UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Memory changes across the adult lifespan: formation of gains and losses Mori, Monica Sachiko 1998

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1999-389472.pdf [ 8.86MB ]
Metadata
JSON: 831-1.0099432.json
JSON-LD: 831-1.0099432-ld.json
RDF/XML (Pretty): 831-1.0099432-rdf.xml
RDF/JSON: 831-1.0099432-rdf.json
Turtle: 831-1.0099432-turtle.txt
N-Triples: 831-1.0099432-rdf-ntriples.txt
Original Record: 831-1.0099432-source.json
Full Text
831-1.0099432-fulltext.txt
Citation
831-1.0099432.ris

Full Text

M E M O R Y C H A N G E S ACROSS THE A D U L T LIFESPAN: F O R M A T I O N OF GAINS A N D LOSSES by M O N I C A SACHIKO MORI B.A., Concordia University, 1985 M . A . , University of Western Ontario, 1989 M.L.I. S., University of Western Ontario, 1989 A THESIS SUBMITTED IN P A R T I A L F U L F I L L M E N T OF THE REQUIREMENTS FOR THE D E G R E E OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES Department of Psychology We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH C O L U M B I A December 1998 © Monica Sachiko Mori, 1998  In presenting  this  thesis in  degree at the University of  partial  fulfilment  of  the  requirements  for  an advanced  British Columbia, I agree that the Library shall make it  freely available for reference and study. 1 further agree that permission for extensive copying of  this thesis for  department  or  by  his  or  scholarly purposes may be granted by the head of my her  representatives.  It  publication of this thesis for financial gain shall not permission.  Department of  ffiflfloU^tf  .  The University of British Columbia Vancouver, Canada  DE-6 (2/88)  is  understood  that  copying  or  be allowed without my written  11  ABSTRACT This experiment investigated memory changes across the adult lifespan and some factors that might be associated with these changes. Adult participants of all ages (16 to 83 years old) were asked to orally describe scenic color photographs, and then following a delay, to redescribe these pictures from memory. Given information is objective, physical objects and their attributes that are depicted in a target picture, whereas beyond information is subjective, personal experiences and inferences that are not depicted in a target picture per se but are associated with a target picture. Chapter 3 examined the content of these picture descriptions for the amount of given and beyond information that was encoded and retrieved about target pictures. The results indicated an age-related decline in memory for given information and preserved memory for beyond information. Chapter 4 examined the relationship between perceptual and verbal ability and memory for given and beyond information. Perceptual ability was assessed by self-report measures of auditory and visual ability and verbal ability was measured by a standardized test. The results indicated that an age-related improvement in verbal ability, but not an age-related decline in perceptual ability, was related to memory for given and beyond information. Chapter 5 explored age-related changes in memory for feminine and masculine information across the adult female l'fespan. Feminine and masculine information is information that would be considered exclusively relevant to young women and men, respectively. The results indicated an age-related increase in memory for feminine information and no age-related change in memory for masculine information. The divergent age-related changes in memory for given and beyond information and for feminine and masculine information were interpreted in terms of a developmental approach to schema theory and the lifespan psychology notions of selective optimization with compensation and loss in the service of growth. The present study suggests an integration between the domains of personality and cognitive psychology as one avenue for future research that could lead to a more complete understanding of memory and aging.  T A B L E OF CONTENTS Contents  Page  Abstract Table of Contents List of Tables List of Figures Acknowledgements  ii iii v vii viii  '.  Chapter One  Introduction and Overview Memory for Given and Beyond Information A Lifespan Approach to Memory Lifespan Developmental Approach to Cognition A Developmental Approach to Schema Theory Factors Related to Memory for Given and Beyond Information Memory for Feminine and Masculine Information Summary of Chapter .. Overview of Remaining Chapters  1 2 6 7 9 13 14 16 20  Chapter Two  Method Participants Materials Design and Procedure Scoring Procedure  Chapter Three  Memory for Given and Beyond Information Results Preliminary Analyses Obtained and Aggregated Items....; Univariate and Multivariate Outliers Primary Analyses Memory Encoding and Retrieval Phase Performance Wordiness Study Instructions Discussion  31 38 39 39 43 44 44 47 63 69 76  Chapter Four  Factors Related to Memory for Given and Beyond Information Method Measurement Tools Results Age, Verbal Ability and Performance on the Obtained and Aggregated Items Age, Perceptual Ability and Performance on the Obtained and Aggregated Items Discussion..  81 82 82 83  22 22 25 .25 28  84 93 99  T A B L E OF CONTENTS (continued) Contents  Page  Chapter Five  Memory for Feminine and Masculine Information Method Items on the Feminine and Masculine Scales Neutral Item Results Discussion  101 103 103 104 107 115  Chapter Six  General Discussion Motivation, and Main Objectives and Findings Theoretical Implications Limitations of this Research Future Research Main Contribution  116 116 124 127 132 134  References Appendix A Appendix B Appendix C  Target and Practice Pictures Scoring Key for Picture Descriptions Testing for floor effects in the encoding phase, retrieval phase and encoding-retrieval match performance on each item on the given and beyond scales (n=162)  135 143 152  167  LIST OF TABLES Table 2.1 2.2 2.3  3.1 3.2 3.3 3.4a 3.4b 3.5a 3.5b 3.6 3.7a 3.7b 3.8a 3.8b  Page Demographic data Target pictures Reliability estimates for encoding phase, retrieval phase and encodingretrieval match performance on each item on the given and beyond scales  23 26  Hypotheses for Chapter 3 Intercorrelations between encoding phase, retrieval phase and encodingretrieval match performance on each item on the given scale Intercorrelations between encoding phase, retrieval phase and encodingretrieval match performance on each item on the beyond scale Influence of age and age squared on retrieval phase performance on each item on the given scale Influence of age and age squared on retrieval phase performance on each item on the beyond scale Influence of age and age squared on encoding phase performance on each item on the given scale Influence of age and age squared on encoding phase performance on each item on the beyond scale Influence of encoding phase performance on retrieval phase performance on each item on the given and beyond scales Influence of age and age squared on encoding-retrieval match performance on each item on the given scale Influence of age and age squared on encoding-retrieval match performance on each item on the beyond scale..... Influence of age and age square on forgetting for each item on the given scale Influence of age and age square on forgetting for each item on the beyond scale... Influence of age and age square on additions made at the retrieval phase for each item on the given scale Influence of age and age square on additions made at the retrieval phase for each item on the beyond scale Influence of age on verbatim repetitions made at the encoding and retrieval phases Influence of verbatim repetitions and age on encoding phase performance on each item on the given and beyond scales Influence of verbatim repetitions and age on retrieval phase performance on each item on the given and beyond scales Influence of study instructions and age on encoding phase performance on each item on the given and beyond scales Influence of study instructions and age on retrieval phase performance on each item on the given and beyond scales  33  ;  3.9a 3.9b 3.10 3.11 3.12 3.13  ;  3.14  30  41 42 45 46 50 51 53 55 56 57 58 59 60 64 66 67 71 72  LIST OF TABLES (continued) Table 3.15  4.1 4.2 4.3 4.4  4.5  4.6 4.7  5.1 5.2 5.3 5.4 5.5 5.6  5.7  5.8  Page Influence of study instructions and age on encoding-retrieval match performance on each item on the given and beyond scales  75  Relationship between age and verbal and perceptual ability 85 Relationships between verbal ability, age and encoding phase performance on each item on the given and beyond scales 87 Relationships between verbal ability, age and retrieval phase performance on each item on the given and beyond scales 88 Relationship between N A A R T and encoding phase performance on each item on the given and beyond scales when the influence of age is partialled out . 89 . Relationship between N A A R T and retrieval phase performance on each item on the given and beyond scales when the influence of age.is partialled out 90 Relationship between self-rating of hearing and encoding phase performance on each item on the given and beyond scales : 95 Relationship between self-rating of hearing and retrieval phase performance on each item on the given and beyond scales 96 Gender differences in performance on items on the given and beyond scales for young adults between 20 and 39 years Intercorrelations between encoding phase, retrieval phase and encodingretrieval match performance on each item on the given and beyond scales... Influence of age and age squared on encoding phase performance on the neutral item and each item on the feminine and masculine scales Influence of age and age squared on retrieval phase performance on the neutral item and each item on the feminine and masculine scales Influence of age and age squared on encoding-retrieval match performance on the neutral item and each item on the feminine and masculine scales Influence of age and age squared on encoding phase performance on each item on the feminine and masculine scales (relative to performance on the neutral item) Influence of age and age squared on retrieval phase performance on each item on the feminine and masculine scales (relative to performance on the neutral item) Influence of age and age squared on encoding-retrieval match performance on each item on the feminine and masculine scales(relative to performance on the neutral item)  105 106 108 109 110  Ill  113  114  LIST OF FIGURES Figure 3.1 3.2 3.3 3.4 3.5  Page Relationship between encoding and retrieval phase performance. Retrieval phase performance on the quantity item by study instruction Encoding-retrieval match performance on the quantity item by study instruction Encoding-retrieval match performance on the absence item by study instruction Encoding-retrieval match performance on the timeframe item by study instruction  48 73 73 73 73  viii  ACKNOWLEDGEMENTS First and foremost, I am grateful to Peter Graf for his dedication in teaching me about cognitive psychology, his patience in reading endless versions of my dissertation, and his conscientiousness in preparing me for a career. I have learned more (given and beyond information) than I could have ever expected from one person. I am indebted to Jerry Wiggins for his generativity (high nurturance and high dominance) in inviting me into his lab and teaching me about personality and interpersonal research. I am grateful to my committee members. Susan Butt for providing her professional support and direction and for helping me to develop my teaching skills. Jim Enns for his clear thinking and for offering his constructive comments through the years. A special thanks to Gloria Gutman for introducing me to gerontology and strongly encouraging me to return to school. I also appreciate all the people who helped at the different stages of data collection and analysis. Nadine Bruce for collecting the data, Jennifer Shapka for her endless hours painstakingly coding the picture descriptions, and Arsalan Ghani for assisting with the reliability estimates (and for being just generally reliable). And Michelle Yik and K i m Barchard for offering their statistical expertise. I am grateful for the financial support that I received from the B C Health Services Resarch Foundation (Student Grant) and the Canadian Association on Geronotology (Donald Menzies- C A G Bursary). And finally, I'd like to thank my family and friends for their years (and years) of emotional support and encouragement. M y Mom for urging me to get an education and my Dad for inspiring me to study storytelling. M y brother, Brian, and sister-in-law, Angel, for keeping me laughing, and my nephew, Aaron, for reminding me about the bigger world. Del Paulhus for his academic stream of consciousness over pizza. And Janet McNern and Carrie Hanson for making sure I always had enough ice cream, chocolate and good food.  1  CHAPTER 1 INTRODUCTION AND OVERVIEW In North America, older persons are viewed as a population in need of resources in order to compensate for declining mental and physical health, yet paradoxically in many other cultures, older persons are considered as information resources whose role is to convey their knowledge and wisdom through storytelling (Blum-Kulka, 1993; Mergler & Goldstein, 1983; Miller. Wiley. Fung & Liang, 1997; Minami & McCabe, 1995; Ramelli & Stella, 1993; Wallace, 1992). This discrepancy between different cultural perspectives on older persons' memory systems is paralleled in the empirical literature. In cognitive psychology, research with an aging population has traditionally sought to identify age-related memory failures or impairments (Hasher & Zacks, 1979; Kane & Hasher, 1994; Light, 1991; Salthouse, 1982), and recently, research has revealed some mental operations that are preserved or enhanced with age. For example, studies have shown that explicit memory declines while implicit memory is preserved with age (Graf, 1990; Light & Singh, 1987). As well, a decline in fluid intelligence has been shown to be accompanied by an increase in crystallized intelligence (Baltes, 1987; Bakes, Dittmann-Kohli & Dixon, 1984; Horn, 1982; Kaufman, Kaufman-Packer, McLean & Reynolds, 1991). Together, cultural and scientific perspectives present a dynamic and complex picture of the aging memory system which involves the coordination of declining and improving mental operations. The present experiment investigated divergent age-related changes in the types of information that was encoded and retrieved about target pictures and then explored some factors that may be related to these age-related changes in memory. The primary focus of the present experiment was to trace memory for given and beyond information across the adult lifespan. Given information is objective, physical attributes that are depicted in a target picture, whereas beyond information is subjective, personal experiences and inferences that are not depicted in a target picture per se but are associated with a target picture. It was hypothesized that there  would be an age-related decline in memory for given information and an age-related increase in memory for beyond information. A secondary focus of the present experiment was to explore two factors, verbal and perceptual ability, that may be related to these divergent age-related changes in memory and to explore the possibility of changes in memory for feminine and masculine information across the adult female lifespan. Feminine and masculine information is information that would be considered exclusively relevant to young women and men, respectively. This is the first experiment to employ target pictures in an investigation of age-related changes in memory for given and beyond information. Previous research employed verbal stimuli (Adams, 1991; Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Gould, Trevithick & Dixon, 1991) and this experiment was designed to replicate and extend these findings to nonverbal stimuli. Picture stimuli were also employed to allow participants rather than the experimenter to define the to-be-remembered information and thereby enhance the opportunity for age-related changes in what is extracted and remembered about target stimuli. Two accounts that follow from a lifespan approach to memory were used to guide my understanding of age-related changes in memory and the factors that may be related to these changes. A lifespan approach to cognition describes the relationship that may exist between divergent age-related changes in cognitive ability. A developmental approach to schema theory describes age-related changes in schema-driven remembering, where the accumulation of experiences can influence schema structure and schemas can direct the encoding and retrieval of different types of information. MEMORY FOR GIVEN AND BEYOND INFORMATION What is remembered about a target stimulus can be conceptualized along a continuum, with given information at one end and beyond information at the other. In the linguistic and cognitive literature with verbal stimuli (Bruner, 1957, 1986), given information is the objective  meaning of a text either in its exact form or in a form that can be directly and logically tied to a text. Beyond information is the subjective meaning that is drawn from a text, such as personal experiences and world knowledge that are associated or inferred from a text, and so this type of information, while not directly tied, is plausibly related to a text. To illustrate this distinction with a non-verbal stimulus, a photograph of a park, given information is the things that are depicted in the picture, such as the people, trees and a lake, and the physical attributes of these things, such as the lake is blue. By contrast, beyond information is personal experiences that are activated by things that are depicted in the picture, such as reminiscing about the last occasion one was in a similar park or inferring that the photograph was taken in autumn because the leaves have fallen. Given and beyond information are distinguished by the degree to which they are tied to a target stimulus and the degree to which they reflect typical or common knowledge (Bruner, 1957, 1986). Given information is the physical details of a stimulus that are objective, and that reflect shared or generic knowledge; beyond information is the subjective interpretations and evaluations of a stimulus that reflect unique personal experiences and world knowledge. An important consideration about this conceptualization of a given-beyond continuum is that information can be both given and beyond. A lake, for example, can be considered given information when it is used to describe a park scene or it can be considered beyond information when it is used to recount a personal experience in a park. The cognitive and gerontological literature reports divergent age-related changes in memory for given and beyond information. A n age-related decline in recall of given information is well documented. When participants are asked to freely recall a list of unrelated words, results typically show that older adults recall fewer targets especially from longer word lists. This pattern of age-related change has been the focus of interest in current research on memory and aging, with numerous studies devoted to accounting for age-related memory decline; these  4  studies reveal that memory decline is related to an age-related slowing in information processing or a disruption in the coordination of processing due to interference (e.g., Hasher & Zacks. 1979: Light, 1991; Salthouse, 1982, 1991, 1993; Salthouse & Kail, 1983). There is some evidence for an age-related increase in memory for beyond information of a target stimulus as demonstrated by research on memory for some types of inferences. Inferences can be differentiated by the types of information they draw on and the purposes they serve (Boden & Bielby, 1983; Gould, Trevithick & Dixon, 1991; Light, Valencia-Laver & Zavis, 1991). Bridging inferences are closely tied to the target stimulus and are required for establishing coherence. For example, the inference that a woman was trained as a secretary is closely connected to the target statement that the woman was a secretary. Elaborative inferences are drawn from personal experiences and can serve as a conversational resource. For example, evaluations that are prefaced by "I like", "sounds like", and "I remember" are drawn from personal experiences. Research has demonstrated that young and older adults are equally likely to remember bridging inferences (Gould, Trevithick & Dixon, 1991). By contrast, age differences are found in remembering elaborative inferences, with older adults showing better memory for elaborative inferences than young adults (Adams, 1991; Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Gould, Trevithick & Dixon, 1991). A clear demonstration of this pattern comes from a study by Gould, Trevithick and Dixon (1991) who asked groups of young 5  and groups of older participants to collaboratively recall a story. Content analyses of their conversations revealed that, while young and older adults expressed an equal number of bridging inferences, older adults produced more elaborative inferences than did young adults. The extant literature, however, does not present unequivocal findings. A number of studies have reported a decrement in memory for bridging inferences with age (Cohen, 1979, 1981; Till, 1985; Till & Walsh, 1980) and no age differences in memory for elaborative inferences (Belmore, 1981; Light, Valencia-Laver & Zavis, 1991). These conflicting results are likely due to a failure to  draw inferences at the time of encoding and/or task demands that exceed working memory capacity (for a review see Light, 1990). Discrepant findings are also found for age-related changes in memory for beyond but not given information depending on the type of task that was employed to measure memory. Studies show an age-related decrement in memory for given information about stories as measured by free recall (Adams, 1991; Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Walsh & Baldwin, 1977; Walsh, Baldwin & Finkle, 1980; Zandi & Gregory, 1988) and recognition tasks (Reder, Wible & Martin, 1986). Conflicting results have been obtained for remembering beyond information with some studies showing an age-related increase (Adams, 1991; Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Zandi & Gregory, 1988) and others showing no age differences (Reder, Wible and Martin, 1986; Walsh & Baldwin, 1977; Walsh, Baldwin & Finkle, 1980). However, these two sets of studies employed different types of memory tests. Memory performance was assessed by free recall tasks in the former set of studies that revealed an agerelated increase in memory for beyond information and by recognition tests in the latter studies that failed to find age-related changes. It is possible that recognition tests place limits on the uniqueness of the beyond information that can be tested. Beyond information that is measured by recognition tests is defined by the experimenter and is limited to typical integrations of separate target sentences, whereas beyond information that is measured by recall tests is defined by the participant, and in addition to common semantic integrations, includes unique associations to personal experiences. Therefore, it may be that age differences were attenuated in the latter studies because they employed recognition versus recall tests and/or because beyond information was experimenter- rather than participant-defined. Thus far, separate lines of research have demonstrated an age-related decline in memory for given information and an increase in memory for beyond information of a target stimulus or event and there have been some attempts to directly investigate these developmental patterns.  6  The gerontological and psychological research demonstrates an age-related decline in m e m o r y for given information and an age-related increase in memory for beyond information about target stories (Adams, 1991; Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Gould, Trevithick & Dixon, 1991). The present experiment was designed to replicate and extend the findings from previous studies that employed verbal stimuli by investigating age-related changes in memory for given and beyond information produced from non-verbal stimuli, that is, color pictures. A LIFESPAN APPROACH TO M E M O R Y Theories of memory and aging have typically focused on explaining age-related declines in cognitive performance. For example, resource theories propose that successful task completion relies on the efficient assembly or coordination of component processes, where efficiency refers to both the amount of mental energy to fuel requisite processes and the speed of information processing (e.g., Vernon, 1987, 1993; Vernon & Mori, 1992). Thus, age-related declines in cognitive performance have been viewed as a breakdown in efficient processing, for example, an age-related slowing in speed of processing or a disruption in the coordination o f processing (Hasher & Zacks, 1979; Light, 1991; Salthouse, 1982, 1991, 1993; Salthouse & Kail, 1983). These resource theories, however, do not account for age-related improvements in memory. The present experiment integrates cognitive psychology and gerontology by adopting a developmental or lifespan approach to memory changes across the adult lifespan which includes an examination of both those mental operations that improve or emerge and those that decline with maturation or the accumulation of experiences. The integration of these two domains offers different descriptions of the aging memory system. A lifespan approach to cognition describes the relationship that may exist between divergent age-related changes in memory. A developmental approach to schema theory describes age-related changes in schema-driven remembering, where the accumulation of experiences can influence schema structure and  7  schemas can direct the encoding and retrieval of different types of information. Lifespan Developmental Approach to Cognition The lifespan notions of selective optimization with compensation and loss in the service of growth are different descriptions of the relationship that may exist between declining and improving mental operations. The notion of selective optimization with compensation is the proposition that certain mental operations develop that can offset the effects of decline in other competencies (Baltes, 1987, 1993; Baltes, Dittmann-Kohli & Dixon, 1984). The notion of loss in the service of growth is the proposition that new cognitive abilities emerge at the expense of other cognitive capacities (Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Labouvie-Vief & Schell, 1982). Selective optimization with compensation. The lifespan developmental notion of selective optimization with compensation is that intellectual development involves an interplay between the growth and decline in cognitive competencies, where certain mental operations develop or emerge that can counterbalance the detrimental effects of decline in other competencies (Baltes, 1987, 1993; Baltes, Dittmann-Kohli & Dixon, 1984). Support for the notion of selective optimization with compensation comes predominately from a dual-process approach to intellectual development, where contrasting developmental patterns are shown for cognitive mechanics and cognitive pragmatics. Cognitive mechanics "reflects] the neurophysiological architecture of the brain as it has developed during evolution" (Baltes. 1993, p. 582) and can be thought of in terms of computer analogues as the hardware of the mind. Cognitive mechanics can be measured by tests of fluid intelligence (e.g., Performance scale of the WAIS-R), and speed and accuracy measures of elementary sensory and mental processes that require discrimination, comparison and categorization (e.g., dichotic listening tasks, chunking in free recall, Sperling's full report paradigm). By contrast, cognitive pragmatics can be thought of as the software of the mind and "reflect[s] the kind of knowledge and information that cultures  offer as "bodies of factual and procedural knowledge" about the world and human affairs and that individuals acquire as they participate in culture-based socialization" (Bakes, 1993. p. 582). Cognitive pragmatics can be measured by tests of crystallized intelligence (e.g., Verbal scale of the WAIS-R) and measures of breadth of knowledge and experience, language comprehension, judgment, quantitative thinking and wisdom (e.g., analogies and remote association tasks). Cognitive mechanics and cognitive pragmatics show different patterns of development across the adult lifespan. Cognitive pragmatics, as measured by tests of crystallized intelligence, is reported to increase across the adult lifespan with accompanying decreases in cognitive mechanics, as measured by tests of fluid intelligence (Bakes, 1987, 1993; Bakes, DittmannKohli & Dixon, 1984; Horn, 1982; Kaufman, Kaufman-Packer, McLean & Reynolds, 1991). These contrasting developmental patterns are thought to be interrelated, where certain developing mental abilities such as crystallized intelligence can compensate for the detrimental effects of decline in other competencies such as fluid intelligence. Loss in the service of growth. According to the notion of loss in the service of growth, there are changes in how information is processed and organized at each life stage, where each stage builds upon the types of processing that were previously acquired. These competencies may emerge, however, at the expense of other cognitive capacities (Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Labouvie-Vief, 1989; Labouvie-Vief, 1985; Labouvie-Vief & Schell, 1982). Labouvie-Vief (1989) proposes that age-related changes in cognitive competencies arise from a reorganization of the cognitive system, where "outer-oriented, objective, and analytic forms of thinking are reintegrated with inner-oriented, subjective and symbolic forms" (Adams et al., 1990, p. P25). As illustrated by a study by Adams et al. (1990), this new cognitive competency may emerge at the expense of other competencies. In their study, young and older adults were presented with either a fable or a non-fable, and then following a delay, were asked to provide a written recall of the story. The results indicated that, in response to the non-fable,  young adults employed a story recall style that was highly text-based and literal,'for example, paraphrases of the propositional level details of the text, whereas older adults exhibited a style that was inferential, such as inferences that integrated events in a story or interpreted the metaphoric significance of a story. The results were interpreted as a maturational shift from outer-oriented analytical processing of the propositional level details of the text to inner-oriented symbolic processing of the metaphoric significance of the text. As well, because aging may involve a general decline in processing resources, this maturational shift may be a compensatory strategy that efficiently uses limited processing resources. Because inferences are interpretations of integrated chunks of essential propositional level details, the global meaning of a text is retained at the expense of retaining more detailed information. A lifespan perspective of cognitive development views the aging memory system as a coordination of declining and improving cognitive operations. The notions of selective optimization with compensation and loss in the service of growth offer different descriptions of the relationship between diverging age-related changes in cognitive performance, where new cognitive competencies can develop that can either compensate for or trade-off other cognitive competencies that show maturational decline, respectively. Hence, while both the notions of selective optimization with compensation and loss in the service of growth address age-related changes in cognitive abilities, they differ in that they view cognitive competencies that improve with age as either a pull towards or a push away from those cognitive competencies that show age-related decline, respectively. A Developmental Approach to Schema Theory A developmental approach to schema theory offers a theoretical account of memory changes across the lifespan. Schemas are also known as schemata (Bartlett, 1932), scripts (Schank & Abelson, 1977), story grammars (Mandler & Johnson, 1977), scene schemas (Mandler & Parker, 1976) and self-schemata (Markus, 1977). Schemas are abstract  10  representations of organized knowledge, where the scope of the knowledge structure can vary from general conceptualizations of a class of events to particular representations of specific instances. These knowledge structures specify the interrelations among its constituents parts which in turn are interrelated to other knowledge structures so that a person's knowledge base consists of a network of schemas that are both intrarelated and interrelated. A important quality of schemas is that they are active and can change with the accumulation of knowledge or experiences and can guide the encoding and retrieval of information. Changes in schema structure. Rumelhart and Norman (1978) proposed that learning can influence schema structure in three ways — schemas can grow in size, new schemas can be created, and old pathways can be strengthened — called accretion, restructuring and tuning, respectively. Accretion refers to the idea that schemas can grow in size as new experiences and information are combined or integrated with previously acquired information. In this case, new experiences are simply added to the knowledge base when they match information that is already stored in an existing schema. Restructuring refers to the idea that new schemas are formed when new information cannot be added to an existing schema, for example, when a person acquires a new skill or learns a new body of information. Restructuring usually occurs when a new schema is patterned after an old one, but a new schema can also be formed by combining recurring patterns of old schemas. Tuning refers to the idea that certain existing interrelations or associative pathways are strengthened upon repeated exposure to similar experiences or types of information, such as the case when a person refines a motor skill or when a child becomes proficient in a language. Tuning is important to performance because it enhances accuracy by specializing the applicability of a schema to limited situations, by generalizing the applicability of a schema to novel situations, and by allowing for inferences or intelligent guesses to fill gaps when information is absent.  11  Schema-driven remembering. Schemas are assumed to influence memory performance by initiating and guiding encoding and retrieval processes. Schemas can direct attention to certain stimuli and therefore bias the likelihood for certain information to be selected for encoding (Arbuckle, Vanderleck, Harsany & Lapidus, 1990; Hess, Donley & Vandermaas, 1989). In their second study, Arbuckle and her colleagues (1990) presented young and older adults with biographies of well known individuals (e.g., Adolf Hitler, Elizabeth Taylor) under two conditions: early and late schema access conditions. In the early schema access condition, biographies presented well-known then little-known information. In the late schema access condition, little-known then well-known information was presented. Both age groups showed superior memory performance when schemas were activated early versus late during encoding. As well, schemas can influence response selection by accessing or activating schema relevant information or by editing-out schema irrelevant information (Hess & Flannagan, 1992; Hess & Slaughter, 1990). For example, Hess and Flannagan (1992) asked young and older adults to read a story from one of two perspectives and then to recall the story on two occasions: first from the same perspective taken at reading and then from an alternative perspective. For both age groups, they found that memory for story information was related to the recall perspective; for both recall conditions, there was superior memory performance for story information that was relevant versus irrelevant to the recall perspective, and in the second recall condition, there was a decrease in memory performance for information that was previously relevant to the first but irrelevant to the second recall perspective. Therefore, schemas can be employed at study or at test which can influence the type of information that is encoded and retrieved about a target stimulus. Research has been primarily concerned with demonstrating the presence of schemadriven remembering. Studies have shown that schemas exist as early as 4 years of age, although there are conflicting reports as to when schemas can be employed spontaneously to guide task  12  performance. One set of studies found that 4- and 5-year-olds are able to employ schemas to plan a pretend shopping trip to the grocery store (Hudson & Fivush, 1991) and to create a story by re-arranging picture sequences (Fivush & Mandler, 1985). Another set of studies showed that it is not until later in childhood that schemas are spontaneously used to guide remembering (Mandler & DeForest, 1979; Schmidt & Schmidt, 1986). For example, Schmidt and Schmidt (1986) found that 8- and 11-year-olds spontaneously generated themes to aid story recall and that 5-year-olds were also able to use themes to facilitate story recall but only with instruction that prompted them to use thematic retrieval cues. Although there is some discrepancy in the age-ofonset of spontaneous schema-driven remembering in childhood, there is substantial evidence that schemas are employed spontaneously to guide remembering throughout adulthood (Arbuckle, Vanderleck, Harsany & Lapidus, 1990; Hess, 1985; Hess, Donley & Vandermaas, 1989; Hess & Flannagan, 1992; Schmidt & Schmidt, 1986). In fact, young and older adults perform equally well on memory tasks that engage schema-driven remembering; when to-be-remembered targets vary in the degree to which they are associated with a schema, there are no age differences in remembering targets that are relevant (Hess, Donley & Vandermaas, 1989; Hess & Flannagan, 1992), typical (Hess, 1985; Hess, Donley & Vandermaas, 1989; Micco & Masson, 1992), or familiar (Arbuckle, Vanderleck, Harsany & Lapidus, 1990). Less attention has been given to the influence of maturation on schema structure and the impact schematic structural change may have on remembering. Following Rumelhart and Norman (1978), a lifespan approach to schema theory states that maturation or the accumulation of experiences leads to a broader range of knowledge (accretion) and to schemas that are more refined, either generalized or specialized in their applicability to new situations (restructuring and tuning). Because schemas can direct what types of information are encoded (Arbuckle, Vanderleck, Harsany & Lapidus, 1990; Hess, Donley & Vandermaas, 1989) and retrieved (Hess & Flannagan, 1992; Hess & Slaughter, 1990) about a target stimulus, maturation may influence  schema-driven remembering. A lifespan approach to memory offers two different, but not incompatible, explanations of the aging memory system. The notions of selective optimization with compensation and loss in the service of growth refer to the idea that new cognitive competencies can develop that can either compensate for or trade-off other cognitive competencies that show maturational decline, respectively. A developmental approach to schema theory describes the influence of maturation on schema-driven remembering, where the accumulation of experiences can influence schema structure and schemas can direct the encoding and retrieval of different types of information. Although these accounts offer different descriptions of the aging memory system, they are not competing approaches. It is possible that schemas may develop in such a way as to guide the encoding and retrieval of certain types of information at the expense of encoding and retrieving other types of information. It is also possible that older adults may rely or focus on certain types of schema-driven remembering in order to compensate for declines in other types of schemadriven remembering. The present experiment employed these two accounts of the aging memory system in order to guide my understanding of age-related changes in memory and to guide the types of questions that I asked about the aging memory system. FACTORS R E L A T E D TO MEMORY FOR GIVEN AND BEYOND INFORMATION The present experiment explored the relationships between verbal and perceptual ability and age-related changes in memory for given and beyond information. In general, it is difficult to examine causal relationships between cognitive operations with lifespan data. For example, it is not possible to draw causal relationships from the present experiment as correlational analyses were employed. As well, strong conclusions cannot be drawn even when relationships between cognitive operations are found as it is possible that the relationships were mediated by another cognitive operation that was not investigated but is related to the cognitive operations that were investigated. For example, if a relationship between verbal ability and memory is found, it is  14  possible that this relationship was due to their common relationship with another cognitive ability, such as fluid intelligence (Vernon, 1987; Vernon & Mori, 1992). These concerns not withstanding, it is still profitable to explore the relationships between cognitive operations as they place age-related changes in memory into a broader context and provides some avenues for future research. This part of the research project explored the possibility that verbal ability may be related to memory for beyond information and that perceptual ability may be related to memory for given information. Previous research has shown an age-related increase in verbal ability (Baltes, 1987; Baltes, Dittmann-Kohli & Dixon, 1984; Horn, 1982; Kaufman, Kaufman-Packer, McLean & Reynolds, 1991) and Adams (1991) has reported that verbal ability was related to the recall of beyond information, such as the moral or gist of a story, but not to the recall of given information, for example, the actions and events of a story. Because beyond information requires a greater degree of inferencing than given information, it is possible that verbal ability may facilitate elaborative processing that is required for the expression or articulation of beyond information. Hence, an age-related increase in verbal ability may be related to an age-related increase in memory for beyond information. A relationship between age-related declines in perceptual ability and an age-related decline in memory for given information was explored because it is possible that a declining ability to perceive given information may restrict further processing. M E M O R Y FOR FEMININE AND MASCULINE INFORMATION A secondary interest of the present experiment was to explore changes in memory for feminine and masculine information across the adult female lifespan. Gender studies in social psychology define feminine and masculine roles by those behaviors and attitudes that have been empirically demonstrated as characteristic of young women and men, and studies on gender and aging entail tracing these gender stereotypes across the lifespan (Gutmann, 1977, 1987;  15  Labouvie-Vief, 1996; Lowenthal et al., 1975; Sinnott, 1984; Turner, 1982). It is possible that feminine and masculine roles may direct attention toward different types of information, and in turn, processing this information may contribute towards further shaping these roles. Although research has been limited to investigations of young adults' memory for stimuli related to stereotyped feminine and masculine roles (Martin & Paulhus, 1985; Signorella, 1992), the present experiment explored the possibility that maturational changes in feminine and masculine roles may show age-related changes in the types of information that is encoded and retrieved about a target stimulus. Following from social psychology and gerontology, feminine and masculine information was defined according to gerontological and psychological research findings that suggest certain types of information may be more relevant to young women and men, respectively, and then traced memory for feminine and masculine information across the female lifespan. Age-related changes in memory for feminine and masculine information were traced across the adult lifespan for women only, and not for men, because there were too few older male participants. Gerontological studies on feminine and masculine behaviors and attitudes reveal two types of developmental patterns: gender blurring and gender reversal. Gender blurring refers to the disappearance of differences between men and women in their behaviors and attitudes as they age. For example, Sinnott (1984) found that older women and men scored equally high on feminine and masculine attributes of the Bern Sex-Role Inventory (Bern, 1974), indicating that people become more androgynous in their behaviors as they age. Gender reversal refers to a cross over in feminine and masculine behaviors and attitudes, where there is an emergence of gender-opposite attitudes and behaviors with age. Support for a gender reversal shift with maturation has been demonstrated for gender roles (Lowenthal et al., 1975), perceived social attitudes (Gutmann, 1977, 1987), and achievement (Labouvie-Vief, 1996). Because feminine and masculine roles may direct attention toward different types of information and processing  16  this information may contribute towards further shaping feminine and masculine roles, maturational changes in feminine and masculine roles may show age-related changes in the type of information that is encoded and retrieved about a target stimulus. Age-related changes in memory for feminine and masculine information were also explored because several studies provide indirect evidence that young women and men may focus on or show preference for different types of information, and hence, they may encode and retrieve different types of information about a target stimulus. For example, a descriptive study of the meanings of personal possessions reveals that women are more likely than men to consider social or interpersonal aspects as salient of desired gifts and treasured possessions (Kamptner, 1991). Because women are attracted to the social and interpersonal qualities of personal possessions, they may also show a similar preference in encoding and retrieval of information regarding people and personal experiences. Similarly, scores on psychometric tests of spatial ability indicate superior performance for males than females (Feingold, 1988; Voyer & Voyer, 1995). Because men show superior spatial ability, they may be more inclined to encode and retrieve information regarding spatial location and position. And finally, a qualitative analysis of adults' verbal reports of stories reveal that females make more references to emotion than males (Adams, 1991). This finding suggests that women may encode and retrieve more information regarding emotion from photographs. Together, studies that provide indirect evidence that young women and men may show preference for different types of information suggest that men may be more inclined to encode and retrieve information regarding position and spatial location whereas women may be more likely to encode and retrieve information regarding emotion, events and people. SUMMARY OF CHAPTER In North America, older persons are viewed as a population in need of resources in order to assist with declining mental and physical health, whereas in other cultures, older persons are  considered as valuable resources who convey personal experiences and world knowledge through storytelling that can benefit the survival of younger generations and their culture (Mergler & Goldstein, 1983; Ramelli & Stella, 1993; Wallace, 1992). This discrepancy between different cultural perspectives on older persons' memory systems is paralleled in the empirical literature, for example, a decline in fluid intelligence has been shown to be accompanied by an increase in crystallized intelligence (Baltes, 1987; Baltes, Dittmann-Kohli & Dixon, 1984; Horn, 1982; Kaufman, Kaufman-Packer, McLean & Reynolds, 1991). Together, cultural and scientific perspectives present a dynamic.and complex picture of the aging memory system which involves the coordination of declining and improving mental operations. The present experiment investigated divergent age-related changes in memory by examining age-related changes in the types of information that were encoded and retrieved about target stimuli and some factors that may be related to these age-related changes. The present experiment adopts a developmental or lifespan approach to memory changes across the adult lifespan which includes an examination of both those mental operations that improve or emerge and those that decline with maturation or the accumulation of experiences. This position integrates cognitive psychology and gerontology. The integration of these domains offers two accounts for these divergent age-related changes in cognitive ability. First, following from the notions of selective optimization with compensation and loss in the service of growth, older persons may rely on those cognitive operations that emerge in order to compensate for, or which results in a trade-off with, other cognitive competencies that show maturational decline, respectively. Second, following a developmental approach to schema theory, certain cognitive competencies may be encouraged because schematic structural changes that arise through the accumulation of experiences may direct what types of information are encoded and retrieved about a target stimulus. Although these approaches offer different accounts of the aging memory system, they are not competing approaches. It is possible that schemas may develop in such a  18  way as to guide the encoding and retrieval of certain types of information at the expense of encoding and retrieving other types of information. It is also possible that older adults may rely or focus on certain types of schema-driven remembering in order to compensate for declines in other types of schema-driven remembering. The present experiment employed these two accounts of the aging memory system to guide my understanding of the age-related changes in memory and the types of questions that I asked about aging memory system. The present experiment investigated age-related changes in the types of information that were encoded and retrieved about target pictures. The primary interest of this experiment was to investigate age-related changes in memory for given and beyond information. The gerontological and psychological research demonstrates an age-related decline in memory for given information about target stories (Adams, 1991; Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Gould, Trevithick Dixon, 1991; Reder, Wible & Martin, 1986; Walsh & Baldwin, 1977; Walsh, Baldwin & Finkle, 1980; Zandi & Gregory, 1988) and an age-related increase in memory for beyond information about target stories (Adams, 1991; Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Gould, Trevithick & Dixon, 1991; Zandi & Gregory, 1988). The present experiment was designed to replicate and extend the findings from previous studies that employed verbal stimuli by investigating age-related changes in memory for given and beyond information produced from non-verbal stimuli, that is, color pictures. Guided by the notions of selective optimization with compensation and loss in the service of growth that focus on the relationships between cognitive operations, the present experiment further examined divergent age-related changes in memory for given and beyond information by exploring their relationships with other cognitive operations. It is possible that an age-related increase in memory for beyond information may be compensating for or trading-off memory for given information, which shows age-related decline, and that different cognitive operations may be related to these divergent age-related changes in memory. The present experiment explored  19  the possibility that an age-related improvement in verbal ability may be related to an age-related increase in memory for beyond information with the idea that verbal ability facilitates the expression of beyond information. The present experiment also explored the possibility that an age-related decline in given information may be related to an age-related decline in perceptual ability with the idea that declining ability to perceive given information may restrict further processing. Unfortunately, the present experiment does not allow for conclusions to be drawn about causal relationships between cognitive operations because it relies on correlational analyses, as well it is possible that memory for given and beyond information were measured on different types of interval scales. This exploration is still profitable, however, as it places agerelated changes in memory into a broader context and provides some avenues for future research. Guided by the idea that there may be maturational changes in schema-driven remembering, a secondary interest of the present experiment was to explore changes in memory for feminine and masculine information across the adult female lifespan. Gender studies in social psychology define feminine and masculine roles by those behaviors and attitudes that have been empirically demonstrated as characteristic of young women and men (for a review see Turner, 1982). Studies on gender and aging trace these behaviors across the lifespan and these studies reveal agi-related changes in feminine and masculine behaviors and attitudes (Gutmann, 1977, 1987; Labouvie-Vief, 1996; Lowenthal et al., 1975; Sinnott, 1984; Turner, 1982). Following a developmental approach to schema theory, it is possible that schemas may direct attention toward different types of information, and in turn, processing this information may contribute towards shaping feminine and masculine attitudes and behaviors. There is some indirect evidence that young women and men may show preference for different types of information with men more inclined to encode and retrieve information regarding position and spatial location and women more likely to encode and retrieve information regarding emotion, events and people.(Adams, 1991; Feingold, 1988; Kamptner, 1991; Voyer & Voyer, 1995). The  present experiment explored the possibility that maturational changes in schema-drivenremembering may show changes in encoding and retrieval of feminine and masculine information across the adult female lifespan. Age-related changes were traced across the adult lifespan for women but not men because there were few older men. OVERVIEW OF REMAINING CHAPTERS The primary goal of the present experiment was to investigate the nature of age-related changes in memory. Some questions that motivated this experiment include, does maturation influence how much and what we remember about an episode or target stimulus? If so, what kinds of factors might be related to these age-related changes? Chapter 2 presents the method that was employed by the present experiment, including description of the participants and the picture description task that was used to assess memory. The section on the picture description task specifies the types of pictures that were used, the design and procedure, and the scoring procedure for assessing given and beyond information. Chapter 3 addresses age-related changes in memory for given and beyond information and different accounts for age-related changes in memory for given and beyond information, including encoding phase performance and study instruction. This part of the experimeni also addressed the possible confounding influence of wordiness on performance on the picture description task. This chapter presents the hypotheses, preliminary and primary analyses, and the findings for this investigation. Chapter 4 places memory for given and beyond information into a larger context and examines the relationships between age-related changes in verbal and perceptual ability and memory for given and beyond information. This chapter describes the findings for this investigation. Chapter 5 explores age-related changes in memory for feminine and masculine information across the adult female lifespan. This chapter describes the scoring procedure for  assessing feminine and masculine information, and reports the findings for this i n v e s t i g a t i o n . Chapter 6 presents a general discussion of the main findings o f this experiment, i n c l u d i a discussion of the theoretical contributions and limitations o f the results of the present experiment. As well, this chapter provides suggestions for future directions in research on memory and aging. This chapter concludes with the specific and global contributions of this experiment.  22  CHAPTER 2 METHOD The present experiment examined a portion of a data set that was collected by the Kerrisdale Memory Across Life Project. In this larger project, adult participants of all ages were given a battery of memory and memory-related tasks; the method, design and procedure of this larger project are described in Graf et al. (in press). Of primary interest to the present experiment was participants' performance on a picture description task and measures of verbal and perceptual ability. Participants A total of 162 participants ranging from 16 to 83 years of age volunteered for this experiment. They were recruited from the community through advertisements in local newspapers, bulletin boards and radio stations. As shown in Table 2.1, the participants were fairly equally distributed across the adult lifespan, except at the youngest and oldest ages. For all ages, more women than men (70.4% females versus 29.6% males) participated in this experiment. In terms of marital and co-habitation status, many participants were single (29.0%), married (20.3%) or divorced (22.8%) and lived alone (46.9%) or with either their spouse (25.9%) or their children (13.0%). Many participants were either employed (44.4%) or retired (35.2%) and most participants (79%) had pursued post-secondary education regardless of age. Regression analysis revealed no significant relationship between age and level of education. Their employment status and high level of educational achievement likely reflects the affluent community where the experiment was conducted. Participants were in good general health, exercised regularly and reported little health trouble. Their general health was assessed by the question "How would you rate your overall  23  A l OS  ©  QO w  Os  II c  /  s  cn cs o II c Os s© •  o © i* s©  O "a!  •aw  Os 1  <u o ID  Q  ON •  ©  o\ •  ©  u-> oo  so  so  m in  so so  OS  cs II c11  m —i oo cs oo in  oo  f- r— cn  — —<  cs  ,^ o  cn II c  so es  -vi-  -3-  cs  "3-  m  oo cn in  Os  cs II c  cn so  Tt^oosor^cn  oooso  cs  cn cn cn cs cs  O  cs r~~  —• r-  cs >n  CS  cn cs II c  00  ^ —H  ^  CS  —-  ^ ^ Os Os  cs 11 VI II c  — I  m cn m  T3  J3  1/1  3 3  I-I  O  V3  I  0>  Q  #  c o  t3 C  o  3  -  5  oo  T3  s1 U  c» c © "-3  O T3 £  5  c 03  6 < #u U  •^  ! i  £  J3  £  O  24  o  NO  ON O  oo o  in  CN  CN  as  •  o ON  o ©1 N©  CN  ON  CN CN O  O  co o o  m o o  o  ON  CO CO in m co r-~ co NO  II  C  00  ON  NO  o o o  CO.  Al O  m o o o  ON  co CN •i M c  M  NO  CN CN  -H  oo r- r-  CO NO TJ- NO —<  o m co  "S 13 o et u •  Q  ON  4? o  1  •t ON  m © T3 U _C •  c o o  ^  /  ON CS  li c  CN  o co II c  in —  • oo  —  O CN  NO  r-  NO  NO — 00 CO NO T t NO CO ^ —-  -i  O  CN  CO  CN  NO  r-  O  NO  °  OO  CN CN  "-^  »-H  r-  NO  CO CN so  ^ ON  CN  ii II  CO CO  CO  m  CN CN CN  O NO  CN 00 CO  CM  ON  CN I i VI II  co  CO  CN  -H CN  t> o —< CN co m —i CO —' >n  NO  c  bO  w  X)  Q  bO  CO  _ • on <0  O  s  et  P-i T3 T3  ^ ^3  p  (U  T3  >> O  ..  O  "Ei "Ei " O  '-s B 6 8  &3 eeeS 6 w w  B  D D  Pd  2 et  o  >  £  —  O U  O OH  2  O  1) w  (U  bO  2 -§ !3 o j> S  S (50 <»—;g) wbo T3 6ffi  €3  3  C  l-l  T3  O  X>  C/3  t  C0  <D  O  o c  V3  = 3  s  1/5  CO  •2  Q  co  00 c c  Q  CN (D  ON  CO NO CN  00  O  (so S «  NO  ||  T3  &  —  C  ed eS  IS  ON  in in  ON CN  ro a ON  ON  W  tOS  -^00  o  C  5 .  1  CO  3 Pi  U  25  health at the present time?" on a 4-point scale with 1 indicating poor, 2 for fair, 3 for good and 4 for excellent health, and health trouble was assessed by the question "Do you have any health trouble that affects the way you do everyday things?" on a 3-point scale with 1 indicating no trouble at all, 2 for a little trouble and 3 for a great deal of trouble. As shown in Table 2.1, regression analyses revealed no significant relationships between age and self-ratings of general health, and age and self-ratings of health trouble. And finally, when asked the yes/no question "Do you regularly participate in vigorous activity or exercise?", most participants reported engaging in regular exercise (82.0%) Materials Six 8 1/2 x 10 inch color photocopies of photographs depicting different scenes served as target stimuli (see Appendix A for black and white photocopies of the stimuli). As shown in Table 2.2, the pictures were selected so as to appeal to young, middle-age and older adults. Agerelevance of the pictures was manipulated because performance has been shown to be influenced by the age-match between the participant and the stimuli (Backman, 1991; Staudinger, Smith & Baltes, 1992). Picture type was also varied to emphasize social interaction since older adults may perform better in testing environments that resemble a social situation. Hence, in order to control for the possibility of an age-related bias in the testing material, the stimuli were varied to maximize task performance for all age groups (Schaie, 1988). Two similar pictures served as practice stimuli: clothes hanging on a clothesline in the backyard of a house, and a boy preparing something to eat in a kitchen. Design and Procedure As described in Graf et al. (in press), participants were tested on a battery of memory, speed-of-processing, attention, perceptual and verbal ability and personality measures that took a total of two and a half hours to complete. Of primary interest to the present experiment was  Table 2.2: Target pictures. Picture Type Age-Relevance  Social  Asocial  Young adult  canoe race with a picnic in the background  a girl looking at herself in a mirror  Middle-age adult  a couple having a conversation in a living room  a woman in a library  Older adult  two older women rolling a skein of wool  a town with a tractor traveling down a main street  27  participants' performance on a picture description task and measures of perceptual and verbal ability. Participants completed the measure of verbal ability within the first 15 minutes of the testing session, immediately followed by the picture description task. Participants completed the measures of perceptual ability approximately an hour and a half into the testing session. Participants were tested on the picture description task in three phases: an encoding phase, a retention interval, and a retrieval phase. At the encoding phase, participants were asked to orally describe each of the scenic color photographs with one of two study instructions: the friend or the camera instructions. The friend instruction asked participants to describe each picture as though they were describing the picture to a friend and the camera instruction asked participants to describe the pictures as though they were a camera viewing the picture. The friend instruction was designed to encourage participants to provide personal experiences and world knowledge in order to make their picture descriptions more interesting and meaningful to someone else, whereas the camera instruction was designed to focus their attention on accuracy and on details. The study instructions were counterbalanced so that half of the participants for each decadel group received the friend instruction and the other half received the camera instruction. Participants were informed that their memory for the pictures would be assessed and they were encouraged to provide as much information as possible; they were given up to 4 minutes to describe each picture. The pictures were randomly presented. A practice picture was given in order to familiarize participants with the encoding procedure. The encoding phase took on average about 15 minutes to complete. Immediately following the encoding phase, participants were asked to complete a battery of tests, including measures of speed of processing, attention, perceptual ability, and personality (as described in Graf et al., in press), in order to prevent rehearsal during the retention interval.  Participants were assigned to either a 15-minute or 1-hour retention interval. Retention interval was counterbalanced so that each delay occurred equally often for each decadel group. At the retrieval phase, participants were reminded that they had previously described a series of pictures with either the friend or the camera instructions and were asked to re-describe from memory as many of the pictures as possible in any order that they chose. As in the encoding phase, participants were given 4 minutes to re-describe each picture. On average, participants took about 15 minutes to complete the retrieval phase. Scoring Procedure Participants' picture descriptions at the encoding and at the retrieval phases were transcribed, coded and analyzed for the number of references to items on the given and beyond scales as described by the Scoring Key in Appendix B. Briefly, the given scale included a total of nine items that count references to things (i.e., people and objects) and to attributes of things (i.e., color, shape, size, quantifiers, passive states, position and spatial location) that are depicted in a picture. For example, in the picture of the two older women rolling a skein of wool, the first reference to the wool would be given a score of one and counted by the object item and a further specification that the wool was yellow would be given a score of one and counted by the color item. The beyond scale included a total of five items that count references to things that go beyond what is depicted in a picture per se, such as events (i.e., life events described from a first or third person point of view) and inferences (i.e., absence, emotion, global setting and time frame). For example, one participant recalled having rolled yarn with his mother; this reference to his boyhood would be counted by the event item from the first person point of view. Four independent raters were asked to classify 17 items as belonging on either the given or beyond scales. A l l of the raters agreed that nine items belong on the given scale and five items belong on the beyond scale, however the raters disagreed on which scale three items belong.  29  Three items -- action, person's age and object's age — were considered as somewhere in between the two ends of the given-beyond continuum; they require a higher level of inference similar to items on the beyond scale yet also refer to common information similar to items on the given scale. These items were retained for further analyses in order to examine whether they behaved more similarly to items on the given or beyond scales, but were not considered within the main discussion of age-related changes in memory for given and beyond information. Three separate counts of the number of references to items on the given and beyond scales were made to assess encoding phase, retrieval phase and encoding-retrieval match performance. To measure encoding and retrieval phase performance, separate counts were made of participants' picture descriptions provided at the encoding and at the retrieval phases, called encoding and retrieval scores, respectively. To measure encoding-retrieval match performance, a third count was made for references that were provided both at the encoding and retrieval phases, and these are called match scores. Reliability. Reliability estimates for all items on the given and beyond scales were obtained by a new coder re-counting a subset of ten protocols. As shown in Table 2.3, correlation coefficients between the two sets of encoding, retrieval and match scores ranged from .70 to 1.00, and thus indicating moderately high reliability for most items on the given and beyond scales. Most of the correlation coefficients were significant at the .01 level, except for the correlation coefficients for some scores on the event and setting items which reached only the .05 level, likely because of the small number of references to these items (see Appendix C).  Table 2.3: Reliability estimates for encoding, retrieval and match scores on each item on the given and beyond scales (n=10). Encoding Given color object passive state people position quantity shape size spatial location Given/Beyond action object's age person's age Beyond absence emotion event setting time frame * p<.05  Retrieval  Match  99 *** 99 * * *  99 *** 99 ***  9g ***  gg ***  g-j ###  95 ***  95 ***  95 *** 97 ***  gg  ***  gg  ***  9g *** 9g ***  97 * * * 94 #** 99 ***  99 * * * 97 99 ***  94 ***  9g *** gg ##*  9g *** 9g *** g9 ***  90 ***  93 ***  92 ***  .84 ** .84 ** .72 * 1.00 ***  94 *#*  95 * * *  .73 * .73 * 94 ***  ** p<.01  *** p<.001  99 ***  .82 ** 97 * * * 93 * * * 95 ***  9g *** gg *** 99 * * *  gg ***  .70* .80 ** i:oo * * *  31  CHAPTER 3 M E M O R Y FOR GIVEN AND BEYOND INFORMATION The primary purpose of this part of the experiment was to investigate age-related changes in memory for references to given and beyond information produced from color photographs across the full adult lifespan. Given information is objective references to things and attributes of things that are depicted in a picture, such as objects and their color and shape. Beyond information is subjective interpretations and evaluations of things that are not depicted in a picture per se but are plausibly related to. something that is depicted in a picture, such as personal experiences and world knowledge that are associated to things that are depicted in a picture. The gerontological and psychological research demonstrates an age-related decline in verbatim memory for phrases in a target story and an age-related increase in memory for inferences drawn from target stories (Adams, 1991; Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Gould, Trevithick & Dixon, 1991). The present experiment was designed to replicate and extend the findings from previous research that employed verbal stimuli by investigating age-related changes in memory for given and beyond information produced from non-verbal stimuli. Following from research employing target stories, it was expected that there would be an agerelated decline in memory for given information and an age-related increase in memory for beyond information about target pictures. The basic procedure was to ask participants to describe a series of color pictures depicting either a social or asocial situation with the instructions to describe the picture either "as though you are describing the picture to a friend" or "as though you are a camera viewing the picture," and then following either a 15-minute or a 1-hour delay, to re-describe the pictures from memory. Participants' encoding and retrieval phase performance was measured by counting the number of references to items on the given and beyond scales that were provided in their picture  32  descriptions at the encoding and at the retrieval phases, called encoding and retrieval scores, respectively. As well, participants' encoding-retrieval match performance was measured by counting the number of references that were provided both at the encoding and retrieval phases, called match scores. The present experiment also investigated two accounts for age-related changes in retrieval phase performance: the influence of encoding phase performance on retrieval phase performance and the influence of study instructions on performance. As well, the present experiment investigated the possible confounding influence of wordiness, or the greater use of words to convey a given amount of information. The following hypotheses were designed to examine agerelated changes in memory for given and beyond information and the factors that may account for age-related changes in memory for given and beyond information. Memory Previous research has shown an age-related decline in verbatim memory for phrases in a target story and an age-related increase in memory for inferences drawn from a target story (Adams, 1991; Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Gould, Trevithick & Dixon, 1991). The present experiment was designed to replicate and extend these findings by investigating age-related changes in retrieving given and beyond information derived from nonverbal stimuli. Hypotheses Hi and H 2 were designed to examine age-related changes in retrieval phase performance on items on the given and beyond scales (see Table 3.1). Relationship Between Encoding and Retrieval Phase Performance Following from the principle of reinstatement of stimulating conditions and the notions of transfer appropriate processing and encoding specificity (Hollingworth, 1928; Morris, Bransford & Franks, 1977; Tulving & Thomson, 1973), this part of the experiment was designed to examine the influence of encoding phase performance on retrieval phase performance. The  Table 3.1: Hypotheses for Chapter 3. Hypothesis Memory Hi Consistent with the previously reported age-related decline in verbatim memory (Adams, 1991; Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Gould, Trevithick & Dixon, 1991), it was expected that retrieval phase performance on items on the given scale would decline with age. H Consistent with the previously reported age-related increase in memory for inferences that draw on general world knowledge.and personal experiences (Adams, 1991; Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Gould, Trevithick & Dixon, 1990), it was expected that retrieval phase performance on items on the beyond scale would increase with age. 2  Relationship Between Encoding and Retrieval Phase Performance H Consistent with the notions of encoding specificity and transfer appropriate processing (Morris, Bransford & Franks, 1977; Tulving & Thomson, 1973), it was expected that the age-related changes in retrieval phase performance that were predicted by hypotheses Hi and H would also be shown for encoding phase performance. H4 Following from the notions of encoding specificity and transfer appropriate processing (Morris, Bransford & Franks, 1977; Tulving & Thomson, 1973), it was expected that encoding phase performance would account for retrieval phase performance. H5 If there are age-related changes in the degree to which encoding and retrieval processes overlap or do not overlap, then it was expected that there would be age-related changes in matches, losses and gains. 3  2  Wordiness H Following studies indicating that older adults provide a greater number of words to convey a given amount of information (Botwinick & Storandt, 1974; Obler, 1980), it was expected that there would be an age-related increase in verbatim repetitions. H If performance is confounded by wordiness, then it was expected that an age-related increase in verbatim repetitions would account for age-related changes in encoding and retrieval phase performance. 6  7  Study Instructions H If young adults can be encouraged to take the perspective of older adults, then it was expected that they would benefit from the friend instruction as revealed by enhanced performance on items on the beyond scale. H If older adults can be encouraged to take the perspective of young adults, then it was expected that they would benefit from the camera instruction as revealed by enhanced performance on items on the given scale. 8  9  34  principle of reinstatement of stimulating conditions and the notions of transfer appropriate processing and encoding specificity state that memory depends on the overlap between encoding and retrieval processes. To illustrate, in their first experiment, Morris and his colleagues asked participants to study a series of words with either a semantic or a rhyme orienting task that required participants to decide if a word was meaningful within a sentence or if a word rhymed with the last word in a sentence, respectively. In the semantic orienting task, for example, participants read the sentence "The  had a silver engine" and then heard either the word  "train" or ""eagle" representing the semantic-yes and the semantic-no conditions, respectively. In the rhyme orienting task, for example, participants read the sentence "  rhymes with  legal" and then heard the word "eagle" or "peach" representing the rhyme-yes and rhyme-no conditions, respectively. Participants' memory was tested by either a standard or a rhyme recognition test that required participants to decide if the test word was one they had studied or if the test word rhymed with one they had studied, respectively. The results indicated superior performance on the standard recognition task for words that were studied with the semantic versus rhyme orienting task and superior performance on the rhyme recognition task for words that were studied with the rhyme versus semantic orienting task, and thereby demonstrating a relationship between the processes that were activated at study and at test. Hence, following from the principle of reinstatement of stimulating conditions and the notions of transfer appropriate processing and encoding specificity, it was expected that the age-related changes in retrieval phase performance that were observed in the previous section would extend to encoding phase performance. Hypothesis H addresses age-related changes in encoding phase performance 3  (see Table 3.1).  '  A quantitative and a qualitative approach were employed to further investigate the relationship or overlap between encoding and retrieval phase performance. The quantitative  35  approach employed regression analyses to partial out the influence of encoding phase performance on retrieval phase performance and the qualitative approach examined age-related changes in the references that were the same at the encoding and retrieval phases, called matches. As well, the qualitative approach included an investigation of the degree to which encoding and retrieval processes did not overlap by examining age-related changes in losses and gains, or the references that were different between the encoding and retrieval phases. Losses are references that were provided at the encoding phase but were not carried over to the retrieval phase and gains are references that were provided at the retrieval phase but were not carried over from the encoding phase. Hypotheses H4 and H5 were designed to examine the overlap between encoding and retrieval phase performance (see Table 3.1). Wordiness This part of the experiment was designed to examine the possibility that age-related changes in encoding and retrieval phase performance may be due to a confounding influence of wordiness. Research on age-related changes in communicative styles has provided some support for the common impression that older adults are more verbose than younger adults (Gold, Andres, Arbuckle & Schwartzman, 1988; Gold, Andres, Arbuckle & Zieren 1993). Verbosity has been defined as copious and prolonged talking and can be further considered as either on- or off-target. On-target verbosity, or talkativeness, is talk that is "focused on a narrative structure or logical sequence of topics" (p. 68-69, Gold, et al., 1993), whereas off-target verbosity "is talk that, although prompted by a specific conversational topic, becomes a series of loosely associated verbalizations increasingly distant from the original topic" (p. 68, Gold et al., 1993). On-target verbosity was the primary interest of the present experiment and was measured by the items on the given and beyond scales; age-related changes in performance on these items were examined in the Memory section.  36  Off-target verbosity was examined in a subset of the protocols (n=53) in order to get an impression of the frequency of off-target occurrences. An off-target reference was defined as a reference that cannot be logically tied to something depicted in a picture. No references to information that would be considered off-target were found in a third of the protocols that were equally distributed across the decadel groups. Related to verbosity is wordiness which has been defined as the greater use of words to convey a given amount of information. For example, Botwinick and Storandt (1974) found that older adults provided more multiword definitions on a vocabulary test than younger adults. Although some studies have shown that older adults are more wordy than younger adults (Botwinick & Storandt, 1974; Obler, 1980), there are a number of other studies that have failed to demonstrate age-related differences in wordiness on a picture description task (Cooper, 1990), in a life-history interview (Gold, et al., 1993), and on written and speech tasks (Kemper, Kynette, Rash, Sprott & O'Brein, 1989). Previous research has measured wordiness as the mean number of words used per unit of imparted information or the total number of words produced (e.g., Botwinick & Storandt, 1974; Cooper, 1990; Obler, 1980). These measures would not serve as satisfactory measures of verbosity in the present experiment, however, as they do not consider the amount of imparted information. The total number of words produced at the encoding and retrieval phases would not be an adequate measure of wordiness in the present experiment as the total number of words is not independent of the number of references to the items on the given and beyond scales; this measure of wordiness cannot be separated from on-target verbosity. The mean number of words per information unit would also not serve as a satisfactory measure of wordiness as there is no reason to assume that the mean number of words per information unit would be the same for  references to items on the given and beyond scales; it is likely that references to items on the beyond scale would require more words than references to items on the given scale. One method for investigating wordiness is to examine the number of verbatim repetitions to on-target information. Verbatim repetitions are repeated references to words or phrases that are exact or near literal reproductions of references to items on the given and beyond scales. Hypotheses H6 and H were designed to examine age-related changes in verbatim repetitions (see 7  Table 3.1). Study Instructions This part of the experiment examined age-related changes in the influence of study instructions in guiding performance on items on the given and beyond scales. Previous research has shown that study instructions can direct the types of information that are encoded and retrieved about target stimuli. For example, the classic Anderson and Pichert experiments revealed that participants recalled different information about a house depending on whether they took a homebuyer or burglar perspective at the encoding phase (Anderson & Pichert, 1978; Pichert & Anderson, 1977). Because research has shown that both young and older adults benefit from study instructions (Arbuckle, Vanderleck, Harsany & Lapidus, 1990; Hess, Donley & Vandermaas, 1989), the present experiment employed study instructions that would encourage young and older adults to take different perspectives. As described in the Method section, the camera instruction was designed to enhance performance on items on the given scale by focusing participants' attention towards the details in a picture. The friend instruction was designed to enhance performance on the beyond scale by encouraging participants to go beyond the picture per se and provide personal experiences and world knowledge that would make their picture descriptions more interesting and accessible to a friend. Because young adults tend to encode and retrieve more given and less beyond information, they may not benefit from the camera  instruction, but they may benefit from the friend instruction. By contrast, because older adults tend to encode and retrieve more beyond and less given information, they may not benefit from the friend instruction, but they may benefit from the camera instruction. In this way, the friend instruction may serve to encourage young adults to take the perspective of older adults whereas the camera instruction may serve to encourage older adults to take the perspective of young adults. Hypotheses Hg and H 9 were designed to address age-related changes in the influence of study instructions on performance on items on the given and beyond scales (see Table 3.1). RESULTS Before investigating age-related changes in memory for given and beyond information, some preliminary analyses were performed on the obtained and aggregated items. Obtained items are simple counts of the number of references that satisfy the specifications described in the Scoring Key in Appendix B. Aggregated items are the sum of the number of references to two or more obtained items. Obtained and aggregated items comprise the given and beyond scales which are also specified in the Scoring Key. First, aggregated items were sought because they facilitate interpreting patterns of age-related changes rather than inspecting age-related changes for each individual obtained item. In order to justify aggregating obtained items on the given and beyond scales, the obtained items were tested for floor effects and then the intercorrelations between obtained items on the given scale and between obtained items on the beyond scale were examined. Second, performance on the obtained and aggregated items were screened for univariate and multivariate outliers, respectively. The primary analyses employed regression analyses in order to examine age-related changes in memory, a possible confounding influence of wordiness on age-related changes in memory, and different accounts for age-related changes in memory, including encoding phase performance, and study instructions. Regression analyses were performed on standardized scores  39  in order to control for the possibility that the variance associated with each variable may contribute differentially to the standardized regression coefficient. For all regression analyses, homogeneity of variance was verified by inspecting the scatterplots of the standardized residuals plotted against the standardized predicted F s; the points were evenly scattered around the mean of the residuals indicating homoscedasticity. For regression analyses that included the interaction term between two variables, computed as the cross product of the two standardized scores, all lower-order terms of the interaction equation were entered into the equation followed by the interaction term between the two variables; this procedure is necessary even in cases where the lower-order terms of the interaction equation are not significant in order for the significance test of the interaction equation to be appropriate (Cohen, 1978). Regression analyses are reported by the standardized regression coefficient (J3), squared multiple correlation (R ) and F value indicating an alpha at .05, .01 or .001 level of significance 2  for a two-tailed test. Because there are multiple comparisons, all conclusions were conservatively drawn by examining effect sizes, computed as the difference between the R's for the young and older adults divided by the standard deviation, rather than by the significance level of the F values. Preliminary Analyses Obtained and Aggregated Items Participants' encoding phase, retrieval phase and encoding-retrieval match performance were assessed as the number of references to each obtained item that was provided at the encoding, at the retrieval and both at the encoding and retrieval phases, respectively. A subset of ten picture descriptions were tested for floor effects and the results revealed that mean performance on 15 of the 19 obtained items were significantly off the floor. The obtained shape item showed a near floor effect for encoding-retrieval match performance  (z btained=l-56, 0  40 Zcriticai=l-64,  j)=.05), but not for encoding (z b  (z btained=2.43) 0  0  ta  ined=l-97)  or retrieval phase performance  and so this item was retained for further analyses. The obtained second person  point of view item showed clear floor effects for encoding phase, retrieval phase and encodingretrieval match performance (all fneans=0). Consequently this item was dropped from further analyses. The obtained third person point of view item showed a near floor effect for encodingretrieval match performance  (z tained=l-63, ob  z j ; i=1.64, £=.05) and the obtained first person cr  t  ca  point of view item showed clear floor effects for retrieval phase and encoding-retrieval match performance (all means=0). Since performance did not always show floor effects on these two obtained items, these items were collapsed into a more general item, called event. The aggregated event item included references from a first or a third person point of view to events not depicted in a picture. Performance on the 16 obtained items on the given and beyond scales and the one aggregated item on the beyond scale for the full set of protocols (n=162) were tested for floor effects. The results indicate that mean performance on all obtained and aggregated items were significantly off the floor (smallest z=8.57, p_<.001, see Appendix C). Intercorrelations between the obtained items on the given scale and between the obtained and aggregated items on the beyond scale were examined in order to justify aggregating items within the given and beyond scales. As shown in Table 3.2, performance on all obtained items on the given scale showed strong, positive intercorrelations, and hence allowed obtained items to be aggregated into a general given item; most intercorrelations were significant at the .001 level, except for three intercorrelations (smallest intercorrelation, r =.14, p_=.057). Unlike the obtained 2  items on the given scale, items on the beyond scale were not strongly intercorrelated and so creating an aggregated beyond item was not warranted. As shown in Table 3.3, almost all of the intercorrelations for retrieval phase performance were significant at the .01 level, however only about half of the intercorrelations for encoding phase and encoding-retrieval match performance  t  4] Table 3.2: Intercorrelations between encoding phase, retrieval phase and encoding-retrieval match performance on each item on the given scale (n= 162).  color Encoding object passive state people position quantity shape size spatial location Retrieval object passive state people position quantity shape size spatial location Match object passive state people position quantity shape size spatial s location * p.<.05  object  passive state  Item people position  quantity  shape  size  .86 *** 77 59 45 7g  *** *** *#* ***  g3 ***  .65 *** 52 ***  75 *** 37 * * *  .24 **  g3 *** 55 * * *  70 *** 52 ***  57 *** 3g * * *  gO * * *  .68 *** 91 ***  53 *** 75 ***  40 *** 50 ***  g7 * * * 71 * * *  77 ***  .63 *** .58 ***  52 *** 45 ***  .63 *** 53 ***  70 *** .58 ***  49 * * *  30 *** 32 *** 54 ***  go ***  51 **# 55 ***  .64  .60 *** 51 ***  .75  4g * * * 53 * * *  .62 ***  71 ***  79 * * *  .42 *** .62 ***  45 * * *  43 ***  .60 ***  44 ***  32 #**  31 ***  79 ***  50 ***  44 ***  39 **#  91 ***  74 ***  .65 ***  52 ***  .56 ***  go *** 7j * * *  77 * * *  39 * * *  40 ***  .65 ***  75 ***  75 *** 79 ***  .65 *** .62 ***  31 *** .62 *** 7j **# 7g *#*  g3 * * * 75 * * *  4g * * * 5] *** 73 * * *  *** p_<.001  52 *** .68 ***  32 *** .63 *** 40 *** 53 ***  gg ***  73 **#  .25 ** 52 *** 2g ***  40 *** 70 ***  35 *** .14 * 24 *** 44 ***  49 * * *  .55 *** 74 ##*  Table 3.3: Intercorrelations between encoding phase, retrieval phase and encoding-retrieval match performance on each item on the beyond scale (n=162).  absence Encoding emotion event setting time frame  .02 .00 .16 * .22 **  Retrieval emotion event setting time frame  .17 * .12 .36 *** 2g ***  Match emotion event setting time frame * 2<.05  .07 .07  Item emotion event  43 ***  .16 * .14  40 *** .18 *  44 **#  42 *** .26 ***  ^2 ***  .38 ** .14  44 ^^fc*  45 **#  28 *** .23 **  31 * * *  43 * * * 34 * * *  .17 *  .08  * * p_<.01  setting  * * * p_<.001  43  reached this criterion. Hence, the following analyses included performance on an aggregated given item but not an aggregated beyond item. Testing obtained items on the given and beyond scales for floor effects and examining intercorrelations between items on the given scale and between items on the beyond scale yielded a total of 16 obtained items and 2 aggregated items for further analyses. More specifically, there were 9 obtained and 1 aggregated item on the given scale, 4 obtained and 1 aggregated item on the beyond scale and 3 obtained items that were not classified on either the given or beyond scales. Univariate and Multivariate Outliers The critical dependent measures were the number of references to each obtained and aggregated item that were provided at the encoding phase, at the retrieval phase and both at the encoding and retrieval phases, called encoding, retrieval and match scores, respectively. The data were screened for univariate outliers by separately regressing age on encoding, retrieval and match scores for each obtained item. Outliers were identified at the 99% confidence interval and high and low outliers were replaced with their respective upper or lower value at the 99% confidence interval. This procedure detected 183 (2.21 %) outliers that were equally distributed across the decadel groups: in the younger than 30 group there were 12 (1.24%), for the 30 year olds there were 28 (2.38%), for the 40 year olds there were 49 (3.31%), for the 50 year olds there were 37 (2.42%), for the 60 year olds there were 25 (1.69%), for the 70 year olds there were 26 (2.22%) and for the over 80 group there were 6 (1.30%) outliers. The data for the aggregated given item were screened for multivariate outliers using a Mahalanobis distance procedure. Mahalanobis distance is considered a sensitive and accurate procedure for identifying multivariate outliers; Mahalanobis distance has been shown to be more sensitive than Cook's distance (Stevens, 1984) and more accurate than Comrey's Dk  44  (Rasmussen, 1988). This procedure detected 11 encoding scores, 5 retrieval scores, and 6 match scores as outliers that were removed (critical £ (9) = 27.877, p_=.001, as recommended by 2  Tabachnick & Fidell, 1989). Primary Analyses Memory As described by hypotheses H i and H 2 (see Table 3.1), the primary focus of this part of the experiment was to investigate age-related changes in memory for given and beyond information about target pictures by examining retrieval phase performance. Before investigating age-related changes in retrieval phase performance, the influence of the 15-minute and 1-hour delay period was examined in order to warrant collapsing the data across delay period. Retrieval scores on the obtained and aggregated items were regressed separately on delay period, age and then the delay period by age interaction term; a relationship between performance and delay period or the delay period by age interaction term would indicate that the data cannot be collapsed across delay period and would therefore require separate analyses for each delay period. The results revealed that delay period and the delay period by age interaction term were not related to performance on any items, and thereby allowing the data to be collapsed across delay period. In order to examine age-related changes in retrieval phase performance, retrieval scores on the obtained and aggregated items were regressed separately on age and then age squared. As shown in Tables 3.4a and 3.4b, the aggregated given item showed a significant decline in performance. Age-squared was not significantly related to performance on the aggregated given item which indicates no non-linear relationship between age and performance on the aggregated given item. By contrast, one item on the beyond scale — event — showed an age-related increase in performance and another item — emotion — showed the same pattern of results but did not  Tf  ON  m  co a  CN  o p  in CO  NO in  m in  o p  co  OO p  NO  in o o p  NO CN O O  Tf  —*  4?'  3 m  oo o O O  < o NO  # Tf  tf  o CO CN  co.  o m CO  ©  u  o  Al  oV  © 00  Jlc  .  (N  N©  ON  '  II  © NO  C  w  ' cv ©  Q  II c  ON  T' ^II  ©  ©  ON  CN VI  c  c  00 T  CO.  ON  in T f — o CN —' O  O  NO  00  ON NOw  CN  CN CO o p  *  * * *ON  * * *CN  * * *CO  ©  *  00  CN CN CN NO © O CO CN CN ~ CO f» CN co CN  CO ON  5  Tf m  Tf  oo ^  ON  II  m m  m o  Tf  **;  C  ON  t  in CN CO CO  oo CO l" rNO  r~; 00 CN  ON  *—<  P  O  o CO o p  NO  00 —< r-- ON t-I co' r - ~*  co  « w  tf  OO T f t-~; ON 00 CN m  in  o CN O p  NO CN r-^ CN  CN OJDL  o o o  o o o p  ON NO O O  ON  O p  * *r -  in co *  r-  CN CN  00 CO  CN CN m  O 00 c-p  NO co CN O  o o ON CO l'  CO  rCO m  ON  rCN  rON CN  CN  — <  _  CN 00  t-H  p  CN CN  ON  Tf  >n od  m  NO O •n p  O  ON  ON  O  CN O CO CO r  m CN CN i'  CO  CO l'  i'  * *  * *T f  l'  t~- ON 00 T f r - co NO, o oo CN NO T f ON i n ON — i NO NO m NO >n T f oo CN CO i n CN 00 —^ Tf' CN d in NO — —• in w  ,  , , CO CN  ON  CN  O co II c  ON CN  s  t~- o  ON  d CN  ON  ON  O  co —  CN  o ON oq o 00 co i n Tf  00 i n ON CO CO ON NO m NO CN CO ON CO d ON  T£  C  f—1  r~ Tf Tf  T f " oo NO T> T f ON  m  ci  NO f t | T f T f T f CN CO 00 ON o ON CN CO m T f ON O CN NO i n ON ON_ CN CN OO ^— j  O o m o  ~  W  ^  N O O N ' - H N O O O T f T f T f i n O O C O C O — ' C N l n O T f O O r ^ ^ T f c o c N T f p c N c N — incNcoNqpcooNcxj c S c > i d c o ^ » n ^ c o i n T f c o c o c o ' c o c N O N d o N C N C N O i n ^ — ' C N — "-"m N  CO  00  CO CN  l > 00 00 Tf  NO  NO ON  CO  CN  in d  W  o  O  ON  NO CO  d  CN  Tf  W  CO  Tf  NO  ON  m  o p  CN  Tf^ CO d CN  o 00 00 CN  CN  NO  m o  d  S a CO  %Q CO  Q  CO  Q  Q  CO  Q  CO  Tf  d  NO  S CU  o o  X?  o  I S §• & v> tu CU Cu  173  o  in CN  Q  CO  CO  c o CU  CN  CN  Q  0  CO  CO  ON ON  m  ^  CO  00 r CN T f CO CN T f CN  ^  ~ ON ON 00w  ON  r  CO CO O ON Tf 00 oo i> NO NO CO CN  ON CN C-» ON NO o o ^ i n O N i n c o o N c o o NO 00 oo r~- — T f c N c N N q r ^ i n o o o o r - ; T f ^ C O O N T f O N C N r ^ i n d d c N c N O N N O i n o d ON CO CN T f I-H  ON NO  w  co CN II  ^  ^  /  00 ON i n 00 ON CN CO CN ON T f T f NO Tf' >n NO i n i—< r- T ^ ^H  CO T f o 00 m CN m m 00 NO CN NO in CN m NO o NO OO T f i n ON NO t— ON T f o CO T f , NO co d co ON NO i n CN CN CN i n CO CN T f CN  CN  o Tf o CN vO CO CN w  — Tf  C C  CT  O-i 1  c/3  (U N  2 u OH  O  46  ON in a  o o  o ©  o o o o  o ©  so m cs  oo in cn  cn cs  00 m  00  cn  ©  cs q  T—f  ©  ©  m cs  Os  TT  •<*  oo rcn  DC <  o SO  *  SO SO  <U wn  <  q rcn CO.  Al O  2o ©  SO  T3  Os V>  O  CS  »—<  cs  ©  •—;  ©  Os  o o  ©  © O ©  _  OS  *cn cs  Os 00  OS  r-OS cs  •sees en in  SO cn cs  oo  m  cn  ©  ©  q  cn Os t-~  o cs  SO o  ©  en  ©  TT  |  cs  ©  Tt  OS  cn  © © O  ©  cs cn  H  o so  m Os Co oo in" Os cn 00 oo i n oo so 00 © 00 q cs i n i n cs cs i n  00 cn m cs 00 © cn cn cs i n en cs cs  © cs  cs cs cs  CS  cn cn © SO so cs so SO cs en C S cs  TICS  o cn  o © m SO cs  oV M  © © in  cs  OS  Os so I  cn cs  SO cs  Os  u o  © m cs  cs  CS CS  r~ cs cs  Tt cs  r - oo m cs" in 00 OS r » cs cs  ?  cs" SO Os" cn Os T ? oo cs cs © Tt © Tt en cs cs i n cn ©  cn m 00 00 SO cn o Os Os o cn SO Tt cs cn C S SO cs  •TI-  Tt cs" m 00 m OS cs cs  SO Tt i n  cn (n cs SO  m cs  ^ .  OS CS  ©  ©  13 c 03  C O 'oo 00  Q v> OS  oo 4) oo  II c  c  Tf cn Os m SO Co" SO OO oo m SO oo SO © cs cs i n Tt, r-^ cn  SO T ? SO cs oo oo oo OS SO SO cs oo so cs C S cs cs  2  cn  <D  -o m SO oo os SO T ? in q © so cs cn cs r~ cn  SO © so OS 00 cs cs cn cs  cn 00 — cs OS o in TI- 00 cs en C S cs  —  oo OS Co SO cn OS OS cs cs cn  o o  V 04  * o> e  Q 00  2 1/3  S 0>  Q 00  S  cn cs  Q 00  o  00  objec age perso  VI  Givei actioi  os  e o  ^  S  SO o o cs © Tt Tt en  o  o  o  o  so cs so  o  o  cn cs OS i n q , —  '  (U <*5  Q oo  o  ^  o V  oo tr  C3 *  c o  a  <u  S a  u  c  |  ,2  So <  V C4  47  reach significance. Age squared was significantly related to performance on two items on the beyond scale — event and setting — which indicates a non-linear relationship between age and performance on these items. No other items on the beyond scale showed any age-related changes in performance. Hence, as predicted by hypothesis H i , the results showed an age-related decline in performance on the aggregated given item, however, support was not found for hypothesis H  2  as an age-related increase in performance was found for only one item —event — on the beyond scale. Together, the results indicate that there were divergent age-related changes in the types of information that were retrieved about target pictures as there was an age-related decline in performance on items on the given scale and preserved performance on items on the beyond scale. Relationship Between Encoding and Retrieval Phase Performance Following from the principle of reinstatement of stimulating conditions and the notions of transfer appropriate processing and encoding specificity that state that memory performance depends on the overlap between encoding and retrieval processes (Hollingworth, 1928; Morris, Bransford & Franks, 1977; Tulving & Thomson, 1973), this section investigated the relationship between encoding and retrieval phase performance. As addressed by hypothesis H , it was 3  expected that the age-related changes in retrieval phase performance that were observed in the previous section would extend to encoding phase performance. As addressed by hypotheses H4 and H (see Table 3.1), the observed age-related changes in performance were further examined 5  by employing a quantitative and a qualitative approach to investigate the degree of overlap between encoding and retrieval phase performance (see Figure 3.1). The quantitative approach employed regression analyses to partial out the influence of encoding phase performance on retrieval phase performance and the qualitative approach examined age-related changes in the references that were the same at. the encoding and retrieval phases, called matches. The  48  Figure 3.1: Relationship between encoding and retrieval phase performance.  Quantitative Approach  Regress retrieval phase performance on encoding phase performance Encoding Phase Performance  Retrieval Phase Performance  retrieval  Losses: proportion of references provided at the encoding phase but was not carried over to the retrieval phase: encoding - match encoding  Gains: proportion of references provided at the retrieval phase but was not carried over from the encoding phase: retrieval - match retrieval  49  qualitative approach further investigated the degree to which encoding and retrieval processes did not overlap by examining age-related changes in the references that were different between the encoding and retrieval phases, called losses and gains. Losses are references that were provided at the encoding phase but were not carried over to the retrieval phase and gains are references that were provided at the retrieval phase but were not carried over from the encoding phase. Encoding Phase Performance Age-related changes in encoding phase performance was examined by separately regressing encoding scores for each obtained and aggregated item on age and then age squared. As shown in Tables 3.5a and 3.5b, the aggregated given item showed an age-related decline in performance. Age squared was not significantly related to performance on the aggregated given item and thereby indicating no non-linear relationship between age and performance on this item. By contrast, none of the items on the beyond scale showed an age-related change in performance; age and age squared was not significantly related to performance on any items on the beyond scale. Together, these results indicate divergent age-related changes in encoding phase performance, where there was an age-related decline in performance on the aggregated given item and no age-related change in performance on items on the beyond scale. These results support hypothesis H (see Table 3.1) as the same pattern of age-related 3  changes in retrieval phase performance were generally found for encoding phase performance, where there was an age-related decline in performance on the aggregated given item and no agerelated change in performance on items on the beyond scale. Only one item did not show the same pattern of age-related changes in performance; the event item showed an age-related increase in retrieval phase performance and no age-related change in encoding phase performance. Hence, consistent with the principle of reinstatement of stimulating conditions and the notions of transfer appropriate processing and encoding specificity (Hollingworth, 1928;  Os in  cn  CN  CN P4  o o Os Tt  o so en oo  61  CO.  Al O  I  O  o Q  CN Tt  Tt m o q  OS Tt o q  o o o q  m o o q  in CN o q  * *  —^  * * *— .  *  en in  m CN  SO  * *CN  SO  en SO  in  —*  Tt-'  cn so oo q  o q  Os 00 CN q  en Tt Tt q  in o o  OS Os SO  SO o  CN CN o o  o o o  o o o  * * Tt  * *00  * *00  Tt  00  en O q #  oo CN m q  en 00  so  00  en  q  q  Os q  00 q  q  OS OS CN CN  Os as  o en O en i'  in in OS CN  en •<t CN  —' —«  CO. CN l'  CN i'  m  00 en Os CN  *—'  Tf  CN  —-c  —< en  CN  en CN II C  cn r - Tt r> i n «— 00 o TJin d d Os SO CN en SO  os cs II c  en Os d CN  o en  O in  OS i n r-SOC-so—<00-^--^CN—'SOCNSD en oo SO SO i n s q i n i n t ^ T t — . e n — - - - c N T t ; — o CN d so en c x J ^ T t e ^ e n o ^ c N c ^ o o i n ^ s o c N —' CN CN oo x  ^ ^  CN  <-<  II c  Tt r-—- 00 SO Tr' CN CN  O  as en  in so  ° ' o3 CN oo Tt ^ en CN  2o  >n — so CN -  "—'  ' ©  II c  os  r-  w  ^ IT vi £  v  f  " '  • n en r Os en OS en <—1 TT' —< i n d in ^—^ CN CN CN  cs so  m i n cn CN CN CN d  w  en — en o — i Tf i> d i n CN w  Os CN  v o r ~ ' - O o o o t ~ - r ~ - T t o o s o o s o c n o s T t O T t « to - ; i n e n s o e j s c N i n e n o o c N O — ' i n o s o s e n c N c N s d ^ r ^ ^ c n T t T t s d s d e n T j - ' — ^ o s d c N enencNso — w ^ w w ^ ^ w —, w ^ ^  CN CN  en O  en CN  c N T f r ^ c N O S ' — ' T f o o o s i n s o ^ o s c N e n c N c N i n i n o s ^ q e n s q r ^ T t r n s o c N c N e n i n r ~ - o i n - - ' s o o o i n r ^ ^ i n ^ e n r ^ i n i n o s c N c N — osensd CN <—i  CN —<  T* OS OS 00 Os CN O en - H  OS c  Tt —  w  _  w -  w j s j w  w  —. ^  en o Os so CN — t ~ - O s — ' e n s o c N c N c N o o i cn en q ej\ - H i n —^ oo so - ' m <t O co T t e n e N c N ^ i n ' e n e n o d s o o o d en CN en so ^ r c N i n m — ^ ^ CN — — w  Q  Q oo  E  ^  i n CN OO en O co oo (N h oo oo -— oo oo h - fv CN i n O O SO c n ^ c n e n ^ c N C N c n T t o s ^ T t od — CN OS Tf d c N s d i n d ' - ' C N c N o s s d i n ^ t l—1 —H i n  H  r-» CN so in  S O r— so so oo — so —. T t O OS i n en CN oo os CN so o m i n m o o — cNtNosoor~-r~-  1  /  00 CN  so  CO  oo q  Tt CN SO TT, OS so CN m  o  ' ©  O q  OS  "3  «  SO oo  O —' q r-; en © oo O  o SO  oo  o  oV M  os  os  en CN  CN  en  e  Q  Q  00  00  Q 00  o  Q  00  00  Q  oo ^ « "i  O  U  n oo Os in in so CN  00  00 C  oo O  (N  Q S Q  00  c o  o o o  Q  w  io  CT  oo  ID N  .2-2 &  oo  O  - i  ON  in ©  Tf  CN  4* , DDI  ro o p  ON  CO  CN O O  p  *  ON  ON  M c  CO CN  o ON NO •  o u o  o  ON  CN  NO o o  ON CO  NO  o  CN  NO  co  *^  o in CN  NO O O  Tf  rco CN O  NO o  m  p  O  O o oo o  Tf IT) »—i  CO p  ON  >—'  ^  NO  —  CN  o CN o o  >n  NO Tf  Tf  O o  *  —«  ON  CN  Tf  CO o  Tf  CN  00 CN CO  co ON CN  NO  ©  ON  © p  CN  Tf  CN  >-  1  ON o  Tf  o  O CN O NO  Tf  Tf  CO  f - CO T f 00 oo m T f oo NO CN CN CO CN  CN—'OOTfCNCNNOCN'-'^  m co o 00 T f CN NO CN CO CN CN ~ H  c o t ~ - o o c o i n t ~ ~ N o i n T f O N c? co i n O N i n T f N O CN^NOTfcOCNr^CN —'  NO 00  r~-r~ooinoN»ncNcocNON — j p i n r ^ O N r ^ i n o o T f  -H NO  — Tf Tf O  Tf  r~  t —  CN T f  00 T f CN T f CN ^ co  CNi  ^  O N o - H O O T f - H o o c o c o r —  pNO^HTfTfinr^cocooo  1  CO  <u oo  NO  "3  ON  es in u CU  ©  Q  o o o  CO p  Tf Tf  Al O oo  CO  00 CN  oo  CO.  CN CO  CN  T—<  tf  CN  ON  ON  o  r*» o  in 00  CN ~  CN  NO O  ON  r~  CN T f  CN CN  o  o o i n c o N O T f t ^ m ^ 00 C O C N O N C N - - ; C O r ~ ; P ON cNCNinTfrofo'odr^rt  c  O m  03  c _o  co co  c ON  <N  <u ON O  00  <—< —i ON CN CO CO NO CO CO CN CNi  r—  ^  100  ONOONOTfONCN—< NO — NO c N p ^ c o — i i n o o r ~ - T f CNCNNOincNCNOOCO — ~  ft  <u  1—  >» X)  ON  ©  Jl  CO T f CN O » - CO 00 00 NO NO ON CN 00 CN T f CO CN —«  O N O W O C N C N O N O O O N — C p >n co p N O p ' — ^ co co T f  COCNTfcOCOTfdcOCN  T3 <D  3 E 8  —  ID ON  ON  cs VI  oo T f m ON T f <n f ; ON CN 00 CN C*^ —> CO CN CN ~  C 1)  oo O N o i n N O O N T f o o i n o o NO O C N T f O N O N [ > 0 0 0 0 CNCNNOTfcNCNOOCNCN-^  Q SQ  N  to  £ CU  * *  £ * co  O  _  u  o V OJ  CD  to  _ o v  CD 00 03  * *  -o § >q <u ° . 00 V <  C4  o  *  52  Morris, Bransford & Franks, 1977; Tulving & Thomson, 1973), the present results indicate that age-related changes in retrieval phase performance were extended to encoding phase performance. Quantitative Approach As addressed by hypothesis H (see Table 3.1) and illustrated by Figure 3.1, the 4  quantitative approach examined a relationship between encoding and retrieval processes by separately regressing retrieval scores for each obtained and aggregated item on encoding scores and then age. As shown in Table 3.6, the results indicate that encoding phase performance was significantly related to retrieval phase performance on all items, with R ranging from .2888 to 2  .6936. As well, when compared to the previous unpartialled results on age-related changes in retrieval phase performance on the obtained and aggregated items, all of the R ' s due to age 2  declined when the influence of encoding phase performance was partialled out (see Tables 3.4a and 3.4b vs. Table 3.6). The results also showed, however, that encoding phase performance did not completely account for retrieval phase performance as age-related changes in retrieval phase performance were revealed when the influence of encoding phase performance was partialled out; an age-related decline in retrieval phase performance on the aggregated given item and an age-related increase in performance on the event item remained when the influence of encoding phase performance was partialled out. Together, these results support hypothesis H that 4  memory performance depends on the overlap between encoding and retrieval processes as encoding phase performance accounted for a significant portion of retrieval phase performance on all of the obtained and aggregated items, although encoding phase performance did not completely account for age-related changes in retrieval phase performance.  53  Table 3.6: Influence of encoding phase performance and age on retrieval phase performance on each item on the given and beyond scales. Study Performance R F ( l , 147) P .8022 .6436 265.48 *** R F ( l , 160) P  3  Age AR .0432 AR .0453 .0265 .0288 .0107 .0041 .0158 .0026 .0088 .0293  a  Item Given  2  2  color object passive state people position quantity shape size spatial location  Page  -.2126  2  2  .7800 .8190 .7525 .6093 .7231 .7551 .8328 .7786 .7720  .6240 .6709 .5664 .3713 .5228 .5701 .6936 .6063 .5961  265.58 *** 326.14 *** 209.00 *** 94,48 *** 175.30 *** 212.22 *** 362.12 *** 243.36 *** 236.13 ***  -.2183 -.1668 -.1783 -.1087 -.0648 -.1317 -.0517 -.0955 -.1753  Given/Beyond action .6585 object's age .7093 person's age .5106  .4337 .5032 .2608  122.53 *** 162.08 *** 56.43 ***  -.0880 .0076 -.0694 . .0048 -.0708 .0049  Beyond absence emotion event setting Time frame  .3542 .5903 .4506 .2888 .3519  87.75 *** 230.58 *** 131.20 *** 64.95 *** 86.89 ***  .5951 .7683 .6712 .5374 .5932  a  F(2, 146) 20.17 *** F(2, 159) 21.82 *** 13.97 *** 11.35 *** 2.77 1.39 6.09 * 1.38 3.67 12.47 ***  2.17 1.54 1.06  -.0585 .0034 .83 .0430 .0018 .70 .1521 .0231 6.98 ** -.0454 .0020 .45 -.0277 .0007 .18 Study performance and age effects were computed by regression analyses. *E<.05 **p_<.01 ***p_<.001  Qualitative Approach As addressed by hypothesis H (see Table 3.1) and illustrated by Figure 3.1, the 5  qualitative approach further investigated the finding from the quantitative approach that encoding phase performance accounted for a significant portion of retrieval phase performance by examining age-related changes in matches. Matches are the number of references that were provided both at the encoding and retrieval phases while controlling for individual differences in the number of references that were provided at the retrieval phase. Matches were computed for each obtained and aggregated item as match scores divided by retrieval scores. Tables 3.7a and 3.7b present match scores by decadel group and Tables 3.4a and 3.4b present retrieval scores by decadel group. As well, the qualitative approach further investigated the finding from the quantitative approach that encoding phase performance did not account for all of retrieval phase performance by examining age-related changes in losses and gains. Losses are the number of references that were provided at the encoding phase but were not carried over to the retrieval phase while controlling for the number of references that were provided at the encoding phase. Losses were computed as forgetting, or encoding minus match scores, divided by encoding scores; Tables 3.8a and 3.8b present forgetting by decadel group and Tables 3.5a and 3.5b present encoding scores by decadel group. Gains are the number of references that were provided at the retrieval phase but were not carried over from the encoding phase while controlling for the number of references that were provided at the retrieval phase. Gains were computed as additions made at the retrieval phase, or retrieval minus match scores, divided by retrieval scores; Tables 3.9a and 3.9b present additions made at the retrieval phase by decadel group and Tables 3.4a and 3.4b present retrieval scores by decadel group. As discussed in the Memory section, the data were examined in order to warrant collapsing across the 15-minute and 1-hour delay periods. Matches, losses and gains were  55  OS in  en  CN  o o  sq  O q  <3  o o o p  00 m o q  m in o q  SO CM  a! < Tt  in  CO.  Al o oo  6v  £  cs  P  H c  II  s  Q  d CN  Pi  in SO 00 q  d  cn  Tt  so CN cn  Tt  SO OS m q  00 CN CN q  CN O m cn  cn  o  cn  in  so cn  os  SO oo  II c  r - oo SO CN in ^ m os  OS CN II c  m  Os Os* m so  cn CN II C  —  OS  Tt  CN  II C  CN  SO CN CN *~t  m d  oo  SO en  o o q  00 CN O q  o CN o q  * *#  *ob  ,—1  cn CN II c  r -  CN CN  q  cn  O m d ~  Os CN  cn CN  Tt  •*  CN oo O en oo Tt O so m Tt en i n i n i n OS Os en CN r-' en  Tt  cn o o q  CN q  *  Os II c  o cn II c  Tt T>  CN  CN CN CN q  m o  O  o q  Tt,  * * Tt  r~*  CN  Tt  Tt  SO*  Tt  o Os  Tt  in q  cn sq cn  Tt  q  00 Os q  so in SO CN  en  en cn en CN  en i—i  O en  Tt  ^  Tt  OS OS T t i n SO o 00 T t en o o SO OS i n T t CN en sq 00 so OS 00 i n o CN q m Os i n en oo OS T t T t , r~" en CN CN 00 in* en* en* d oo* m en CN CN  C  -  Tt  CN  d  CN  Os en OS T t en oo o SO O CN so en CN SO T t O sd SO en Os CN CN CN* OS* i n t - en  o en  CN  r-  Tt  so cn CN oo q 00 en* CN cn* CN* Tt  CN  00 Os" O OS o CN r~ 00 o i n cn O cn CN" Tt r- Tt sq T t en m CN en O CN sq 00 en 00 CN^ CN CN, en* T t so Tt CN CN 3 , Tt  CN  w  os  ^  ii  °- s •  M c  os  oo  vi  c  o  cn CN CN  in  OS CN CN  * *i n  Os m oo os os oo r-~ cn  CN so SO so CN ~ T t SO  •  * * in  CO.  II c  T  o q  cn  ' O SO  • O in  Tt  * * * in  •*  CN Os  in  SO rCN cn  Tt Tt  CN cn ^ O V 00 cn so  w  * * *T t  cn ro  os  "3 O  o SO  so  2 ©  * *  oo oo  DC  '  Os O o O  SO q  m  r-  oo T t os t -  oo en o o SO SO O OS T t sO SO 00 o r-; o SO Os Os q i n en so 00 i n sq oq i n CN so en 1—H en* r-" 00 r~* Os T t i n Os en CN >n T t Os d  CN  00 OS 00 i n SO O Os" i n en en" oo i n Os OS" SO' in" 00 CN q so 00 SO en en en r - en T i - I> Os i n o SO i n CN i n en OS CN, CN CN T t so CN T t cn* d in* cn r - en 1  CN CN  in in Os i n T t en CN  00  S  CN  Tt  00  SO OS T t OS 00 O T t CN so o en r - i n CN Tt r-- Os 00 q CN q oo O i n oo i n m Tt i n Os en OS en en en* en* SO* i n CN, en* SO en OS en  ^ 00  o  0>  o o  o  Q  00  ^ 00  13 & co 03 a,  4) o-  ^  Q S Q S Q S Q 00  00  00  c o t/3  o  Ci  Q  oo  00  C  c 3  oo  "3 .2 ° 2 w 00  —  56  oo Tf  CN CN  CN  cq  m  r-  ON Tf  —;  Tf  in CO  Tf Tf  Tf  Ml  <  CD  CO  ON  00  13 o  o p  o ©  Tf  O  o p  O O  o CO © ©  ©  o  O O  CN ©  ON  NO  rCN © ©  CO  T3  * *  C  o >> <u  *  ON ON  X>  <D  in  NO  X! •t-t  r-  Tf  NO  m CN co  Tf  co  _  in co  co  co CN  Tf  ON  CN  ON  c  o  ON  tf  CD  O  NO Tf  CD C  CO.  o <u o c  03  Al O oo  oV M c  ^  c"o  ' ©  II c  NO  ON  '  II  S  c  CN  CN  © p  CO p  m CN O  r~  NO NO  ' ON  l'  i'  Tf  Tf  Tf NO  O CN  O l'  i*  Tf  CN ON CO NO ON 0 ? CN CO CO NO OO i> ^ H —' T f CN —<  ON ON  ©  CO CN ©  co  _  CN ©  m  Tf  Tf NO Tf  in  NO O Tf O  N  NO i n  in p CO CO  CO  ^  oo  Tf  © CN  Tf  ON  i  r  NO T f " 00 Tf O T f CO  ON  O-  CM  x: o  03  E  Tf  NO CO ON CN >n T f i n oq  C-; i n CO CN  ,—i  13 >  "5 o u  c CD  s o  ^-^  CN NO oo ON oq CO CN  CN  ^  CD  00 c  NO CO NO NO 00 CO © ON ON CN NO ON CN  o  o « Q  Tf  p in  CN ON T f ON «n CO ON 00 NO oq ° - J <—! CN  O O  oo © ON i n CO © CN 00 T f © CN CN CN O - J CN <N ,—1 i n CN  T?  © CD co >.  w  NO ON  ^  ©  B  «?' 8II IT)  © O in  ON NO ON m CN T f T f o  CN —*  co  co m  o T i n oo co T p CN CN T f <-H ON CN T f CN ON ON i n CN ~* CN CN NO ON"  CN*  w  ON  3 o<  i  03 CO  CD 00 CO  T3 C od CD 00 c3 CD CD c CD 3  ON CN  © Tf  ON NO — 00 NO T f NO T f CO 00 r~; CN — < w  T f o o N O c o - ^ o o T f - H O N i n CN oo i n T f c • co ON NO oo -H cN CN in CN  ~~  in  X) OJ  ON •  CO CN  ©  © m —> © co N O r-- i n co oq - H p i n CN CN —' CNi ^  m N o ©  —  cNOsTfONNONo  N q i n r ^ N o o c o o c N m c N  3  o v 2  ON  <?N  VI  c  " IT  t-~ T f r~ m O N r~T f O T f CO © ~ H m CN  t— 0O ON O N CN O  Q S QS Q CO  CO  E  Tf  T f . NO CO  T f N O o o i n N O O N oT o - c N ~O N f r —«  co CD  co  "B  "0  5 « S « « o a n  CD  B K  B o CD  ©  «3 v C4  c o c o c o c o c o  CO  04  CD  £  C/5  ^  E o CD  X)  1  co CD u. 00  T3  c  co CD  O  CO  T3  2  13 B3 C B  CD »  T00 3 * C3  g  00  I > CD  I CD  I i c  CD • 00 V  <!  a  1. .0092  NO CN*  ON  #  Tf  II  c  ON  m  in  ON  CO Tf*  O  oo  ON l>  o  ON NO CN NO ON" Tf O i CN ii CN Tf* © II ON m NO  ON  CN oo ON © CN oo m CO II i n ON NO o II  i  a> O  geb  Q  <  CN II II  o  V)  ON  TT l  o  in"  CN ll II C  Tf  ON ro  o  CN II C  ON P" CN  VI  II  II c,  NO ON 00  Tf  ON d co  ro  00  ON  0\  CN 11 l CN o II Tf  00  ON  co  CO o m o CN T f in d  d  NO  VI  i  II  CN  1—<  co  ^  o r-  NO NO CN*^ r-"  co T f " 00 o  00 ON  ON  Tf  ON  o  co"  -  CO  Tf  Tf*  ON CN  c  00  co*  r-  NO CN  ON  Tf* T f  00  Tf  O  in  Tf"  p in  in  00  in  CO  -.0631 .0039  r~ m  NO  Tf^  Tf Tf  CO 1  o  NO  CN © CO  CO CN*  Tf* T f  o  /—^ in" co co ON ON  oo  CO  Tf*  CO  ON  ©  CN ON CN CO* CN  CO  in o  Tf  in"  co d  Tf  NO*  ~  <  in  CO CO CN CN  CN CN co in  00  ON NO Tf'  in* co*  in  NO Tf  T f " ON T f 00  C-*  Tf  Tf 00  in  ON"  in  CN  T f ON" NO CN NO" O o oq NO CN oo r-* i n co CN  00  o  CN  ON p d  G"  oo CN" in  NO NO co TF CO ON i n t  00  m o  r^" T f  NO Tf  Tp Tf Tf  oq  CN* CN  in* r-"  t  o  o  oo CN  CN" CN CN NO  co in i>  CO CO ON ON ON ON 0 0 O T f CO*  in" i n O co  Q § Q S Q  CO  ON  in  CO CN  ON  CN  CN CN" oo © OO in" i n p 00 NO i n o T f CN CO* d co* co" CN CN m CN CO CN Tf CO  o  -.0705 .0050  -.0109 .0001  -.1622 .0263  Tf Tf  w  m CO CO*  ON ON" CN CN" ON I  co*  CN NO CN 0 0  CN co T f in T f © CN ON" CN Tf CO r-- NO CO ON CO CN co* co* d ON ON* CO CN *~'  Tf  CN  T? Tf  co  IT)  ON CO NO NO CO • CN CN CN d © II O NO CO  Given  IT)  Item  :adel  a NO oo s 1 CN II o II i* o O NO  co  CN  CN CN" NO T ? CN CN T f NO T f T f " oo T ? O CN NO T f NO o ON i n T f NO i n oo CN i n m T f CO CN* CN CO CO d NO m CO  color  o  CN  (n=23)  l>  -.0925 .0085  -.0385 .0015 co  Tf  o  ON  p  oo O in  CN CN CO  NO NO* T?  o  Tf*  ON"  O  00  c o ^ t o ' ^ t o ' ^ t o ' ^ c o ^ c o  spatial . location  00  o CN  NO  Tf  size  II  Tf  NO  oq  shape  CO  00  Al ON  o  o  p  co  quantity  Tf  -.0859 .0074  in  CN  *~*  position  II C  ~*  -.1268 .0160  Al ON"  O  CN tf  object  CO.  CN  -.0815  tf  -.0478 .0022  differe nee score0 for each:  <U  p  passive state people  cn co  CO  in  2.  3. .0223  rco  ON  o  .0155  2. .0157  co  o  .0049  1  m  .0080  O in  .0004  o  m  1  <  9900'  )f age and age s;quai ed on forgisttin g(en codirlg-m;  tf  o co  Tf  o  Table 3.8a: Inifluei  CN  o  p  .0062  3  tf < NO  00  u  2159)  b  u  (D  CN  .0220  M)  «<  2,145)  n on the given scale.  co  00  0000*  n »s  en Tt  o o  o o  CN in  CN o o  o o .o o  o o o o  Tt  © O  o ©  CN O  CO  cn o  O  p  in  CN O o  o o o p  o p  o  m ON o o  CN in oo O  Tt  O NO o 00 i n T t CM  ON CN  ON  o  o o  CN OO  ON  o o o  o o o o  Tt  1—H  m p  CN  cn CN o  NO NO  Pi O o p  CQ.  Al  ©  ON  II  Tt  o o  ©  o  in oo © ON  00  o\ I  cn CN  ©  11  II  c  ON ON" NO CN  ©  i o  II II  C  r00 CN  Tt  NO  Tt  in  T ?  ON  NO" NO  cn  in  CN  T ?  °  ON  cn in  ON  in oq  CN •  P"  r-  CN OO 00 © ii n n r~- 00 CN CN CN  ro o p  Tt  o p  _  00 r~  CN  Tt  CN o  o  NO o  Q\ CN NO NO en  en en CN i n CN CN T t ,  °  00 cs" Tt  cn o o p  ON" i n o f© l > ON i n cn 00 © CN CN en CN ^"-^ en CN Tt  NO  i  ©  Q in ON Tt  o ON  m ©i cn ON CN  VI  NO  in  "—' 03  ON" ON T t NO  3/  Tt  CN NO CN T t ON" CN en © oq ON ON CN ON en CN CN, T t  3,  C  03  C  3, _o  N© ON IT)  w  Lo  © cn ii ii c  CN CN  CN  NO  00 © <n CN oo p CN CN  cn  NO  r-  Tt  3/  CD  /-—s T t ON Tt  © T ? <n i n " en OO © en en CN CN en en Tt  60 <U  3, T3  ON"  CN H H  £ cn CN II c  O  ON  00  O  NO"  m  ON  CN  00 oo  m Tt  Tt  in © en T t  °  oo en ON" t-- 00 r O ON CN p en en en CN Tt  3-  o ©  £3  Di E o o  i n ON" NO CN  ON i n en en NO 00 © r~ oq ON CN CN, CN CN, T t CN  O ,W  J*© Os 00  ON" Tt  CN  o"  3/  cn ON" CN m 00 O ON cn cn cn CN CN  6N"  IF  ON"  3/ CN  NO  r-  3  Tt  CN en en  Q CO  00  en ON en en m 00 NO en CN CN en  Q  Q  CO  CO  Tt  c O  U  O  X)  03  <u 03  ~  ~  D  © ©  C  c  CU  o  BQ  cs  o  e  c  >  ON  in" oq  3  w  Q  Q  CO  CO  Q  oo c C/3  s  03  3  CO  cr <u 60  o3 T3  o5  g  u 60  • V  *  59  ON  m  CN CN  a  CN  OX  o  Tt  o  ON  ©  O  q  CN in  o o  <  * *  o NO  CN  PH <MD|  04  o q  CO-  ©  * * *C N  * * *T t  tf  ©  O  u Q  os II c  en  en  CN  CN  Al  o  ON  N  m  CN  " I I © p  NO ON , CN  O  T f  *  B  CN  o  CN r-  CN  00  ZL .  O  T  ^  ®  B  00  CN  C  ©  N©  w  <—<  Os  ^  1  NO O N  O  oo  CN  •  Tt  Tf  Tt  cn CN r~- en  o  II  S  ?5 ©  c  00  Tt  ©  CN CN Tt  ©  en Tt  ON  ^ VI  in en  Tt  in  O  CN  o  oo en  ©  in  ON  00  NO  CN  CN  IF  '  CN  oo en cn r~; oq en Tt  o  Tt  in r-* o m m d CN  NO  Tt  s  en en  NO  ©  _;  CN  00 © CN  cn  '  CN  r-  CN  in  —'  NO  en  CN  o  Tt  ©  q  OO  Tt  ON  NO  Tt  in q  en  CN  NO  o q 00  CN CN  q  r  Tt Tt  00  i'  NO  00  cn  NO  en  C/ ^  m  o  CN  i'  1*  en oo  in  ON  Tt  '—I 00 en en  in  W  ON ON  O r-~  CN  CN  CN  NO  en  en 00 00 o Tt Tt i> en •—< cn' O N N O —' —  en r - N O en m en 00 q m  in  in  :  Tt  en  CN CN  Tt  Tt  q  CN  NO  oo  q  CN  _;  CN  ©  Tt  ©  Tt  ©  s  Tt  ON  en roo  en © en © cn in en o  o  ON  o q  ©  * * *  en  •  CN CN O  o © q  q  * *  Tf Tt  r-  ©  —i  ©  CN  ©  Tt  ;  Tt  NO ON  Tt,  in  ON  ON  Tt  o  in •—; in in  NO  CN  d  cncN©NO'— 0 N r ^ 0 N N 0 T t T t 0 N O N 0 0 © N 0 O e N i n c T N C ^ O N i n © T t o q N q i n o q ^ T t N q T t O N — • NO M K J - ' D ^ N c s - - ^ ^ o d i o - - i n ^ - d  CN Os © oo" CN O N N O N O O N O N CN cn CN —~ ON N O N O en q ^ o o c N c n ^ T t e n i n q q i n T t N o r ^ r n i n i n -^oddin'encN --—• c n c n O N o — — « N O i n — CN -  — ' N o o N r - ~ r ~ < n © — r ~ N o r - ~ O N O N c N c N © ' — t~N o q q o N ^ o o q c N i n N O N O o q e N T t O N e n o N N q c x J N O ^ o > e n ^ ^ ^ e n c N o 6 i n ^ ^ i n T t d ' - < ^ C N ^ w  w  w  w  w  t ^ T t N o c n o o ^ © c ^ r ^ q o q - ^ . q e n O N r - ; O N ' - ;  c  Q  Q  CO  CO  o O  o  Q  o  CO  CO  CO ^  CO  Q  CO  CO  a «  O  a  O  Q  CO c  c o  > ^  O  Q  Q  P  CO  s  >  CN CN  CN  Tt  W  cn  cn »•  Tt  ' I I c  "? 8  CN  VI  .  w  ~.  r-  Os  ON  NO Tt  Tt  ©  NO ON • , CN O N  NO  m  o  C  00  NO m || NO CN © c  ON  ©  CN  ON  00  ©  ON CN  m o  II co  Tt  ON  ON CN CN II C  o\ irj  in en  en cn  en  NO  oo  oo.  O N CN 00 O N  © ©  NO  ©  i'  o o o ©  O  ON  en  CO  o  00  N©  NO  ©  CN  *N O  ON  oo  CN  Tt Tt  2 O ©'  ©  m o o ©  O  t  O 00  o  Tt  ©  p f  <  Al  ON  3-S N  CT  oo  5 CJ cu o oo —  60  JZ  * *ON  ON  in en  co O  CN  O  O ©  CN  m q  NO NO  NO  d  CN  WD  tf <1  CN O  OO NO  O O  CO  o O  O o o o  in  o o o  o  CO CO  oo r-;  in  ON  o o  CO  NO  o  o  CN  O  o  *—  o  NO NO  q  CN q  o o q  o o q  Tf  o  Tf  ©  CN CN  NO CO  oo  o o  Tf  CO.  00 Tf  ON"  00  c  o  OS  "  CO  CN  ON NO  2 o  ON  CN  00 r--  q  l"  NO  q  o m o  ON  in Tf  ON O  co CN CN q  o  CN O q  00  Tf  ON NO  ©  Al  o CN CN o  o o o.  Tf  o  ON —«  ©  i'  <o CN T f 00 CO CN T f ON  CN  oo Tf  CO  co  in"  NO  Tf  CN  Tf  00  CO CO  CN  s  0O 00  CN  r Tf Tf o i n co CN T f  i n NO r~~ oo  Tf  O m  O  Tf  Tf  —;  NO  CN CN  Tf CN  O C-;  i n co ON" oo ON T f oq T f >n CN i n CN  CN  T f " NO CO ON  q  ' '  CO  CO  ON  NO  Tf  CN" NO  1  CN  CN  O NO  CN  in in  Tf  CO NO  Tf  ^""^  m  ON  in q  CO  oo  CN  CN  o  '  °  I*  O  ON IT)  o co  o  O  Tf  o  oo  in  Tf  00 oo CN" 00 ON" i n oq NO NO T f  ON Tf I  ON  CN  o  /—s  o  r-- oo T f oo ~* r-NO T f CN T f T f >n  ON  NO  'oo oo r— co oo T f T f co CO CN T f  CO  CN  CN  <N  NO  oo 00 CO  Tf ON  ©  CO ON  CN VI  03  C  o  IT)  O  oo U oo >-> "3 C  °  oo £>  X) T3  CJ  o <n" 00 in Tf co  ;~. NO  O  C— NO  CO  co  ^  oo  m  oo  Tf r-  ,  c  ""  CO  r-  in" co  ON  Tf  NO CO  T? NO  TT  m q  -H  oq  co T f  NO  NO  o oq  ^  S eu  V  e  w ON  ^ 8 o o  1 1  H  a *  -  c  O cj  O  "  Tf  oq  J)  a.  ON  00 i n q  CN  CO  e co o «U  CQ  o c D  CO  -s  *^  CO  Tf  CO  ~  ON" NO CO CN  in  •5  Tf  H  ^  CO  ^  Q CO  ^u V« <L)  00 03  *  II *  T3  C  o  CN ,  Q S Q S  CO  03  CO CN  oo S£ So >>  .5? 03  o  co © NO i n  a > D  00 c  to co  00  V  61  regressed separately on delay period, age and then the delay period by age interaction term. Matches, losses and gains were not significantly related to delay period or the delay period by age interaction term for most obtained and aggregated items, except for losses for the event item which showed a significant delay period by age interaction, J3deiay=--2638, f3 =. 1939, age  &nteract=-.7537, R =.0271, F(3, 158)=4.43, p<.05. Consequently, the data were collapsed across 2  delay period for all items except for losses on the event item which was analyzed separately for each delay period. Matches. Age-related changes in matches were examined by separately regressing matches for each obtained and aggregated item on age and then age squared. The results indicate that matches were not significantly related to age or age squared for any of the items, and thereby indicating that there were no age-related changes in the proportion of references that were provided both at the encoding and retrieval phases. Losses. Age-related changes in losses were examined by separately regressing losses for each obtained and aggregated item on age and then age squared. The results indicate that there were no age-related changes in losses for most items, except for two items on the given scale — quantity and size — which showed an age-related increase {quantity: J3=l .4366, R =.0276, F ( l , 2  160)=4.55, p<.05; size: £ = . 7 5 5 0 , R =.0241, F ( l , 160)=3.95, p<.05). As well, the results indicate 2  that losses were not related to age squared for any of the items and thereby indicating no nonlinear relationships between losses and age. Because the event item showed a delay period by age interaction, losses for the event item were analyzed separately for each delay period; for each delay period, losses on the event item were regressed on age and then age squared. The results indicate that there were no age-related changes in losses for the event item as losses were not related to age or age squared for the 15-minute or the 1-hour delay periods. Together, these  results indicate that there were generally no age-related changes in the proportion of references that were provided at the encoding phase but were not carried over to the retrieval phase. Gains. Age-related changes in gains were examined by separately regressing gains for each obtained and aggregated item on age and then age squared. The results indicate that gains were not related to age or age squared for most items; gains on one item on the beyond scale — event — was related to age squared, and thereby indicating a non-linear relationship between age and gains on thisitem (event: [j =-2.0948, [4^2=2.1510, R =.0472, F(2, 159)=7.87, p_<.01). 2  age  Together, these results indicate that there were generally no age-related changes in the proportion of references that were provided at the retrieval phase but were not carried over from the encoding phase. Summary Together with the previous section, the results indicate divergent age-related changes in encoding and retrieval phase performance, where there was an age-related increase in performance on the aggregated given item and no age-related change in performance on most items on the beyond scale. These results support hypothesis H3 (see Table 3.1) as the same pattern of age-related changes were revealed for encoding and retrieval phase performance. A quantitative and a qualitative approach (see Figure 3.1) were employed to investigate the relationship or overlap between the observed age-related changes in encoding and retrieval phase performance. As predicted by hypothesis H , the results of the quantitative approach showed that 4  encoding phase performance accounted for a significant portion of retrieval phase performance, and thereby indicating that retrieval phase performance was related to encoding phase performance; when retrieval scores were regressed on encoding scores and then age, the results showed that encoding scores were positively related to retrieval scores for all of the obtained and aggregated items. There was little support for hypothesis H , however, as there were generally 5  63  no age-related changes in matches, losses and gains, and thereby indicating that there were no age-related changes in the degree to which encoding and retrieval processes overlap or do not overlap; when matches, losses and gains were regressed separately on age and age squared, the results indicate that matches, losses and gains were not related to age or age squared for most of the obtained and aggregated items. Wordiness As addressed by hypotheses PL; and H (see Table 3.1), this section investigated the 7  influence of wordiness, or the greater use of words to convey a given amount of information, on age-related changes in encoding and retrieval phase performance. Wordiness was measured by the verbatim repetition item which counted the number of repeated references to words or phrases that were exact or near literal reproductions of references to the obtained and aggregated items (see the Scoring Key in Appendix B for a full description of the verbatim repetition item). Before investigating age-related changes in verbatim repetitions, the data for the verbatim repetition item were screened for univariate outliers using the same regression procedure described in the Univariate and Multivariate Outliers section. Four outliers were detected in encoding phase performance and three in retrieval phase performance and they were replaced by their respective upper or lower value at the 99% confidence interval. In order to investigate age-related changes in verbatim repetitions, the number of references to the verbatim repetition item at the encoding and retrieval phases were regressed separately on age. No support was found for hypothesis H6; there was an unexpected age-related decline in verbatim repetitions produced at the encoding and retrieval phases (see Table 3.10). Although there are some reports that older adults provide a greater number of words to convey a given amount of information (Botwinick & Storandt, 1974; Obler, 1980), the present results revealed that young rather than older adults were more wordy on a picture description task. It  o NO  * *  *  00 CN  NO  Tf  ON  CU  in oo O  ON  oc|< tf  00 O CN  cn  CN  CN cn  ON  NO OO  Al O 00  oo — r-; T t NO cn CN ON cn cN cn ~ * cn cn  oC M c  m —< r - ON cn ON NO CN ON wn m CN m T t i n TJ-  ON  o NO  ,  3* o t- ©N© O  « u Q ft  ON  ON  cn oq cn* un  ON  CN C  w .  t— — i CN r-; T t CN © T t NO cn NO cn  o cn ON r - © vn ~-'< CN* CN* NO cn r Tt r-  ^  CN  © IT)  TT . J.  NO  00 ON CN t~~ © r~;  cn*  NO  NO  Tt  Tt  ON  CN II  Tt* ON f» Tt  oo cn cn r> r - ON —< r -  ' © cn  II C  ON CN  ON *->  Tt  —  ON  ©  O O  ON ©  ON  NO* ON  in  oo cn  w  Tt  00  VI  s  Q  | S Q tS cu  co  CO  a  CU  tf  £ CU  S oo « .£  •e ° « >  c W  Tt  >  tf  65  may be that while verbatim repetitions result in more wordy picture descriptions, they may also serve as a communicative style or as a memory strategy. Young adults employed a communicative style that typically used verbatim repetitions to unfold information. For example, "It looks like a canoe race. There's three canoes. Each canoe manned by two people. Must be men because they don't have shirts on. They're racing down the river. The river is blue." While young adults used verbatim repetitions to connect or to integrate information, older adults employed a communicative style that also presented information in an integrated manner but without the use of verbatim repetitions. For example, "This is a picture of three canoes racing down a beautiful river. Two men in each canoe." While both young and older adults provided a cohesive, integrated presentation of information, older adults used fewer verbatim repetitions and thus made a more economical use of words resulting in picture descriptions that were more informationally dense. Verbatim repetitions may also have been employed by young adults as a memory strategy, where verbatim repetitions may have served as a rehearsal strategy that is used at the encoding phase to commit more information to memory or as a retrieval cue employed at the retrieval phase. As addressed by hypothesis H , in order to examine the influence of verbatim repetitions 7  on encoding and retrieval phase performance, the number of references to the verbatim repetition item at the encoding and retrieval phases were regressed separately on encoding and retrieval scores, respectively. As shown in Table 3.11, for the encoding phase, the number of verbatim repetitions was positively related to performance on the aggregated given item and on some items on the beyond scale — absence and setting. When compared to the unpartialled age-related changes in performance (see Table 3.5a and 3.5b vs. Table 3.11), the age-related decline in performance on the aggregated given item disappeared when the influence of the verbatim repetition item was partialled out, and thereby indicating that the verbatim repetition item  Table 3.11: Influence of verbatim repetition and age on encoding phase performance on each item on the given and beyond scales. Verbatim Repetition Item Given  P  R  .8575  .7354  P  R  color object passive state people position quantity shape size spatial location  .7412 .8288 .7648 .5373 .4786 .7659 .6849 .6672 .8489  Given/Beyond action object's age person's age Beyond absence emotion event setting time frame  2  Age  a  F ( l , 149) 414.11 ***  P age  .8538  AR  a  2  F(2, 148)  .0002  .1 1 F(2, 159)  F ( l , 160)  P age  .7354 .6869 .5849 .2888 .2290 .5867 .4690 .4452 .7206  195.13 *** 351.08 *** 225.49 *** 64.97 *** 47 52 *** 227.08 *** 141.34 *** 128.42 *** 412.75 ***  -.0085 .0281 -.0917 -.1543 .0138 -.0812 .1040 .0230 .0353  .0007 .0007 .0077 .0218 .0002 .0061 .0099 .0005 .0011  .11 .37 3.01 5.04 * .04 2.36 3.03 .70 .65  .4381 .3551 .3795  .1920 .1261 .1440  38.01 *** 23.09 *** 26.92 ***  -.0751 .0235 -.0496  .0052 .0005 .0023  1.02 .09 .42  .2967 . .1271 .0418 .1910 .1464  .0880 .0161 .0017 .0364 .0214  15.44 *** 2.62 .28 6.06 * 3.51  -.0201 .1749 .0576 -.0947 -.1003  .0003 .0281 .0030 .0082 .0092  .06 4.67 * .48 .24 .21  2  AR  2  Verbatim repetition and age effects were computed by regression analyses. *p_<.05 **p_<.01 ***p<.001  Table 3.12: Influence of verbatim repetition and age on retrieval phase performance on each item on the given and beyond scales. Verbatim Repetition  F ( l , 155)  P age  .6676  311.34 ***  -.0971  P  R  F ( l , 160)  P age  color object passive state people position quantity shape size spatial location  .7248 .8077 .7134 .5809 .5561 .7430 .6186 .7278 .8200  .5253 .6524 .5090 .3375 .3093 .5520 .3826 .5298 .6724  177.10*** 300.29 *** 165.88 *** 81.49 *** 71.64 *** 197.18 *** 99.15 *** 180.25 *** 328.42 ***  -.1602 -.0746 -.1740 -.0987 .0330 -.1094 .0772 .0163 -.0674  .0228 .0049 .0269 .0087 .0000 .0106 .0053 .0002 .0040  Given/Beyond action object's age person's age  .4828 .3830 .4025  .2332 .1467 .1618  48.65 *** 27.51 *** 30.90 ***  -.0613 .0004 -.0173  .0033 .0000 .0002  Beyond absence emotion event setting time frame  .4648 .2182 .1370 .4025 .2429  .2160 .0476 .0188 .1620 .0590  44.08 ***  .0377 .2956 .2518 .0133 -.0299  .0012 .0493 .0565 .0002 .0008  Item Given  P  R  .8171  Age  a  2  2  7 99 **  3.06 30.94 *** 10.03 **  AR  2  F(2, 154)  .0084 AR  2  4.01 * F(2, 159) 8.06 ** 2.30 9.24 ** 2.11 .22 3.88 1.37 .08 1.99 .69 .00 .05 .26 8.68 ** 9.72 ** .03 .13  Verbatim repetition and age effects were computed by regression analyses. *p_<.05 **p_<.01 ***p<.001  68  accounted for an age-related decline in performance on the aggregated given item. As shown in Table 3.12, for the retrieval phase, the number of verbatim repetitions was positively related to performance on the aggregated given item and most items on the beyond scale; only performance on one item on the beyond scale ~ event — was not related to the verbatim repetition item. When compared to the unpartialled age-related changes in performance (see Table 3.4a and 3.4b vs. Table 3.12), age-related changes in performance remained when the influence of the verbatim repetition item was partialled out. These results indicate that the verbatim repetition item did not completely account for age-related changes in performance. Together, the results support hypothesis H as encoding and retrieval phase performance on the verbatim repetition item was 7  related to performance on most of the obtained and aggregated items and encoding phase performance on the verbatim repetition item accounted for an age-related decline in performance on the aggregated given item. Summary As addressed by hypotheses H6 and H (see Table 3.1), this section investigated the 7  influence of wordiness, as measured by the verbatim repetition item, on age-related changes in encoding and retrieval phase performance. No support was found for hypothesis rL; as the results indicate an unexpected age-related decline in verbatim repetitions made at the encoding and retrieval phases rather than an age-related increase. Support was found for hypothesis H as 7  encoding and retrieval phase performance on the verbatim repetition item was related to performance on most of the obtained and aggregated items and encoding phase performance on the verbatim repetition item accounted for an age-related decline in performance on the aggregated given item. These results indicate that performance on the obtained and aggregated items was related to verbatim repetitions and that encoding phase performance on the aggregated given item was confounded by verbatim repetitions.  69  Study Instructions As addressed by hypotheses Hg and H9 (see Table 3.1), this section examined age-related changes in the influence of study instructions in guiding performance. The basic idea was to examine if study instructions could encourage young adults to perform like older adults and older adults to perform like younger adults. As described in the Method section, the camera instruction was designed to enhance performance on items on the given scale by directing participants' attention to the details in the picture. The friend instruction was designed to enhance performance on the beyond scale by encouraging participants to go beyond the picture per se and to provide references to personal experiences and world knowledge in order to make their picture descriptions more accessible and interesting to a friend. Because young adults tend to encode and retrieve more given and less beyond information, they may not benefit from the camera instruction, but they may benefit from the friend instruction. By contrast, because older adults tend to encode and retrieve more beyond and less given information, they may not benefit from the friend instruction, but they may benefit from the camera instruction. In order to examine the influence of study instructions on performance, encoding, retrieval and match scores for each obtained and aggregated item was regressed separately on study instruction, age and then the age by study instruction interaction term. Encoding-retrieval match performance was examined as a measure of retention. The influence of study instructions on performance was investigated by examining the relationship between study instruction and performance. Because the camera instruction was dummy coded as 0 and the friend instruction coded as 1, a negative standardized regression coefficient (fj instruct) indicates that performance was enhanced by the camera instruction whereas a positive standardized regression coefficient means that performance was enhanced by the friend instruction. The influence of study instructions on performance was also examined by comparing age-related changes in  70  performance when the influence of study instructions was partialled out with unpartialled agerelated changes in performance. Hypotheses H and H were investigated by examining the age 8  9  by study instruction interaction. Encoding Phase Performance As shown in Table 3.13, the camera instruction influenced encoding phase performance on some items on the given scale and the absence item on the beyond scale and the friend instruction enhanced performance on some items on the beyond scale. When the influence of study instructions was partialled out, however, age-related changes in performance were comparable to the previously reported age-related changes in performance when the influence of study instructions was not partialled out, and thereby indicating that study instructions did not account for performance (see Table 3.5a and 3.5b vs. Table 3.13). As well, although not reported in Table 3.13, an age by study instruction interaction was not significant for any of the items. Together these results indicate that the camera instruction influenced performance on some items on the given scale and the friend instruction influenced performance on some items on the beyond scale, however, there were no age-related changes in benefit from study instructions on performance. Retrieval Phase Performance As shown in Table 3.14, the camera instruction influenced retrieval phase performance on some items on the given scale and the absence item on the beyond scale and the friend instruction enhanced performance on some items on the beyond scale. When the influence of study instructions was partialled out, however, age-related changes in performance were comparable to the previously reported age-related changes in performance when the influence of study instructions was not partialled out, and thereby indicating that study instructions did not account for performance (see Tables 3.4a and 3.4b vs. Table 3.14). As well, an age by study  71  Table 3.13: Influence of study instructions and age on encoding phase performance on each item on the given and beyond scales.  Item  Study Instruction R F ( l , 149) ^ instruct  Given  .1431  .0204  P instruct  R  .2100 .1822 .2412 .4027 .1081 .1061 .0345 .0779 .2308  .0440 .0331 .0582 .1622 .0117 .0112 .0011 .0061 .0532  Given/Beyond action .0629 .0041 object's age .1885 person's age  .0040 .0000 .0355  Beyond absence emotion event setting time frame  .0567 .1249 .0511 .0031 .0105  Age  a  color object passive state people position quantity shape size spatial location  2  .2383 .3535 .2263 .0563 .1028  2  3.11 F ( l , 160) 7.38 ** 5.49* 9.88 ** 30.98 *** 1.89 1.82 .19 .98 9.00 **  .64 .00 5.90* 9.63 ** 22.85 *** 8.63 ** .51 1.70  AR  2  F (2, 148)  3 age -.2314  .0535  Page  AR  -.2082 -.1805 -.2389 -.4004 -.1072 -.1039 -.0337 -.0766 -.2291  -.2183 -.2099 -.3012 -.2924 -.1235 -.2930 -.1003 -.1693 -.2087  .0476 .0441 .0907 .0855 .0152 .0859 .0101 .0286 .0435  -.0614 -.0035 -.1873  -.1939 -.0800 -.1527  .0376 .0064 .0233  6.23 * 1.02 3.94 *  -.2374 .3525 .2259 .0573 .1038  -.1016 .1215 .0392 -.1421 -.1348  .0103 .0148 .0015 .0201 .0181  1.75 2.73 .25 3.28 2.98  3 instruct  -.1455 P instruct  Study instruction and age effects were computed by regression. *p_<.05 **p_<.01 ***p<.001  2  8.56 ** F(2, 159) 8.34 ** 7.60 ** 16.95 *** 18.07 *** 2.49 15.12 *** 1.61 4.72 * 7.67 **  72  Table 3.14: Influence of study instructions and age on retrieval phase performance on each item on the given and beyond scales. Study Instruction R F ( l , 155)  R  Item  instruct -.0716  .0051  P instruct  R  J  Given color object passive state people position quantity * shape size spatial location  F ( l , 160)  5 instruct -.0643 P instruct  R  age -.3494  .1220  P age  AR  F ( 2 , 154)  2  21.53 * * * F ( 2 , 159)  2  27.27 * * *  -.0969  .0094  1.52  -.0939  -.3808  .1450  -.0745 -.1685  .0055 .0284  .89  -.0718  -.3320  .1102  19 81 * * *  4.68 *  -.1656  -.3888  .1512  -.2582  .0667  11.43 * * * -.2561  -.2773  .0768  29.29 * * * 14 27 * * *  -.0617  .0038  .61  -.0605  -.1532  .0235  .0013  .0000  .00  .0039  -.3422  .1171  -.0147 -.1043  .0002  .03  -.0136  -.1348  .0181  .0108  1.76  -.1026  -.2243  .0503  8.52 * *  -.1014  .0103  1.66  -.0989  -.3293  .1084  19 57 * * *  .0012  .19  -.0332  -.2135  .0456  7.60 * *  .0002  .05  .0181  .0158  2.56  .0215  3.59  .0139  2.32  Given/Beyond action -.0349 object's age .0171 person's age .1617 Beyond absence emotion event setting time frame  2  .79  Age AR a  a  3.83 21.09 * * * 2.94  .0262  4.30*  -.1606  -.1259 -.1467  -.1708  .0292  4.80*  .2021  -.1181  .2031  .0412  6.88 * *  .1545  .1362  .0186  3.14  .1558  .0242  3.98 *  .0956  .1782  .0317  5.35 *  .0948  .0089  1.45  .1875  -.1214  .0147  2.40  .1866  .0348  5.77 *  -.1081  .0117  1.95  Study instruction and age effects were computed by regression. * quantity: study instruction by age interaction effect: j3instruct=-0039, [J =-.3333, age  fiinstructage^-1466, AR =.0212, F(3,l58)=3.91, <.05. 2  E  *E<.05  **p<.01  ***p<.001  73  Figure 3.2: Retrieval phase performance on the quantity item by study instruction.  Figure 3.3: Encoding-retrieval match performance on the quantity item by study instruction.  4or  Camera  Camera  Friend  Friend  CO <D O 3Cf  CO CU  CO CU  CU  DC O 2Gt cu  01 O 20f  o c  c  CU CU XI  _n  E  E  3  *  <30  30-39 40-49 50-59 60-69 70-79  >80  10t  <30  Age by Decadel Group  Figure 3.4: Encoding-retrieval match performance on the absence item by study instruction.  30-39  40-49  50-59 60-69  70-79  >80  Age by Decadel Group  Figure 3.5: Encoding-retrieval match performance on the time frame item by study instruction.  Camera Friend  in cu  CO CU  o c  o c  CU  -S  CU  CU  CU  01  CU X)  CU X)  E  E  3  3  <30  30-39  40-49  50-59  60-69  70-79  >80  Age by Decadel Group  <30  30-39 40-49 50-59 60-69 70-79  Age by Decadel Group  >80  74  instruction interaction was not significant for most items; as shown in Table 3.14. only the quantity item showed an age by study instruction interaction. As illustrated in Figure 3.2. the friend instruction enhanced performance on the quantity item for young adults and the camera instruction enhanced performance on this item for middle-aged and older adults. Together, these results indicate that the camera instruction influenced performance on some items on the given scale and the friend instruction enhanced performance on some items on the beyond scale, and that there were age-related changes in the influence of study instructions on performance on the quantity item. Encoding-Retrieval Match Performance As shown in Table 3.15, the camera instruction influenced encoding-retrieval match performance on some items on the given scale and the friend instruction enhanced performance on some items on the beyond scale. When the influence of study instructions was partialled out, however, age-related changes in performance were comparable to the previously reported agerelated changes in performance when the influence of study instructions was not partialled out, and thereby indicating that study instructions did not account for performance (see Tables 3.7a and 3.7b vs. Table 3.15). As well, an age by study instruction interaction was not significant for most items, except for the quantity, absence and timeframe items. As illustrated in Figures 3.3, 3.4 and 3.5, the camera instruction enhanced performance on these items for young adults and the friend instruction generally enhanced performance on these items for middle-aged and older adults. Together, these results indicate that the camera instruction enhanced performance on some items on the given scale and the friend study instruction enhanced performance on some items on the beyond scale, and that there were age-related changes in benefit from study instructions on performance on the quantity, absence and time frame items.  75  Table 3.15: Influence of study instructions and age on encoding-retrieval match performance on each item on the given and beyond scales. Agel  Study Instruction Item  R  5 instruct  Given  -.1097  F (1,154) 1.88  .0120 R  P instruct  color object passive state people position quantity * shape size spatial location  2  2  F ( l , 160) 3.37  AR  -.1089  3445  .1186  20.88 ***  P instruct  P age  AR  F ( 2 , 159)  -. 3265 -. 2933 -. 3486  .1066  19.43 *##  .0860  15.14 ***  .1215 .0586  23.04 #** 10.74 **  -.1410  -.1436  .0206  -.0999  .0099  -.1999 -.2707  .0399 .0733  -.1033  .0107  1.72  -.1022  -.0134  .0002  .03  -.0105  -.0145  .0002  .03  1.61 6.68 * 12.65 * * *  F ( 2 , 153)  3 age  5 instruct  -.0977 -.1972 -.2688  -.2422  2  2  .0225  -.0133  -. 1502 -. 3642 -. 1489  .0701  12.14 *** 17.72  .1326 .0222  3.71 24.32 *** 3.60  -.1120  .0127  2.07  -.1109  -.2647  -.1631  .0266  4.37 *  -.1607  -. 3125  .0976  Given/Beyond -.0338 action .0076 object's age -.0722 person's age  .0011 .0001  .18 .00  -.0322 .0082  -. 2044 -.0648  .0418 .0042  6.94 ** .67  .0052  .84  -.0708  -. 1760  .0309  5.11 *  .0166  2.72  -.1280  -. 1583  .0251  1389 1447  .0193  4.16 * 3.38  .0209  3.59  -. 1557 - 1198  .0242  4.00 *  .0143  2.33  Beyond absence * -.1292 .2743 emotion .2296 event .1063 setting time frame * .0850  .0753  13.02 * * *  .2732  .0527  8.90 * *  .2284  .0113  1.83 1.16  .1076  .0072  .0859  Study instruction and age effects were computed by regression. .0106,B_ =-.3524, * quantity: study instruction by age interaction effect: J3  a  age  ^instructage=-.1923, AR =.0368, F(3,158)=7.00 p<.01. Z  instruct"• -  i  * absence: study instruction by age interaction effect: r3i  nstruct  =-.1280, f3 =-. 1489, age  fiinstructa e=-.1539, AR =.0236, F(3,158)=3.99, p<.05. 2  g  F-.0858, fj =-.1099, * timeframe: study instruction by age interaction effect: J3 instruct age  f3instructage=-.1640, AR =.0267, F(3,158)=4.45, <.05. 2  E  * p<.05  * * E<-01  * * * E<-001  76  Summary As addressed by hypotheses H and H (see Table 3.1), this section examined age-related g  9  changes in the influence of study instructions in guiding performance. The basic idea was to examine if the friend instruction could encourage young adults to perform like older adults by enhancing performance on items on the beyond scale and if the camera instruction could encourage older adults to perform like young adults by enhancing performance on items on the given scale. As expected, the results confirmed that the camera instruction enhanced performance on some items on the given scale and the friend instruction enhanced performance on some items on the beyond scale, although study instructions did not account for performance. There were, however, generally no age-related changes in benefit from study instructions. Agerelated changes in benefit from study instructions were found for performance on only the absence, quantity, and timeframe items, where young adults benefited from the friend instruction and middle-aged and older adults generally benefited from the camera instruction. Hence, these results provided little support for hypotheses Hg and H9 that study instructions could encourage young adults to perform like older adults and older adults to perform like young adults as both young and older adults benefited equally from study instructions. DISCUSSION The results provided partial support for the hypothesized age-related changes in memory. As expected, there was an age-related decline in retrieval phase performance on the aggregated given item and an age-related increase in performance on one item — event — on the beyond scale. There was, however, no age-related change in performance on most items on the beyond scale which is inconsistent with previous research that showed an age-related increase in memory for inferences that draw on world knowledge and personal experiences (Adams, 1991; Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Gould, Trevithick & Dixon, 1991). The discrepancy  77  between the results of the present and previous research may be due to task differences. In the Adams' studies (Adams, 1991; Adams, Labouvie-Vief, Hobart & Dorosz, 1990), for example, participants' performance was assessed on only one target story whereas the present experiment employed a total of six target pictures. It is possible that encoding more target stimuli led to greater interference at the retrieval phase, and thereby resulting in preserved rather than improved memory for beyond information. One interpretation for the observed divergent age-related changes in memory for given and beyond information follows from the lifespan notions of selective optimization with compensation (Baltes, 1987, 1993; Baltes, Dittmann-Kohli & Dixon, 1984) and loss in the service of growth (Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Labouvie-Vief & Schell, 1982) that state that certain cognitive competencies can develop that can counterbalance or tradeoff other cognitive competencies that show age-related decline, respectively. Consistent with these notions, one interpretation of the present results is that people may come to rely on memory for beyond information in order to compensate for, or results in a trade-off with, memory for given information which shows an age-related decline. The results also showed that the age-related changes in retrieval phase performance were extended to encoding phase performance; both encoding and retrieval phase performance showed an age-related decline in performance on the aggregated given item and, with the exception of the event item, no age-related change in performance on items on the beyond scale. One interpretation for these results follows from a developmental approach to schema theory that states that schemas can change with experience, either by restructuring, tuning or accretion (Rumelhart & Norman, 1978), and schemas can direct the types of information that is encoded and retrieved about a target stimulus (Arbuckle, Vanderleck, Harsany & Lapidus, 1990; Hess, Donley & Vandermaas, 1989: Hess & Flannagan, 1992; Hess & Slaughter, 1990). Following  7s; from a developmental schema approach, the present results suggest that maturation may influence the tuning or the refinement of associative pathways, where the associative pathways that lead to the encoding and retrieval of beyond information may be strengthened, as witnessed by preserved performance on items on the beyond scale, and some of the associative pathways that lead to the encoding and retrieval of given information may become weaker or pruned, as demonstrated by an age-related decline in performance on items on the given scale. It is possible that the associative pathways between beyond information may be strengthened either because of repeated exposure to the same beyond information or because an emerging role as storyteller in older adulthood (Mergler & Goldstein, 1983; Ramelli & Stella, 1993; Wallace, 1992) directs attention toward the encoding and retrieval of beyond information. It is also possible that the pathways between beyond information are not strengthened per se, but that maturation leads to the accumulation of more beyond information and thereby enhancing the likelihood for beyond information to be selected and communicated. One exception to the above pattern of findings was found for performance on the event item which showed an age-related increase in retrieval phase performance and no age-related change in encoding phase performance. One explanation is that these results are due to selfinitiated processing (Craik, 1986). At the encoding phase, participants' performance was guided by the target pictures, or in other words, their performance was environmentally-driven. At the retrieval phase, however, their performance was self-initiated and it is possible that older participants may have resorted to recalling more beyond information. Although the lifespan notions and a developmental approach to schema theory offer different interpretations of the present results, they are not competing approaches. It is possible that the strengthening of certain associative pathways can be viewed as growth that arises at the expense of pruning other associative pathways.  79  Consistent with the principle of reinstatement of stimulating conditions and the notions of transfer appropriate processing and encoding specificity (Hollingworth, 1928; Morris, Bransford & Franks, 1977; Tulving & Thomson, 1973), the results of the quantitative and qualitative approaches showed that age-related changes in retrieval phase performance were related to encoding phase performance. The quantitative approach revealed that encoding phase performance accounted for a significant portion of retrieval phase performance and the qualitative approach further revealed that there were no age-related changes in the degree to which encoding and retrieval phase performance overlapped and did not overlap. These results encourage memory research to consider encoding processes. For example, future research should examine the role of encoding processes, such as sensory and perceptual processes, on retrieval processes. More importantly, these results raise the question that an examination of encoding processes may be interchangeable with an investigation of memory. Although,the present experiment provided some insight into age-related changes in memory, the results also revealed some curious findings to be followed up by future research. First, although previous research typically finds an age-related increase in forgetting (e.g., Light, 1991), the present experiment generally failed to find age-related changes in forgetting. The results indicate that there were no age-related changes in losses for most items, except for two items on the given scale — quantity and size — which showed the expected age-related increase. As well, performance was not influenced by delay period; participants performed equally well following the 15-minute versus the 1-hour delay period. It is possible that the failure to observe age-related changes in forgetting may be due to the length of the delay period or superior memory for pictorial stimuli (Park, Puglisi & Sovacool, 1983; Winograd, Smith & Simon, 1982). Second, there were generally no age-related changes in the influence of study instructions in guiding performance. Although study instructions enhanced performance on some of the  .  8  0  obtained and aggregated items, age-related changes in benefit from study instructions were found for only a few items; as expected, young adults benefited from the friend instruction and middleaged and older adults generally benefited from the camera instruction for the absence, quantity, and timeframe items. Similarly, Adams (1991) found that study instructions to read a story either for enjoyment or with the intention of answering questions did not influence young and older adults' story recall for given and beyond information. It is possible that the processing that is required to perform the picture description and story recall tasks may override study instructions. Perhaps performance may be influenced by a more compelling testing environment, for example, instructions to describe the pictures to a young versus older experimenter.  81  CHAPTER 4 FACTORS R E L A T E D TO MEMORY FOR GIVEN AND BEYOND INFORMATION A secondary interest of the present experiment was to explore the relationships between verbal and perceptual ability and age-related changes in memory for given and beyond information that were observed in Chapter 3. In general, it is difficult to examine causal relationships between cognitive operations with lifespan data. For example, it is not possible to draw causal relationships from the present experiment as correlational analyses were employed. As well, strong conclusions cannot be drawn even when relationships between cognitive operations are found as it is possible that the relationships were mediated by another cognitive operation that was not investigated but is related to the cognitive operations that were investigated. For example, if a relationship between verbal ability and memory is found, it is possible that this relationship was due to their common relationship with another cognitive ability, such as fluid intelligence (Vernon, 1987; Vernon & Mori, 1992). These concerns not withstanding, it is still profitable to explore the relationships between cognitive operations as they place age-related changes in memory into a broader context, allow for theory testing, and provide some avenues for future research. The question "Are there relationships between age, verbal and perceptual ability, and performance on the obtained and aggregated items?" was addressed by this part of the experiment Because the results of Chapter 3 indicate that retrieval phase performance was related to encoding phase performance, this part of the experiment included an examination of both encoding and retrieval phase performance. Relationships between age, verbal ability and performance on items on the beyond scale were explored because previous research has shown an age-related increase in verbal ability (Baltes, 1987; Baltes, Dittmann-Kohli & Dixon, 1984; Horn, 1982; Kaufman, Kaufman-Packer, McLean & Reynolds, 1991) and Adams (1991) has  82  reported that verbal ability was related to the recall of beyond but not given information. Because beyond information requires a greater degree of inferencing than given information, it is possible that verbal ability may facilitate elaborative processing that is required for the expression or articulation of beyond information. Relationships between age, perceptual ability and performance on items on the given scale were explored because it is possible that a declining ability to perceive given information (e.g., Fozard, 1990) may restrict further processing. METHOD Participants' performance on a picture description task and on self-report measures of verbal and perceptual ability was examined. The procedure for these tasks was described in Chapter 2. Participants' performance on the picture description task was examined in Chapter 3. Of particular interest to the present research was their encoding and retrieval phase performance on the obtained and aggregated items. Verbal ability was assessed by a standardized test, the North American Adult Reading Test. Perceptual ability was assessed by self-ratings of visual and auditory ability and by the Color Vision Screening Inventory and the Hearing Screening Inventory. Measurement Tools Verbal ability. Verbal ability was measured by the North American Adult Reading Test (NAART; Blair & Spreen, 1989). The N A A R T requires participants to pronounce increasingly more difficult words (e.g., cellist, synecdoche) and is considered to be a measure of verbal ability; the correlation between N A A R T scores and performance on the WAIS-R VIQ is .83 (Blair & Spreen, 1989). The N A A R T has an interscorer reliability of .99 and internal consistency (coefficient alpha) of .94 (Blair & Spreen, 1989). Auditory ability. Auditory ability was assessed by the Hearing Screening Inventory (HSI; Coren & Hakstian, 1992) and by a self-rating. The HSI is an 11-item questionnaire that  83  asks participants to rate their ability to discriminate sounds on a 5-point scale, with 1 indicating never and 5 for always (e.g., Can you hear the telephone ring when you are in the same room in which it is located?). The HSI has been shown to be a reliable and valid measurement tool, with an internal consistency coefficient (alpha) of .89, a test-retest stability coefficient of .88, and a correlation of .81 with pure-tone hearing thresholds (Coren & Hakstian, 1992). One participant did not complete this test because of time constraints and an HSI score could not be computed for another three participants because of missing data. Auditory ability was also measured by asking participants to rate the question, "How is your hearing with corrective aid?", on a 4-point scale with 1 indicating poor, 2 for fair, 3 for good and 4 for excellent hearing. Visual ability. Visual ability was measured by the Color Vision Screening Inventory (CSI; Coren & Hakstian, 1988) and by one self-rating. The CVSI is a 10-item questionnaire that asks participants to rate their ability to discriminate between colors on a scale of 1 for never to 5 for always (e.g., Do you have difficulty discriminating between yellow and orange?). The CVSI has been shown to be a valid measurement tool, with an internal consistency coefficient (alpha) of .91, and has been cross-validated with the Hardy Rand Rittler pseudo-isochromatic plates (Coren & Hakstian, 1988). One participant did not complete this test because of time constraints. Visual ability was also measured by asking participants to rate the question, "How is your eyesight with corrective lenses?", on a 4-point scale with 1 indicating poor, 2 for fair, 3 for good and 4 for excellent eyesight. RESULTS The critical dependent measures were encoding and retrieval scores on the obtained and aggregated items and participants' scores on measures of verbal and perceptual ability. Participants performance on the obtained and aggregated items were screened for univariate and multivariate outliers as described in Chapter 3. The same regression procedure that was  84  described in Chapter 3 was used to screen participants' performance on the measures of verbal and perceptual ability for univariate outliers. This procedure detected no outliers on the N A A R T , 1 outlier on the eyesight self-rating, 2 outliers on the hearing self-rating, 3 outliers on the HSI, and 5 outliers on the CVSI. Outliers were replaced with their respective upper or lower value at the 99% confidence interval. Regression analyses were employed to explore the relationships between age-related changes in performance on the obtained and aggregated items and verbal and perceptual ability, including the relationships between age and verbal and perceptual ability, verbal ability and performance on the obtained and aggregated items, and perceptual ability and performance on the obtained and aggregated items. As described in Chapter 3, regression analyses were performed on standardized scores. Regression analyses are reported by the standardized regression coefficient (fj), squared multiple correlation (R ) and F value indicating an alpha at .05, .01 or 2  .001 level of significance for a two-tailed test. Age, Verbal Ability and Performance on the Obtained and Aggregated Items This section investigated the question ~ "Are there relationships between age, verbal ability and performance on the obtained and aggregated items?" ~ with the idea that an agerelated increase in verbal ability may facilitate the expression of beyond but not given information. This question was answered by examining the relationships between age and verbal ability and between verbal ability and performance on the obtained and aggregated items. Age and Verbal Ability Consistent with previous research (Baltes, 1987; Baltes, Dittmann-Kohli & Dixon, 1984; Horn, 1982; Kaufman, Kaufman-Packer, McLean & Reynolds, 1991), it was hypothesized that verbal ability would increase with age. An age-related increase in verbal ability, as measured by the N A A R T , was investigated by regressing N A A R T scores on age and then age squared. As  85  o  NO  * *  * o o  m o  ON  Of)  tf  cn cn in o  cn m CN o  ON CO.  Al  oo  O cn CN  in  cn  c  ON  T£  cn CN  NO ON  CN  ON  r-  in  ON NO  o u O -a U  CN  •a OO  NO ON  i o  ^  cn  II  cu  Q  ON ^  I  4? O ON  cn i  cn Tf  ON  cn  ^  cn CN  O  rCN Tf  O  ©  oo  w! IT c  cn CN  00  o  NO  cn m  NO  ^  CN © cn CN NO CN cn  CN q cn  Tf" CN cn q NO CN cn  T f t-- ON" t- q in T f cn  ON  00 O cn" in Tf q cn cn  rcn d CN  oo NO, cn  CN cn  °  Tf 00 cn CN NO T f cn  cn  r~"  °  cn © q cn NO i n cn CN cn ON  oo ON 00 q i n Tf i n cn "—  Q  co  CN CN cn  ON  X  cn T f m CN 00 Tf d CN- cn CN  s  Tp 00 CN cn cn  oo  Tf o Tf CN  sS  cn  NO  m oq cn  CN  f> ' o  ©  NO  cn  Xl Tf  © Tf  ON  Tf o  o cn CN oo cn oo T f NO CN i n cn CN  O in  in cn oo Tf  © H cn c  VI  cn  NO  ^  ON  CN  00 cn o o  cn ON" © 00 T f cn i n CN CN  CN d Tf  o  00 CN cn  CN ON cn 00 i n oo cn i n T f CN cn CN  Tf  ON  NO  V  ON  o  NO  ON  NO  ON  CN  m Tf CN  CO  S  NO  00 CN  s  Q  CO  cn" CN NO cn NO <N cn "—'  c o <D 00 «•>  S  a  Q  CO  S Q  CO  "O  5 3  6  CN CN  II  E o  p >  < CU  u S  cn  "3 <|  OX)  O  •O  00  c  c 03  03  ^  CO CO  CO  s >U  •cn S3  CO  c  o  V OH *  CN  II  <Si  tj o  CN <D CN «> M < C U  IS  o o V  X)  C  m  NO  *  in o  V CL,  *  shown in Table 4.1, age was positively related to N A A R T scores (£=.2309, R =0533, 2  F(l,160)=9.00, p_<.01), and thereby confirming that there was an age-related increase in verbal ability. Although not reported in Table 4.1, age squared was not related to N A A R T scores, and thereby indicating that there was no non-linear relationship between age and verbal ability. Verbal Ability and Performance on the Obtained and Aggregated Items The relationship between verbal ability and performance on the obtained and aggregated items was investigated by regressing encoding and retrieval scores for each obtained and aggregated item separately on N A A R T scores. As shown in the first three columns of Tables 4.2 and 4.3, verbal ability was positively related to performance on most obtained and aggregated items; verbal ability was not related to encoding phase performance on the absence, event and timeframe items. Hence, these results indicate that there was a relationship between verbal ability and performance on the obtained and aggregated items. Because the results of Chapter 3 indicate that age was related to performance on items on the given scale and one item — event — on the beyond scale, and because the results of this section revealed that age was related to verbal ability, it is possible that the relationship between verbal ability and performance on these items was mediated by age. In order to rule out this possibility, encoding and retrieval scores for each obtained and aggregated item were regressed separately on age and then N A A R T scores. When compared to the previous unpartialled relationship between verbal ability and performance on the obtained and aggregated items, the relationship between performance on the obtained and aggregated items and verbal ability remained when the influence of age was partialled out (see the first three columns of Tables 4.2 and 4.3 vs. Tables 4.4 and 4.5). These results provide further support for the finding that there was a relationship between verbal ability and performance on the obtained and aggregated items as the results indicate that this relationship was not mediated by age.  Table 4.2: Relationships between verbal ability, age and encoding phase performance on each item on the given and beyond scales. NAART F ( l , 149) R a  Item Given  2  P NAART  .2354  .0527  P NAART  R  2  8.64 ** F ( l , 160)  Age AR P age a  B NAART  .2965 B NAART  -.2919  2  F(2, 148)  .0857  15.33 ***  AR  F(2, 159)  P age  2  color object passive state people position quantity shape size spatial location  .1453 .0211 .2435 .0593 .1934 .0373 .0968 .0094 .2374 .0564 .1371 .0188 .1657 .0275 .2580 .0665 .2225 .0495  3.45 10.08 ** 6.21 * 1.51 9.55 ** 3.06 4.51 * 11.41 *** 8.33 **  .2071 .3087 .2782 .1743 .2811 .2165 .1995 .3139 .2864  -.2677 -.2827 -.3673 -.3357 -.1891 -.3438 -.1466 -.2424 -.2767  .0678 .0756 .1277 .1067 .0338 .1119 .0203 .0556 .0724  11.84 *** 13 90 *** 24.32 *** 19.20 *** 5.92 * 20.47 *** 3.40 10.07 ** 13 12 ***  Beyond absence emotion event setting time frame  .1429 .2064 .0968 .2732 .1865  3.34 7.12 ** 1.51 12 91 *** 5.77 *  .1762 .1877 .0923 .3232 .2297  -.1441 .0809 .0196 -.2163 -.1870  .0197 .0062 .0004 .0443 .0331  3.26 1.04 .06 7.99 ** 5.65 *  .0204 .0426 .0094 .0746 .0347  N A A R T and age effects were computed by regression analyses. * p< .05 ** p<.01 *** p<.001  a  88  Table 4.3: Relationships between verbal ability, age and retrieval phase performance on each item on the given and beyond scales. NAART R F ( l , 157)  Age  a  Item Given  2  B NAART  .2503 .0528 B NAART  R  2  8.76 ** F ( l , 160)  B NAART  .3494 BNAART  AR  B age -.4311  2  .1674 AR  B age  2  F(2, 156) 33.50 *** F(2, 159)  color object passive state people position quantity shape size spatial location  .1446 .2398 .1644 .1258 .2918 .1878 .1692 .2633 .2129  .0209 . 3.42 .0574 9.76 ** .0270 4.44 * .0158 2.57 .0851 14.89 *** .0353 5.85 * .0286 4.71 * .0693 11.92 *** .0454 7.60 **  .2458 .3343 .2687 .2009 .3457 .2819 .2117 .3330 .3054  -.4383 -.4097 -.4521 -.3256 -.2335 -.4073 -.1837 -.3020 -.4007  .1818 .1589 .1935 .1004 .0516 .1570 .0320 .0863 .1519  36.28 *** 32.25 *** 39.48 *** 18.07 *** 9.50 ** 30.91 *** 5.41 * 16.26 *** 30.11 ***  Beyond absence emotion event setting time frame  .2322 .2174 .2745 .3144 .2007  .0539 .0472 .0754 .0988 .0403  .2744 .1960 .2462 .3615 .2380  -.1827 .0925 .1225 -.2041 -.1616  .0316 .0081 .0142 .0395 .0247  5.49 * 1.36 2.48 7.28 ** 4.20 *  9.12 ** 7 94  **  13.03 *** 17.55 *** 6.71 *  N A A R T and age effects were computed by regression analyses. *p<.05 **p<.01 ***p<.001  a  89  Table 4.4: Relationship between N A A R T and encoding phase performance on each item on the given and beyond scales when the influence of age is partialled out.  Item Given  P  age  -.2919 P  age  P  NAART AR NAART  2  F (2, 148)  .2965  .0840  14.41 ***  P NAART  AR  F(2, 159)  2  color object passive state people position quantity shape size spatial location  -.2677 -.2827 -.3673 -.3357 -.1891 -.3438 -.1466 -.2424 -.2767  .2071 .3087 .2782 .1743 .2811 .2165 .1995 .3139 .2864  .0406 .0903 .0732 .0287 .0748 .0443 .0377 .0933 .0777  7.09 ** 16.58 *** 13.95 *** 5.18 * 13.07 *** 8.11 ** 6.29* 16.90 *** 14.07 ***  Beyond absence emotion event setting time frame  -.1441 .0809 .0196 -.2162 -.1870  .1762 .1877 .0923 .3232 .2297  .0294 .0333 .0080 .0989 .0499  4.86* 5.57 * 1.29 17.84 *** 8,52 **  N A A R T effects were computed by regression analyses. * p< .05 ** p<.01 *** p<.001  a  90  Table 4.5: Relationship between N A A R T and retrieval phase performance on each item on the given and beyond scales when the influence of age is partialled out.  Item Given  P  age  -.4311 P  age  NAART AR PNAART  2  F(2, 148)  .3494  .1157  23.40 ***  P NAART  AR  F(2, 159)  2  color object passive state people position quantity shape size spatial location  -.4383 -.4097 -.4521 -.3256 -.2335 -.4073 -.1837 -.3020 -.4007  .2458 .3343 .2687 .2009 .3457 .2819 .2117 .3330 .3054  .0572 11.41 *** .1058 21 47 *** .0683 13 95 *** 6.87 ** .0382 .1131 20.84 *** .0752 14.81 *** .0424 7.18 ** .1050 19 77 *** .0883 17.50 ***  Beyond absence emotion event setting time frame  -.1827 .0925 .1225 -.2041 -.1616  .2744 .1960 .2462 .3615 .2380  .0713 .0364 .0574 .1237 .0536  12.40 *** 6.12* 10.02 *** 22.83 *** 9.12 **  N A A R T effects were computed by regression analyses. *p<.05 **p<.01 ***p<.001  a  91  Together with the results of the previous section, these results support the idea that an age-related increase in verbal ability was related to the expression of beyond information, but also indicate that, an age-related improvement in verbal ability was related to the expression of given information. It is possible that these results reflect the relationship between performance on items on the given and beyond scales and different aspects of verbal ability. For example, it is possible that vocabulary size may be important to naming given information and verbal fluency may be important to the expression of beyond information. Hence, these results encourage future research with more direct measures of the different components of verbal ability. Additional analyses were performed to examine the consequences of the findings of the present section on the findings that were obtained in Chapter 3. Thus far, the results indicate two types of relationships ~ a direct and an indirect relationship ~ between age and performance on the obtained and aggregated items. The results of Chapter 3 indicate a direct relationship between age and performance on the aggregated given item and the event item. The results of the present section indicate that there was also an indirect relationship between age and performance on the obtained and aggregated items that was mediated by verbal ability; there was a relationship between age and verbal ability and between verbal ability and performance on the obtained and aggregated items. It is possible, however, that these are not separate types of relationships, where the relationship between age and performance on the obtained and aggregated items that was observed in Chapter 3 may be due to verbal ability. In order to investigate this possibility, the relationship between age and performance on the obtained and aggregated items was examined when the influence of verbal ability was partialled out. Encoding and retrieval scores for each obtained and aggregated item were regressed separately on N A A R T scores and then age. When compared to the previous unpartialled analyses, the present results showed that age-related changes in performance on most of the obtained and aggregated  92  items remained when the influence of verbal ability was partialled out. except for one item on the beyond scale -- event (see the last four columns of Tables 4.2 and 4.3 vs. Tables 3.4a, 3.4b, 3.5a and 3.5b). For the event item, the results indicate that verbal ability accounted for an age-related increase in retrieval phase performance. This finding provides further support for the conclusion that was drawn in Chapter 3 that there was no age-related change in memory for beyond information. Together, the results of Chapter 3 and the present section indicate that there was a direct relationship between age and performance on items on the given scale and an indirect relationship between age and performance on items on the given scale that was mediated by verbal ability. As well, these results indicate that there was a relationship between age and verbal ability and between verbal ability and performance on items on the beyond scale, but there was no relationship between age and performance on items on the beyond scale. Summary This section explored the relationships between performance on the obtained and aggregated items, verbal ability and age with the idea that an age-related increase in verbal ability may facilitate the expression of beyond but not given information. This idea was explored by examining the relationships between age and verbal ability and between verbal ability and performance on the obtained and aggregated items. The results confirmed that there was an agerelated increase in verbal ability; when N A A R T scores were regressed on age, the results indicate a significant positive relationship between age and verbal ability. As well, the results showed that there was a positive relationship between verbal ability and performance on the obtained and aggregated items; when encoding and retrieval scores for each obtained and aggregated item were regressed on N A A R T scores, there were significant positive relationships for most obtained and aggregated items. Because the present results indicate a relationship between age and verbal ability, and because the results of Chapter 3 indicate a relationship between age and performance  93  on the obtained and aggregated items, further analyses were performed in order to rule out the possibility that the observed relationship between verbal ability and performance on the obtained and aggregated items was not due to a common relationship with age. The results indicate that the relationship between verbal ability and performance on the obtained and aggregated items was not due to a common relationship with age; the relationship between performance on the obtained and aggregated items and verbal ability remained when the influence of age was partialled out. Hence, together, the results indicate that an age-related improvement in verbal ability facilitated the expression of beyond information, as well as, facilitated the expression of given information. Age, Perceptual Ability and Performance on the Obtained and Aggregated Items This section investigated the question — "Are there relationships between age, perceptual ability and performance on the obtained and aggregated items?" — with the idea was that an agerelated decline in ability to perceive given information may restrict further processing. This question was answered by examining the relationships between age and perceptual ability and between perceptual ability and performance on the obtained and aggregated items. Age and Perceptual Ability Consistent with previous research (e.g., Fozard, 1990), it was hypothesized that perceptual ability would decline with age. Perceptual ability was measured by two measures of auditory ability, the HSI and a self-rating of hearing, and two measures of visual ability, CVSI and a self-rating of eyesight. A n age-related decline in perceptual ability was examined by separately regressing each of the perceptual ability scores on age and then age squared. As shown in Table 4.1, age was negatively related to auditory ability scores, and hence, as expected, there was an age-related decline in auditory ability. The results also showed that age was not related to visual ability scores, and thereby indicating no age-related change in visual ability. As  94  well, although not reported in Table 4.1, age squared was not related to any of the perceptual ability scores, and thereby indicating no non-linear relationship between age and perceptual ability. Together, these results provided mixed support for the hypothesis as an age-related decline in perceptual ability was found for auditory but not visual ability. Because these measures asked participants to rate their visual and auditory ability with corrective aid, it is possible that auditory ability was a more sensitive measure than visual ability of age-related declines in perceptual ability as auditory difficulties may go undetected whereas visual problems are commonly detected and corrected. Auditory Ability and Performance on the Obtained and Aggregated Items The relationship between auditory ability and performance on the obtained and aggregated items was explored by separately regressing encoding and retrieval scores for each obtained and aggregated item on each of the auditory ability scores. Hearing Screening Inventory (HSI) The results indicate that performance on most of the obtained and aggregated items was not related to HSI scores. Encoding phase performance on one item — shape — on the given scale was positively related to HSI scores, and thereby indicating that enhanced performance was related to better hearing ability (shape item: [^,=.1955, R =.0383, F(l,156)=6.20, p<.05). 2  Retrieval phase performance on two items — setting and timeframe — on the beyond scale were inversely related to HSI scores, and thereby indicating that enhanced performance was related to poorer hearing ability (setting item, B_ si =-.1720, R =.0296, F(l,156)=4.76, p<.05; timeframe 2  H  item, fJHsi =-.1986, R =.0394, F(l,160)=6.40, p<.05). 2  Self-Rating of Hearing As shown in Tables 4.6 and 4.7, the results indicate that the self-rating of hearing was related to performance on some of the obtained and aggregated items. Retrieval phase  95  Table 4.6: Relationship between self-rating of hearing and encoding phase performance on each item on the given and beyond scales. Hearing R F ( l , 149) P hearing a  Item  2  .1340  .0180  P hearing  R  color object passive state people position quantity shape size spatial location  .0522 .0809 .1003 .0766 .0769 .0627 .0087 .0276 .0660  .0027 .0065 .0101 .0059 .0059 .0039 .0001 .0008 .0044  Beyond absence emotion event setting time frame  .0904 .0067 .0362 .2584 .1546  .0081 .0000 .0013 .0667 .0239  Given  2  2.72 F ( l , 160) .44 1.05 1.63 .94 .95 .63 .01 .12 .70 1.31 .00 .21 11.45 *** 3.92*  Self-rating of hearing effects were computed by regression analyses. *p<.05 **p<.01 ***p<.001  a  96  Table 4.7: Relationship between self-rating of hearing and retrieval phase performance on each item on the given and beyond scales.  Item Given  color object passive state people position quantity shape size spatial location Beyond absence emotion event setting time frame  Hearing a R F ( l , 155) P hearing 2  .2247  .0505  8.24 **  P hearing  R  F ( l , 160)  .1667 .1852 .1350 .1160 .1009 .0717 .0509 .0861 .1.549  .0278 .0342 .0182 .0134 .0101 .0051 .0025 .0074 .0240  .1259 .0389 .0821 .2151 • .2427  2  4.57 * 5.68 * 2.97 2.18 1.64 .82 .41 1.19 3.93 *  .0158 2.58 .24 .0015 .0067 1.09 .0463 7.76 ** .0589 10.01 **  Self-rating of hearing effects were computed by regression analyses. * p< .05 ** p<.01 *** p<.001  a  97  performance on the aggregated given item was positively related to the self-rating of hearing and performance on two items on beyond scale — setting and timeframe — were positively related to the self-rating of hearing, and thereby indicating that enhanced performance was related to better hearing ability. Because the results of Chapter 3 indicate that age was related to retrieval phase performance on the aggregated given item and because the earlier results indicate that age was related to the self-rating of hearing, it is possible that the relationship between the self-rating of hearing and retrieval phase performance on the aggregated given item was mediated by a common relationship with age. In order to rule out this possibility, the retrieval score on the aggregated given item was regressed on age and then the self-rating of hearing. When compared to the previous unpartialled analysis, the R due to age declined when the influence of hearing 2  ability was partialled out (see Tables 3.4a vs. Table 4.7), but an age-related decline in retrieval phase performance on the aggregated given item remained when the influence of hearing ability was partialled out (&hearin =-1076, £ ^ = - . 3 1 0 3 , R =.0825, F(2,154)=14.65, e<.01 V S . Table 3.4a). 2  g  Together, these results indicate that age was related to retrieval phase performance on the aggregated given item and to the self-rating of hearing, and performance on the aggregated given item was related to the self-rating of hearing. Visual Ability and Performance on the Obtained and Aggregated Items Although the previous results indicate a relationship between age and auditory but not visual ability, this section further explored the possibility that visual ability is related to performance on the obtained and aggregated items as it places performance on the obtained and aggregated items into a broader context. The relationship between visual ability and performance on the obtained and aggregated items was explored by separately regressing encoding and retrieval scores for each of the obtained and aggregated items on each of the visual ability scores.  98  Color Vision Screening Inventory (CVSI) The results indicate that performance on all of the obtained and aggregated items was not significantly related to CVSI scores. Hence, these results failed to find a relationship between color vision and performance on the obtained and aggregated items. Self-Rating of Eyesight The results indicate that performance on most of the obtained and aggregated items was not related to the self-rating of eyesight. Encoding phase performance on the shape item and performance on the emotion item were inversely related to the self-rating of eyesight, and thereby indicating that performance was related to poorer eyesight on these items (encoding phase performance on the shape item: j3_evesight-=  1634, R =.0266, F(l,160)=4.38, D<.05, encoding phase 2  performance on the emotion item, fj esight=--1732, R =.0300, F(l,160)=4.95, p_<.05, and retrieval 2  eV  phase performance on the emotion item, fievesight=--2078, R =.0432, F(l,160)=7.22, p_<.01). 2  Summary This section explored the relationships between age, perceptual ability and performance on the obtained and aggregated items with the idea that a declining ability to perceive given information may restrict further processing. This idea was explored by examining the relationships between age and perceptual ability and between perceptual ability and performance on the obtained and aggregated items. There was mixed support for an age related decline in perceptual ability, as assessed by two measures of auditory ability, the HSI and a self-rating of hearing, and two measures of visual ability, the CVSI and a self-rating of eyesight. The results indicate an age-related decline in auditory but not visual ability; when each of the perceptual ability scores was regressed separately on age, there was a significant negative relationship between age and auditory ability scores and no relationship between age and visual ability scores. As well, when encoding and retrieval scores for each of the obtained and aggregated items were  \  99  regressed separately on each of the perceptual ability scores, the results failed to show a relationship between perceptual ability scores and performance on most of the obtained and aggregated items. Together, these results provided no support for the idea that an age-related decline in ability to perceive given information is related to performance on the obtained and aggregated items. DISCUSSION The present results indicate that an age-related improvement in verbal ability facilitated the expression of given and beyond information. These results encourage future research with other cognitive abilities as well as more direct measures of the different components of verbal ability. As mentioned previously, it is possible that the relationships between age, verbal ability and performance on the obtained and aggregated items may be due to other cognitive operations that were not investigated. For example, it is possible that the relationship between verbal ability and performance oh the obtained and aggregated items was mediated by another cognitive . operation, such as fluid intelligence, which has been shown to be related to verbal ability and to memory (Vernon, 1987; Vernon & Mori, 1992). As well, it is possible that performance on items on the given and beyond scales may be related to different subcomponents of verbal ability. For example, it is possible that vocabulary size may be important to naming given information and that verbal fluency may be important to the expression of beyond information. The results failed to support the idea that an age-related decline in perceptual ability is related to further processing given or beyond information. Although previous research has shown an age-related decline in auditory and visual ability (e.g., Fozard, 1990), the present results found the expected age-related decline for auditory but not visual ability. As well, the results failed to find a relationship between perceptual ability and performance on most of the obtained and aggregated items. It is possible that an age-related decline in perceptual ability was  attenuated in the present experiment because self-report measures of perceptual ability were employed. Future research with, for example, behavioral measures of perceptual ability, are required to further investigate the possible relationships between age, perceptual ability and memory.  101  CHAPTER 5 M E M O R Y FOR FEMININE AND MASCULINE INFORMATION A secondary interest of the present experiment was to survey changes in memory for feminine and masculine information across the adult female lifespan. Gender studies in social psychology define feminine and masculine roles by those behaviors and attitudes that have been empirically demonstrated as characteristic of young women and men, and studies on gender and aging entail tracing these gender stereotypes across the lifespan (for a review see Turner, 1982). It is possible that feminine and masculine roles may direct attention toward different types of information, and in turn, processing this information may contribute towards further shaping feminine and masculine attitudes and behaviors. The present experiment explored the possibility that maturational changes in feminine and masculine roles may show age-related changes in the types of information that is encoded and retrieved about a target stimulus. Following from social psychology and gerontology, feminine and masculine information was defined according to gerontological and psychological research findings that suggest certain types of information may be more relevant to young women and men, respectively, and then traced memory for feminine and masculine information across the female lifespan. Age-related changes in memory for feminine and masculine information were traced across the adult lifespan for women but not men because there were fewer older male participants. Although previous research has investigated only young adults' memory for stimuli related to stereotyped gender roles (Martin & Paulhus, 1985; Signorella, 1992), age-related changes in memory for feminine and masculine information were explored because of two sources of indirect evidence. First, age-related changes in memory for feminine and masculine information were explored because gerontological research has shown maturational changes in feminine and masculine behaviors and attitudes. These studies reveal two types of  102  developmental patterns: gender blurring and gender reversal. Gender blurring refers to the disappearance of gender differences in behaviors and attitudes, where people become more androgynous in their behaviors as they age (Sinnott, 1984). Gender reversal refers to a cross over in feminine and masculine behaviors and attitudes, where there is an emergence of genderopposite attitudes and behaviors with age (Gutmann, 1977, 1987; Labouvie-Vief, 1996; Lowenthal et al, 1975; Turner, 1982). This part of the experiment was motivated by these sources of evidence, but was not designed to distinguish between these two types of developmental patterns. Second, age-related changes in memory for feminine and masculine information were explored because several studies provide indirect evidence that young women and men may focus on or show preference for different types of information. For example, a descriptive study of the meanings of personal possessions reveals women are more likely than men to indicate social or interpersonal aspects of desired gifts and treasured possessions (Kamptner, 1991). Because women are attracted to the social or interpersonal qualities of personal possessions, they may also show a similar preference for encoding and retrieving information regarding people and personal experiences. Similarly, scores on psychometric tests of spatial ability indicate superior performance for males than females (Feingold, 1988; Voyer & Voyer, 1995). Because men show superior spatial ability, they may be more inclined to encode and retrieve information regarding spatial location and position. And finally, a qualitative analysis of adults' verbal reports of stories reveal that females make more references to emotion than males (Adams, 1991). This finding suggests that women may encode and retrieve more information regarding emotion. Together, these studies provide indirect evidence that young women and men may show preference for different types of information, where men may be more inclined to encode and  103  retrieve information regarding position and spatial location whereas women may be more likely to encode and retrieve information regarding emotion, events and people. METHOD This part of the research project re-examined a portion of the data that were transcribed, coded and analyzed for Chapter 3. The data set in Chapter 3 included men and women's performance on obtained and aggregated items, and this part of the research project employed only women's performance on some of the items. Of the data on female participants, some of the obtained and aggregated items were re-coded and analyzed as items on the feminine and masculine scales and as the neutral item. Items on the feminine and masculine scales were identified by research findings that suggest certain types of information may be more relevant to young adults of one gender than the other. The neutral item was composed of highly intercorrelated items that were not included on either the feminine or masculine scales and served as a baseline from which to compare performance on items on the feminine and masculine scales. Items on the Masculine and Feminine Scales Gerontological studies on feminine and masculine behaviors define feminine and masculine roles by those behaviors and attitudes that have been empirically demonstrated as characteristic of young women and men and trace these gender stereotypes across the lifespan (for a review see Turner, 1982). Similarly, the present experiment defined feminine and masculine information according to previous gerontological and psychological research findings, as well as findings from the present experiment, that suggest certain types of information may be more relevant to young adults of one gender than the other, and then traced memory for feminine and masculine information across the adult female lifespan. Items on the feminine and masculine scales were identified according to previous research findings that suggest young men and women may show preference for different types of  104  information. As discussed in the previous section, these studies indicate that young men may be more inclined to encode and retrieve information regarding passive states, position and spatial location and women may be more likely to encode and retrieve information regarding emotion, events and people (Adams, 1991; Feingold, 1988; Kamptner, 1991; Voyer & Voyer, 1995). Together, these findings indicate that the feminine scale includes the people, emotion and event items and the masculine scale includes the position and spatial location items. Items on the feminine and masculine scales were also identified by inspecting gender differences in young adults' performance on the obtained and aggregated items. One-way A N O V A ' s compared women and men between the ages of 20 and 39 years for their encoding and retrieval phase performance on the obtained and aggregated items. As shown in Table 5.1, there was a significant difference due to sex on four items. Men provided more references than women to the absence, passive state, position, and spatial location items, and hence, these items were identified as items on the masculine scale. Together, previous research and the present experiment identified 5 items on the masculine scale and 2 items on the feminine scale. Two of these items — position and spatial location — were identified by both sources and thus provided converging evidence that these items belong on the masculine scale. Neutral Item The neutral item is an aggregated item composed of highly intercorrelated obtained items that were not re-coded as items on the feminine and masculine scales. Table 5.2 shows the intercorrelations between items that were not re-coded as items on the feminine and masculine scales. Performance on four items — object, shape, size and setting — that were not re-coded as items on the feminine and masculine scales showed strong, positive intercorrelations, and hence allowed these items to be aggregated into a neutral item.  Table 5.1: Gender difference in performance on each item on the given and beyond scales for young adults between 20 and 39 years.  Women (n=22)  Study Men (n=18)  M SD M SD M SD M SD M SD M SD M SD M SD M SD  31.45 (23.34) 120.27 (54.12) 11.59 (5.93) 12.54 (4.31) 5.73 (5.21) 25.68 (10.48) 2.1.8 (1.99) 10.82 (7.14) 50.59 (25.46)  38.61 (21.79) 147.83 (49.07) 14.05 (5.18) 12.22 (2.95) 11.27 (5.14) 28.27 (8.18) 2.83 (2.43) 14.65 (8.41) 70.17 (23.57)  M SD M SD M SD M SD M SD  2.27 (2.05) 5.06 (3.48) 3.37 (4.25) 8.73 (2.98) 2.09 (1.26)  3.56 (2.58) 4.61 (3.48) 2.55 (1.85) 10.50 (3.32) 2.55 (1.24)  Item  Given color object passive state people position quantity shape size spatial location Beyond absence emotion event setting time frame  F(l,38)  a  .98 2.78 1.91 .07 11.33 ** .73 .87 2.43 6.25 *  Women (n=22)  Test Men (n=18)  29.82 (19.03) 99.09 (37.53) 10.82 (3.88) 11.00 (2.94) 5.36 (4.88) 20.22 (7.41) 2.14 (2.36) 10.00 (5.71) 48.72 (23.25)  32.27 (17.69) 121.77 (38.44) 13.17 (4.96) 11.55 (3.23) 10.77 (5.36) 29.00 (10.38) 3.11 (2.51) 12.94 (7.27) 64.42 (23.41)  2.36 (2.42) 3.59 (2.38) 1.64 (2.43) 6.68 (2.69) 1.54 (1-18)  3.89 (2.49) 3.55 (3.01) 1.22 (1.31) 8.00 (2.84) 2.00 (1.19)  3.09 .46 .58 3.11 1.34  Gender differences were computed by one-way A N O V A ' s . * p_< .05 ** p_< .01 * p = .0579  a  F ( l , 38)  .17 3.54 2.82 .32 11.14 ** 9.70 ** 1.59 2.05 4.48 *  3.82 * .00 .43 2.25 1.45  a  106  Table 5.2: Intercorrelations between encoding phase, retrieval phase and encoding-retrieval match performance on each item on the given and beyond scales (n=162).  color Study object shape size setting time frame Memory object shape size setting time frame Match object shape size setting time frame *p_<.05  .86 *** .65 *** 59 * * *  .04 .09  object  Item shape  .68 *** .68 *** ,12 .17  67 *** .15 .07  .24 * .21 *  .43  .60 *** g2 *** 41 *** .23 *  64 *** .27 ** .17 .  42 *** .21 *  .50 ***  .36 *** .14  .36 ***  size  setting  gg *** 57 #** 77 * * *  22 *** • 14 g2 *** 47 ##*  69 *** 27 **  .08 **_p=.01  49 *** 72 *** 40 *** .16 ***p_<-001  57 **#  2\ *** .18*  107  RESULTS The critical dependent measures were feminine and masculine scores which was computed as scores on items on the feminine and masculine scales divided by scores on the neutral item. Tables 5.3, 5.4, and 5.5 present encoding, retrieval and match scores for the items on the feminine and masculine scales and for the neutral item by decadel group. Regression analyses were employed to explore age-related changes in memory for feminine and masculine information. As described in Chapter 3, regression analyses were performed on standardized scores. Regression analyses are reported by the standardized regression coefficient (JV), squared multiple correlation (R ) and F value indicating an alpha at 2  .05, .01 or .001 level of significance for a two-tailed test. Age-related changes in memory for feminine and masculine information was explored by examining encoding phase, retrieval phase and encoding-retrieval match performance on items on the feminine and masculine scales. Encoding phase performance was included because the results of Chapter 3 indicate that memory depends on the overlap between encoding and retrieval processes, and encoding-retrieval phase performance was included as it is a measure of retention. Feminine and masculine scores were regressed separately on age and then age squared. Study Performance As shown in Table 5.6, there were no age-related changes in encoding phase performance on most items on the feminine and masculine scales; an age-related decline in performance was found for one item on the masculine scale ~ quantity. As well, age squared was not related to performance on any items, and thereby indicating no non-linear relationships between age and performance.  3  O  r-o  ON  —  rx  oo  DC  <  cn © o  r-- •  ON  m oq  NO ON  o o  CN  o  Tt  O  rcn O  r-.  o o  —  o o  o o op  p  >n en  NO  _  CN  CN  O  p  p  cn O  cn  o  in O  oo  CN  cn  OJD  <  cn cs  tf  CO-  p  m  r~ o o o  CN  Tt  NO  in  c^cs  en O  fx  oo  O  NO  o o  cn  CN  (x-  Tt  cn o  T—1  _  *in  CN O  CN  00  oo CN o  r~ oo  *  NO  ON  o o o  o  o  o  00  cn O  o o  rx-  Tt  ON  en  as  o  •  ON  o o  NO o  en  NO NO  ON  Tt  cs  as  ©  QO  o  fx P cs O  Al Cn M  O  C  in  O  ON  "  £  2  ©  u  o  en  £ 2  II  o  NO rx  rx  cn vo  NO  Tt .  ' ©  en  00 oq  CN  c  Tt  w  cn cs  P Tt  II c  £ cs  en  NO NO  ON  O in Tt  CO  in  Z  fx" oq  O  CN  en  Tt  d  ON Tt  in  m  NO cs CN CN  Tt  en  Tt  O O ON" o  en i n CN CN d  en in  oo r-  ON  NO  "—'  o  Tt  CN NO CN CN  en en  en NO" cn cn 00 T t T t in d en  e .2 o fa E  Q CO  O  in  O  Tt Tt  en  CN ON cs  NO  rx" CN NO rx o cs ON rx NO i n CN CN ON oo CN Tt CN  ON en T t NO p CN p en en i n i n CN  oo oq  o NO O in fx- —<  Tt  oo in Tt  CN  en  ON Tt  Tt Tt  en  CN" m  o  d  oo rx O  in  00 r^ m i n ON Tt  o  O  w  o  en  C N — H c n i n r x N O N o c N N o ' d  Q  CO a» B  «-> o  r-  -  o  CN  w  CO  CN" O Cn CN p ON NO  rCN T t NO NO ON_ oo i n en CN CN CN ON rx  o 00 00 in o en en en T t T t ,  Q  ©  in CN en cn  CN en o ON" o NO" oo 00 CN CN en OO o rx en en ON NO en CN CS* CN ON T t , NO in, ON T t CN  o" NO NO in en  o NO rx  ON  d CN  00  Tt  Tt  r ^ 00 fx" o en CN  Tt,  in  O © rx r - O o NO en cn o NO O en NO CS NO cs cs d NO NO in in rx ON, rx CN en  en  in  NO T f  B  3  as  fx^ T t Tt  n cn > in ON in cn NO Q  c  E  O oq rx  CN ir,  cs  o  x^  NO  oq  II c  o  o m NO ON 00 p CN CS p NO T t en en d en CN  Cn ly-)  Q  Tt  m NO en en NO in O m in Tt,  m  IT)  o  3 ^  "3 «  oo  NO CN NO T t ,  VI  O a  u oo  X)  ^  w  w  Q  Q  Q  CO  CO  CO c  > 00 03  o •)->  00 O  jsi ^  un  CN  Q CO  2?  C •S  03  3  cr  B • 73 .2 03  109  c  o o  * *  o p  CN NO  Tt  r-'  NO NO  Tt  Tt  q  00 ON  CN  «—i .  *-  1  6/0 Ml  o  o o o o  *NO  ON  NO 00  o  CN  Tt  ON NO CN O  NO  00  i-*  Tt  o NO  ©  ON V©  Tt  CN  NO ~-  »—i  2 o O u  O "3  « u  ©  coN  Tt  in o o  O O  NO CO oo O  ON ©  in  >  CN ©  NO  NO  in r - ON <N NO  CO  CN  m  O  CN 0 0 —«  in  00  o o  *  NO NO  O  NO  CO  ©  o  ©  N O Tt  in  oo  ON  o  CN  o  CN NO m o  in  ©  o r— CO CN  m CN CN  © C N O T t O c N O — © m N O i n q c N O c N N O r t C N c o ^ ^ c ^ c N ^ T t ^ c o o o ^ d i n  —<  00  m  in in  CN CN  CN CN o CN CN d CN NO ©  00  *CN  q  © ©  NO  00  CN  o  ©  o o  r> T t CN m in Tt  CO T t co ON i n CN i n m i n CN CN d co NO i n  Tt  NO  r ~ NO NO CN NO ON T t 00 i n CO  ©  in  ON  oo  ON  ON,  r-  ©  © ©  CN in CN  CN CO  Tt  w  ON  in in  -  00 NO  O  CO CN ON  ON  ©  Tt,  CN oc CN CN  Tt  O  in  NO  in  ©  NO NO  ON  ON  CN in m in CN CN in Tt  ©  NO NO  OO CN TF  NO  NO  r-- © in  ON  ON  ON 00  CN CO  NO  OO m in CO NO  ©  CO  ON  CN  ©  CO  i n co 0 0 i—< r-« ON — ; NO CO CN CO CN CN co d CN  ©  •—^  Tt  CN T t  CO CO OO ON CN NO T t CN CO CN CN ON CN i n CN  0 0 oo 00 CO © CN NO oq CN CN O N co CO CN, ON i n co T t CN  Tt  i—(  ON  CN i n  ON  NO  00  Tt  , ^ ©  C O cn cn  , ^ ON 1—» o m T t T t ON Tt i n T t , - H i-* d  TT  <ii c  CN  o  CN ©  o  _  CN  ON  ©  oo 1—1 CN  —<  o o  00  co co ON ON m co CN T t NO T t CN CN ON co  CN co  4?©T'  CN q  m o  NO  ON  (13  It  NO  NO — i in Tt 0 0 ON O CO  © c Q in w <u  ©  NO NO CO  «n  N©  II  00  Tt  Tt  '  O  NO  oq © co CN ON in  CN II C  ON  t>  00  Tt  oq oq oo d co  ©  NO  1  in  O  f: in  CN  I—  CN  o  Al in M  CO  CO  o  O  m o o  Tt  Tt  CD-  •  ©  CN  tf  CO  Tt  NO o  #  ON  Pi  "  CN  -a a, £ o o  !H  ON  O  ,<U  ON CN CN r ;  VI  O 00  NO 00  oo r-  d co  ^  i n ON CO — CN NO' CN ON NO — T t CN T t , CN ^ CN CO  S  CU  •tmi  s  cu  v  2 Q 2 Q ^ Q  CO  CO  CO  CO  CU  u  CO ON oo i n r CN NO CO NO r - i n q ^ oo ON q CN CN ci * N h '—^ "—' cN — ' i n CN  .2 •a S ° 3  Q  a c > (U  "a, o <u Cu  co  cu  u cn  Q  CO  CO  a  Q CO  CO  O  Gt in (U  03  cu  cn  cn  O CU  c  3  03  3 cr  3  V,  00 03  *  "O  C  o c  a3  Q  2 P  d  cn  C  c  03  5  00  O —  <  © V  ,  ol  110  ) -a c o3  oo  <  cn CN  r-x  CN  cn m  tf <  CN  o  ON  ON  O o  o  en  Tt  e n CN  NO  OO  oo oo  O  o  in en  e On  '—I  1—<  O  o  * en en NO  o  Tt  1  CN  o o  CN  en  en  cn Tt  00  m  ON  Tt  O  CN  U4  Pi-  in en  CN  ON  o cn o  Tt  r~  ON  i' Al  Cn  c  o  o Tt rx  o M  tx  O  rx  •  o  ON  r~  Tt,  00  o  oo  Tt  O  00  5  O CN O o  in  —i  en  m  en  q  en CN  o  o CN  q  !*  NO  O Tt  o  Tt  o q o o o  en in o  Tt Tt  en oq N O T t en en •—< - J oo - J  NO  •—<  Tt  o o d  CN ON O  ,  i n CN  OS  CN CN  NO  NO ON  Tt  ^ ^  Tt  o  CN  rx  Tt  q  O CN  m  ca.  oo  en en  o  CN  <  *  o d  o NO  NO CN  ON  00  o  o in o  in  NO  in CN CN r  fx00  O  q  i'  i'  *ON  ^  CN i'  o  NO  rx o Tt in o o o i n NO r x NO m in o oo 0 0 o O N w rx Tt, r x r^ T t CN  en  NO  W  cn CN  O  ON  CN  ,—;  m  ON  NO  in  rx  ON  o Tt  oo en CN  NO  00  en  ON  NO  ON  CN  q  _:  x^  O r x en r x CN en oq —• T t i n i n — CN r x rx  NO  rTf  oo  NO  en CN ON i n  tx.  2 ©  cs w  Q  /-v  *-H  en m — ' CN  , CN  © NO  C  w  II c  § cn  4?© '  II  3  CN  , CN  cn  rm  CN  C3N  s  ON  Tt  rx  o  CS  Tt  rx  51  00  NO CN,  in  cn Cn ON oq oq CN  P  i n oo N O o O N O CN T t r x oo q CN oo T t N O N O N O T t i n N O en CN CN O N i n _: CN  oo  fx"  r-;  OO O  ON  00  c  S  to  (U  c  ~Cl,  >  P-  ci>  O <D  ^  rx!  Tt  00  en  CN o ON q  c  C  CN  CN  rx cs.  o cn m CN cn  00  Tt  en en  m cn m en CN 0CN0  NO ON cn  o °^  00  Z  oq  m O en Co rx NO r-- O  —i Tt  »—;  Cn in m  en  m CN  « u  en  CN  en  CN oo  NO  00  q  in rx  <—;  CN oo o m q CN in T t ~ H — ' 0 0 CN  Tt  ON  Tt  en  CN CN CN  NO  NO en  o cn ON  oo  Tt  CN CN  CN O CN  NO  ' ©  ON  SD  0\ N©  "3  0  o CN NO ON  <—*  Tt  cn o i  cs  rx en h  d  CN oo  ON ON in en w CN  CN d -H  oo CN NO" in in  in  Co 0 0 Cn en Tt en rx d ON en  in  o Cn 0 0 en o cn en m in n oq in ON q in e NO en en in NO CN in  , ^  , ^  ,  en en O ON rx T t m ON oq T t CN ON NO CN q 0 0 en CN CN CN cn in rx cs  0  0  0  >  (U  03  oo  0 C O  o PH  0  0  0  0  0 C  c 03  03 cj Q< o oo —  Table 5.6: Influence of age and age squared on encoding phase performance on each item on the feminine and masculine scales (relative to performance on the neutral item). Age  Age  3  (3  R  Feminine emotion event people  .1642 .1162 .0079  .0269 .0135 .0000  3.10 1.53 .00  .0133 .0102 .0050  .03 1.15 .55  Masculine absence passive state position quantity spatial location  .0567 -.1526 .0710 -.3169 -.0849  .0032 .0233 .0050 .1004 .0072  .36 2.67 .57 12.50 *** .81  .0002 .0001 .0006 .0118 .0069  .01 .01 .07 1.47 .77  2  F(l,112)  AR  2a  Age and age squared effects were computed by regression. *** p_<.001  a  2  F(2,lll)  112  Memory Performance As shown in Table 5.7, no age-related changes in retrieval phase performance were found for any of the items on the masculine scale; performance on items on the masculine scale was not related to age or age squared. By contrast, an age-related increase in performance was shown for two items on the feminine scale ~ emotion and event. Age squared was significantly related to performance on the event item, and thus indicating a non-linear relationship between age and performance on this item. Encoding-Retrieval Match Performance As shown in Table 5.8, there were no age-related changes in encoding-retrieval match performance on most items on the masculine scale; one item on the masculine scale — quantity — showed an age-related decline in performance. Age squared was not related to any items on the masculine scale, and thereby indicating no non-linear relationships between age and performance on these items. By contrast, an age-related increase in performance was found for two items on the feminine scale — emotion and people. As well, age squared was related to performance on the event item, and thereby indicating a non-linear relationship between age and performance on this item. Summary Together these results indicate there were generally no age-related changes in encoding phase performance on items on the feminine and masculine scales, but divergent age-related changes in retrieval and encoding-retrieval match performance were found. There was an agerelated increase in retrieval and encoding-retrieval match performance on items on the feminine scales, whereas there was generally no age-related change in performance on items on the masculine scales.  113  Table 5.7: Influence of age and age squared on retrieval phase performance on each item on the feminine and masculine scales (relative to performance on the neutral item). Age  Feminine emotion event people Masculine absence passive state position quantity spatial location  Age  3  p  R  .2797 .2549 .1352  .0782 .0649 .0182  9.51 ** 7.78 ** 2.09  .0195 .1223 .0034  2.41 16.71 .39  .1188 -.0696 .1372 -.0709 -.1395  .0141 .0049 .0188 .0050 .0194  1.60 .57 2.15 .56 2.22  .0033 .0010 .0002 .0072 .0131  .38 .10 .02 .80 1.50  2  F(l,112)  AR  Age and age squared effects were computed by regression. **p_<.01 ***E<.001  a  23  2  F(2,l 11)  1 14  Table 5.8: Influence of age and age squared on encoding-retrieval match performance on each item on the feminine and masculine scales (relative to performance on the neutral item). Age  Feminine emotion event people Masculine absence passive state position quantity spatial location  Age  3  AR  R  .2338 .1282 .2029  .0546 .0164 .0412  6.48 * 1.87 4.81 *  .0175 .0352 .0086  2.10 4.12 * 1.00  .0324 -.1089 .0645 -.1920 -.1814  .0010 .0118 .0041 .0368 .0329  .11 1.34 .47 4.28 * 3.81  .0059 .0006 .0067 .0104 .0037  .66 .07 .75 1.21 .42  Age and age squared effects were computed by regression. * rj_< .05  a  F(2,lll)  P  2  F(l,112)  2 3  2  115  DISCUSSION This part of the research project revealed that women showed no age-related changes in the encoding of information and divergent age-related changes in the retrieval and retention of feminine and masculine information, where there was an age-related increase for feminine information and no age-related change for masculine information. These results can be interpreted by a developmental approach to schema theory which states that schema structure may change with maturation through restructuring and tuning (Rumelhart & Norman, 1978) and thereby influence the types of information that is encoded and retrieved about target stimuli. The finding that there were no age-related changes in encoding of feminine and masculine information suggest that by young adulthood, schemas are already structured or organized in such a way so as to store both feminine and masculine information and that there is no further reorganization with age that would influence age-related changes in encoding of feminine and masculine information. The finding that there were divergent age-related changes in retrieval and retention of feminine and masculine information suggest that there may be age-related changes in tuning. It is possible that certain associative pathways are strengthened because socialization encourages women to provide feminine rather than masculine information. This communicative style may have been further encouraged in this experiment because of a female experimenter; it is possible that female participants may be more likely to provide feminine information to a female rather than male experimenter. Hence, the findings from the present experiment encourages further research with participants and experimenters of both sexes in order to gain a better understanding of age-related changes in memory for feminine and masculine information.  116  CHAPTER 6 GENERAL DISCUSSION This chapter includes five sections. The first section reviews the motivation and research objectives and the main hypotheses and findings of this experiment. The second section discusses the theoretical implications of these findings and the following section presents some factors that limit the generalizability of these findings. The fourth section presents some avenues for future research on a more complete understanding of memory and aging. The final section summarizes the general and specific contributions of the present experiment. MOTIVATION, MAIN OBJECTIVES AND FINDINGS In North America, older persons are viewed as a population in need of resources to assist with declining mental and physical health, whereas in other cultures, older persons are considered as information resources who convey valuable personal experiences and wisdom through storytelling (Blum-Kulka, 1993; Mergler & Goldstein, 1983; Miller, Wiley, Fung & Liang, 1997; Minami & McCabe, 1995; Ramelli & Stella, 1993; Wallace, 1992). This discrepancy between different cultural perspectives on older persons' memory systems is paralleled in the empirical literature, for example, a decline in fluid intelligence has been shown to be accompanied by an increase in crystallized intelligence (Baltes, 1987; Baltes, DittmannKohli & Dixon, 1984; Horn, 1982; Kaufman-Packer, McLean & Reynolds, 1991). Together, cultural and scientific perspectives present a dynamic and complex picture of the aging memory system which involves the coordination of declining and improving mental operations. The purpose of the present experiment was to investigate the nature of age-related changes in memory. The primary focus was to examine changes in memory for given and beyond information across the adult lifespan. A secondary focus was to explore some factors that may be related to age-related changes in memory for given and beyond information, including verbal and perceptual ability, and to investigate changes in memory for feminine and masculine  1 17  information across the adult female lifespan. Memory for Given and Beyond Information The first important finding was that there were divergent age-related changes in memory for given and beyond information. It was hypothesized that retrieval phase performance on items on the given scale would decline with age, whereas performance on items on the beyond scale would increase with age. The first hypothesis was confirmed as there was an age-related decline in performance on the aggregated given item. These results replicated previous research demonstrating an age-related decline in verbatim memory for verbal targets (Adams. 1991; Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Gould, Trevithick & Dixon, 1991; Reder. Wible & Martin, 1986; Walsh & Baldwin, 1977; Walsh, Baldwin & Finkle, 1980; Zandi & Gregory, 1988) and extended these findings to target pictures. The second hypothesis was not confirmed as there was generally no age-related change in performance on items on the beyond scale. These results are inconsistent with previous research that showed an age-related increase in memory for inferences that draw on world knowledge and personal experiences from verbal stimuli (Adams, 1991; Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Gould, Trevithick & Dixon, 1991). It is possible that the discrepancy between the results of the present and previous research is due to task differences. In the Adams' studies (Adams, 1991; Adams, Labouvie-Vief, Hobart & Dorosz, 1990), for example, participants' performance was assessed on only one target story whereas the present experiment employed a total of six target pictures. It is possible that encoding more target stimuli led to greater interference at the retrieval phase and hence resulted in preserved rather than improved memory for beyond information. Hence, the results of the present experiment indicate that there were divergent age-related changes in the retrieval of given and beyond information, where there was an age-related decline for given information and no age-related change for beyond information. The second important finding was that the observed age-related changes in retrieval  118  phase performance were related to encoding phase performance as demonstrated by both a qualitative and a quantitative approach (see Figure 3.1). The quantitative approach examined the overlap between encoding and retrieval processes by employing regression analyses to partial out the influence of encoding phase performance on retrieval phase performance. Consistent with the notions of transfer appropriate processing and encoding specificity that state that memory performance depends on the overlap between encoding and retrieval processes (Tulving & Thomson, 1973; Morris, Bransford & Franks, 1977), it was hypothesized that encoding phase performance would account for retrieval phase performance. As expected, the results showed that encoding phase performance accounted for a significant portion of retrieval phase performance, and thereby indicating that encoding and retrieval phase performance were related; when retrieval scores were regressed on encoding scores, the results showed that encoding scores were positively related to retrieval scores. The qualitative approach examined age-related changes in the degree to which encoding and retrieval processes do and do not overlap by examining references that were the same at the encoding and retrieval phases, called matches, and references that were different between the encoding and retrieval phases, called losses and gains. Losses are references that were provided at the encoding phase but were not carried over to the retrieval phase and gains are references that were provided at the retrieval phase but were not carried over from the encoding phase. If there are age-related changes in the degree to which encoding and retrieval processes overlap and do not overlap, it was hypothesized that there would be age-related changes in matches, losses and gains. The results generally showed that there were no age-related changes in matches, losses and gains, and thereby indicating that there were no age-related changes in the degree to which encoding and retrieval processes do or do not overlap; when matches, losses and gains were regressed separately on age and age squared, the results indicate that matches, losses and gains were not related to age or age squared for most obtained and aggregated items. Hence, the results of the quantitative approach indicate that  1 19  encoding and retrieval phase performance were related and the results of the qualitative approach further indicate that there were no age-related changes in the degree to which encoding and retrieval phase performance did and did not overlap. The third important finding was that there were generally no age-related changes in the influence of study instructions in guiding performance. Such changes might be expected if study instructions could encourage young adults to perform like older adults and older adults to perform like younger adults. The camera instruction was designed to enhance performance on items on the given scale and the friend instruction was designed to enhance performance on the beyond scale. Because young adults encode and retrieve more given and less beyond information, they may not benefit from the camera instruction, but they may benefit from the friend instruction. By contrast, because older adults remember more beyond and less given information, they may not benefit from the friend instruction, but they may benefit from the camera instruction. Hence, if study instructions can encourage young adults to perform like older adults and older adults like young adults, it was hypothesized that young adults would benefit from the camera instruction and older adults would benefit from the friend instruction. The results provided little support for this hypothesis, however, as there were generally no agerelated changes in benefit from the two types of study instructions. A n age-related change in. benefit from study instructions was found for only the absence, quantity, and timeframe items, where, as expected, young adults benefited from the friend instruction and middle-aged and older adults generally benefited from the camera instruction. The general pattern of findings were that, for both young and older adults, the camera study instruction enhanced performance on items on the given scale and the friend study instruction enhanced performance on some items on the beyond scale; the camera instruction was significantly related to performance on some items on the given scale and the friend instruction was related to performance on some items on the beyond scale. Hence, these results indicate that study instructions generally did not encourage  120  young adults to perform like older adults and older adults to perform like young adults; rather, both young and older adults benefited equally from study instructions. To summarize, the present results revealed three important findings. The most important finding was that there were divergent age-related changes in the types of information that were encoded and retrieved about target pictures, where there was an age-related decline for given information and no age-related change for beyond information. Second, age-related changes in the types of information that were retrieved depended on the overlap with the types of information that were encoded, and there were no age-related changes in the degree of overlap and non-overlap. Third, there were no age-related changes in benefit from study instructions; rather, young and older adults benefited equally from study instructions. Factors Related to Memory for Given and Beyond Information The present experiment explored the relationships between perceptual and verbal ability and memory for given and beyond information in order to place memory for given and beyond information in a broader context. Relationships between age, verbal ability and memory for beyond information were explored with the idea that an age-related improvement in verbal ability may facilitate the expression of beyond information. This idea was explored by examining the relationships between age and verbal ability and between verbal ability and encoding and retrieval phase performance on the obtained and aggregated items. Consistent with previous research (Baltes, 1987; Baltes, Dittmann-Kohli & Dixon, 1984; Horn, 1982; Kaufman, Kaufman-Packer, McLean & Reynolds, 1991), the present results showed an age-related increase in verbal ability, as measured by North American Adult Reading Test (NAART; Blair & Spreen, 1989); when N A A R T scores were regressed on age, the results, indicate a significant positive relationship. As well, the results indicate that there was a positive relationship between verbal ability and performance on the obtained and aggregated items; when encoding and retrieval scores for each obtained and aggregated item were regressed on N A A R T scores, there were  121  significant positive relationships between age and performance on most obtained and aggregated items. Because the present results indicate a relationship between age and verbal ability, and because the results of the previous section indicate a relationship between age and performance on the obtained and aggregated items, further analyses were performed in order to rule out the possibility that the observed relationship between verbal ability and performance on the obtained and aggregated items was not due to their common relationship with age. The results indicate that the relationship between verbal ability and performance on the obtained and aggregated items was not mediated by age; the relationship between performance on the obtained and aggregated items and N A A R T scores remained when the influence of age was partialled out. Hence, together, these results are consistent with previous research (Adams, 1991) that indicate that an age-related improvement in verbal ability facilitates the expression of beyond information. As well, the results indicate that and an age-related improvement in verbal ability facilitates the expression of given information. Additional analyses were performed to examine the consequences of these findings on the findings that were reported in the previous section. Thus far, the results indicate two types of relationships — a direct and an indirect relationship ~ between age and performance on the obtained and aggregated items. The results of the previous section indicate a direct relationship between age and performance on the aggregated given item and the event item. The present results indicate that there was also an indirect relationship between age and performance on the obtained and aggregated items that was mediated by verbal ability; there was a relationship between age and verbal ability and between verbal ability and performance on the obtained and aggregated items. It is possible, however, that these are not separate types of relationships, where the relationship between age and performance on the obtained and aggregated items that was observed in the previous section was due to verbal ability. In order to investigate this possibility, the relationship between age and performance on the obtained and aggregated items  122  was examined when the influence of verbal ability was partialled out. Encoding and retrieval scores for each obtained and aggregated item were regressed separately on N A A R T scores and then age. When compared to the previous unpartialled analyses, the present results showed that age-related changes in performance on most of the obtained and aggregated items remained when the influence of verbal ability was partialled out, except for one item on the beyond scale ~ event. For the event item, the results indicate that verbal ability accounted for an age-related increase in retrieval phase performance. This finding provides further support for the conclusion that was drawn in the previous section that there was no age-related change in memory for beyond information. Together, the results of previous and present section indicate that there was a direct relationship between age and performance on items on the given scale and an indirect relationship between age and performance on items on the given scale that was mediated by verbal ability. As well, these results indicate that there was a relationship between age and verbal ability and between verbal ability and performance on items on the beyond scale, but there was no relationship between age and performance on items on the beyond scale. Relationships between age, perceptual ability and performance on the obtained and aggregated items were explored with the idea that a declining ability to perceive given information may restrict further processing. This idea was explored by examining the relationships between age and perceptual ability and between perceptual ability and encoding and retrieval phase performance on the obtained and aggregated items. There was mixed support for an age related decline in perceptual ability, as assessed by two measures of auditory ability, the Hearing Screening Inventory (HSI; Coren & Hakstian, 1992) and a self-rating of hearing, and two measures of visual ability, the Color Vision Screening Inventory (CSI; Coren & Hakstian, 1988) and a self-rating of eyesight. The results indicate an age-related decline in auditory ability, which is consistent with previous research (e.g., Fozard, 1990), but no age-related change in visual ability; when each of the perceptual ability scores was regressed separately on age,  there was a significant negative relationship between age and auditory ability scores, but not visual ability scores. As well, when encoding and retrieval scores for each of the obtained and aggregated items were regressed separately on each of the perceptual ability scores, the results failed to show a relationship between perceptual ability and performance on most of the obtained and aggregated items. Together, these results provided no support for the idea that an agerelated decline in perceptual ability is related to further processing given or beyond information. To summarize, the present results revealed two important findings. First, an age-related improvement in verbal ability facilitated the expression of both given and beyond information; there were positive relationships between age and verbal ability and between verbal ability and performance on the obtained and aggregated items. Second, an age-related decline in perceptual ability did not restrict further processing of given or beyond information; there was a negative relationship between age and auditory but not visual ability, and there was no relationship between perceptual ability and performance on most of the obtained and aggregated items. Memory for Feminine and Masculine Information A secondary purpose of the present experiment was to explore changes in memory for feminine and masculine information across the adult female lifespan. Gerontological studies on feminine and masculine behaviors define feminine and masculine roles by those behaviors and attitudes that have been empirically demonstrated as characteristic of young women and men and then trace these gender stereotypes across the lifespan (for a review see Turner, 1982). . Similarly, the present experiment defined feminine and masculine information according to previous gerontological and psychological research findings that suggest certain types of information may be more relevant to young adults of one gender than the other. As well, feminine and masculine information was defined by inspecting gender differences in young adults' performance on the obtained and aggregated items. Memory for feminine and masculine information was then traced across the adult female lifespan. The results indicate that there were  124  generally no age-related changes in encoding feminine and masculine information; age and age squared was not related to encoding phase performance on most items. There were, however, divergent age-related changes in retrieval and retention of feminine and masculine information, where there was an age-related increase for feminine information and no age-related change for masculine information; there was an age-related increase in retrieval phase and encodingretrieval match performance on items on the feminine scale and no age-related change for items on the masculine scale. To summarize, this part of the research project revealed that women showed no agerelated changes in the encoding of feminine and masculine information and divergent age-related changes in the retrieval and retention of feminine and masculine information, where there was an age-related increase for feminine information and no age-related change for masculine information. THEORETICAL IMPLICATIONS Theories of memory and aging have typically focused on explaining age-related declines in cognitive performance. For example, resource theories state that successful task completion relies on the efficient assembly or coordination of component processes, where efficiency refers to both the amount of mental energy needed to fuel requisite processes and the speed of information processing (e.g., Vernon, 1987, 1993; Vernon & Mori, 1992). Thus age-related declines in cognitive performance have been viewed as a breakdown in efficient processing, for example, an age-related reduction in speed of processing or a disruption in the coordination of processing due to interference (e.g., Hasher & Zacks, 1979; Salthouse, 1982, 1991, 1993; Salthouse & Kail, 1983). These resource theories, however, do not easily account for divergent age-related changes in memory, especially age-related improvements in memory and the preservation of memory. The idea of a central executive, a system that controls the allocation of resources to different operations, must be adopted for resource theories to accommodate these  age-related memory changes. A lifespan or developmental approach focuses on both age-related decline and improvement in cognitive abilities, and hence, offers a more parsimonious explanation of the aging memory system. Because the present experiment revealed the full spectrum of age-related changes in cognitive function, from an age-related decline in memory for given information, to preserved memory for beyond information, to an age-related improvement in verbal ability, these results are more consistent with a lifespan approach than supportive of resource theories. The present results were discussed in terms of two lifespan approaches to cognition: the lifespan developmental psychology notions of selective optimization with compensation and loss in the service of growth and a developmental approach to schema theory. The developmental notions of selective optimization with compensation (Baltes, 1987. 1993; Baltes, Dittmann-Kohli & Dixon, 1984) and loss in the service of growth (Adams, Labouvie-Vief, Hobart & Dorosz, 1990; Labouvie-Vief & Schell, 1982) state that certain cognitive competencies can develop that can counterbalance or trade-off other cognitive competencies that show age-related decline. Following from these notions, one interpretation of the present results is that people may come to rely on memory for beyond information in order to compensate for, or results in a trade-off with, memory for given information which shows an age-related decline. A developmental approach to schema theory proposes that maturation may influence schema-driven remembering. Schemas can change with experience by accretion, restructuring and tuning (Rumelhart & Norman, 1978). Accretion refers to the idea that schemas can grow in size as new experiences and information are combined or integrated with previously acquired information. Restructuring refers to the idea that new schemas are formed when new information cannot be added to an existing schema, for example, when a person acquires a new skill or learns a new body of information. Tuning refers to the idea that certain existing interrelations or associative pathways are strengthened upon repeated exposure to similar  126  experiences or types of information, such as the case when a person refines a motor skill or when a child becomes proficient in a language. Because schemas can direct what types of information is encoded (Arbuckle, Vanderleck, Harsany & Lapidus, 1990; Hess, Donley & Vandermaas, 1989) and retrieved (Hess & Flannagan, 1992; Hess & Slaughter, 1990) about a target stimulus, it is possible that maturation may influence schema-driven remembering. Following from a developmental approach to schema theory, the results of present experiment suggest that maturation may influence the tuning or refinement of associative pathways, where the accumulation of experiences may strengthen certain associations between stored information and other associations become relatively weakly connected or pruned. The observed age-related changes in memory for given and beyond information can be interpreted as a strengthening of associative pathways that lead to encoding and retrieving beyond information, as witnessed by preserved memory for beyond information, whereas the associative pathways that lead to encoding and retrieving given information may be weakly associated or pruned, as demonstrated by an age-related decline in memory for given information. The finding that there were no age-related changes in encoding feminine and masculine information suggest that by young adulthood, schemas are already structured or organized in such a way so as to store both feminine and masculine information and that there is no further reorganization with age that would influence age-related changes in encoding feminine and masculine information. The finding that there were divergent age-related changes in memory for feminine and masculine information suggest that there may be age-related changes in the strengthening of associative pathways for feminine but not masculine information. The associative pathways for feminine information may be strengthened because socialization encourages women to provide feminine rather than masculine information. This communicative style may have been further encouraged in this experiment because of a female experimenter; it is possible that female participants may be more likely to provide feminine information to a female rather than male experimenter.  127  The lifespan developmental psychology notions of selective optimization with compensation and loss in the service of growth and a developmental approach to schema theory are not conflicting interpretations of the present findings. The strengthening of certain associative pathways can be viewed as growth that arises at the expense of weakening or pruning other associative pathways. LIMITATIONS OF THIS RESEARCH The generalizability of the results of the present experiment are limited by the characteristics of the participants, including the possible influence of a select sample, personality factors and cohort differences on memory performance, and the types of measures that were employed, including the picture description method, and measures of verbal and perceptual ability. Subject Variables Participants were not representative of the general population in several important ways, including gender, health, educational achievement, and openness to experience. The majority of participants were female (70.4% women vs. 29.6% men). Participants reported that they were in good general health and had little health trouble; there were no relationships between age and self-ratings of health and health trouble. This is noteworthy as older participants were likely healthier and more mobile than their cohorts. As well, participants were well educated, with most (79%) having pursed post-secondary education. This is particularly striking for the older participants as older adults are generally less well educated than young adults; there was a failure to find a relationship between age and educational achievement, and thereby indicating that older and young participants were equally well educated. And finally, participants exhibited high levels of openness to experience (i.e., curiosity, imagination, intellect) as evidenced by high levels of educational achievement and by virtue of having volunteered for the present experiment. Because women may be more verbally skilled (Kaufman, Kaufman-Packer,  128  McLean & Reynolds, 1991), performance on a verbally-based task by a sample of mostly women, who are healthy, highly educated and motivated, and high in openness, may portray only the best picture of memory and aging. Personality traits, particularly the personality dimension of extraversion and introversion, may be related to performance on the picture description task. Extraversion is the personality trait to be active, assertive, energetic and enthusiastic whereas introversion is the personality trait to be passive, compliant, shy and lethargic. There is some evidence indicating an age-related increase in extraversion from early to middle adulthood (Nann, Millsap & Hartka, 1986) and an age-related increase in introversion in later adulthood (Costa et al., 1986). Hence young adults may be more likely than older adults to express their thoughts on a picture description task. As well, there is some evidence that the personality trait of dominance, which is related to extraversion, is positively related to performance on intelligence tests, and the personality trait of avoidance, which is related to introversion, is negatively related to performance on intelligence tests (Harris, Vernon & Jang, in press). Hence, extraversion and introversion may be related to memory performance on a picture description task. The results of the present experiment are subject to cohort effects as a cross-sectional design was employed. Cohort differences in having experienced major historical events may be important to the set size of beyond information that is available for recall. Older adults in this experiment have experienced several social upheavals, such as the Great Depression, World War II and the Baby Boom, which may have resulted in a large pool of significant life experiences and events from which to draw on. It is possible that subsequent generations may experience fewer social upheavals and perhaps resulting in a smaller pool of significant life experiences and events that can be recounted. As well, cohort differences in social attitudes, for example, changing attitudes toward family and social relationships (e.g., Perry-Jenkins & Crouter, 1990; Stewart & Healy, 1989), may direct attention toward encoding and retrieving different types of  129  information about pictures of social situations. Hence, the possibility for cohort effects warrants further studies with longitudinal or sequential designs before strong conclusions can be drawn. Picture Description Method The verbal report method is considered a measure of the types of processing that are being carried out at a particular time, and as such, has been described as a window into the mind (Crutcher, 1994; Ericsson & Simon, 1993; Payne, 1994; Wilson, 1994). The verbal report method asks participants to verbally describe their thoughts, "to think aloud," either while they are performing a problem solving task or immediately following a thinking episode (Ericsson & Simon, 1993). The picture description method that was used in this experiment is a version of both types of the verbal report method as participants were asked to verbally describe their thoughts while they, were viewing a series of pictures, and then following a delay, were asked to describe what they remembered about the pictures. The verbal report method is limited, however, as it is possible that people may be mentally processing more information than is verbally reported. For example, in the picture description task, it is likely that participants chose to edit out more detailed given information because it was assumed when they verbalized a more integrative or global piece of information and they may have edited out beyond information because it was considered socially inappropriate. Hence, by analogy, the window might present only a portion of the whole scene. As addressed by Ericsson and Simon (1993), there is some concern as to whether the data obtained from the verbal report method can be treated as objectively as behavioral data. The present experiment attempted to reduce the likelihood for subjectively interpreting what was provided in participants' picture descriptions by developing a coding scheme for different types of information and by obtaining inter-coder reliability. Although high inter-coder reliability was obtained, with correlation coefficients ranging from .70 to 1.00, it is nonetheless impossible to rule out subjective interpretation. This may be more evident when coding for beyond than given  130  information as coding for beyond information requires a coder's interpretation and evaluation of a participant's interpretation and evaluation of a target picture. And finally, there is some concern that performance on the picture descriptions task was influenced by the ability to re-code perceptual information into a verbal code. Because older adults may suffer from a reduction in processing resources (Hasher & Zacks, 1979; Light, 1991; Salthouse, 1982, 1991, 1993; Salthouse & Kail, 1983), this may present a more difficult task for older than younger adults. As well, although Ericsson and Simon (1993) have argued that performance on decision making tasks are similar under verbalizing and silent conditions, other studies have shown that verbalizing alters performance on tasks that require processes that are difficult to put into words, for example, remembering faces or solving insight problems (Schooler & Engstler-Schooler, 1990; Schooler, Ohlsson & Brooks, 1993). Hence, it is possible that young adults processed more beyond information than was provided in their picture descriptions because they lack the skills to articulate this type of information. Measures of Verbal and Perceptual Ability Verbal ability was assessed by the North American Adult Reading Test ( N A A R T ; Blair & Spreen, 1989). The results of the present experiment suggest that the N A A R T may be a broad indicator of verbal ability as an age-related improvement in verbal ability was related to memory for both given and beyond information. It is possible that memory for given and beyond information are related to different aspects of verbal ability. For example, it is possible that vocabulary size may be important to naming given information and that verbal fluency may be important to the expression of beyond information. As well, previous research indicates that the N A A R T may be a broad indicator of verbal ability as a correlation of .83 has been found between performance on the N A A R T and the WAIS-R VIQ (Blair & Spreen, 1989). The results of the present experiment encourage future research with more direct measures of the different components of verbal ability.  131  The present experiment employed self-report measures of perceptual ability, including self-ratings of eyesight and hearing, the Color Vision Screening Inventory (Coren & Hakstian. 1988) and the Hearing Screening Inventory (Coren & Hakstian, 1992). The results of the present experiment suggest that more sensitive measures of perceptual ability are required as an expected age-related decline in perceptual ability (e.g., Fozard, 1990) was found for auditory but not visual ability. Because these measures asked participants to rate their visual and auditory ability with corrective aid, it is possible that auditory ability was a more sensitive measure than visual ability of age-related declines in perceptual ability as auditory difficulties may go undetected whereas visual problems are commonly detected and corrected. The present experiment relied on correlational analyses to investigate the relationships between age, verbal and perceptual ability, and memory for given and beyond information. Correlational analyses do not permit for strong conclusions as it is possible that the observed relationships were due to a third variable. For example, it is possible that the observed relationship between verbal ability and performance on the obtained and aggregated items was mediated by another cognitive operation, such as fluid intelligence, which has been shown to be related to verbal ability and to memory (Vernon, 1987; Vernon & Mori, 1992). As well, correlational analyses do not allow for interpretations about causal relationships. Other experimental methods are required to draw causal relationships between cognitive operations. For example, a simulation method that adjusts young adults perceptual ability so as to mimic the perceptual ability of older adults (e.g., wearing specially designed glasses) while performing a cognitive task is one possible method for examining a causal relationship among perceptual and cognitive abilities. Hence, the results of the present experiment provide a basis for future research.  132  FUTURE RESEARCH One avenue for future research is to integrate personality and cognitive psychology. A method for bridging the two domains is to integrate the verbal report procedure that was employed in this experiment with Wiggins' circumplex approach to personality and interpersonal behavior. Following the circumplex tradition, personality is assessed along two orthogonal dimensions: dominance and nurturance (Wiggins, 1979, 1995; Wiggins, Trapnell & Phillips, 1988). Dominance is one's ability to forcefully take charge in a social situation or one's tendency to submissively comply with another's demands. Nurturance is one's ability to provide warmth and understanding towards others or one's tendency to be cold and hostile towards others. The Interpersonal Adjective Scale (IAS; Wiggins, 1995) provides a measure of dominance and nurturance, and based on these scores, provides a location in interpersonal space. Following from Wiggins' circumplex approach to interpersonal behavior, one's location in interpersonal space allows for predictions regarding interpersonal behavior. Basically, people who are high in dominance tend to get along with people who are high in nurturance, and vice versa, people who are high in nurturance tend to get along with people who are high in dominance. People who are low in dominance tend to get along with people who are low in nurturance, and vice versa, people who are low in nurturance tend to get along with people who are low in dominance (for a complete description see Wiggins, 1995). Employing the circumplex approach to personality and interpersonal behavior as a framework for the verbal report method entails asking participants to read and later to verbally recall a series of stories about people who are high or low in dominance or nurturance. Participants' IAS scores can be compared with the amount of given and beyond information that was provided in their verbal reports. Of particular interest is a personality and interpersonal match between the participant and the target person. The proposed method investigates the role of personality on memory, or who remembers what, by examining the amount of information  133  that is recalled about a target person who is similar in personality to the participant. This method also places memory into an interpersonal context, or who remembers what about whom, by examining the amount of information that is recalled about a target person who is an interpersonal match with the participant. As well, the proposed method can be used to investigate memory for feminine and masculine information and wisdom. Memory for Feminine and Masculine Information The proposed method can be used to further investigate the present finding that women showed divergent age-related changes in memory for feminine and masculine information. Wiggins has demonstrated that the Bern Masculinity and Femininity scales (Bern, 1974) are equivalent to and are among the best markers of the measures of dominance and nurturance (Wiggins & Broughton, 1985; Wiggins & Holzmuller, 1981). Hence, feminine information would be defined as information provided in stories about target people who are high or low in nurturance and masculine information would be defined as information about people who are high or low in dominance. Assessing men and women's memory for these target stories across the lifespan is one attempt to replicate and extend the present results. Wisdom The present experiment investigated memory for beyond information but did not evaluate beyond information for wisdom. One method for investigating wisdom is to integrate the proposed method with a measure of wisdom that has been developed by Staudinger (Staudinger, Smith & Baltes, 1992; Staudinger, Maciel, Smith & Baltes, 1998). Her wisdom task involves asking participants for advice regarding different life dilemmas and then coding their responses for wisdom. The proposed method could employ stories about life dilemmas of people who are high or low in dominance or nurturance. Participants would be asked to recall the stories as well as to provide advice for the target person. Hence, this method would provide some insight into memory as well as wisdom.  Memory for Given and Beyond Information Future research might consider employing different types of target pictures to enhance the likelihood of producing beyond information. The present experiment employed target pictures that varied in type (i.e., depictions of social and asocial situations) and age-relevance in order to maximize test performance for all age groups. The selection of target stimuli successfully controlled for the possibility of age-related bias in the testing material as the results indicated preserved memory for beyond information. Hence, these results indicate that the verbal report procedure can be used as an age-free memory test. More references to beyond information might be provided, especially from older adults, by employing personal photographs, along with the friend instruction. As well, more references to beyond information might be obtained from older adults from collectivist versus individualistic cultures. MAIN CONTRIBUTION The present experiment makes a general and specific contribution to the study of memory and aging. With respect to its specific contribution, the present experiment demonstrated divergent age-related changes in memory, with an age-related decline in memory for given information and preserved memory for beyond information. This extends our understanding of the aging memory system as research has traditionally focused on age-related declines in memory. The results of the present experiment were discussed in terms of a developmental approach to schema theory and the lifespan developmental psychology notions of selective optimization with compensation and loss in the service of growth. With respect to its general contribution, the present experiment demonstrated the importance of encoding processes on retrieval processes. As well, the present experiment illustrated the importance of employing participant-rather than experimenter-defined to-be-remembered targets in memory research as this consideration has lead to a more complete understanding of memory and aging.  135  REFERENCES Adams, C. (1991). Qualitative age differences in memory for text: a life-span developmental perspective. Psychology and Aging, 6(3), 323-336. Adams, C , Labouvie-Vief, G., Hobart, C. J. & Dorosz, M . (1990). Adult age group differences in story recall style. Journal of Gerontology: Psychological Sciences, 45(1), P17-P27. Allen, P. A . & Madden, D. J. (1993). Influence of age and processing stage on visual word recognition. Psychology and Aging, 8(2), 274-282. Anderson, J. R. (1981). Cognitive skills and their acquisition. Hillsdale, NJ: Erlbaum. Anderson, J. R. (1982). Acquisition of cognitive skill. Psychological Review, 89, 369406. Anderson, R. C. & Pichert, J. W. (1978). Recall of previously unrecallable information following a shift in perspective. Journal of Verbal Learning and Verbal Behavior, 17, 1-12. Arbuckle, T. Y . & Gold, D. P. (1993). Aging, inhibition and verbosity. Journals of Gerontology. 48(5), P225-P232. Arbuckle, T. Y., Vanderleck, V . F., Harsany, M . & Lapidus, S. (1990). Adult age differences in memory in relation to availability and accessibility of knowledge-based schemas. Journal of Experimental Psychology: Learning. Memory and Cognition. 16(2), 305-315. Backman, L . (1991). Recognition memory across the adult life span: the role of prior knowledge. Memory & Cognition, 19(1), 63-71. Baltes, P. B . (1987). Theoretical propositions of life-span developmental psychology: On the dynamics between growth and decline. Developmental Psychology, 23(5), 611-626. Baltes, P. B., Dittmann-Kohli, F. & Dixon, R. A . (1984). New perspectives on the development of intelligence in adulthood. Toward a dual-process conception and a model of selective optimization with compensation. In P. B . Baltes & O. G. Brim, Jr. (Eds.), Life-span developmental psychology and behavior (Vol. 6, pp. 61-87). N Y : Academic. Bartlett, F. C. (1932). Remembering: An experimental and social study. Cambridge, England: Cambridge University Press. Belmore, S. M . (1981). Age-related changes in processing explicit and implicit language. Journal of Gerontology, 36, 316-322. Bern, S. L . (1974). The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 42. 155-162. Blair, J. R. & Spreen, O. (1989). Predicting premorbid IQ: a revision of the National Adult Reading Test. Clinical Neuropsychologist, 3, 129-136.  1 36  Botwinick, J. & Storandt, M . (1974). Vocabulary ability in later life. Journal of Genetic Psychology, 125, 303-308. Blum-Kulka,,S. (1993). "You gotta know how to tell a story": telling tails, and tellers in American and Israeli narratives at dinner. Language in Society, 22(3), 361-402. Bruner, J. S. (1957). Going beyond the information given. In J. S. Bruner, E. Brunswick, L. Festinger, F. Heider, K. F. Muenzinger, C. E. Osgood, D. Rapaport (Eds.), Contemporary approaches to cognition: a symposium held at the University of Colorado (pp. 4174). Cambridge: Harvard University Press. Bruner, J. S. (1986). Actual minds: possible worlds. Cambridge: Harvard University Press. Cohen, G. (1979). Language comprehension in old age. Cognitive Psychology, 11,412429. Cohen, G. (1981). Inferential reasoning in old age. Cognition. 9, 59-72. Cohen, J. (1978). Partialed products are interactions; partialed powers are curve components. Psychological Bulletin, 85, 858-866. Cooper, P. (1990). Discourse production and normal aging: performance on oral partial description tasks. Journal of Gerontology. 45, 210-214. Coren, S. & Hakstian, R. (1992). The development and cross-validation of a self-report inventory to assess pure-tone threshold hearing sensitivity. Journal of Speech and Hearing Research, 35, 921-928. Coren, S. & Hakstian, R. (1988). Color vision screening without the use of technical equipment: scale development and cross-validation. Perception & Psychophysics, 43, 115-120. Costa, P. T., Jr., McCrae, R. R., Zonderman, A. B., Barbano, H . E., Lebowitz, B., & Larson, D. M . (1986). Cross-sectional studies of personality in a national sample: 2. Stability in neuroticism, extraversion, and openness. Psychology and Aging, 1, 144- 149. Crutcher, R. J. (1994). Telling what we know: the use of verbal report methodologies in psychological research. Psychological Science, 5(5), 241-244. Ericsson, K. A . & Simon, H. A. (1993). Protocol analysis: Verbal reports as data. Cambridge, M A : MIT Press. Feingold, A . (1988). Cognitive gender differences are disappearing. American Psychologist, 43(2), 95-103. Fivush, R. (1995). Gendered narratives, gendered lives. Paper presented at the first meeting of the Society of Applied Research in Memory and Cognition, Vancouver, B.C. Fivush, R. & Mandler, J. M . (1985). Developmental changes in the understanding of temporal sequence. Child Development, 56, 1437-1446.  137  Fozard, J. L. (1990). Vision and hearing in aging. In J. E. Birren & K . W. Schaie (Eds.), Handbook of the psychology of aging (pp. 150-170). N Y : Academic Press. Gold, D. P., Andres, D., Arbuckle, T. & Schwartzman, A . (1988). Measurement and correlates of verbosity in elderly people. Quantification and correlates. Journal of Gerontology. 43(2), 27-34. Gold, D. P., Andres, D., Arbuckle, T. & Zieren, C. (1993). Off-target verbosity and talkativeness in elderly people. Canadian Journal on Aging, 12, 67-77. Gould, O. N . , Trevithick, L. & Dixon, R. A . (1991). Adult age differences in elaborations produced during prose recall. Psychology & Aging, 6, 93-99. Graf, P. (1990). Life-span changes in implicit and explicit memory. Bulletin of the Psvchonomic Society, 28(4), 353-358. Graf, P., Uttl, B., Mori, M . , Birt, A., Shapka, J., Ghani, A . & Olofsson, U . Cognition across the adult lifespan: the Kerrisdale project, overview, method, and sample description with normative data, (manuscript in preparation) Gutmann, D. L , (1977). The cross-cultural perspective: notes toward a comparative psychology of aging. In J. Birren & K. W. Schaie (Eds.), Handbook of the psychology of aging. N Y : Van Nostrand Reinhold. Gutmann, D. L . (1987). Reclaimed powers: Towards a new psychology of men and women in later life. New York: Basic. Harris, J. A., Vernon, P. A . & Jang, K . L . (in press). A multivariate genetic analysis of relationships between intelligence and personality. Developmental Neuropsychology. Hasher, L. & Zacks, R. T. (1979). Automatic and effortful processes in memory. Journal of Experimental Psychology: General, 108(3), 356-388. Hess, T. M . (1985). Aging and context influences on recognition memory for typical and atypical script actions. Developmental Psychology, 21, 1139-1151. Hess, T. M . , Donley, J. & Vandermaas, M . (1989). Aging-related changes in the processing and retention of script information. Experimental Aging Research, 15(2), 89-96. Hess, T. M . & Flannagan, D. A . (1992). Schema-based retrieval processes in young and older adults. Journal of Gerontology: Psychological Sciences, 47(1), P52-P58. Hess, T. M . & Slaughter, S. J. (1990). Schematic knowledge influences on memory for scene information in young and older adults. Developmental Psychology, 26(5), 855-865. Hollingworth, H. L . (1928). Psychology: Its facts and principles. New York: Appleton. Horn, J. L . (1982). The theory of fluid and crystallized intelligence in relation to concepts of cognitive psychology and aging in adulthood. In F. I. M . Craik & S. Trehub (Eds.), Aging and cognitive processes (pp. 237-278). New York: Plenum. i  Hudson, J. A . & Fivush, R. (1991). Planning in the preschool years: the emergence of plans from general event knowledge. Cognitive Development. 6, 393-415. Hudson, J. & Nelson, K. (1983). Effects of script structure on children's story recall. Developmental Psychology, 19, 625-635. Hudson, J. & Nelson, K. (1986). Repeated encounters of a similar kind: effects of familiarity on children's autobiographical memories. Cognitive Development, I, 253-272. Jung, C. (1933). Modern man in search of a soul. N Y : Harcourt Brace Jovanovich. Kamptner, N . L . (1991). Personal possessions and their meanings: a life-span perspective. Journal of Social Behavior and Personality, 6(6), 209-228. Kane, M . J. & Hasher, L. (1994). Inhibitory attentional mechanisms and.aging. Psychology & Aging, 9(1), 103-112. Kaufman, A. S., Kaufman-Packer, J. L., McLean, J. E. & Reynolds, C. R. (1991). Is the pattern of intellectual growth and decline across the adult life span different for men and women? Journal of Clinical Psychology, 47(6), 801-812. Kemper, S., Kynette, D., Rash, S., Sprott, R. & O'Brein, K. (1989). Lifespan changes to adults' language: effects of memory and genre. Applied Psycholinguistics, 10, 49-66. Labouvie-Vief, G. (1985). Intelligence and cognition. In J. E. Birren & K. W. Schaie (Eds.), Handbook of the psychology of aging (2nd ed.) (pp. 500-530). N Y : Van Nostrand Reinhold. Labouvie-Vief, G. (1996). Knowledge and the construction of women's development. In P. B . Baltes & U . M . Staudinger (Eds.), Interactive minds: Life-span perspective on the social foundation of cognition (pp. 109-130). N Y : Cambridge University Press. Labouvie-Vief, G. & Schell, D. A . (1982). Learning and memory in later life. In B . Wolman & G. Strieker (Eds.), Handbook of developmental psychology (pp. 828-846). Englewood Cliffs, NJ: Prentice-Hall. Light, L. L . (1990). Interaction between memory and language in old age. In J. E. Birren & K. W. Schaie (Eds.), Handbook of the psychology of aging (3rd ed.) (pp. 275-290). N Y : Academic. Light, L . L. (1991). Memory and aging: Four hypotheses in search of data. Annual Review of Psychology. 42, 333-376. Light, L. L . & Singh, A. (1987). Implicit and explicit memory in young and older adults. Cognition. 13(4), 531-541. Light, L., Valencia-Laver, D. & Zavis, D. (1991). Instantiation of general terms in young and older adults. Psychology & Aging. 6(3), 337-351.  139  Mandler, J. M . & DeForest, M . (1979). Is there more than one way to recall a story? Child Development, 50. 886-889. Mandler, J. M . & Johnson, N . S. (1977). Remembrance of things parsed: Story structure and recall. Cognitive Psychology, 9, 111-151. Mandler, J. M . & Parker, R. E. (1976). Memory for descriptive and spatial information in complex pictures. Journal of Experimental Psychology: Human Learning and Memory, 2, 3848. Markus, H. (1977). Self-schemata and processing information about the self. Journal of Personality and Social Psychology, 35, 63-78. Martin, C. L. & Paulhus, D. L . (1985). Bipolar-biasing effects of sex-role extremity on memory for traits. Sex Roles, 13,463-474. McCrae, R. R. & Costa, P. T., Jr. (1989). The structure of interpersonal traits: Wiggins's circumplex and the five-factor model. Journal of Personality and Social Psychology, 56, 586595. Mergler, N . L. & Goldstein, M . D. (1983). Why are there old people?: Senescence as biological and cultural preparedness for the transmission of information. Human Development, 26, 72-90. Miller, G . A . (1956). The magical number seven, plus or minus two: some limits on our capacity for processing information. Psychological Review, 63, 81-97. Miller, P. J., Wiley, A. R., Fung, H. & Liang, C-H. (1997). Personal storytelling as a medium of socialization in Chinese and American families. Child Development, 68(3), 557-568. Minami, M . & McCabe, A. (1995). Rice balls and bear hugs: Japanese and North American family narrative patterns. Journal of Child Language, 22(2), 423-445. Morris, C D . & Bransford, J. D. & Franks, J. J. (1977). Levels of processing versus transfer appropriate processing. Journal of Verbal Learning and Verbal Behavior, 16, 519-534. Nann, N . , Millsap, R. & Hartka, E. (1986) As time goes by: Change and stability in personality over fifty years. Psychology & Aging, 1_, 220-232. Obler, L. K . (1980). Narrative discourse style in the elderly. In L. K . Obler & M . L . Albert (Eds.), Language and Communication in the Elderly (pp. 75-90). Lexington, M A : Heath. Obler, L. K . & Albert, M . L. (1981). Language and aging: a neurobehavioral analysis. In D. S. Beasley & G. A . Davis (Eds.), Aging: Communication Processes and Disorders (pp. 107121). New York: Grune & Stratton. Park, D. C , Puglisi, J. T. & Sovacool, M . (1983). Memory for pictures, words, and spatial location in older adults: evidence for pictorial superiority. Journal of Gerontology, 38(5), 582-588.  140  Payne, J. W. (1994). Thinking aloud: Insights into information processing. Psychological Science. 5(5), 241, 245-248. Perry-Jenkins, M . & Crouter, A . C. (1990). Men's provider-role attitudes. Implications for household work and marital satisfaction. Journal of Family Issues. 11, 136-156. Pichert, J. W. & Anderson, R. C. (1977). Taking different perspectives on a story. Journal of Educational Psychology, 69, 309-315. Ramelli, E. & Stella, S. (1993). The old-time's memory. New Trends in Experimental and Clinical Psychiatry, 9(4), 145-148. Rasmussen, J. C. L . (1988). Evaluating outliner identification tests: Mahalanobis D squared and Comer Dk. Multivariate Behavioral Research, 23, 189-202. Reder, L . M . , Wible, C. & Martin, J. (1986). Differential memory changes with age: exact retrieval versus plausible inference. Journal of Experimental Psychology: Learning, Memory and Cognition, 12(1), 72-81. Rumelhart, D. E. & Norman, D. A . (1978). Accretion, tuning, and restructuring: Three modes of learning. In J. W. Cotton & R. Klatzky (Eds.), Semantic factors in cognition (pp. 3753). Hillsdale, N . J.: Erlbaum. Salthouse, T. A . (1982). Adult cognition: An experimental psychology of human aging. N Y : Springer-Verlag. Salthouse, T. A . (1991). Theoretical perspectives on cognitive aging. Hillsdale, N J : Erlbaum. Salthouse, T. A . (1993). Speed mediation of adult age-differences in cognition, Developmental Psychology, 29, 722-738. Salthouse, T. A . & Kail, R. (1983). Memory development throughout the life span: The role of processing rate. Life-Span Development and Behavior, 5, 89-116. Schaie, W. (1988). Ageism in psychological research. American Psychologist, 43(3), 179-183. Schank, R. C. & Abelson, R. P. (1977). Scripts, plans, goals and understanding. Hillsdale, NJ: Erlbaum. Schmidt, C. R. & Schmidt, S. R. (1986). The use of themes as retrieval cues in children's memory for stories. Journal of Experimental Child Psychology, 42, 237-255. . Schooler, J. W. & Engstler-Schooler, T. Y . (1990). Verbal overshadowing of visual memories: Some things are better left unsaid. Cognitive Psychology, 22, 36-71. Schooler, J. W., Ohlsson, S., Brooks, K . (1993). Thoughts beyond words: When language overshadows insights. Journal of Experimental Psychology: General, 122, 166-183.  141  Signorella, M . L. (1992). Remembering gender-related information. Sex Roles, 27(3-4), 143-156. Sinnott, J. D. (1984). Older men, older women: Are their perceived sex roles similar? Sex Roles, 10(11-12), 847-856. Staudinger, U . M . , Maciel, A. G., Smith, J. & Baltes, P. B . (1998). What predicts wisdom-related performance? A first look at personality, intelligence and facilitative experiential contexts. European Journal of Personality, 12(1), 1-17. Staudinger, U . M . , Smith, J. & Baltes, P. B . (1992). Wisdom-related knowledge in a life review task: Age differences and the role of professional specialization. Psychology & Aging, 7(2), 271-281. Stevens, J. P. (1984). Outliers and influential data points in regression analyses. Psychological Bulletin. 95(2), 334-344. Stewart, A. J. & Healy, J. M . , Jr. (1989). Linking individual development and social change. American Psychologist, 44, 30-44. Tabachnick, B. A . G. & Fidell, L. S. (1989). Using multivariate statistics (2nd ed.). New York: Harder Colleens Publishers. Till, R. E. (1985). Verbatim and inferential memory in young and elderly adults. Journal of Gerontology. 40(3), 316-323. Till, R. E. & Walsh, D. A . (1980). Encoding and retrieval factors in adult memory for implicational sentences. Journal of Verbal Learning and Verbal Behavior, 19, 1-16. Tipper, S. P. (1991). Less attentional selectivity as a result of declining inhibition in older adults. Bulletin of the Psvchonomic Society, 29( 1), 45-47. Tulving, E. & Thomson, D. M . (1973). Encoding specificity and retrieval processes in episodic memory. Psychological Review, 80, 352-373. Turner, B . F. (1982). Sex-related differences in aging. In B. B . Wolman & G. Strieker (Eds.), Handbook of developmental psychology (pp. 912-936). Englewood Cliffs, NJ: Prentice Hall. Vernon, P. A . (1987). Speed of information-processing and intelligence. Norwood, NJ: Ablex. Vernon, P. A . (1993). Biological approaches to the study of human intelligence. Norwood, NJ: Ablex. Vernon, P. A . & Mori, M . (1992). Intelligence, reaction times, and peripheral nerve conduction velocity. Intelligence, 16, 273-288. Voyer, D. & Voyer, S. (1995). Magnitude of sex differences in spatial abilities: A meta analysis and consideration of critical variables. Psychological Bulletin, 117(2), 250-270.  142  Wallace, J. B . (1992). Reconsidering the life review: The social construction of talk about the past. Gerontologist, 32(1), 120-125. Wiggins, J. S. (1979). A psychological taxonomy of trait-descriptive terms: The interpersonal domain. Journal of Personality and Social Psychology. 37, 395-412. Wiggins, J. S. (1991). Agency and communion as conceptual coordinates for the understanding and measurement of interpersonal behavior. In W . Grove & D. Cicchetti (Eds.), Thinking clearly about psychology: Essays in honor of Paul E. Meehl (Vol. 2, pp. 89-113). Minneapolis: University of Minnesota Press. Wiggins, J. S. (1995). Interpersonal Adjective Scales: Professional Manual. Odessa, FL: Psychological Assessment Resources. Wiggins, J. S. & Broughton, R. (1985). The interpersonal circle: A model for the integration of personality research. Perspectives in Personality, 1, 1-47. Wiggins, J. S. & Holzmuller, A. (1981). Further evidence on androgyny and interpersonal flexibility. Journal of Research in Personality, 15, 67-80. >  Wiggins, J. S. & Trapnell, P. D. (1996). A dyadic-interactional perspective on the fivefactor model. In J. S. Wiggins (Ed.). The five-factor model of personality: Theoretical perspectives (pp. 88-162). New York: Guilford Press. Wiggins, J. S., Trapnell, P., & Phillips, N., (1988). Psychometric and geometric characteristics of the Revised Interpersonal Adjective Scales (IAS-R). Multivariate Behavioral Research, 23, 517-530. Wilson, T. D . (1994). The proper protocol: Validity and completeness of verbal reports. Psychological Science, 5(5), 249-252. Wilson, T. D. (1994). The proper protocol: validity and completeness of verbal reports. Psychological Science, 5(5), 249-252. Winograd, E., Smith, A. D. & Simon, E. W. (1982). Aging and the picture superiority effect in recall. Journal of Gerontology, 37(1), 70-75. Zandi, T. & Gregory, M . (1988). Making inferences in adulthood: falling leaves means it's fall. Educational Gerontology, 14(3), 203-210.  APPENDIX A Target and Practice Pictures There were six target pictures: 1) boat race 2) couple in living room 3) women with yarn 4) girl in bedroom 5) woman in library 6) empty street There were two practice pictures: 1) clothes hanging on a clothesline 2) boy preparing something to eat  WW'  3  '  •l  152  APPENDIX B Scoring Key for Picture Descriptions There were a total of six pictures: boat race, couple in living room, women with yarn, woman in bedroom, woman in library and empty street (see Appendix A). Each picture description at the encoding and retrieval phases were scored individually for each of the items in the five sections of this key: things, attributes, inferences, point of view and verbatim repetitions. There are a total of 20 items which are distributed among the five sections. Items in Sections 1 to 4 — things, attributes, inferences and point of view — count the first reference to different types of information that were provided in the picture descriptions and the item in Section 5 — verbatim repetition — counts the additional or repeated references to items in Sections 1 to 4. Three separate counts are made for items in Sections 1 to 4. Items in Sections 1 and 2 — things and attributes — count the number of instances to words or phrases that refer to something that is present in a picture and items in Sections 3 and 4 — inferences and point of view — count the number of instances to words or phrases that refer to something that is not present but inferred from the picture. Items are scored by adding up the number of instances of words and phrases that meet the defined criteria for each picture description at the encoding and at the retrieval phase. As well, items are scored for the number of instances of words and phrases that meet the defined criteria that were provided in the picture descriptions at both the encoding and retrieval phases. A match is considered when the same idea or information is conveyed at the encoding and retrieval phases; the information does not have to be conveyed in exactly the same words. Two separate counts are made for the item in Section 5 -- verbatim repetition. This item counts the additional or repeated references to items in Sections 1 to 4 that were provided in the picture description at the encoding and at the retrieval phases.  Chapters 3 and 4 Participants' verbal reports are coded for two types of information: given and beyond information. Given information includes the items in Sections 1 and 2 — things and attributes — which count the number of instances of words or phrases that refer to something that appears in a picture. Beyond information includes the items in Sections 3 and 4 — inferences and point of view — which count the number of instances of words or phrases that refer to something that is not present but inferred from the picture. Chapter 5 Participants' verbal reports are coded for two types of information: feminine and masculine information. Feminine information includes items 1, 12, 17 and 19 which count the number of instances or phrases that refer to people, emotion and life events. Masculine information includes items 4, 5, 6, 9 and 10 which count the number of instances or phrases that refer to passive state, position, quantity, spatial location and absence. Note 1: The practice picture is not scored. Note 2: Words or phrases are counted only once by an item and are not counted by more than one item. Note 3: Words or phrases that re-describe something either by clarifying or further specifying something are not counted by these items. Section 1: Things This section describes the procedure for analyzing and summarizing the encoding and retrieval of two things ~ people and objects ~ that occur in each picture. Each objective item is scored separately for the number of instance of words or phrases that correctly refer to a thing that is present in a picture for each picture description at the encoding and retrieval phases. Only  154  the first reference to a thing is counted; additional words or phrases that are repetitions or alternative labels for the same thing are counted separately by Item 20 - Verbatim Repetition. Item 1: People. A count of the number of words or phrases that refer to people in each picture description. To obtain this count, add up the number of references to people by means of nouns (e.g., girl, woman, guy, men) and pronouns (e.g., she, he, they) for each picture description. A person can be counted separately when they are mentioned as an individual and as part of a group. Examples receiving a score of 2: There's a group of people, some of them are... They are...he is... However, a person is counted only once when alternative labels are assigned to the same person or when a person is referred to on more than one occasion. In these cases, only the first instance is counted. Examples receiving a score of 1: she is...her... The woman...she was... Note: Do not count references to inferred roles or relationships. Item 2: Objects. A count of the number of words or phrases that refer to objects in each picture description. To obtain this count, add up the number of references to whole objects (including animals), parts of an object and collections of objects for each picture description. A n object can be counted more than once when it is mentioned individually and as part of a collection. Examples receiving a score of 2: there's things...a comb, a... [only when a list of things is given, otherwise comb  155  is considered to be an alternative label for thing] there are a group of houses and one off to the left - there are three canoes...the one in front the filing cabinets... the drawers... there's a s/i/? on the back of the chair [back is scored by Item 9 — Spatial location] However, reference to an object is counted only once when it is a clarification, elaboration or further specification of the same object or when it is referred to on more than one occasion. In these cases, only the first instance is counted. Examples receiving a score of 1: a painting of a man there's an antenna, possibly a t.v. antenna Section 2: Attributes This section describes the procedure used for analyzing and summarizing the encoding and retrieval of properties of things that appear in a picture. Each objective item is scored separately for the number of instances of words or phrases that correctly refer to an attribute of a thing that is present in a picture for each picture description at the encoding and retrieval phases. Only the first reference to an attribute of a thing is counted; additional words or phrases that are repetitions or alternative labels for the same attribute are counted separately by Item 20 -Verbatim Repetition. Item 3: Color. A count of the number of words or phrases that refer to color names (e.g., red, blue, green), shades (e.g., darkish, brownish) or other color descriptives (e.g., it's the color of nursing homes) for each picture description. To obtain this count, add up the number of references to each color of a thing. Separate points are given when more than one color or shade is provided for the same thing.  156  Examples receiving a score of 2: the cat is brown and white the canoes are green and white However, only one point is given for words or phrases that further specify or clarify a color, for hyphenated words, and for repeated reference to the same color of the same thing. Examples receiving a score of 1: the yarn is brownish yellow the cushion is blue-green the curtains are yellow or brown Item 4: Passive state. A count of the number of words or phrases that describe passive actions or activities of people for each picture description. To obtain this count, add up the number of references to each passive action. Only one point is given for words or phrases that further specify or clarify a passive action and for repeated reference to the same passive action. Examples receiving a score of 1: the girl is looking at herself in the mirror he is wearing a shirt, pants and dirty running shoes Note: Do not count references to inferred or ongoing actions that are presented in a picture; these are counted separately in Item 11 — Actions. Item 5: Position. A count of the number of words or phrases that refer to the position of a thing in relation to the picture for each picture description. The person need not explicitly state that the reference is the picture, but it must be implied or inferred from the person's mention of position that the reference is to the picture. To obtain this count, add up the number of references to each position of a thing. Separate points are given for repeated reference to the same position when it is a description of the position for different things.  Examples receiving a score of 2: at the bottom of the picture there's a lake...at the bottom [of the picture] there are three canoes However, only one point is given for words or phrases that further specify or clarify the position of a thing or for repeated reference to the position of the same object. Examples receiving a score of 1: at one end, at the back, of the picture in the corner [of the picture] there's a room, the corner room... Note: Do not count references to the orientation of a thing in relation to other things in a picture; these are scored separately in Item 9 — Spatial Location. Item 6: Quantity. A count of the number of words or phrases that refer to the quantity (i.e., the number or amount) of a thing for each picture description. To obtain this count, add up the number of references each the quantity of a thing. More than one point is given when a quantifier is used to distinguish a thing from a group of things. Examples receiving a score of 2: there are many houses, one house is...[when one is used as a quantifier and not an identifier] However, only one point is given to words or phrases that further specify or clarify the quantity of a thing, for counting out loud and for repeated reference to the quantity of a thing. Examples receiving a score of 1: there's a couple of people, two people... there are many people there are one, two, three canoes  Item 7: Shape. A count of the number of words or phrases that refer to the shape of a thing for each picture description. To obtain this count, add up the number of references to each shape of a thing. Only one point is given for words or phrases that further specify or clarify the shape of a thing and for repeated reference to the shape of a thing. Examples receiving a score of 1: her necklace is in a pyramid design the lines are arranged in a rainbow shape, a semi-circle or arc  Item 8: Size. A count of the number of words or phrases that refer to the volume or size of a thing for each picture description. To obtain this count, add up the number of references to each size of a thing. Separate scores are given when the size of a thing is described in different ways. Examples receiving a score of 2: tall pillars, about 12 feet or more However, only one point is given to words and phrases that further specify or clarify a size and for repeated reference to the size of a thing. Examples receiving a score of 1: the river isn't very wide, in fact it's quite narrow there are tall pillars...the tall pillars... Item 9: Spatial location. A count of the number of words or phrases that refer to the orientation or position of a thing relative to other things in the picture for each picture description. To obtain this count, add up the number of references to each spatial location of a thing. Separate points are given for repeated reference to the same spatial location when it is a description of the spatial location for different things. Examples receiving a score of 2:  159  on top of the dresser is a comb...there's also a picture on the dresser However, only one point is given for words or phrases that further specify or clarify the spatial location of a thing and for repeated reference to the spatial location of the same thing. Examples receiving a score of 1: towards the back of the library is another room on one side, on the left hand-side, of the street there are some buildings in one corner of her room there is a closet... in the corner closet... Note: The orientation of a thing in relation to the picture is counted in Item 5 - Position. Section 3: Inferences This section describes the procedure for analyzing and summarizing the types of inferences that are made at encoding and at retrieval regarding properties of things that appear in a picture. Each subjective item is scored by counting the number of instances of words or phrases that refer to an attribute that is not present but inferred from a picture for each picture description at the encoding and retrieval phases. Only the first reference to an inferred attribute of a thing is counted; additional words or phrases that are repetitions or alternative labels for the same inferred attribute are counted separately by Item 20 - Verbatim Repetition. Item 10: Absence. A count of the number of words or phrases that refer to the absence of people or objects. Examples receiving a score of 1: There's nobody at the reception desk. I can't see anyone in this scene. There doesn't seem to be any mud puddles. There are no trees by the house.  160  An additional point is given if an absence of people or objects is remarked upon at more than one location. Example receiving a score of 2: Nobody is behind the trees...Nobody is on the street. Item 11: Action. A count of the number of words or phrases that describe inferred or ongoing actions or activities of things, but does not include passive states (e.g., she is wearing a dress, they are sitting on a couch) for each picture description. To obtain this count, add up the number of references to each action. Separate points are given when the same action is used to describe a thing and a part of a thing. Examples receiving a score of 2: the couple is having a conversation, the man is speaking they're canoeing, one is moving faster However, only one point is given for words or phrases that further specify or clarify an action and for repeated reference to the same action. Examples receiving a score of 1: they're rolling a ball of wool...they're winding a ball of yarn they're conversing...maybe they're talking about... Note: Do not count references to passive actions; these are scored separately in Item 4 — Passive State. Item 12: Emotion. A count of the number of words or phrases that infer a person's emotion, facial expression or state of mind for each picture description. To obtain this count, add up the number of references to each emotion. Separate points are given when the same emotion is described in different ways. Examples receiving a score of 2:  she seems happy although she isn't smiling However, only one point is given for words or phrases that further specify or clarify an emotion and for repeated reference to the same emotion. Examples receiving a score of 1: they aren't smiling they seem content, they're enjoying each other's company Item 13: Object's age. A count of the number of words or phrases that refer to the age of an object or place for each picture description. To obtain this count, add up the number of references to each age of an object or place. Only one point is given for words or phrases that further specify or clarify an object's age and for repeated reference to the same age. Examples receiving a score of 1: the architecture follows traditional lines she's wearing old shoes, they're from the 40's Item 14: Person's age. A count of the number of words or phrases that refer to a person's age for each picture description. To obtain this count, add up the number of references to each person's age. Separate points are given when the same person or group of people are described at different ages. Examples receiving a score of 2: a teenager...when she was a baby maybe the two elderly women were childhood friends However, only one point is given for words or phrases that further specify or clarify a person's age and for repeated reference to the same age. Examples receiving a score of 1: the couple is in their 50's, they're middle-aged  162  Item 15: Setting. A count of the number of words or phrases that refer to the global arrangement of things or location for the scene of a picture for each picture description. To obtain this count, add up the number of references to each global arrangement, place (including inside and outside areas) and architecture (including rooms) for each picture description. Setting can be counted more than once when it is mentioned as a whole and as part of an area. Examples receiving a score of 2: in the park there is a picnic area in the library, there are a number of smaller rooms Only one point is given for words or phrases that further specify or clarify the global arrangement of things or location for the scene of a picture and for repeated reference to the global arrangement of things or location for the scene of a picture. Examples receiving a score of 1: clothes are scattered about...there, are clothes everywhere people are gathered in a park or picnic area this is a town, a rural town this is a library, maybe a public library or an information center Note: Do not count the specific arrangement or orientation of a thing; these are scored separately in Item 5 — Position and Item 9 — Spatial Location. Item 16: Time frame. A count of the number of words or phrases that refer to the period or season for each picture description. To obtain this count, add up the number of references to each time frame. Only one point is given for words or phrases that further specify or clarify a time frame and for repeated reference to the same time frame. Examples receiving a score of 1: this could be in the 60's or 70's  16?  it looks like a town during the depression, or a recession it looks like fall or late September Note: Do not include direct references to time frames from the subject's own life or someone else's life; they are counted separately in Section 5 — Point of View. Section 4: Point of View This section describes the procedure used for analyzing and summarizing the perspective or point of view taken by the participant when describing past or possible life events or experiences that are associated or inferred from a target picture. Items are scored by counting the number of instances of words or phrases that describe past or possible life events from different points of view for each picture description at the encoding and retrieval phases. Only the first reference to a past or possible life event is counted; additional words or phrases that are repetitions or alternative labels for the same life event are counted separately by Item 20 Verbatim Repetition. Item 17: First person point of view. A count of the number of words or phrases that explicitly refer to and focus on the participant's own life events for each picture description. To obtain this count, add up each life event that focuses on the first person point of view (I, me, my). Separate scores are given when words or phrases that focus on separate life events in the participant's life. Examples receiving a score of 2: she could be my grandmother, or my great aunt this could be the town that I grew up in or a place I take my family to for vacation However, only one point is given for words or phrases that further specify or clarify a life event from the participant's life and for repeated reference to the same life event.  164  Examples receiving a score of 1: This is how I wish I could keep my room but my room is always kept neat. This could be my daughter's room, my youngest daughter's room. Item 18: Second person point of view. A count of the number of words or phrases that explicitly refer to and focus on the life events of a person that is familiar to the participant (e.g., a friend, a spouse, an offspring) for each picture description. To obtain this count, add up each life event that focuses on a second person's point of view. Separate scores are given to words and phrases that focus on separate life events in a familiar person's life or for the same life event but from different familiar people's lives. Examples receiving a score of 2: my aunt and my grandmother use to enjoying rolling balls of wool However, only one point is given for words or phrases that further specify or clarify a life event from a familiar person's life and for repeated reference to the same life event. Examples receiving a score of 1: my grandmother, my mother's mother, enjoyed doing this my daughter always kept a messy room Item 19: Third person point of view. A count of the number of words or phrases that explicitly refer to and focus on the life events of a person that is unfamiliar to the subject for each picture description. To obtain this count, add up each life event that employs a third person point of view. Separate scores are given to words or phrases that focus on separate life events in an unfamiliar person's life. Examples receiving a score of 2: one might recognize this from their youth or their children's youth  165  However, only one point is given for words or phrases that further specify or clarify a life event from an unfamiliar person's life and for repeated reference to the same life event. Examples receiving a score of 1: this could remind someone of hard times, like the depression the bed is off the ground, someone could fit under it Section 5: Verbatim Repetitions Item 20: Verbatim repetition. A count of the number of repeated references to words or phrases that are exact or near literal reproductions of references to most items, except for Item 5 --Position and Item 9 -- Spatial location, that were previously described for each picture description at the encoding and retrieval phases. In the example, "She is sitting on a bed. The woman sitting on the bed has gray hair", the references to "woman", "sitting", and "bed" are all counted by the verbatim repetition item, but the reference to "on the" is not counted because it refers to position. A near literal reproduction is a word or phrase that is synonymous to words or phrases that have been coded by an item that was previously described. Examples receiving a score of 1: The library has high, high ceilings. There's a tractor going down a main street. The tractor is green. She's really disorganized. She's really messy. Do not count references to words or phrases that are clarifications to items that were previously described. Clarifications are words or phrases that provide additional information about an item. On the other hand, an exact or near literal reference to a clarification should be counted by the verbatim repetition item. Example receiving a score of 1: It's a library or information centre. In the information centre...  166  Do not count references to people that are used within a reference to a past or possible life event as these references are counted by items in Section 4 — Point of View, but do count repeated references to items in Section 4 — Point of View. Do not count references to people that are used in a possessive form or that designate a point of view, but do count repeated references to people (and objects) in a pronoun form. Examples of references that should not be counted: She's looking at herself in the mirror. She's getting ready to go out. [She's getting ready to go out is coded by Item 19 — Third person point of view.] She's sitting on the bed. Her hair is gray. [Her is a possessive form of "she"] She has some things hanging in her closet. She has a lamp and a picture on her dresser, [both instances of "she" are used to designate a point of view} Examples of references that should be counted: She's getting ready for a date. She's getting ready to go out with her boyfriend. The woman is sitting on the stairs. She's wearing a pyramid shaped necklace. The house is red brick. There's a tree next to it.  167  -a a o <D X)  T3 C  CN  03  en*  c u >  1—1  N  o  Tt Tf* CN  O  Tt  d  CN  cn  rx.  Tt  Os  oq NO en i n T t — i oo CN d oo en d 1 CN CN cn  "5b x:  J3  •4—1  c o  cn 00 en o CN en fx* T t cn —<  Q  CO  6  —1 NO NO NO i—H r x ON en NO 00 i n CN CN CN NO en NO  00 ON i n oo m i n ON © i n cn ON ON T t T t en T t CN NO 00 CN r - H Tt* -J T-H CN *—i r-  3 <u  s  in  Tt  in  CN  T t ON NO ON CN  cn CN  _ _ Tt CN  in  in  NO NO ON —^  en oo  ON NO oq T t »—«' — i  Tt  en  ON T t CN CN  CN CN  Tt Tt CN  o r-- t~ON O ON  ^_ CN  ON  en o  l  ^J, CN  ON  00  oq o  in*  en O  o u o  B  03  B o o  u  OH  ON T t ON 00 i n i n NO i n CN CN  oo N  CJ  03 > <D  00. 00 oo i n 00 m 00 00 en oo oo i n CN i n _< d in* en* Tt  CN  m O  Q  oo  CO  ON Cx NO NO OO o ON r x T t cn i n d  in  CN  en oo CN  Tt CN  Tt  rx  in  p  CN  NO m ON T t CN ON T t in* CN CN  o  r~ r x o r x m CN CN* CN  — .  00 CN  o  r-;  cn en o ON  ON  cn cn d  CN  rx  CN  Tt  O  Tt ON  CN  Tt  ^J. en  en  CN  Tt Tt  o  CN  00  c  •3  NO  <u  CS  en m ON m cn oq ON en i n en en 00 ON d NO d  o o c  in  T3  —  CN  NO r x NO ON en CN* ON* NO* Tt  NO O m NO* en CN  NO Tt NO 00 NO CN* Tt*  Tt ON  ^  NO*  Tt  oo en  1  c  03  V oo  03 X! OH  Tt  o  "o3 >  'E to  s Q CO  03 XI OH  rx  cn  CN  Tt CN CN  CN i n d Tt Tt  o  O  o  NO i n i n 00 NO ON CN CN r x  Tt  en ON en >n  CJ oo N  00  NO NO ON  ON T t en NO CN tx- ON CN d i n —^ CN CN  en T t i n r x i n en i n CN m  00  Tt ON CN  en cn T t m O CN i n od CN en ON  CN NO CN  CN 00  p en - J  ON NO  CN  Tt  Tt  CN NO Tt*  en oo i n CN cn cn en*  O  O 00 O NO en* NO CN* en r x en  CN  I*  C  "o3 o c u  O en o CN 00 r~ O o- en i n oq i n p en p NO r x r x 00* 00 d NO en CN* d d < CN *— CN m 1  o  oo.  a  \3  <D  <  UH  °  1  oo  *H-* 03 O O  03 00  & 1 w 3 t  CN  00 (x. oq CN* NO* en* 00  in  CN  CN  NO  -1—» cd  c  03  O  bp  vo  - 1  CN NO  'oo C  C  oo CN m oq r x en*  V 04  '2  o <H-C  U X  O O  £  CD  >  o  '6 °  00  ^H  C  O H O H O H C T O O O O  £ o o  oo N  QJ  • • ^ o o . O o o o 3 c " T S o 3 O  £ 03 00  SO  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0099432/manifest

Comment

Related Items