests. After completing the

memory tests, participants completed a computerized priming task. The priming

task exposed each participant to either positive or negative words, related to

either wisdom or memory decline. The words flashed quickly on a computer

screen, so the participants were not aware of having seen them. This task was

designed to activate participants' internal stereotypes about aging and memory.

Levy (1996) found that the priming activity was successful and influenced

memory performance, attitudes toward aging, and beliefs about memory in older









adults. The older adults primed in the negative condition performed more poorly

and reported more negative attitudes about aging and memory. The older adults

primed in the wisdom condition, performed better and reported more positive

attitudes toward aging and memory. The results indicate that older adults

respond to stereotypes directly and without knowing that they are doing so. This

has important implications regarding the effects cultural stereotypes about aging

and memory have on older adults and the aging process they will expect and

experience.

Expectations of memory decline and negative age-related social

stereotypes reinforce one another. These beliefs affect the judgments we make

about others in everyday life, and bias us to look for instances that verify these

negative beliefs. Consequently, memory failures will become particularly salient,

especially when experienced by an older individual. Such general beliefs and

biases become particularly relevant as we interact with older adults, and as we,

ourselves age.

Erber (1989) explored age differences in the assessment of memory

failures. In the first study, participants between the ages of 18 and 29 were given

a booklet containing 30 memory failure episodes. Half of the episodes were

designed to show serious memory failures (e.g., forgetting the name of someone

the protagonist has known for years) and half were designed as nonserious

errors (e.g., forgetting the name of someone the protagonist just met). The

subject in each episode was always a 30 year-old woman or a 70 year-old

woman. Participants rated the seriousness of each episode on a 7-point Likert









scale, and consistently rated the memory failures of the 70 year-old woman as

more serious than those same failures made by the 30 year-old woman (Erber,

1989).

In a second study, older adults between 62 and 82 years of age completed

the same exercise. The older adults did not rate the seriousness of the memory

failures any differently for the 30 year-old and the 70 year-old. In addition, older

adults rated the serious failures less seriously than the younger adults had in the

previous study. Thus, younger adults rate the memory failures of older adults

more harshly than failures experienced by younger people. Further, older adults

are more moderate in their opinions regarding the seriousness of a memory

failure, regardless of age. This is most likely due to a greater acceptance of

memory decline as normal and inevitable among older adults (Erber, 1989).

Thus, expectations of memory deterioration and negative age-related

stereotypes do contribute to different perceptions of memory ability. Compared

to older adults, younger adults believe that memory failure is more serious.

Younger adults are also more concerned when memory failures occur among

older individuals. This bias reflects a belief that memory failures may indicate

early stages of serious cognitive impairment among older individuals. However,

with increasing age, adults consider memory failure a less serious offense,

because memory decline is an expected part of life. These general expectations

and beliefs about memory affect each of us in important ways. Not only do they

affect our perceptions of others, they influence our personal beliefs about our

own memory.









Personal Beliefs About Memory

Personal beliefs about memory are self-referent beliefs or beliefs that refer

or relate directly to the self. Self-referent beliefs about memory include memory

self-efficacy beliefs, memory control beliefs, and perceived changes in individual

memory abilities.

Self-efficacy is an individual's personal judgment of his or her capability to

perform a specific task or reach a specific goal. Self-efficacy beliefs are

important for several reasons. First, self-efficacy beliefs affect performance.

These beliefs guide which tasks and goals will be approached and which will be

avoided (Berry & West, 1993). Second, self-efficacy beliefs influence the amount

of effort and persistence an individual sustains once engaged in a task (Berry &

West, 1993). Thus individuals with high levels of self-efficacy will be more

motivated to continue working toward a related goal.

Self-efficacy beliefs are also domain specific (Berry & West, 1993). It is

therefore reasonable to expect each individual to have different self-efficacy

beliefs over multiple domains. Memory self-efficacy beliefs are specific

judgments related to an individual's capability to engage in and successfully

complete memory tasks. Given the general beliefs we hold regarding memory

decline, it is also reasonable to expect memory-related self-efficacy beliefs to

decline with age.

Berry et al. (1989) developed a measure to assess memory self-efficacy

beliefs in older and younger adults. Participants were between 18 and 80 years

of age. Each participant completed the Memory Self-efficacy Questionnaire and

eight memory tasks. Memory tasks were divided among familiar, everyday









memory activities like remembering phone numbers and less familiar, laboratory

tasks such as recalling word lists. These tasks were used to estimate the

accuracy of memory self-efficacy judgments. In general, younger adults had

higher self-efficacy scores on all memory tasks. Further, older adults efficacy

ratings were more accurate for everyday tasks than laboratory tasks. Thus, the

authors concluded that task familiarity is an important factor in accurate

estimates of self-efficacy (Berry et al., 1989).

In addition to memory self-efficacy beliefs, memory control beliefs are also

important, as both types of beliefs affect behavior. Memory control beliefs are

internal or external attributions regarding memory loss. If a person believes

memory loss is inevitable, it is viewed as completely out of ones control. If, on

the other hand, one believes memory loss is preventable or controllable, loss can

be minimized through individual efforts to maintain and improve upon ones

abilities. Control beliefs and self-efficacy beliefs are related, however they are

not the same. Self-efficacy beliefs are personal judgments about ones ability to

complete a task or goal. An individual may believe he or she is able to complete

a memory task, but may still feel that memory loss is an inevitable part of aging

(Hertzog et al., 1999).

Lachman et al. (1995) developed a measure to assess beliefs about

memory control and memory ability. They also examined how these beliefs

relate to actual memory performance in older and younger adults. Participants

were between 20 and 90 years of age. Each participant completed the Memory

Controllability Inventory and several different memory tasks. Participants who









felt secure about their current memory ability also felt they had more control over

their memory ability. These participants indicated that they believed making an

effort could improve their memory, and reported that they could find strategies to

do so. Participants who agreed with the inevitability of memory loss with age felt

less secure about their current memory ability and less able to make

improvements. With increasing age, people became more likely to report that

memory deterioration is inevitable and also became more likely to doubt their

own memory ability (Lachman et al., 1995).

Participants who predicted they could handle their daily memory tasks

independently, without depending on others, also rated their current memory

ability more positively. These individuals believed in their ability to improve their

memory, endorsed the use of effort, and did not believe that memory loss was

inevitable. In contrast, participants who were worried about developing

Alzheimer's disease rated their current memory ability less positively. These

individuals also believed in the inevitability of memory loss with age (Lachman et

al., 1995).

Interestingly, control beliefs were related to actual memory performance.

Beliefs in inevitable memory loss were associated with poorer performance on

the memory tasks. Positive beliefs about current memory ability and the ability to

improve memory were associated with better performance on the memory tasks.

Clearly, believing one has control over the maintenance of memory ability affects

behavior.









In addition to differences in memory self-efficacy and memory control

beliefs, older and younger adults have different perceptions regarding when and

how they will experience changes in their memory abilities. The personal

changes that older and younger adults report, generally coincide with the generic

expectations we have for the population, however, individual variation does

occur.

McDonald-Miszczak, Hertzog, and Hultsch (1995), conducted two

longitudinal studies in an attempt to uncover age-related changes in memory

beliefs and abilities. The first study looked at 231 adults between 22 and 78

years of age, over a period of two years. Participants completed several

metamemory questionnaires (Metamemory in Adulthood, Memory Functioning

Questionnaire, Memory Self-efficacy Questionnaire). In addition, participants

completed questions regarding perceived changes in memory ability and several

recall tasks. This study found considerable stability over the two-year period;

however, memory was reported as getting worse over time (McDonald-Miszczak

et al., 1995).

The second study looked at adults between 55 and 86 years of age, at

three points in time over a period of six years. At the end of the 6-year period,

234 of the initial 477 participants remained. Participants completed similar

measures as in the first study. Memory ability and self-efficacy both declined

over time, whereas anxiety about memory increased, as did reliance on external

aids and memory strategies (McDonald-Miszczak et al., 1995).









Hultsch et al. (1987) also explored age differences in beliefs about memory

change in two different communities. Participants were between 20 and 78 years

old, and each completed the Metamemory in Adulthood Scale and the Memory

Functioning Questionnaire. In general, when compared to younger adults, older

adults perceived themselves as having less memory capacity. Older adults also

reported that their memory had declined over time; further, they believed they

could do very little to improve their current memory abilities or to prevent memory

decline in the future.

Lastly, Lineweaver and Hertzog (1998) examined general beliefs about

memory and personal beliefs about memory in both younger and older adults, to

determine if personal beliefs and general beliefs differ, and if adults of different

ages have different beliefs. Participants were between 18 and 93-years-old.

Each participant completed two questionnaires; each designed to measure either

general or personal beliefs about memory.

Personally, older adults reported lower memory self-efficacy, and less

control over memory than both middle-aged and younger adults. Participants

also reported on perceived memory change. When comparing current memory

ability to past memory ability, younger adults reported improvement, middle-aged

adults reported little change, and older adults reported significant memory loss.

Participants also reported anticipated change between current and future

memory ability. In the future, younger adults anticipated improvement for

themselves, but not for the general population. Middle-aged adults anticipated









small declines, and older adults expected substantial decline (Lineweaver &

Hertzog, 1998).

Thus, general beliefs and personal beliefs are related, but not identical.

Personal beliefs appear responsive to life experience. Although older adults

report declines in memory ability, memory control, and memory self-efficacy, they

also report a later and more gradual decline than younger adults. Presumably

this is due to the effects of personal experience on general beliefs.

In conclusion, the results reviewed here confirm that memory decline is

seen as an inevitable and normative developmental process by adults of all ages

(Heckhausen et al., 1989; Lineweaver & Hertzog, 1998; Ryan, 1992). This

general belief results in age-related stereotypes and biases regarding memory

and memory failure. Older adults internalize these general beliefs and

stereotypes. Consequently, they learn to anticipate memory deterioration as a

normal part of the aging process (Levy, 1996). As a result, older adults tend to

downplay instances of memory failure as a normal part of life, whereas younger

adults tend to magnify instances of memory failure among older adults (Erber,

1989). Further, younger adults may use memory failure as a signal of potential

cognitive impairment among older adults.

Given the general beliefs we hold, it is not surprising that older adults also

report lower levels of self-efficacy than younger adults (Berry et al., 1989; Hultsch

et al., 1987), and less control over memory than younger adults (Hultsch et al.,

1987; Lachman et al., 1995).









Older adults also report experiencing declines in memory ability (Erber,

1989; Hultsch et al., 1987; Lachman et al., 1995; Ryan, 1992). However, the

decline for older adults occurs later in life and is more gradual than the decline

expected by younger adults. This indicates that personal experience can

influence personal beliefs, resulting in slight deviations from the generic path of

memory loss.

Thus, for older adults, memory beliefs become of notable significance, or

more salient, as they experience normative age-related declines and respond to

social stereotypes (Hultsch et al., 1987; Cavanaugh, Feldman, & Hertzog, 1998).

As a consequence, these age-related opinions and beliefs about memory should

be reflected in the possible selves of older adults.

Memory-related Possible Selves


Possible selves and memory beliefs fit together well. Sehulster (1981ab)

proposed a self-theory of memory, in which beliefs about personal memory

ability, and memory in general, are considered part of the beliefs we have about

our selves. Our memory-related beliefs and experiences create a memory self-

schema or identity. Further, these beliefs and experiences play a role not only in

our current identity, but also in how we view the trajectory of our future memory

abilities. More specifically, our current memory self-schema outlines a future

memory self. Like health, memory is believed to become more salient with age

(Berry & West, 1993; Cavanaugh et al., 1998; Hultsch, et al., 1987);

consequently, memory-related possible selves should become more apparent in

an older adults repertoire of possible selves. These memory-related selves and









beliefs should also affect behavior, such as persistence, effort, goal setting, and

strategy use on memory tasks (West & Berry, 1994).

Methods Used to Measure Possible Selves

Several methods have been used to measure possible selves. Initially

possible selves were measured by using lists of nouns and adjectives.

Participants indicated whether each item from the list of possibilities was self-

descriptive. Markus and Nurius (1986) gave participants a list of 150

possibilities. For each item, participants indicated if the item described them

currently, or had described them in the past. Participants were also asked if they

had ever considered the item as a possible self, and if so, how probable the self

was and how much they would like the self to be true for them. This approach

has several advantages and disadvantages. It may be possible to determine if

particular selves in specific domains are consistently endorsed or not; however,

individuals may have important selves that are not included, and consequently

not discovered.

Possible selves have also been measured by open-ended questionnaires

(Cameron, 1999; Cross & Markus, 1991; Frazier et al., 2000; Hooker, 1992;

Hooker & Kaus, 1992; Hooker & Kaus, 1994; Ruvolo & Markus, 1992). In these

studies, participants were asked to spontaneously list all of the possible selves

they would like to achieve, and all those that they would like to avoid.

Participants selected the two or three most important feared and most important

hoped-for selves and then explained why each was important. They also rated

the likelihood of attaining or avoiding these selves, and listed any steps actively

taken to achieve or avoid these selves. Although this method allows each









individual to share their unique set of selves, it may be more time consuming and

difficult for participants to understand (Hooker, 1992; Hooker, 1999).

Ryff (1991) asked participants to answer several questions from various

psychological well-being measures. The questions were answered multiple

times, in terms of each participant's current self, ideal self, past self, and future

self. Possible selves were not directly assessed in this study, but were used as a

manipulation "to determine perceptions of improved or worsened functioning over

time" (Ryff, 1991, p. 287).

Possible selves have also been assessed using latency response

measures (Cross & Markus 1994, Ruvolo & Markus, 1992). These measures

indicate which selves are most active in an individual's working self-concept after

some experimental manipulation. Ruvolo and Markus (1992) asked participants

to imagine themselves in the future. Participants were then instructed to imagine

their own success or failure as a direct outcome of their own actions and write

about what it would be like. After this imagery manipulation, participants

completed a self-description task, rating positive, negative, and control words as

either possible for me or not possible for me. Response latencies for each word

indicated whether or not a corresponding self was active or accessible in an

individual's working self-concept.

The current study plans to investigate a specific domain of possible selves.

An open-ended format can provide a glimpse into the most salient selves of each

individual (Hooker et al., 1996); however, a closed format allows participants to

endorse memory-related responses that are not prominent enough to lead to






28


spontaneous mention. The combination of both methods in this investigation

allowed us to fully tap the possibilities existing within each individual.














CHAPTER 2
STUDY 1 INTRODUCTION

Memory-related possible selves were examined in the current study by

administering both open-ended and closed-ended possible selves questionnaires

to younger and older adults. Within the older adult group, two age groups were

created: a young-old group with participants between 50 and 69 years of age,

and an old-old group with participants 70 years of age and older. This was done

to examine possible differences and similarities between the types of selves

generated by older adults at different points in life. In addition to the possible

selves measures, all participants completed several questionnaires assessing

their beliefs about memory to examine the relationship between age, possible

selves, and a number of memory beliefs, including locus of control, anxiety, and

beliefs about recent memory performance. Lastly, each participant completed a

health survey and a participant information form to provide basic demographic

information.

Pilot data had indicated the younger adults were unlikely to spontaneously

mention memory-related selves. Therefore, the current study also examined

whether younger adults might be more likely to spontaneously generate memory-

related selves when exposed to specific memory activities prior to completing the

possible selves questionnaires. Research by Levy (1996) suggested that

exposure to social stereotypes about aging and memory can affect memory

beliefs and memory performance of older adults on subsequent memory tasks.









Further, Rahhal, Hasher, and Colcombe (2001) have demonstrated that

instructions can activate the different self-schemas older adults have related to

aging and memory. The current study planned to examine if the working self-

concepts of younger adults could be influenced in a similar manner, by exposing

primed individuals to a memory task and several memory beliefs measures

before the possible selves questionnaires.

Hypotheses

As health becomes more important later in life, this is reflected in the self-

conceptions of older adults (Hooker, 1992). Older adults reliably report health-

related possible selves, whereas younger adults rarely report selves in this

domain (Frazier et al., 2000; Hooker, 1992; Hooker & Kaus, 1992; Hooker &

Kaus, 1994). Memory is also believed to increase in importance with age, as

age-related declines in performance must be managed, and social stereotypes

come to affect self-conceptions. As a consequence, we expected that older

adults would generate memory-related possible selves, especially feared

memory selves, and that the number of reported memory-related selves would

increase with age. Further, we expected these selves to be interrelated with a

constellation of memory beliefs.

1. Older adults would report a greater number of memory-related
selves than younger adults.

2. In particular, the old-old would report a greater number of memory-
related selves than the young-old.


3. Given the pattern of age-related changes in memory beliefs, older
adults are expected to primarily generate feared memory-related
selves.






31


4. In keeping with the experimental evidence on possible selves, we
expected to successfully manipulate the relevance of memory
selves through priming. Younger adults were expected to be more
likely to report memory-related possible selves when primed to do
so.

5. Lastly, beliefs about memory (control, anxiety, achievement,
memory value, fear of Alzheimer's disease and memory self-
evaluation) would be related to whether a person reported a
memory-related possible self, and the content of that self as
hoped-for or feared.














CHAPTER 3
STUDY 1 METHODS

Participants

Participants included both younger adults and older adults. Forty-five

younger adults, ages 18 to 22 (M= 18.84, SD= .852) were recruited from the

University of Florida's research participant pool. Eighty-five older adults between

the ages of 52 and 88 (M= 71.69, SD= 7.52) were recruited from local

communities. Older adults were split into two groups, a young-old and an old-old

group. The young-old group included 36 individuals between the ages of 50 and

70 years old (M= 64.94, SD= 5.06). The old-old group included 49 individuals

between the ages of 70 and 88 years of age (M= 76.65, SD= 4.59). Two older

adults were excluded from the study. One older adult was excluded for

depression, and another was excluded for failing to answer the questionnaire in a

consistent manner. Education and health were examined as a function of age.

The young-old group had significantly more years of education than both the old-

old group and the younger adults, F (2, 129) = 8.43, p < .000, eta2 = .117. No

significant differences were found in self-rated health between any of the three

age groups, F (2, 123) < 1, p > .25. See Table 3.1.









Measures

Possible Selves

To thoroughly examine possible selves, both closed and open-ended

measures were used. Initially, participants completed the open-ended

questionnaire developed by Cross and Markus (1991). This questionnaire was

designed to spontaneously elicit both hoped-for and feared possible selves.

Participants were first asked to list all of their hoped-for selves. Once they

completed this list, they were then asked to list their most important, hoped-for

self. Next participants answered two questions to assess self-efficacy and

perceived control over attaining their most important hoped-for self, and listed

any goal-oriented activities they had undertaken to accomplish this hoped-for

self.

Next, participants were asked to list all of their feared selves. After

completing this list, they were asked to list their most dreaded, feared self.

Participants then answered two questions to assess self-efficacy and perceived

control over avoiding their most dreaded self, and listed any goal-oriented

activities they had undertaken to avoid this feared self.

The open-ended questionnaire was followed by an adaptation of the closed-

ended questionnaire developed by Markus and Nurius (1986). The closed-ended

questionnaire consisted of 73 specific descriptors, considered to be negative

(unpopular, paralyzed), or positive (long-lived, admired). The majority of the

items used in this study were taken directly from the original questionnaire;

however, 14 memory-specific descriptors were added. The memory-specific

descriptors included negative items such as "Alzheimer's patient" and "unable to









remember," and positive items such as "mentally alert all my life" and "sound

mind." The memory-related items were selected as either positive or negative

based on pilot data. Each participant was asked to complete four versions of this

questionnaire, endorsing each item as either descriptive or not descriptive of their

current self, their past self, their future hoped-for self, and their future feared self.

The current study focused solely on the future hoped-for and feared selves. The

current and past selves were included to help participants think about themselves

in relation to a personal timeline.

Memory Beliefs

The Metamemory in Adulthood scale developed by Dixon, Hultsch, and

Hertzog (1988), was used to assess beliefs about memory ability. This measure

was selected because it is a standard measure of memory beliefs that has shown

consistent validity and reliability. The following three subscales were used:

achievement, anxiety, and locus of control. These scales were selected because

of their expected relationship to hoped or feared memory selves. For instance, a

participant listing a feared memory-related self may also report greater anxiety

over memory-related tasks and less control over their memory ability. These

subscales included items like the following: "It is important to me to have a good

memory," "I get upset when I cannot remember something," and "I have little

control over my memory ability." Each item was rated on a 5-point Likert scale (1

= agree strongly, 5 = disagree strongly). Achievement, anxiety and locus were

calculated by averaging an individual's responses within each subscale. Internal

consistency reliability was good for the three scales: achievement (alpha = .77),

anxiety (alpha = .86), and locus (alpha = .71).









Questions specifically related to Alzheimer's disease were included from

the Memory Controllability Inventory (MCI) developed by Lachman et al. (1995).

The current study expected a greater number of feared memory-related selves

among older adults, therefore, it was important to also assess the degree to

which individuals feared the most common type of memory loss in late life.

Typical items on this scale included: "When I forget something I am apt to think I

have Alzheimer's disease," and "I think there's a good chance I will get

Alzheimer's disease." Each item was also rated on a 5-point Likert scale (1 =

agree strongly, 5 = disagree strongly). The four responses were averaged to

create a scale measuring fear of Alzheimer's disease. This MCI scale had good

internal consistency reliability (alpha = .81).

Four additional items assessed how each individual values memory. The

memory value items were adapted from the Health Value questionnaire used by

Hooker (1992), and developed by Lau, Hartman, and Ware (1986). These items

were included to determine whether individuals reporting memory-related selves

valued memory more than those who did not report future selves related to

memory. Individuals responded on a 5-point Likert scale (1 = strongly agree, 5 =

strongly disagree) to the following items: (1) "If you don't have your memory you

don't have anything;" (2) "There are many things I care about more than my

memory;" (3) "A good memory is of only minor importance in a happy life;" (4)

"There is nothing more important than a good memory." A memory value score

was calculated by averaging an individual's answers across these four questions.

The memory value scale had good internal consistency reliability (alpha = .75).









Opinions about recent memory performance were assessed with four

questions developed by West, Thorn, and Bagwell (2003). These questions

were included to examine whether participants reporting memory-related selves

rated their recent memory performance differently than those who did not.

Participants were first primed to think about their everyday memory experience

by answering this item: "How important has it been to you to perform well on

memory activities in your everyday life?" (1 = not important at all, 7 = very

important). This was followed by the three critical scale items on a 7-point Likert

scale: "How have you performed on most memory tasks you have done

recently?" (1 = very poor, 7 = very good); "How do you think your memory

compares with most other people your age?" (1 = much worse, 7 = much better);

"How satisfied are you with your recent memory performance?" (1 = very

satisfied, 7 = very unsatisfied). The responses to these three questions were

summed to create one memory evaluation scale. The memory self-evaluation

scale had good internal consistency reliability (alpha = .88).

Lastly, the SF-36 short-form health survey developed by Ware and

Sherbourne (1992) assessed the degree to which health and depression affected

an individual's participation in daily activities. This survey was used because it

provides information on eight comprehensive subscales in a relatively short and

simple format. The SF-36 is composed of these subscales: General Health,

Physical Limitations, Physical Functioning, Bodily Pain, Vitality, Emotional

Limitations, Social Functioning, and Mental Health. General Health included four

questions rated on a 5-point Likert scale, such as "In general, would you say your









health is..." (1 = excellent, 5 = poor) and "I seem to get sick a little easier than

other people" (1 = definitely true, 5 = definitely false). Physical limitations were

examined by four yes-no questions, such as, "During the past 4 weeks, have you

had any of the following problems with your work or other regular daily activities

as a result of your physical health: Accomplished less than you would like"

(yes/no). Physical functioning was measured using ten questions rated on a 3-

point Likert scale, such as "Does your health now limit you in climbing several

flights of stairs?" (1 = yes, limited a lot, 3 = no, not limited at all). Bodily Pain was

assessed by two questions rated on a 6-point Likert scale, such as "How much

bodily pain have you had during the past 4 weeks?" (1 = none, 6 = very severe).

Vitality was measured by four questions rated on a 6-point Likert scale, such as,

"How much of the time during the past 4 weeks did you feel full of pep?" (1 = all

of the time, 6 = none of the time). Emotional limitations were assessed using

three yes/no questions, such as "During the past 4 weeks, have you had any of

the following problems with your work or other regular daily activities as a result

of any emotional problems (such as feeling depressed or anxious): Cut down the

amount of time you spent on work or other activities?" (yes/no). Social

functioning was measured by two questions rated on a 5-point Likert scale, such

as "During the past 4 weeks, to what extent has your physical health or emotional

problems interfered with your normal social activities with family, friends,

neighbors, or groups?" (1 = not at all, 5 = extremely). Last, mental health was

assessed with five questions rated on a 6-point Likert scale, such as "How much

of the time during the past 4 weeks have you been a nervous person?" (1 = all of









the time, 5 = none of the time). The responses within each scale were summed

for each individual (Ware, 1993). Internal consistency reliability was good for

each of the eight scales: general health (alpha = .83), physical limitations (alpha

= .81), physical functioning (alpha = .91), bodily pain (alpha = .79), vitality (alpha

= .83), emotional limitations (alpha = .85), social functioning (alpha = .74), and

mental health (alpha = .82).

To examine the spontaneously-generated possible selves, four categories

of selves were created: health, dependency, memory, and cognitive selves.

Health-related selves included direct statements such as "maintaining or losing

my health," "avoiding illness," and "remaining physically active." Health-related

selves also included more specific statements such as "becoming a cancer

patient," "having a stroke or heart attack," and "losing the ability to see."

Dependency-related selves included general statements like "maintaining my

independence" or "becoming dependent on another," but also included such

statements as becoming a "burden on my family" and "moving into a nursing

home." Memory-related selves included "becoming an Alzheimer's patient,"

"senile," or "demented," and also included positive items such as "improving my

memory ability" and remembering specific items, like vocabulary and dance

steps. Lastly, cognitive selves included statements such as "staying mentally

alert or active," or "losing my mind" and also included cognitive activities such as

"learn to use a computer" or "learn to play an instrument." Although many of

these cognitive selves refer to the mind and to memory, selves that fit this









category were counted separately to ensure that memory-related selves would

be defined as conservatively as possible.

The spontaneously-generated selves were scored into the four categories

by two independent raters. The overall agreement between the two raters was

93%. Table 3.2 shows the percentage of agreement, by category of self, for

hoped-for and feared selves.

Procedure

Interviews were conducted in groups of 3 to 20 individuals. Participants

were instructed to complete the questionnaire packet, in order, without looking

back at any of their previous answers. Experimenters were present at all times

to answer any questions.

Each packet began with the open-ended portion of the possible selves

questionnaire, followed by the closed portion of the possible selves

questionnaire. Participants then completed the remaining items concerning

memory beliefs. Last, participants completed a Participant Information form, in

order to provide us with basic demographic information and the SF-36 health

survey. Due to experimenter error, the majority of the older adults completed the

SF-36 one week after completing the questionnaire packet.






40


Table 3-1. Mean Education and Health Ratings


Standard Standard

Age group Mean education deviation Mean health deviation

Young 13.22 .765 15.37 2.85

Young-old 14.75 2.99 16.06 3.14

Old-old 13.08 2.11 15.11 3.58









Table 3-2. Percentage of Agreement Between Independent Raters


Type of selves Percentage agreement

Hoped-for

Health 85.9%

Dependency 94.5%

Memory 100%

Cognitive 88.3%

Feared

Health 89%

Dependency 93.7%

Memory 99.2%

Cognitive 93.8%














CHAPTER 4
STUDY 1 RESULTS

Analyses of variance were conducted to examine spontaneous memory-

related selves as a function of age group. In addition to the analyses of variance,

cross tabs were also conducted because the actual number of individuals

spontaneously reporting memory-related selves was low. These additional

analyses examined whether the number of individuals listing these selves

differed significantly by age group. Significant group differences were identified

in each case by using Tukey's post hoc tests at the .05 level.

Spontaneous Selves

An analysis of variance was conducted to examine whether the three age

groups (young, young-old, and old-old) differed by the total number of memory-

related selves each group spontaneously listed. Memory-related selves included

items such as: "improving or losing my memory," and "becoming an Alzheimer's

patient." In general, older adults spontaneously reported significantly more

memory-related selves than younger adults, F (2, 127) = 7.82, p < .000, eta2 =

.110, in the open-ended questionnaire, in support of Hypothesis 1. In fact,

younger adults did not spontaneously report any memory-related selves.

Contrary to the prediction in Hypothesis 2, no significant differences were found

in the number of memory-related selves reported by the old-old and young-old.

To further explore the spontaneous reports of the older adult groups, older

participants were divided into those who spontaneously mentioned a memory self









and those who did not. A cross-tabulation of the two older groups (young-old

and old-old) listing spontaneous memory-related selves was not significant,

Pearson X2 (N = 85, df = 1) = .17, p > .60, further supporting the notion that these

two older groups did not differ in their spontaneous report of memory selves.

See Table 4.1.

Further, a closer look at the types of memory-related selves (none, hoped-

for, or feared) reported by the young-old and the old-old groups revealed a

significant difference in the types of memory-related selves reported by each age

group, F (2, 129) = 7.468, p < .001, eta2 = .105, such that the old-old and young-

old groups both reported a greater number of feared memory-related selves than

hoped-for memory-related selves as predicted in Hypothesis 3. To further

examine the types of memory-related selves reported by the older adults,

participants were categorized into: those who primarily reported hoped-for

memory-related selves, those who primarily reported feared memory-related

selves, and those who did not report any memory-related selves. Because the

majority of participants' spontaneous memory selves represented fears, any

participant who mentioned a hoped-for memory self was categorized as hoped-

for. Those with feared selves and no hoped-for selves mentioned spontaneously

were classified as feared. A cross-tabulation of the two older adult groups

(young-old, old-old) by spontaneous self-categorization (none, hoped-for, feared)

was not significant: Pearson X2 (N = 85, df = 2) = -.04, p > .70. In the two older

groups, there were about the same number of individuals reporting hoped-for and

feared memory selves. See Table 4.2.









In addition to memory-related selves, participants also included items such

as remaining mentally alert, improving mental capacity and losing mental

faculties on the open-ended portion of the questionnaire. To be conservative

about the memory category, these items were not initially coded as memory-

related selves. These cognitive items were later combined with the memory-

related selves above to form a general cognitive category. An analysis of

variance was conducted examining the number of cognitive selves reported by

the three age groups (young, young-old, and old-old). The younger adults

reported fewer cognitive selves than both the young-old and the old-old, F (2,

129) = 9.02, p < .000, eta2 = .045. No significant differences were found between

the number of cognitive selves reported by the young-old and old-old.

Individuals were divided into two categories, those who spontaneously

mentioned a cognitive and/or memory-related self and those who did not. Again,

a cross-tabulation of the three age groups (young, young-old, old-old) showed

significant differences when including the spontaneously reported cognitive

selves with the memory-related selves: Pearson X2 (N=130, df = 2) = 16.025, p<

.000. The younger adults were less likely to report cognitive selves than the two

older adult groups. See Table 4.3.

Selves Endorsed on Closed-ended Questionnaire

A mixed analysis of variance was conducted to examine the total number

and type of memory-related selves (hoped-for or feared) endorsed on the closed-

ended questionnaire, by age group (young, young-old, and old-old). Age group

was the between-subjects factor and type of memory self was the within-subjects

factor. Overall, no significant age differences were revealed in the number of









selves endorsed on the closed-ended questionnaire by the three age groups, F

(1, 127) = 1.58, p > .20. However, all three groups endorsed more feared than

hoped-for memory-related selves on the closed-ended questionnaire, F (1, 127)

= 82.68, p < .000, eta2= .394. See Table 4.4.

An additional mixed analysis of variance was conducted to examine the

total number and type of memory-related selves (hoped-for or feared) endorsed

on the closed-ended questionnaire by age group (young-old and old-old) and by

type of memory self spontaneously reported (none, hoped-for, or feared) on the

open-ended portion of the questionnaire. The analysis confirmed a main effect

for closed selves, F_(1, 79) = 30.19, p < .000, eta2 = .277. Again, the young-old

and old-old endorsed more feared selves than hoped-for selves on the closed

portion of the questionnaire. No other significant differences were found.

Next both closed and open memory-related selves were examined together.

Individuals were categorized into two groups: those with predominately hoped-for

selves and those with predominately feared selves. This categorization utilized

both the open and closed item responses. This involved a two-step process.

Most individuals on the open and closed portions of the possible selves

questionnaires tended to mention feared memory-related selves. Therefore,

anyone reporting a spontaneous feared memory-related self, but no spontaneous

hoped-for self, was categorized as "feared." Next, individuals were categorized

as "hoped-for" if they reported a spontaneous hoped-for memory-related self and

endorsed an equal or greater number of hopes than fears about memory on the

closed portion of the questionnaire. A cross-tabulation of the three age groups









(young, young-old, old-old) by self-categorization (hoped-for or feared) did not

reveal any significant differences: Pearson's X2 (N = 130, df = 1) = .13, p = .937

(See Table 4.8).

The memory beliefs measures were highly related to each other, both

theoretically and empirically as shown by the age-partialled correlations in Table

4.5. Previous research has shown that older and younger adults answer memory

beliefs questions differently. Therefore, age was partialled out to ensure that the

correlations reflected the relationships between the variables of interest

independent of age. A multivariate analysis of variance was conducted to

compare the three age groups (young, young-old, old-old) across the following

dependent variables: achievement, locus, anxiety, memory value, MCI, and

memory self-evaluation. Age differences were significant, F (12, 230) = 4.77, P <

.000, eta2 = .199, and this significant result was further explored in univariate

tests

Significant age differences were found for memory value, F (2, 119)=

24.59, p < .000, eta2 = .292, due to the fact that older adults in both the old-old

and young-old groups placed a greater value on memory than younger adults.

The young-old and old-old did not significantly differ from one another, as shown

in Table 8. Significant age differences were also found for the MCI scale which

measured fear of Alzheimer's disease, F (2, 121) = 5.99, p < .003, eta2 = .091,

with older adults in both groups reporting greater fear of Alzheimer's disease

than younger adults. The old-old also reported greater levels of anxiety in

response to memory-related tasks than both the young-old and young adults, F









(2, 121) = 4.731, p < .01, eta2 = .074. Next, significant age differences were

found for achievement, F (2, 121) = 3.57, p = .03, eta2 = .057. For achievement,

the old-old placed significantly greater importance on performing well on memory

tasks than did younger adults. The young-old did not significantly differ from the

old-old or the younger adults. No significant age differences were found for

Locus, F (2, 121) = 1.56, p > .20, suggesting that participants in each age group

believed they had similar amounts of control over their memory ability. Further,

no significant age differences were found on the Memory Self-Evaluation scale, F

(2, 121) < 1, p > .50. Thus young adults, the young-old and the old-old all

reported similar levels of ability and satisfaction regarding their recent memory

performance. See Table 4.6.

An additional multivariate analysis of variance examined the memory

beliefs measures as a function of age and the type of memory-related selves

primarily reported in the open-ended questionnaire, using the categorization

described previously (none, hoped-for, feared), and including only the young-old

and old-old (because the younger adults reported no memory selves on the

open-ended questionnaire). The dependent variables were achievement, locus,

anxiety, memory value, MCI, and memory self-evaluation. No significant age

differences in memory beliefs were found between the young-old and old-old

groups, F (6, 66) = .77, p > .50. Type of memory self spontaneously reported

was close to significance, F (12, 134) = 1.73, p > .06, so univariate analyses

were conducted. Significant differences were found for anxiety, F (2, 76) = 5.54,

p = .006, eta2 = .135, due to the fact that participants who spontaneously listed









hoped-for memory-related selves also reported greater levels of anxiety when

faced with memory-related activities than participants who did not list any

memory-related selves. Individuals reporting feared memory selves did not

significantly differ from those with hoped-for selves or with no memory selves at

all. A significant difference was also found for the MCI scale, F (2, 76) = 4.92, p

= .01, eta2 = .122. Participants spontaneously reporting hoped-for memory-

related selves also reported a greater fear of Alzheimer's disease than

participants who reported no memory selves. Again, individuals reporting feared

memory selves did not differ from those with hoped-for selves or with no memory

selves. Last, a significant difference was observed for the memory self-

evaluation scale, F (2, 76) = 5.83, p = .003, eta2 = .141. Participants reporting

hoped-for memory-related selves evaluated their recent memory performance

less favorably than participants with feared memory-related selves and those

reporting none at all. All other memory beliefs measures showed no variation as

a function of type of spontaneous memory self (See Table 4.7).

Additional Analyses

Several additional analyses were conducted to compare the current results

with prior research on possible selves. Although this study focused on memory-

related selves, participants were asked to generate as many future selves as

they could think of, both hoped-for and feared. We compared the mean number

of hoped-for and feared selves spontaneously generated by age group. Overall,

the old-old spontaneously reported significantly fewer hoped-for selves than the

younger adults, F (2, 129) = 6.83, p < .002, eta2 = .097. The young-old did not

significantly differ from either the old-old or the younger adults in the number of









hoped-for selves they generated. The old-old also spontaneously reported

significantly fewer feared selves than both the young-old and younger adults, F

(2, 129) = 10.15, p < .000, eta2 = .14. See Table 4.9.

Participants were also asked to identify their most important hoped-for self

from the list of possible selves they generated. Next they were asked to indicate

how capable they felt of accomplishing this most important self, and how likely it

is that this self would come true. We have compared these mean capability and

likelihood ratings for the most important hoped-for self by age group. The old-old

and young-old rated themselves as significantly less capable of accomplishing

their most important hoped-for self than the younger adults, F (2, 125) = 11.85, p

< .000, eta2 = .16. Further, the old-old and young-old rated their most important

hoped-for self as significantly less likely to come true than the younger adults, F

(2, 125) = 15.84, p < .000, eta2 = .21. See Table 4.10.

Participants were also asked to identify their most dreaded feared self from

the list they generated. Again, they were asked to indicate how capable they felt

of preventing this self and how likely it is that this self would come true. We have

compared these mean capability and likelihood ratings for the most dreaded

feared self by age group. The old-old and young-old rated themselves as

significantly less capable of preventing their most dreaded feared self than the

younger adults, F (2, 123) = 8.00, p < .001, eta2 = .12. The old-old and young-

old also reported that their most dreaded feared self was more likely to come true

than did the younger adults, F (2, 123) = 5.89, p < .004, eta2 = .09. See Table

4.11.









Last, participants were asked to list the number of goal-oriented activities

recently undertaken to attain their most important hoped-for self and to avoid

their most dreaded feared self. The old-old reported significantly fewer goal-

oriented activities recently undertaken to attain their most important hoped-for

self than both the young-old and younger adults, F (2, 129) = 9.38, P < .000, eta2

= .13. No significant differences were found between the three age groups for

the number of goal-oriented activities undertaken to avoid their most dreaded

feared self, F (2, 129) = 2.96, p > .05 (Table 4.12).

These results are generally consistent with previous research on possible

selves. In general, older adults report fewer possible selves than younger adults

(Cross & Markus, 1991). Further, previous research has also shown that older

adults feel less capable of attaining or avoiding their most important or most

dreaded selves than younger adults (Cross & Markus, 1991). The current results

are not consistent with previous research showing that older adults report

undertaking a greater number of goal-oriented activities to attain or avoid their

most important or most dreaded selves than younger adults (Cross & Markus,

1991).












































'0 10
















0-r
0 CO C






0 CY) C(
















o o -
-- (D r-














-0
0
0 ) -0

0 0 -
>- >- 0



















































LO (Y-





10i Cb











CN
bi



















~-0






0

S0
0 -
>- 0















































UY) CN

CN l- ICO
























a: U

co 0'




0

S 0
0 0 -
>- >- 0













































Y) 'IT 0













CO D N















CY)O
rl- CY- (D






















-0s
0


o o
>- >- 0)


































CY 10
G)N-


1O 0
Cr- -
C r-1II
I-- -


0
(0o-
^.

CY 0.


0 ^
0 (










O)
0

-_


CN
Nt- ^-





LO I"




0


CYC










0


CD
0
CO~ 0










0
N-

LO
I II
CY.










(0w-





-LU
.D









-J


N- CN






0




O 0
7r--











S]- O

0'-









0") 1O

* N-11-


0

NO-
II


0
O -O
r I '-
h 1^ 1














O C
c














0 O0
Cu





















e >
-C




















cp c





0
a,
E
U,























_i
-c

S>
al)


U) -C






















c


T- (D 1O
Cem










0 C CO)

CO CO O
















CY)O 0
co CT) o











CO OD -D
aCO O) C-












c c ,



C C0 0
-5
0 -
o 0
>->O 0


4 I
0
a.







c



<
CD


0






Qo >
0 a,









c










.
_0
C -c


, >


0
E
I "


r- CN

- cN\


0
-5


0 0
>- >- 0


r- (D 0
CN 0 r-
cd cq








0 r1 -













CN t- CN
1n CY) co






N CO) CO
'T COY) '
c\i c cd










CO OT 0
-COc






































IO O 0 I -
0 v- I T- T- 0
0 0 0 0 0 0





V r- CT) Y- O (D
co 0o c co rY
d d d d d d





CO 0 CO CO )
CY CO CO (D C) 0
d d T- d d d


C

E


c o 0 >
< < -1 2 > LU


-D

-o
0


76
-6
0
6)



0-
CD
a
0










































0- 0 0 v- 0 0-






(O (D 0 I- O 10
0 1O 0 (O 0
0 d d d d 0





' CO C N (C C)









N N N N N N










E


S 0
c < o i > >
< < -J 2 > Lu


-7

0
4-
c7

0



O






E
-0




4-

0

o
I
a






H-
L,






59


Table 4-8. Type of Memory Selves Reported By Age Group


Age Group Feared selves Hoped-for selves



Young 35.7% 32.6%

Young-old 27.4% 28.3%

Old-old 36.9% 39.1%









Table 4-9. Mean Number of Selves by Age Group


Mean Standard

Age group selves deviation

Hoped-for

Young 6.13 2.75

Young-old 5.47 2.77

Old-old 4.29 1.85

Feared

Young 4.29 1.66

Young-old 3.92 2.10

Old-old 2.76 1.44






61



Table 4-10. Mean Capability and Likelihood Ratings: Feared


Age group

Young

Young-old

Old-old


Mean

capability

5.73

4.54

4.61


Standard

deviation

1.19

1.80

1.63


Mean

likelihood

2.93

4.09

3.91


Standard

deviation

1.51

1.56

1.87






62



Table 4-11. Mean Capability and Likelihood Ratings: Hoped-for


Age group

Young

Young-old

Old-old


Mean

capability

6.22

5.60

5.04


Standard

deviation

.82

.98

1.50


Mean

likelihood

6.31

5.00

4.93


Standard

deviation

.76

1.35

1.61









Table 4-12. Mean Number of Goal Activities


Mean Standard

Age group activities deviation


Hoped-for

Young 2.09 1.38

Young-old 1.06 .90

Old-old 2.03 1.56

Feared

Young 1.98 .21

Young-old 2.31 .24

Old-old 1.55 .21














CHAPTER 5
STUDY 2 INTRODUCTION

The current study also sought to examine whether younger adults could be

primed to spontaneously list memory-related possible selves. In Study 1,

younger adults did not spontaneously report any memory-related possible selves,

either hoped-for or feared. These initial results suggested that memory, and

memory loss specifically, was not salient to younger adults. This study examined

whether or not memory could be brought into focus for younger adults, thus

prompting them to spontaneously report memory-related selves. Prior research

suggests that specific testing conditions can serve as primes to activate belief

schemas during memory testing. Levy (1996) examined the effects of age-

related stereotypes in a series of two studies. In the first study a priming task

exposed each participant to either positive or negative words, related to either

wisdom or memory decline, after a series of memory tests. The words flashed

quickly on a computer screen, so the participants were not aware of having seen

them. This task was designed to activate participants' internal stereotypes about

aging and memory. After the priming task, a third of the participants immediately

completed a second series of memory tests and questionnaires. Levy (1996)

found that the priming activity was successful and influenced memory

performance, attitudes toward aging, and beliefs about memory in older adults.

The older adults primed in the negative condition performed more poorly and

reported more negative attitudes about aging and memory. In contrast, the older









adults primed in the wisdom condition, performed better and reported more

positive attitudes toward aging and memory. The results indicated that priming

was effective in accessing the different beliefs older adults hold toward memory

and aging.

Rahhal et al. (2001) also manipulated beliefs in relation to memory testing.

They examined the effects of instructional manipulations (memory-emphasis and

memory-neutral instructions). In the memory-emphasis condition, participants

were told that the activity was a memory test, and the need to remember

information was stressed. The memory-neutral condition focused on the

participant's ability to learn new facts and did not mention memory or

remembering. In each condition, participants were presented with 60 trivia

questions labeled with no answer or labeled as true or false. During the recall

phase, 12 new items were mixed in with the original 60 questions. Participants

identified each statement as new or old and as true, false, or blank. For both

studies, the authors found that when instructions emphasized memory and

remembering facts, older adults performed worse than younger adults on the

recall task. However, when the instructions did not emphasize memory, and

instead focused on learning new information, no significant age differences were

obtained in recall. These studies suggest that the type of instruction presented to

older adults can influence their performance by reinforcing age-related beliefs

and stereotypes about memory and memory performance (Rahhal et al., 2001).

The previous studies suggest that memory testing under particular

instructional conditions can prime beliefs and stereotypes in older adults. There









is also considerable literature showing that beliefs and feelings can be altered in

memory studies with younger adults as well. Bower and colleagues (e.g., Bower,

1994) have repeatedly shown that mood can be manipulated in college students,

with a subsequent effect of memory. Further, such manipulations affect the

types of past memories recalled, judgments of personal competence, and

judgments about the likelihood of future events (Bower, 1994). Taken together,

the studies above suggest that memory testing under certain circumstances can

affect thoughts about the self in relation to memory. That is the effect we expect

here. The current study planned to use memory tasks and questionnaires in a

similar fashion as the studies described above. Exposing younger adults to a

challenging memory task should access their memory-related beliefs, making it

more likely that younger adults might spontaneously mention a memory-related

possible self.

In the current study, priming was accomplished in two ways. First,

participants in the primed condition completed a challenging memory activity.

Second, they completed several memory beliefs questionnaires. Both of those

activities preceded their completion of the open-ended and closed possible

selves questionnaires. By having the primed group complete the memory activity

and memory questionnaires before the possible selves questionnaires, memory

should be more salient to these individuals. Consequently, these younger adults

should be more likely to generate memory-related possible selves on the open-

ended questionnaire than the younger adults in the unprimed condition.














CHAPTER 6
STUDY 2 METHODS

Participants

Of 97 younger adults recruited from the University of Florida's research

participant pool, 52 were primed and 45 were unprimed. Age, health and

education were examined as a function of priming condition. No significant

differences in age, F (1, 96) = 2.019, p > .10; health, F (1, 94) < 1, p > .25; or

education, F (1, 94) < 1, D > .50 were found between primed and unprimed

young adults. See Table 6.1.

Procedure

Young adults in the primed condition received the same questionnaire

packet used in Study 1, but in an altered order. First each participant studied a

15-item shopping list created by West, Welch, & Thorn (2001). Participants were

given 1 minute to study the list and a maximum of 4 minutes to recall the list.

Next, primed participants completed the memory belief measures, followed by

the open-ended and closed possible selves questionnaires, respectively. Lastly,

primed participants completed the participant information form and SF-36 health

survey.








68




















0C
0
m
N 1


-rs
m O
2 CN C

Sr-




< C
C) 0
O C
o0 Co Co

E

a l

00 N- CO

-0 -

E






0 C)
i_
L0
ac 0 -
0 0
10 co* N:
0- U E





0 E a
E








0- E D0
CO E
U- 0 D














CHAPTER 7
STUDY 2 RESULTS

The younger adults in the unprimed condition did not spontaneously report

any memory-related possible selves as expected. Further, the majority of primed

younger adults did not report memory-related selves (only two younger adults in

the primed condition listed a memory-related self), therefore comparisons

between the number of spontaneous memory-related selves generated by the

primed and unprimed groups were not conducted.

An analysis of variance was conducted to examine the total number of

memory-related selves (spontaneously generated plus those endorsed on the

closed-ended questionnaire) by priming condition (primed and unprimed). No

significant difference was observed in the number of memory-related selves

endorsed by primed and unprimed younger adults, F (1, 96) < 1, p > .50, (primed,

M = 11.52, SD= 2.07; unprimed, M = 11.76, SD= 2.24).














CHAPTER 8
DISCUSSION

Memory-related possible selves are important for two main reasons. First,

these selves identify what a person hopes to become or fears becoming. Cross

and Markus (1994) stated that possible selves can help describe the direction of

development an individual is anticipating. Thus the presence or absence of

memory selves will provide a more complete picture of how individuals view their

own cognitive development toward the latter half of the lifespan. Next, the

presence or absence of these selves may ultimately affect behavior. Hooker has

shown that the presence of health-related possible selves are related to health

behaviors such as seeking preventative medical treatment and exercising.

Specifically, individuals with a health-related self were more likely to report

participating in health behaviors than were individuals that placed a high value on

health alone (Hooker & Kaus, 1992). From this perspective, possible selves act

as goals, and may help motivate individuals into action. These future selves

provide a measure to which current selves can be compared and evaluated

(Cross & Markus, 1991). When the discrepancy between the current self and a

future hoped-for self becomes too great (or the distance between the current self

and a future feared self becomes too close), something must be done.

From a developmental perspective, Cross and Markus (1991) have

suggested that possible selves may actually become more motivating with age.

Older adults experience less conflict among their competing future selves.









Further, their future selves are more closely tied to their current selves, and are

therefore more specific, well defined, and personally meaningful. As such, the

steps required to avoid or achieve a certain self may be more clearly defined and

motivating, as one gets older.

This study was the first to look specifically at memory-related possible

selves. The results confirmed that the types of possible selves spontaneously

generated, that is, those selves that are most important or central to an

individual's self-concept, differ with age. Younger adults did not spontaneously

report any memory-related possible selves. Even after cognitive and memory-

related selves were combined, younger adults spontaneously reported fewer

cognitive selves than older adults. From this we can conclude that younger

adults are not concerned about their memory, nor are they particularly focused

on their future cognitive abilities. The future hopes and fears of younger adults

are focused elsewhere.

In contrast, older adults did report memory-related selves. Further, the

memory selves reported by older adults were primarily feared selves, like

dementia. Both older and younger adults expect memory decline to occur with

age, and the greatest decline is expected to occur in the later years of life

(Hertzog et al., 1999; Lachman et al., 1995; Lineweaver & Hertzog, 1998; Ryan,

1992). Older adults also face social expectations of memory decline. Episodes

of forgetting among the elderly are rated more seriously than when the same

episodes occur for a younger person (Erber, 1989). Further, when forgetting

occurs among older adults, it is more likely to be viewed as indicative of a need









for medical attention or cognitive evaluation (Erber, 1989). These social

expectations and stereotypes contribute to the views older adults have about

themselves and their memory ability. On a personal level, older adults also

report lower levels of control over improving or maintaining their memory abilities

(Hultsch et al., 1987; Lachman et al., 1995) and feel less capable of completing

memory tasks (Berry et al., 1989; Hultsch et al., 1987). With personal and social

expectations of memory loss looming on the horizon, it is not surprising to find

that older adults have incorporated these beliefs and expectations into their self-

concepts.

Cavanaugh et al. (1998) have hypothesized that memory is schematic for

older adults. Individuals are said to be schematic for a particular domain or trait

when they consider it both highly important to and highly descriptive of

themselves (Cavanaugh et al., 1998; Cross & Markus, 1994). Thus, for older

adults, memory and memory loss have more personal importance (Cavanaugh et

al., 1998). The current results support this idea. If an individual spontaneously

generates a possible self in a specific domain, like health or memory, that

domain is of central importance to their self-concept. For these individuals,

memory or health are important in general but also define who they are in some

way. The older adult data in this research may reflect this kind of schematicity.

When memory and cognitive selves were combined, 63.5% of older adults

spontaneously listed at least one memory or cognitive-related self. This

suggests the prominence of cognition as a key element in the self-concept of the

older group.









As part of the self-concept, these memory-related selves may play an

important role, guiding behavior in several ways. These selves may create shifts

in information processing (Kato & Markus, 1993), allowing information related to

a particular self to be processed more quickly and efficiently (Cross & Markus,

1994; Ruvolo & Markus, 1992). For instance, an episode of forgetting will be

differentially attended to and may take on a different meaning for an individual

with a memory-related possible self. Next, the existence of a memory self within

the self-concept can also result in changes in levels of motivation and self-

efficacy (Cameron, 1999; Cross & Markus, 1991; Hooker, 1992). For example,

Hooker has shown that individuals with health-related selves were more likely to

report engaging in health-related behaviors and felt more capable of attaining

health-related goals than individuals who reported valuing health, but listed no

health-related selves (Hooker & Kaus, 1992). Last, these selves may affect the

level of effort and persistence an individual is willing to put forth on activities

related to a specific self (Ruvolo & Markus, 1992). For instance, individuals with

a memory-related self may be more likely to seek out challenging cognitive

activities, further once engaged in a task, they may be more likely to stick with it.

Interestingly, the three age groups did not significantly differ in the number

of memory-related selves endorsed on the closed portion of the questionnaire.

Thus, when presented with positive and negative future memory outcomes, both

groups identified and endorsed similar items. However, all three groups

endorsed a greater number of feared items than hoped-for items on the closed

portion of the questionnaire. When asked about future selves they would like to









avoid, both groups were likely to circle memory-related items like "Alzheimer's

patient" or "senile." However, when asked about future selves they would like to

attain, they were less likely to circle the positive memory-related items such as

"mentally alert all my life" or "sound mind." There may be several reasons for

these differences. First, negative memory outcomes may be more salient to both

groups. Previous research has shown that younger and older adults expect

memory decline to occur as they age (Hertzog et al., 1999; Lachman et al., 1995;

Lineweaver & Hertzog, 1998; Ryan, 1992). The current results appear to confirm

a shared stereotype among older and younger adults regarding aging and

memory. Second, we know what it means to have a poor memory, to become

senile or demented. We have terms to describe and diagnose these problems in

a very concrete way. It may be that these terms are simply easier to identify than

the more general positive memory-related items.

The three age groups differed across several of the memory beliefs

measures, verifying results from earlier research examining age differences in

memory beliefs. In general, the results indicate that the old-old may be more

concerned about their memory. The old-old reported a greater fear of

Alzheimer's disease than younger adults, however the young-old did not

significantly differ from the old-old or the younger adults. Further, the old-old

reported greater levels of anxiety in response to memory tasks than both the

young-old and younger adults. Consequently the old-old may fear that

participating in a memory activity might confirm expected memory loss. Again,

these results are consistent with previous studies showing that older adults report









greater levels of anxiety and a greater fear of Alzheimer's disease than younger

adults (Lachman et al., 1995) and suggest that memory and memory loss are

more salient to older adults than to younger adults.

Along these lines, the old-old and the young-old both valued memory more

than the younger adults. Thus both the old-old and the young-old placed more

importance on performing well on memory tasks than did younger adults. For

older adults, performing well on a memory task might act as evidence that their

memory is intact. It may also be that the older adults place a greater value on

performing well in general, regardless of the task, because of heightened

concerns about perceived age-related losses. The younger adults, in contrast,

may accept their cognitive abilities as strong and stable. For them, memory

performance may not be as important because it has less diagnostic value

(Hess, Auman, Colcombe, & Rahhal, 2003).

Contrary to what may have been expected, individuals who reported only

feared memory-related selves did not appear to differ from those with only a

hoped-for memory self or those with no memory selves at all, on the memory

self-evaluation scale. These individuals listed a memory-related fear, but still

reported feeling confident about their recent memory ability. Further, individuals

with a feared memory self or no memory self at all were less anxious and

reported less fear of Alzheimer's disease than individuals with a hoped-for

memory self. Cross and Markus (1991) have discussed this phenomena in terms

of defensive elaboration. Individuals spontaneously listing a memory-related fear

may be experiencing declines in memory ability or increased episodes of









forgetting, resulting in the presence of a memory fear. However, these

individuals may be able to feel content with their current memory ability by

comparing current ability to a future that appears far worse (Cross & Markus,

1991). As a result they are less anxious and have less immediate fear because

their current memory abilities appear intact when compared to the severe

impairment expected in the future. This view has not been extensively examined

in the literature, and may be a useful concept for future investigation.

Last, in Study 2, priming did not elicit any memory-related selves from

younger adults. We expected that younger adults in the primed condition would

be more likely than unprimed younger adults to generate memory-related

possible selves. The memory task included in the priming activities may not

have been sufficiently challenging to activate memory-related possible selves in

younger adults. The primed younger adults correctly recalled approximately 11

out of 15 shopping list items (M = 11.17, SD = 1.72). A more difficult memory

task might be more successful in eliciting these types of selves from younger

adults in the future. On the other hand, it may be the case that memory is so low

in salience for younger adults that they will not spontaneously generate memory-

related selves, even after a taxing memory activity. A primed older adult group

would be interesting to examine to determine whether more memory selves

would be generated when older adults are confronted with a challenging memory

task. Stereotype threat research suggests that older adults, for whom memory is

self-defining, may react differently when faced with a memory task (Hess et al.,

2003). Specifically, their performance may suffer as they anxiously strive to









debunk the negative stereotypes surrounding aging and memory (Hess et al.,

2003). More importantly, after experiencing anxiety over a challenging memory

task, thoughts and stereotypes about the self, aging, and memory become more

accessible. If this is the case it should be apparent in the possible selves

generated by these older adults after priming.

We must be cautious when interpreting these results due to the fact that

only 28 percent of the older participants spontaneously reported memory-related

hopes and fears. In particular, we must point out that the group of individuals

who reported a hoped-for memory self was small. Further, half of these

individuals endorsed both a hoped-for and a feared memory self. Although

memory may be particularly salient for these individuals with both a memory-

related hope and fear, further investigation is warranted before any conclusion

can be made.

The cross-sectional nature of the current study also limits the conclusions

we can currently draw. From the current data, we cannot conclude that memory-

related selves increase with age, however the current results warrant further

exploration. A longitudinal study of memory-related possible selves will be a

necessary and exciting endeavor, and will help us explore when memory selves

begin to appear, and if the nature of these selves as hoped-for or feared change

over time. Along the same lines, the current study did not recruit a middle-aged

sample. It will be important to examine this age group in the future, to

understand when memory-related possible selves begin to appear in the selves

spontaneously generated by older adults. Greater cultural diversity would also









extend this research, and help determine whether memory selves are equally

apparent across different racial and cultural backgrounds.

A baseline test of memory ability was not included in the current study.

Although the memory evaluation scale showed that younger and older adults in

the current sample reported similar levels of performance and satisfaction

regarding their recent memory performance, a direct measure of actual ability

may be useful. A direct measure of memory performance would allow us to

examine how individuals with hoped-for or feared memory-related selves actually

performed in comparison to individuals not reporting a memory-related self. It

might be the case, for instance, that feared selves are more strongly related to

actual performance than to other beliefs about memory.

Last, group testing may have limited the amount of data collected on the

open-ended portion of the possible selves questionnaire. Most of the previous

research on possible selves has been conducted using individual interview

sessions. This was done because the questionnaire could potentially be

confusing to older adults (Hooker, 1992; Hooker & Kaus, 1992; Hooker, 1999).

However, our results are consistent with those obtained from individual

interviews, suggesting that this methodological change was not problematic. On

average, 86% of the older participants in these prior studies reported health-

related selves, which included both physical and cognitive-related selves such as

having a heart attack or becoming an Alzheimer's patient. Consistent with these

previous results, in the current sample, 83.5% of our participants reported a

health-related self when health and memory-related selves were combined.









Although it is true that we may have elicited richer descriptions of possible selves

in a one-on-one setting, participants in the current study reported a diverse and

rich array of future selves in a group setting. The groups were small and an

interviewer was always present to answer any questions.

The relationship between memory-related possible selves and subsequent

behavior is not clear. Stereotypes about aging and memory create a culture in

which memory decline is an expected part of growing old. In such an

environment, instances of forgetting may become more salient to older adults,

acting as proof of inevitable decline. As a consequence, older adults may feel

less capable of and less control over maintaining or improving their memory

abilities. These expectations and experiences may be of even greater

importance for older adults who place a high value on memory ability (Hess et

al., 2003). Individuals who place a high value on memory ability may be more

likely to have a memory-related possible self. Further, an individual who has a

memory self as a central feature of their self-concept may be more likely to

actively pursue or avoid this self than someone for whom memory is less

schematic. When stereotypes, decreased feelings of efficacy and control, and

actual age-related declines in memory performance are combined with the

knowledge that the causes of Alzheimer's disease are largely unknown,

attempting to take the necessary actions to avoid a feared memory-related

possible self may be seen as futile for aschematic individuals.

This has been the first study to examine memory-related possible selves.

Although memory beliefs have been examined extensively, the relationship






80


between these beliefs and an individual's self-concept, especially in terms of

possible selves, is not well understood. Older adults are more likely to

incorporate a memory-related self into their self-concept. Further these memory-

related selves are more often feared selves, not hoped-for. The ultimate goal of

future investigations is to determine how these selves affect older adults, by

exploring how the presence of these memory-related hopes and fears motivate

or debilitate behavior.














APPENDIX A
CLOSED-ENDED POSSIBLE SELVES MEASURES

Circle all the items that describe you now:


Happy
Senile
Confident
Depressed
Intelligent
Good-looking
Sound mind
Blind
Rich
Not able to fit in
Successful
Important
Alzheimer's patient
Competent
In good shape
Have lots of friends
Owner of a business
Paralyzed
Make own decisions
Manipulate people
Trusted
Unimportant
Remember what I need to


Offensive
Welfare recipient
Sexy
Brain dead
Spouse or child abuser
Athletic
Wrinkled
Active social life
Being health conscious
Remember every detail
Good friend
Not in control
Likely to die young
Admired
Mentally alert all my life
Unwanted
A good parent
Alone
Interesting
Adjusted
Underachiever
Incurable memory loss
Optimistic









Circle all the items that describe you now:


Alcohol dependent
Able to fix things
Secure
Reasonably alert
Able to influence people
Unable to remember
Knowledgeable about art
Failing memory
Being appreciated
Loved
Unpopular
Able to cook well
Respected
A cancer victim


Long-lived
Nonaggressive
A failure
Have a heart attack
Not mentally impaired
Win high honors
Artistic
Stupid
Attractive
Keen memory
Independent
Have a nervous breakdown
Fuzzy memory









Circle all the items that described you in the past:


Happy
Senile
Confident
Depressed
Intelligent
Good-looking
Sound mind
Blind
Rich
Not able to fit in
Successful
Important
Alzheimer's patient
Competent
In good shape
Have lots of friends
Owner of a business
Paralyzed
Make own decisions
Manipulate people
Trusted
Unimportant
Remember what I need to


Offensive
Welfare recipient
Sexy
Brain dead
Spouse or child abuser
Athletic
Wrinkled
Active social life
Being health conscious
Remember every detail
Good friend
Not in control
Likely to die young
Admired
Mentally alert all my life
Unwanted
A good parent
Alone
Interesting
Adjusted
Underachiever
Incurable memory loss
Optimistic









Circle all the items that described you in the past:


Alcohol dependent
Able to fix things
Secure
Reasonably alert
Able to influence people
Unable to remember
Knowledgeable about art
Failing memory
Being appreciated
Loved
Unpopular
Able to cook well
Respected
A cancer victim


Long-lived
Nonaggressive
A failure
Have a heart attack
Not mentally impaired
Win high honors
Artistic
Stupid
Attractive
Keen memory
Independent
Have a nervous breakdown
Fuzzy memory











Circle all the items that describe a future self that you would like to be:


Happy
Senile
Confident
Depressed
Intelligent
Good-looking
Sound mind
Blind
Rich
Not able to fit in
Successful
Important
Alzheimer's patient
Competent
In good shape
Have lots of friends
Owner of a business
Paralyzed
Make own decisions
Manipulate people
Trusted
Unimportant
Remember what I need to


Offensive
Welfare recipient
Sexy
Brain dead
Spouse or child abuser
Athletic
Wrinkled
Active social life
Being health conscious
Remember every detail
Good friend
Not in control
Likely to die young
Admired
Mentally alert all my life
Unwanted
A good parent
Alone
Interesting
Adjusted
Underachiever
Incurable memory loss
Optimistic









Circle all the items that describe a future self that you would like to be:


Alcohol dependent
Able to fix things
Secure
Reasonably alert
Able to influence people
Unable to remember
Knowledgeable about art
Failing memory
Being appreciated
Loved
Unpopular
Able to cook well
Respected
A cancer victim


Long-lived
Nonaggressive
A failure
Have a heart attack
Not mentally impaired
Win high honors
Artistic
Stupid
Attractive
Keen memory
Independent
Have a nervous breakdown
Fuzzy memory











Circle all the items that describe a future self that you would NOT like to be:


Happy
Senile
Confident
Depressed
Intelligent
Good-looking
Sound mind
Blind
Rich
Not able to fit in
Successful
Important
Alzheimer's patient
Competent
In good shape
Have lots of friends
Owner of a business
Paralyzed
Make own decisions
Manipulate people
Trusted
Unimportant
Remember what I need to


Offensive
Welfare recipient
Sexy
Brain dead
Spouse or child abuser
Athletic
Wrinkled
Active social life
Being health conscious
Remember every detail
Good friend
Not in control
Likely to die young
Admired
Mentally alert all my life
Unwanted
A good parent
Alone
Interesting
Adjusted
Underachiever
Incurable memory loss
Optimistic









Circle all the items that describe a future self that you would NOT like to be:


Alcohol dependent
Able to fix things
Secure
Reasonably alert
Able to influence people
Unable to remember
Knowledgeable about art
Failing memory
Being appreciated
Loved
Unpopular
Able to cook well
Respected
A cancer victim


Long-lived
Nonaggressive
A failure
Have a heart attack
Not mentally impaired
Win high honors
Artistic
Stupid
Attractive
Keen memory
Independent
Have a nervous breakdown
Fuzzy memory














APPENDIX B
MEMORY VALUE QUESTIONNAIRE

MEMORY VALUE
On this page, there are some questions asking for your opinions. To answer
each question, you should circle the number that best indicates your opinion.
Please read each question carefully before you decide how to answer. There are
no right or wrong answers on these questions.

Please circle the number that best represents your opinion.

1. If you don't have your memory you don't have anything.
1 2 3 4 5 6 7
Strongly Strongly
disagree agree



2. There are many things I care about more than my memory.
1 2 3 4 5 6 7
Strongly Strongly
disagree agree



3. A good memory is of only minor importance in a happy life.
1 2 3 4 5 6 7
Strongly Strongly
disagree agree



4. There is nothing more important than a good memory.
1 2 3 4 5 6 7
Strongly Strongly
disagree agree















LIST OF REFERENCES


Berry, J.M. & West, R.L. (1993). Cognitive self-efficacy in relation to personal
mastery and goal setting across the life span. International Journal of
Behavioral Development, 16, 351-379.

Berry J.M., West, R.L. and Dennehey, D.M. (1989). Reliability and validity of the
memory self-efficacy questionnaire. Developmental Psychology, 25, 701-
713.

Bower, G. (1994). Temporary emotional states act like multiple personalities. In
R. M. Klein & B. K. Doane (Eds.), Psychological concepts and dissociative
disorders (pp. 207-234). Hillsdale, NJ: Lawrence Erlbaum Associates.

Cameron, J.E. (1999). Social identity and the pursuit of possible selves:
Implications for the psychological well-being of university students. Group
Dynamics: Theory, Research, and Practice, 3, 179-189.

Cavanaugh, J. C., Feldman, J. M., and Hertzog, C. (1998). Memory beliefs as
social cognition: A reconceptualization of what memory questionnaires
assess. Review of General Psychology, 2, 48-65.

Cross, S., & Markus, H. (1991). Possible selves across the life span. Human
Development, 34, 230-255.

Cross, S.E. & Markus, H.R. (1994). Self-schemas, possible selves, and
competent performance. Journal of Educational Psychology, 86, 423-438.

Dixon, R. A. & Hultsch, D. F. (1999). Intelligence and cognitive potential in late
life. In J. C. Cavanaugh & S. K. Whitbourne (Eds.), Gerontology: An
interdisciplinary perspective (pp. 213-237). London: Oxford University
Press.

Dixon, R. A., Hultsch, D. F., & Hertzog, C. (1988). The Metamemory in Adulthood
(MIA) Questionnaire. Psychopharmacology Bulletin, 24, 671-688.

Erber, J. T. (1989). Young and older adults' appraisal of memory failures in
young and older adult target persons. Journals of Gerontology, 44, P170-
P175.









Frazier, L.D., Hooker, K., Johnson, P.M., & Kaus, C.R. (2000). Continuity and
change in possible selves in later life: A 5-year longitudinal study. Basic and
Applied Social Psychology, 2, 237-243.

Heckhausen, J., Dixon, R. A., & Baltes, P. B. (1989). Gains and losses in
development throughout adulthood as perceived by different adult age
groups. Developmental Psychology, 25, 109-121.

Hertzog, C., & Dixon, R. A. (1994). Metacognitive development in adulthood and
old age. In J. Metcalfe & A. P. Shimamura (Eds.), Metacognition: Knowing
about knowing. Cambridge, MA: The MIT Press, pp. 227-251.

Hertzog, C., Lineweaver, T. T., & McGuire, C. L. (1999). Beliefs about memory
and aging. In T. M. Hess & F. Blanchard-Fields, (Eds.) Social cognition and
aging. San Diego, CA: Academic Press, Inc. pp. 43-68.

Hess, T. M., Auman, C., Colcombe, S. J., & Rahhal, T. A. (2003). The impact of
stereotype threat on age differences in memory performance. Journal of
Gerontology, 58, P3-11.

Hooker, K. (1992). Possible selves and perceived health in older adults and
college students. Journal of Gerontology, 47, P85-95.

Hooker, K. (1999). Possible selves in adulthood: Incorporating teleonomic
relevance into studies of the self. In T. M. Hess & F. Blanchard-Fields
(Eds.) Social cognition and aging (pp. 97-122). San Diego, CA: Academic
Press.

Hooker K., Fiese, B.H., Jenkins, L., Morfei, M.Z., & Schwagler, J. (1996).
Possible selves among parents of infants and preschoolers. Developmental
Psychology, 32, 542-550.

Hooker, K. & Kaus, C.R. (1992). Possible selves and health behaviors in later
life. Journal of Aging and Health, 4, 390-411.

Hooker, K. & Kaus, C.R. (1994). Health-related possible selves in young and
middle adulthood. Psychology and Aging, 9, 126-133.

Hultsch, D. F., Hertzog, C., & Dixon, R. A. (1987). Age differences in
metamemory: Resolving the inconsistencies. Canadian Journal of
Psychology, 41, 193-208.

Kato, K. & Markus, H. R. (1993). The role of possible selves in memory.
Psychologia, 36, 73-83.

Lachman, M. E., Bandura, M., Weaver, S. L., & Elliott, E. (1995) Assessing
memory control beliefs: The Memory Controllability Inventory. Aging &
Cognition, 2, 67-84.











Levy, B. (1996). Improving memory in old age through implicit self-stereotyping.
Journal of Personality and Social Psychology, 71, 1092-1107.

Lineweaver, T. T. & Hertzog, C. (1998) Adults' efficacy and control beliefs
regarding memory and again: Separating general from personal beliefs.
Aging, Neuropsychology & Cognition, 5, 264-296.

Markus, H.R. & Herzog, A.R. (1992). The role of self-concept in aging. In K. W.
Schaie& M. P. Lawton (Eds.), Annual Review of Gerontology and Geriatrics
(Vol 11, pp. 110-143). New York: Springer.

Markus, H. & Nurius P. (1986). Possible selves. American Psychologist, 41, 954-
969.

Markus, H. & Wurf, E. (1987). The dynamic self-concept: A social psychological
perspective. Annual Review of Psychology, 38, 299-337.

McDonald-Miszczak, L., Hertzog, C., & Hultsch, D. F. (1995). Stability and
accuracy of metamemory in adulthood and aging: A longitudinal analysis.
Psychology & Aging, 10, 553-564.

Rahhal, T. A., Hasher, L., & Colcombe, S. J. (2001). Instructional manipulation
and age differences in memory: Now you see them, now you don't.
Psychology and Aging, 16, 697-706.

Ruvolo, A.P., & Markus, H.R. (1992). Possible selves and performance: The
power of self-relevant imagery. Social Cognition, 10, 95-124.

Ryan, E. B. (1992). Beliefs about memory changes across the adult lifespan.
Journals of Gerontology, 47, P41-P46.

Ryff, C.D. (1991). Possible selves in adulthood and old age: A tale of shifting
horizons. Psychology and Aging, 6, 286-295.

Selhulster, J. R. (1981a). Phenomenological correlates of a self theory of
memory. American Journal of Psychology, 94, 527-537.

Sehulster, J. B. (1981b). Structure and pragmatics of a self-theory of memory.
Memory & Cognition, 9, 263-276.

Schaie, K. W. (1994). The course of adult intellectual development. American
Psychologist, 49, 304-313.

Smith, J. & Freund, A. M. (2002). The dynamics of possible selves in old age.
Journal of Gerontology, 57B, P492-P500.